Channel state information reporting

Abstract

A method of wireless communication includes determining, by a first communication device, channel state information CSI corresponding to P ports, where P is a positive integer; and transmitting, by the first communication device to a second communication device, the CSI that includes at least one set of a precoding matrix indicator PMI, Q sets of channel quality indicators CQIs, or R values of rank indicator RI. The PMI indicates C4 sets of precoding matrices corresponding to C4 first type of time domain units, the Q sets of CQIs corresponds to Q second type of time domain units, the R values of RI corresponds to R third type of time domain units, and Q, R, C4 are positive integers.

Description

CHANNEL STATE INFORMATION REPORTING TECHNICAL FIELD This document is directed generally to wireless communications. BACKGROUND Mobile telecommunication technologies are moving the world toward an increasingly connected and networked society. In comparison with the existing wireless networks, next generation systems and wireless communication techniques will need to support a much wider range of use-case characteristics and provide a more complex and sophisticated range of access requirements and flexibilities. Long-Term Evolution (LTE) is a standard for wireless communication for mobile devices and data terminals developed by 3rd Generation Partnership Project (3GPP). LTE Advanced (LTE-A) is a wireless communication standard that enhances the LTE standard. The 5th generation of wireless system, known as 5G is committed to supporting higher data-rates, large number of connections, ultra-low latency, high reliability and other emerging business needs. SUMMARY Methods, apparatuses and computer readable media are described. The disclosed techniques may be used by embodiments for receiving reference signals or providing a feedback report based on received reference signals. In one example aspect, a wireless communication method is disclosed. The method includes determining, by a first communication device, channel state information CSI corresponding to P ports, where P is a positive integer; and transmitting, by the first communication device to a second communication device, the CSI that includes at least one set of a precoding matrix indicator PMI, Q sets of channel quality indicators CQIs, or R value of rank indicator RI. The PMI indicates <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> sets of precoding matrices corresponding to <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time domain units, the Q sets of CQIs corresponds to Q second type of time domain units, the R values of RI corresponds to R third type of time domain units, and <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>R<annotation encoding="application / x-tex">R< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> are positive integers. In another aspect, another method is disclosed. The method includes receiving, by a first communication device from a second communication device, signal transmissions comprising transmissions of a first reference signal and a second reference signal according to a pattern and transmitting, by the first communication device to the second communication device, channel state information according to the first reference signal and the second reference signal. In yet another example aspect, another wireless communication method is disclosed. The method includes receiving, by a second communication device from a first communication device, the CSI that includes at least one set of a precoding matrix indicator PMI, Q sets of channel quality indicators CQIs, or R value of rank indicator RI. The PMI indicates <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> sets of precoding matrices corresponding to <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time domain units, the Q sets of CQIs corresponds to Q second type of time domain units, the R values of RI corresponds to R third type of time domain units, and <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>R<annotation encoding="application / x-tex">R< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> are positive integers. In another aspect, another method is disclosed. The method includes transmitting, by a second communication device to a first communication device, signal transmissions comprising transmissions of a first reference signal and a second reference signal according to a pattern and receiving, by the second communication device from the first communication device, channel state information which is got by the first communication device according to the first reference signal and the second reference signal. In yet another aspect, a wireless communication apparatus is disclosed. The apparatus includes a processor configured to implement a disclosed method. In yet another aspect, a computer readable medium having program code stored thereon is disclosed. The program code, upon execution by a processor, causes the processor to implement a method disclosed in the present document. These, and other, aspects are described throughout the present document. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1-2 are timing diagrams depicting example of one or more transmission occasions of channel state information (CSI) reference signal and time duration of the CSI. FIG. 3 is a timing diagram for example of a second type of time units. FIG. 4 is a flowchart for an example method of wireless communication. FIG. 5 is a flowchart for an example method of wireless communication. FIG. 6 shows an exemplary block diagram of a hardware platform that may be a part of a network device or a communication device. FIG. 7 shows an example of wireless communication network including a base station (BS) and user equipment (UE) based on some implementations of the disclosed technology. FIG. 8 is a flowchart for an example method of wireless communication. FIG. 9 is a flowchart for an example method of wireless communication. DETAILED DESCRIPTION In the following description, headings may be used to improve clarity without limiting the combinations of the various disclosed features. Furthermore, terminology specific to certain industry standards (e.g., 3GPP, LTE or 5G) is used for ease of explanation only and the disclosed techniques may be used in other wireless systems implementing different communications protocols. Last few years have seen a steady growth in the number of wireless devices deployed in wireless communication networks. More and more, users are expecting availability of high bandwidth communication connections in many settings including during high-speed travel. Existing wireless communication standards are finding it difficult to provide high quality of service and meet user expectations in scenarios in which wireless communication channel between network-side device and user devices (sometimes called user equipment UE) may be unreliable or fast-changing due to high-speed movement of user devices during use. The present document provides various techniques that, in one example aspect, may be used for transmission, reception or processing of reference signals that allows wireless devices to adapt quickly to the time-varying nature of a wireless communication channel and providing reliable and high quality wireless connectivity to user devices. The wireless device feedback one set of PMI (precoding matrix indicator). The one set of PMI indicate <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> set of precoding matrices corresponding to <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time domain units. The one set of PMI applies for each precoding matrix of <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> set of precoding matrices. It captures the frequency domain feature and time domain feature of the channel. All precoding matrices share same information indicated by the one set of PMI. It reduces overhead of CSI and the second communication device can get more accurate CSI. The present document also reports RI / CQI for multiple time domain units in one CSI-reporting. It allows the second communication device to schedule with high flexibility based on the reported CSI Example 1 The UE determines P of CSI-RS ports according to at least one of a received signaling, or reported CRI (CSI-RS resource indicator). The UE measures the P of CSI-RS ports and gets CSI (channel state information) based on the measurement. The UE reports the determined CSI to gNB. P is the number of CSI-RS ports. The CSI includes at least one of Q sets of CQI (channel quality indicator), R values of RI (rank indicator), or one set PMI (precoding matrix indicator). Following is the procedure of determining CSI. The Q sets of CQI are for Q second type of time domain units. Each of the Q sets is for one of the Q second type of time domain units. The R values of RI correspond to R third type of time domain units. Each of the <com:Math><semantics>R<annotation encoding="application / x-tex">R< / annotation>< / semantics>< / com:Math> values of RI corresponds to one of the <com:Math><semantics>R<annotation encoding="application / x-tex">R< / annotation>< / semantics>< / com:Math> third type of time domain units The one set of PMI indicates <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> sets of precoding matrices for <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time domain units. Each of the <com:Math><semantics>C4<annotation encoding="application / x-tex">^{C_4}< / annotation>< / semantics>< / com:Math> sets of precoding matrices correspond to one of the <com:Math><semantics>C4<annotation encoding="application / x-tex">^{C_4}< / annotation>< / semantics>< / com:Math> first type of time domain units. Each of the <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> sets of precoding matrices includes <com:Math><semantics>N3<annotation encoding="application / x-tex">N_3< / annotation>< / semantics>< / com:Math> precoding matrices. In another implementation, the number of precoding matrices in different sets of the <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> sets of precoding matrices can be different. For example, the number of precoding matrices in a precoding matrix set corresponds to one first type of time domain unit including at least one of transmission occasion of the P CSI-RS ports is larger than the number of precoding matrices in a precoding matrices set corresponds to one first type of time domain unit including no transmission of the P CSI-RS ports. The CSI is for one CSI reporting. As shown in FIG. 1 or FIG. 2, there are <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> CSI-RS transmission occasions of the P of CSI-RS ports during <com:Math><semantics>T1<annotation encoding="application / x-tex">T_1< / annotation>< / semantics>< / com:Math>. The UE gets channel <com:Math><semantics>ht<annotation encoding="application / x-tex">h_t< / annotation>< / semantics>< / com:Math> on each occasion <com:Math><semantics>t=0,1,...,C1<annotation encoding="application / x-tex">t = 0,1,...,C_1< / annotation>< / semantics>< / com:Math> of the <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> CSI-RS transmission occasion for each first frequency domain unit <com:Math><semantics>k1=0,1,...N4−1<annotation encoding="application / x-tex">k_1 = 0,1,...N_4 - 1< / annotation>< / semantics>< / com:Math>, wherein <com:Math><semantics>ht<annotation encoding="application / x-tex">h_t< / annotation>< / semantics>< / com:Math> includes R rows and P columns. R is the number of receiving antennas at the UE side. Then the UE gets <com:Math><semantics>Hk1=[h0h1⋯hC1−1]<annotation encoding="application / x-tex">H_{k_1} = \begin{bmatrix} h_0 & h_1 & \cdots & h_{C_1-1} \end{bmatrix}< / annotation>< / semantics>< / com:Math> for each first frequency domain unit <com:Math><semantics>k1=0,1,…N4−1<annotation encoding="application / x-tex">k_1 = 0, 1, \dots N_4 - 1< / annotation>< / semantics>< / com:Math>. The UE gets a precoding matrix for each second frequency domain unit <com:Math><semantics>k=0,1,...N3−1<annotation encoding="application / x-tex">k = 0,1,...N_3 - 1< / annotation>< / semantics>< / com:Math> and for each first type of time domain unit belong to a set of <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time domain units, for example <com:Math><semantics>C4=C3<annotation encoding="application / x-tex">C_4 = C_3< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>t=0,1,...,C3<annotation encoding="application / x-tex">t = 0,1,..., C_3< / annotation>< / semantics>< / com:Math> during <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math> based on <com:Math><semantics>N4<annotation encoding="application / x-tex">N_4< / annotation>< / semantics>< / com:Math> of <com:Math><semantics>Hk1<annotation encoding="application / x-tex">H_{k_1}< / annotation>< / semantics>< / com:Math>. <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math> can be one of <com:Math><semantics>T3,1,T3,2,T3,3,T3,4,T3,5<annotation encoding="application / x-tex">T_{3,1}, T_{3,2}, T_{3,3}, T_{3,4}, T_{3,5}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>T3,6<annotation encoding="application / x-tex">T_{3,6}< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>T1<annotation encoding="application / x-tex">T_1< / annotation>< / semantics>< / com:Math> as shown in FIG. 1 or FIG. 2. <com:Math><semantics>T3,1<annotation encoding="application / x-tex">T_{3,1}< / annotation>< / semantics>< / com:Math> equals one period of CSI-RS resource 1 including the <com:Math><semantics>P<annotation encoding="application / x-tex">P< / annotation>< / semantics>< / com:Math> of CSI-RS ports and starts from a first OFDM symbol of a slot wherein the first symbol of a burst of CSI-RS resource 1 locates, or starts from the first OFDM symbol of the first transmission occasion of CSI-RS resource 1 in a burst. <com:Math><semantics>T3,2<annotation encoding="application / x-tex">T_{3,2}< / annotation>< / semantics>< / com:Math> equals one period of CSI-RS 1 minus <com:Math><semantics>T1<annotation encoding="application / x-tex">T_1< / annotation>< / semantics>< / com:Math> which equals time duration of a burst of CSI-RS 1. The length of <com:Math><semantics>T3,3<annotation encoding="application / x-tex">T_{3,3}< / annotation>< / semantics>< / com:Math> also equals to one period of CSI-RS 1 but it starts from a symbol after the end of a burst of CSI-RS 1. <com:Math><semantics>T3,4<annotation encoding="application / x-tex">T_{3,4}< / annotation>< / semantics>< / com:Math> equals to one period of CSI-RS 1 plus one burst of the CSI-RS 1 in adjacent period. The duration of <com:Math><semantics>T3.5<annotation encoding="application / x-tex">T_{3.5}< / annotation>< / semantics>< / com:Math> is smaller or equal to <com:Math><semantics>T3,4<annotation encoding="application / x-tex">T_{3,4}< / annotation>< / semantics>< / com:Math>. The gNB can informs its starting position and / or length of <com:Math><semantics>T3,5<annotation encoding="application / x-tex">T_{3,5}< / annotation>< / semantics>< / com:Math>. The starting OFDM symbol of <com:Math><semantics>T3,6<annotation encoding="application / x-tex">T_{3,6}< / annotation>< / semantics>< / com:Math> is before the first OFDM symbol of a slot wherein the first symbol of a burst of CSI-RS resource 1 locates, or before the first OFDM symbol of the first transmission occasion of CSI-RS resource 1 in a burst of the first OFDM symbol. The gNB can also inform information about the time duration <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math> of the reported CSI by signaling. For example, the gNB informs starting position and / or length of the time duration <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math>. One period of CSI-RS 1 includes one burst. One burst includes one or more transmission of CSI-RS1. There are <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> transmission occasions of CSI-RS 1 in one burst of CSI-RS 1. The CSI-RS 1 is simple name of CSI-RS resource 1 including the P of CSI-RS ports. One period of the P CSI-RS ports includes one burst. One burst includes one or more transmission of the P CSI-RS ports. There are <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> transmission occasions of the P CSI-RS ports in one burst of CSI-RS 1. The first frequency domain unit can be same or different from the second frequency domain unit. <com:Math><semantics>N3≥N4<annotation encoding="application / x-tex">N_3 \ge N_4< / annotation>< / semantics>< / com:Math>. The difference between FIG. 1 or FIG. 2 is the ending position of <com:Math><semantics>T1<annotation encoding="application / x-tex">T_1< / annotation>< / semantics>< / com:Math>. There is no first type of time domain unit of the <com:Math><semantics>C3<annotation encoding="application / x-tex">C_3< / annotation>< / semantics>< / com:Math> first type of time domain units <com:Math><semantics>t=0,1,...,C3<annotation encoding="application / x-tex">t = 0,1,...,C_3< / annotation>< / semantics>< / com:Math> between the end of the CSI-RS burst and the start of <com:Math><semantics>T2<annotation encoding="application / x-tex">T_2< / annotation>< / semantics>< / com:Math> in FIG. 1 or FIG. 2. The ending position of <com:Math><semantics>T1<annotation encoding="application / x-tex">T_1< / annotation>< / semantics>< / com:Math> can be last transmission occasion of one CSI-RS burst or end of a slot where the last occasion of one CSI-RS burst locates. The last transmission occasion of the P CSI-RS ports of the n1 period belongs to <com:Math><semantics>T1<annotation encoding="application / x-tex">T_1< / annotation>< / semantics>< / com:Math> of the n1 period instead of <com:Math><semantics>T2<annotation encoding="application / x-tex">T_2< / annotation>< / semantics>< / com:Math> of the n1 period. For each second frequency domain unit <com:Math><semantics>k=0,1,...N3−1<annotation encoding="application / x-tex">k = 0,1,...N_3 - 1< / annotation>< / semantics>< / com:Math> and for each a first type of time domain unit <com:Math><semantics>t=0,1,...,C3<annotation encoding="application / x-tex">t = 0,1,..., C_3< / annotation>< / semantics>< / com:Math>, the UE gets a precoding matrix. These <com:Math><semantics>N3*C3<annotation encoding="application / x-tex">N_3 * C_3< / annotation>< / semantics>< / com:Math> precoding matrices are based on three types of vector sets. The first type of vector set includes L first type of vectors each of which includes <com:Math><semantics>P2<annotation encoding="application / x-tex">\frac{P}{2}< / annotation>< / semantics>< / com:Math> elements, wherein the <com:Math><semantics>P2<annotation encoding="application / x-tex">\frac{P}{2}< / annotation>< / semantics>< / com:Math> elements have same amplitude and each of the <com:Math><semantics>P2<annotation encoding="application / x-tex">\frac{P}{2}< / annotation>< / semantics>< / com:Math> elements has a respective phase, or only one of <com:Math><semantics>P2<annotation encoding="application / x-tex">\frac{P}{2}< / annotation>< / semantics>< / com:Math> elements has value 1 and remaining <com:Math><semantics>P2<annotation encoding="application / x-tex">\frac{P}{2}< / annotation>< / semantics>< / com:Math>-1 elements have value 0. The PMI indicate one first type of vector set which shared by all layers. The PMI indicates one or v second type of vector sets. If it indicates one second type of vector set, the set is shared by all layers. If it indicates v second type of vector sets, each of the v second type of vector sets corresponds to one layer of v layers. v is indicated by the gNB, or reported by the gNB. If the CSI includes R values of RI(rank indicator), the v is one of the first value of the R values of RI, the maximal value of the R values, the maximal of values of RI corresponds to time duration <com:Math><semantics>T1<annotation encoding="application / x-tex">T_1< / annotation>< / semantics>< / com:Math>. Each second type of vector set includes <com:Math><semantics>M<annotation encoding="application / x-tex">M< / annotation>< / semantics>< / com:Math> second type of vectors each of which includes <com:Math><semantics>N3<annotation encoding="application / x-tex">N_3< / annotation>< / semantics>< / com:Math> elements each of which corresponds to one frequency domain unit <com:Math><semantics>k<annotation encoding="application / x-tex">k< / annotation>< / semantics>< / com:Math> (i.e., the second frequency domain unit). In some implementation, different of the second type of vector sets can includes different number of second type of vectors. The PMI indicates D third type of vector sets, wherein D is larger than 1. Each of D third type of vector set includes <com:Math><semantics>X<annotation encoding="application / x-tex">X< / annotation>< / semantics>< / com:Math> third type of vectors each of which includes <com:Math><semantics>C3<annotation encoding="application / x-tex">C_3< / annotation>< / semantics>< / com:Math> elements, wherein each of the <com:Math><semantics>C3<annotation encoding="application / x-tex">C_3< / annotation>< / semantics>< / com:Math> elements corresponds to one first type of time domain unit <com:Math><semantics>t<annotation encoding="application / x-tex">t< / annotation>< / semantics>< / com:Math>. <com:Math><semantics>L<annotation encoding="application / x-tex">L< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>M<annotation encoding="application / x-tex">M< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>X<annotation encoding="application / x-tex">X< / annotation>< / semantics>< / com:Math> is an integer larger than 0. In some implementation, the sum of <com:Math><semantics>L<annotation encoding="application / x-tex">L< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>M<annotation encoding="application / x-tex">M< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>X<annotation encoding="application / x-tex">X< / annotation>< / semantics>< / com:Math> should be larger than 1. In some implementation, the L should be larger than 1. L, M and X is based on at least one of received signaling, <com:Math><semantics>P<annotation encoding="application / x-tex">P< / annotation>< / semantics>< / com:Math>, the total number of layers <com:Math><semantics>v<annotation encoding="application / x-tex">v< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>C3<annotation encoding="application / x-tex">C_3< / annotation>< / semantics>< / com:Math>, or <com:Math><semantics>N3<annotation encoding="application / x-tex">N_3< / annotation>< / semantics>< / com:Math>. In some implementation, the <com:Math><semantics>L<annotation encoding="application / x-tex">L< / annotation>< / semantics>< / com:Math> should be larger than 1, the sum of <com:Math><semantics>L<annotation encoding="application / x-tex">L< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>M<annotation encoding="application / x-tex">M< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>X<annotation encoding="application / x-tex">X< / annotation>< / semantics>< / com:Math> should be larger than 4. In some implementation, the <com:Math><semantics>L<annotation encoding="application / x-tex">L< / annotation>< / semantics>< / com:Math> should be larger than 1, the sum of <com:Math><semantics>M<annotation encoding="application / x-tex">M< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>X<annotation encoding="application / x-tex">X< / annotation>< / semantics>< / com:Math> should be larger than 1. In some implementation, <com:Math><semantics>L≤X≤M<annotation encoding="application / x-tex">L \le X \le M< / annotation>< / semantics>< / com:Math>. In some implementation, the X is larger than 2. In some implementation, above restriction about <com:Math><semantics>L<annotation encoding="application / x-tex">L< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>M<annotation encoding="application / x-tex">M< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>X<annotation encoding="application / x-tex">X< / annotation>< / semantics>< / com:Math> applies <com:Math><semantics>L<annotation encoding="application / x-tex">L< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>M<annotation encoding="application / x-tex">M< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>X<annotation encoding="application / x-tex">X< / annotation>< / semantics>< / com:Math> of each layer. Each of the <com:Math><semantics>N3*C3<annotation encoding="application / x-tex">N_3 * C_3< / annotation>< / semantics>< / com:Math> precoding matrices are determined according to the one set of first type of vector set, the one or v second type of vector sets, and the D third type of vector sets. Each of the v second type of vector sets is used to determine a column of each of the <com:Math><semantics>N3*C3<annotation encoding="application / x-tex">N_3 * C_3< / annotation>< / semantics>< / com:Math> precoding matrices and the column corresponds to one layer of the v layers. The <com:Math><semantics>lth<annotation encoding="application / x-tex">l^{th}< / annotation>< / semantics>< / com:Math> column <com:Math><semantics>Wk,tl<annotation encoding="application / x-tex">W_{k,t}^{l}< / annotation>< / semantics>< / com:Math> of one precoding matrix <com:Math><semantics>Wk,t<annotation encoding="application / x-tex">W_{k,t}< / annotation>< / semantics>< / com:Math> of the <com:Math><semantics>N3*C3<annotation encoding="application / x-tex">N_3 * C_3< / annotation>< / semantics>< / com:Math> for each second frequency domain unit <com:Math><semantics>k=0,1,...N3−1<annotation encoding="application / x-tex">k = 0,1,...N_3 - 1< / annotation>< / semantics>< / com:Math>, for each a first type of time domain unit <com:Math><semantics>t=0,1,...,C3<annotation encoding="application / x-tex">t = 0,1,...,C_3< / annotation>< / semantics>< / com:Math>, and for each layer <com:Math><semantics>l=0,1...,v−1<annotation encoding="application / x-tex">l = 0,1...,v-1< / annotation>< / semantics>< / com:Math> has one of format shown in formula (1 to 4) [Image disponible dans le document PDF, Image available in the PDF document] (1) [Image disponible dans le document PDF, Image available in the PDF document] (2) [Image disponible dans le document PDF, Image available in the PDF document] (3) [Image disponible dans le document PDF, Image available in the PDF document] (4) wherein <com:Math><semantics>vi,i∈{0,1,...,L−1}<annotation encoding="application / x-tex">v_i, i \in \{0,1,...,L-1\}< / annotation>< / semantics>< / com:Math> is the first type of vector in the first type of vector set, <com:Math><semantics>yk,l(f)<annotation encoding="application / x-tex">y_{k,l}^{(f)}< / annotation>< / semantics>< / com:Math> is <com:Math><semantics>kth<annotation encoding="application / x-tex">k^{th}< / annotation>< / semantics>< / com:Math> element of a second type of vector with index <com:Math><semantics>f<annotation encoding="application / x-tex">f< / annotation>< / semantics>< / com:Math> in the second type of vector set, <com:Math><semantics>Sl,tx<annotation encoding="application / x-tex">S_{l,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,f,tx<annotation encoding="application / x-tex">S_{l,f,t}^x< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>Stx<annotation encoding="application / x-tex">S_t^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,i,f,tx<annotation encoding="application / x-tex">S_{l,i,f,t}^x< / annotation>< / semantics>< / com:Math> are an element associated with <com:Math><semantics>t<annotation encoding="application / x-tex">t< / annotation>< / semantics>< / com:Math> of a third type of vector with index <com:Math><semantics>x<annotation encoding="application / x-tex">x< / annotation>< / semantics>< / com:Math> in the third type of vector set. The phase and amplitude of <com:Math><semantics>al,i,f,j<annotation encoding="application / x-tex">a_{l,i,f,j}< / annotation>< / semantics>< / com:Math> should be feedback to gNB. The amplitude of <com:Math><semantics>al,i,f,j<annotation encoding="application / x-tex">a_{l,i,f,j}< / annotation>< / semantics>< / com:Math> is smaller than or equals to 1. The one set of PMI includes at least one of: information about the one of first type of vector set, information about the one or v the second vector sets, information about the D third type of vector sets, information about <com:Math><semantics>al,i,f,j,i∈{0,1,...,2L−1}<annotation encoding="application / x-tex">a_{l,i,f,j}, i \in \{0,1,...,2L-1\}< / annotation>< / semantics>< / com:Math> (e.g., the weighted coefficients). In some implementation, the first type of vector <com:Math><semantics>vi<annotation encoding="application / x-tex">v_i< / annotation>< / semantics>< / com:Math> can be got according to following formula [Image disponible dans le document PDF, Image available in the PDF document] (5) [Image disponible dans le document PDF, Image available in the PDF document] (6) The information about the one first vector set includes the index of <com:Math><semantics>e,m<annotation encoding="application / x-tex">e, m< / annotation>< / semantics>< / com:Math> for each <com:Math><semantics>i=0,1,...,L−1<annotation encoding="application / x-tex">i = 0,1,...,L-1< / annotation>< / semantics>< / com:Math> . For example, <com:Math><semantics>ei=O1n1i+q1,n1i∈{0,1,...N1−1},q1∈{0,1,...O1−1}<annotation encoding="application / x-tex">e_i = O_1 n_{1i} + q_1, n_{1i} \in \{0,1,...N_1-1\}, q_1 \in \{0,1,...O_1-1\}< / annotation>< / semantics>< / com:Math> , <com:Math><semantics>mi=O2n2,i+q2,n2,i∈{0,1,...N2−1},q2∈{0,1,...,O2−1}<annotation encoding="application / x-tex">m_i = O_2 n_{2,i} + q_2, n_{2,i} \in \{0,1,...N_2 - 1\}, q_2 \in \{0,1,...,O_2 - 1\}< / annotation>< / semantics>< / com:Math> for <com:Math><semantics>vi<annotation encoding="application / x-tex">v_i< / annotation>< / semantics>< / com:Math>, wherein <com:Math><semantics>vi=vei,mi<annotation encoding="application / x-tex">v_i = v_{e_i,m_i}< / annotation>< / semantics>< / com:Math>. The PMI of the one first vector set includes information of <com:Math><semantics>n1,i,n2,i<annotation encoding="application / x-tex">n_{1,i}, n_{2,i}< / annotation>< / semantics>< / com:Math> for each <com:Math><semantics>i=0,1,...,L−1<annotation encoding="application / x-tex">i = 0,1,...,L-1< / annotation>< / semantics>< / com:Math> and information about <com:Math><semantics>q1<annotation encoding="application / x-tex">q_1< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>q2<annotation encoding="application / x-tex">q_2< / annotation>< / semantics>< / com:Math> which are same for all <com:Math><semantics>i<annotation encoding="application / x-tex">i< / annotation>< / semantics>< / com:Math>. In some implementation, only one element with index <com:Math><semantics>mimod⁡P2<annotation encoding="application / x-tex">m_i \mod \frac{P}{2}< / annotation>< / semantics>< / com:Math> of <com:Math><semantics>vi<annotation encoding="application / x-tex">v_i< / annotation>< / semantics>< / com:Math> is value 0, remaining <com:Math><semantics>P2<annotation encoding="application / x-tex">\frac{P}{2}< / annotation>< / semantics>< / com:Math>-1 elements are value 0, the CSI includes information about <com:Math><semantics>mi∈{0,1,...,P2−1}<annotation encoding="application / x-tex">m_i \in \{0,1,...,\frac{P}{2}-1\}< / annotation>< / semantics>< / com:Math> for each <com:Math><semantics>vi<annotation encoding="application / x-tex">v_i< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>i=0,1,...L−1<annotation encoding="application / x-tex">i = 0,1,...L-1< / annotation>< / semantics>< / com:Math>. In some implementation, one second frequency domain unit includes one subband or half of one subband, a second type of vector <com:Math><semantics>Yif<annotation encoding="application / x-tex">Y_i^f< / annotation>< / semantics>< / com:Math> has following format <com:Math><semantics>Y⋅f=[1.ej2πn3,lfN3,ej2π2n3,lfN3...,ej2πn3,lfkN3...,ej2πn3,lf(N3−1)N3]T,n3,lf∈{0,1,...,N3−1}The yk,l(f)<annotation encoding="application / x-tex">Y_{\cdot}^{f} = \begin{bmatrix} 1.e^{j2\pi \frac{n_{3,l}^{f}}{N_{3}}}, & e^{j2\pi \frac{2n_{3,l}^{f}}{N_{3}}}..., e^{j2\pi \frac{n_{3,l}^{f}k}{N_{3}}}..., e^{j2\pi \frac{n_{3,l}^{f}(N_{3}-1)}{N_{3}}} \end{bmatrix}^{T}, n_{3,l}^{f} \in \{0,1,...,N_{3}-1\} \quad \text{The } y_{k,l}^{(f)}< / annotation>< / semantics>< / com:Math> <com:Math><semantics>kth<annotation encoding="application / x-tex">k^{th}< / annotation>< / semantics>< / com:Math> element of the second vector <com:Math><semantics>Yif<annotation encoding="application / x-tex">Y_i^f< / annotation>< / semantics>< / com:Math>. The information about one second vector set included in the CSI includes information about <com:Math><semantics>n3,lf<annotation encoding="application / x-tex">n_{3,l}^f< / annotation>< / semantics>< / com:Math> for each second type of vector in the second vector set. In some implementation, the information about the second vector set included in the CSI doesn't include absolute <com:Math><semantics>n3,lf<annotation encoding="application / x-tex">n_{3,l}^f< / annotation>< / semantics>< / com:Math> for each second type of vector in the second vector set. It includes <com:Math><semantics>n3,lf<annotation encoding="application / x-tex">n_{3,l}^f< / annotation>< / semantics>< / com:Math> after remapping <com:Math><semantics>n3,lf=(n3,lf−n3,lf*)mod⁡N3<annotation encoding="application / x-tex">n_{3,l}^f = (n_{3,l}^f - n_{3,l}^{f^*}) \mod N_3< / annotation>< / semantics>< / com:Math> and it only includes <com:Math><semantics>n3,lf<annotation encoding="application / x-tex">n_{3,l}^f< / annotation>< / semantics>< / com:Math> for <com:Math><semantics>M−1<annotation encoding="application / x-tex">M - 1< / annotation>< / semantics>< / com:Math> second type of vectors in the second vector set and it doesn't includes <com:Math><semantics>n3,lf<annotation encoding="application / x-tex">n_{3,l}^f< / annotation>< / semantics>< / com:Math> for a second vector with <com:Math><semantics>n3,lf*=0<annotation encoding="application / x-tex">n_{3,l}^{f^*} = 0< / annotation>< / semantics>< / com:Math> after remapping. In some implementation, the second vector set is specific to one layer. The CSI includes information about a second vector set for each layer. That is the CSI includes v second type of vector sets. Each of the v second type of vector sets is used to determine a column of each of the <com:Math><semantics>N3*C3<annotation encoding="application / x-tex">N_3 * C_3< / annotation>< / semantics>< / com:Math> precoding matrices. Different layers respectively corresponds to one piece of information about a second type of vector set and different layers respectively corresponds to different second vector sets. In another implementation, the second type of vector set are shared by all layers. The CSI includes information about the one second type of vector set shared by all layers, then <com:Math><semantics>yk,l(f)<annotation encoding="application / x-tex">y_{k,l}^{(f)}< / annotation>< / semantics>< / com:Math> in formula (1-3) can be replace with <com:Math><semantics>yk(f)<annotation encoding="application / x-tex">y_k^{(f)}< / annotation>< / semantics>< / com:Math>. <com:Math><semantics>Ylf<annotation encoding="application / x-tex">Y_l^f< / annotation>< / semantics>< / com:Math> can be replaced with <com:Math><semantics>Yf<annotation encoding="application / x-tex">Y^f< / annotation>< / semantics>< / com:Math>. In some implementation, each first type of time domain unit includes one OFDM symbol, t is index of OFDM symbol index during <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math>. In some implementation, each first type of time domain unit includes one or more OFDM symbols, <com:Math><semantics>t<annotation encoding="application / x-tex">t< / annotation>< / semantics>< / com:Math> is index of OFDM symbol group index during <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math>, that is t is the index of the first type of time domain unit. One OFDM symbol group corresponds to one first type of time domain unit. Different first type of time domain units with different index t include same or different number of OFDM symbols. In some implementation, one first type of time domain unit includes up to one transmission occasion of the P of CSI-RS ports. In some implementation, one first type of time domain unit includes to one or more slots. In some implementation, the period of reporting the CSI is equals to period of the P of CSI-RS ports. For example, the period of reporting the CSI can be not configured. It equals to the period of the P CSI-RS. Time where PUSCH / PUCCH (physical uplink shared channel or a physical uplink control channel) including the CSI including PMI is located is based on the position of the last transmission occasion of the P CSI-RS ports of one burst. It is after the last transmission occasion of the P of CSI-RS ports of one burst as shown in FIG. 1 or FIG. 2. The gNB can inform gap between the last transmission occasion of the P of CSI-RS ports of one burst and starting time where PUSCH / PUCCH including the CSI is located. The information about the gap includes at least one of a quantity of slot, or a quantity of OFDM symbol. In some implementation, the number of OFDM symbols in one first type of time domain unit depends on at least one of the number OFDM symbols in one CSI-RS 1 transmission occasion, or the OFDM symbol gap between the starting OFDM symbol and the ending OFDM symbol of one CSI-RS1 transmission occasion, maximal of <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> OFDM symbol gaps each of which is OFDM symbol gap between the starting OFDM symbol and the ending OFDM symbol of one CSI-RS1 transmission occasion of <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> CSI-RS1 transmission occasion, received value, <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> a maximal value of <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math>, a maximal value of <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math>, a value reported by the UE, or a capability reported by the UE, wherein <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math> is doppler shift(or doppler spread) associated with a third type of vector with index <com:Math><semantics>x<annotation encoding="application / x-tex">x< / annotation>< / semantics>< / com:Math> and the unit of <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> is Hz, the unit of <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> is 1 or one sub carrier space. For example, the number of OFDM symbols in one first type of time domain unit is equal to or larger than at least one of the number OFDM symbols in one CSI-RS 1 transmission occasion, or the OFDM symbol gap between the starting OFDM symbol and the ending OFDM symbol of one CSI-RS1 transmission occasion, or the maximal of <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> OFDM symbol gaps. Each third type of vector is based on at least one of doppler frequency shift <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math>,, <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> sub-carrier space <com:Math><semantics>Δf<annotation encoding="application / x-tex">\Delta f< / annotation>< / semantics>< / com:Math>, or length of a OFDM symbols. For example, any one of the time-domain vector <com:Math><semantics>stx<annotation encoding="application / x-tex">s_t^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>sl,tx<annotation encoding="application / x-tex">s_{l,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>sl,f,tx<annotation encoding="application / x-tex">s_{l,f,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>sl,i,f,tx<annotation encoding="application / x-tex">s_{l,i,f,t}^x< / annotation>< / semantics>< / com:Math> (e.g., the third type of vector) for a first type of time domain unit t can have at least one of following formats: [Image disponible dans le document PDF, Image available in the PDF document] time-domain vector are based on parameter <com:Math><semantics>n3,lf<annotation encoding="application / x-tex">n_{3,l}^f< / annotation>< / semantics>< / com:Math> of frequency domain vector (e.g., the second type of vector) . In addition, because the reported <com:Math><semantics>n3,lf<annotation encoding="application / x-tex">n_{3,l}^f< / annotation>< / semantics>< / com:Math> for the second type of vector with index <com:Math><semantics>f<annotation encoding="application / x-tex">f< / annotation>< / semantics>< / com:Math> is a value after remapping, that is the reported <com:Math><semantics>n3,lf=n3,l,af<annotation encoding="application / x-tex">n_{3,l}^f = n_{3,l,a}^f< / annotation>< / semantics>< / com:Math> wherein <com:Math><semantics>n3,l,af=(n3,l,bf−n3,l,bf*)mod⁡N3<annotation encoding="application / x-tex">n_{3,l,a}^f = (n_{3,l,b}^f - n_{3,l,b}^{f^*}) \mod N_3< / annotation>< / semantics>< / com:Math>. But if <com:Math><semantics>n3,lf<annotation encoding="application / x-tex">n_{3,l}^f< / annotation>< / semantics>< / com:Math> used to generate the third type of vector is a value before remapping <com:Math><semantics>n3,l,bf<annotation encoding="application / x-tex">n_{3,l,b}^f< / annotation>< / semantics>< / com:Math>, the third type of vector can match well with <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math> especially when <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math> is larger. For example the relationship between the third type of vector <com:Math><semantics>Sf,t,bx<annotation encoding="application / x-tex">S_{f,t,b}^x< / annotation>< / semantics>< / com:Math> using <com:Math><semantics>n3,l,bf<annotation encoding="application / x-tex">n_{3,l,b}^f< / annotation>< / semantics>< / com:Math> value before remapping and the third type of vector <com:Math><semantics>sf,t,ax<annotation encoding="application / x-tex">s_{f,t,a}^x< / annotation>< / semantics>< / com:Math> using <com:Math><semantics>n3,l,bf<annotation encoding="application / x-tex">n_{3,l,b}^f< / annotation>< / semantics>< / com:Math> value after remapping has following format [Image disponible dans le document PDF, Image available in the PDF document] <com:Math><semantics>Sf,t,ax<annotation encoding="application / x-tex">S_{f,t,a}^{x}< / annotation>< / semantics>< / com:Math> adds a new factor <com:Math><semantics>e−j2πFxn3,l(f*)<annotation encoding="application / x-tex">e^{-j2\pi F_{x}n_{3,l}^{(f^{*})}}< / annotation>< / semantics>< / com:Math> compared with <com:Math><semantics>Sf,t,bx<annotation encoding="application / x-tex">S_{f,t,b}^{x}< / annotation>< / semantics>< / com:Math> which matches the channel better. If we use <com:Math><semantics>n3,lf<annotation encoding="application / x-tex">n_{3,l}^f< / annotation>< / semantics>< / com:Math> after remapping to generate the third type of vector, such as <com:Math><semantics>sl,f,tx=sl,f,t,ax<annotation encoding="application / x-tex">s_{l,f,t}^x = s_{l,f,t,a}^x< / annotation>< / semantics>< / com:Math>. The factor <com:Math><semantics>e−j2πFxn3,l(f*)<annotation encoding="application / x-tex">e^{-j2\pi F_x n_{3,l}^{(f^*)}}< / annotation>< / semantics>< / com:Math> need <com:Math><semantics>al,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x}< / annotation>< / semantics>< / com:Math> to offset. It increases the number of bits for reporting <com:Math><semantics>al,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x}< / annotation>< / semantics>< / com:Math> especially when <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math> is large. So the <com:Math><semantics>n3,lf<annotation encoding="application / x-tex">n_{3,l}^{f}< / annotation>< / semantics>< / com:Math> before remapping should be reported. For example, the UE reports one value <com:Math><semantics>n3,l,bf*∈{0,1,...,N3−1},f*=0<annotation encoding="application / x-tex">n_{3,l,b}^{f^*} \in \{0,1,...,N_3-1\}, f^*=0< / annotation>< / semantics>< / com:Math> before remapping and M-1 values <com:Math><semantics>n3,l,af∈{0,1,...,N3−1},f=1,...,M−1<annotation encoding="application / x-tex">n_{3,l,a}^f \in \{0,1,...,N_3-1\}, f=1,...,M-1< / annotation>< / semantics>< / com:Math> after remapping. <com:Math><semantics>n3,lf*<annotation encoding="application / x-tex">n_{3,l}^{f^*}< / annotation>< / semantics>< / com:Math> corresponds to strongest value of <com:Math><semantics>al,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x}< / annotation>< / semantics>< / com:Math> Then the <com:Math><semantics>n3,l,bf=(n3,l,af+n3,l,bf*)mod⁡N3,f=0,1,...,M−1<annotation encoding="application / x-tex">n_{3,l,b}^f = (n_{3,l,a}^f + n_{3,l,b}^{f^*}) \mod N_3, f = 0,1,...,M-1< / annotation>< / semantics>< / com:Math>, the value before remapping is used to generate the third type of vector. That is <com:Math><semantics>Sl,f,x=Sl,f,x,b<annotation encoding="application / x-tex">S_{l,f,x} = S_{l,f,x,b}< / annotation>< / semantics>< / com:Math>. The value before remapping or the value after remapping is used to generate the second type of vector. The second type of vector using a value before remapping and the second type of vector using a value after remapping is same in terms of performance. In another implementation, the UE reports M values <com:Math><semantics>n3,l,bf∈{0,1,...,N3−1},f=0,...,M−1<annotation encoding="application / x-tex">n_{3,l,b}^f \in \{0,1,...,N_3-1\}, f = 0,...,M-1< / annotation>< / semantics>< / com:Math> before remapping. The <com:Math><semantics>M3,l,bf<annotation encoding="application / x-tex">M_{3,l,b}^f< / annotation>< / semantics>< / com:Math> value can be directly used to generate the second type of vector and the third type of vector. But the strongest value of <com:Math><semantics>al,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x}< / annotation>< / semantics>< / com:Math> doesn't always correspond to <com:Math><semantics>f=0<annotation encoding="application / x-tex">f=0< / annotation>< / semantics>< / com:Math>. The <com:Math><semantics>f<annotation encoding="application / x-tex">f< / annotation>< / semantics>< / com:Math> corresponding to the strongest value of <com:Math><semantics>al,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x}< / annotation>< / semantics>< / com:Math> should be feedback to gNB by the UE. The remapping operation refers to <com:Math><semantics>n3,l,af=(n3,l,bf−n3,l,bf*)mod⁡N3<annotation encoding="application / x-tex">n_{3,l,a}^f = (n_{3,l,b}^f - n_{3,l,b}^{f^*}) \mod N_3< / annotation>< / semantics>< / com:Math> Alt 2: <com:Math><semantics>Stx<annotation encoding="application / x-tex">S_t^x< / annotation>< / semantics>< / com:Math> isn't specific to each second type of vector because only one <com:Math><semantics>n3,ly<annotation encoding="application / x-tex">n_{3,l}^y< / annotation>< / semantics>< / com:Math> is used to determine the third type of vector. Then <com:Math><semantics>Sf,tx<annotation encoding="application / x-tex">S_{f,t}^{x}< / annotation>< / semantics>< / com:Math> is replaced with <com:Math><semantics>Sfx<annotation encoding="application / x-tex">S_{f}^{x}< / annotation>< / semantics>< / com:Math>, <Formule mathématique disponible dans le document PDF, Math available in the PDF document display="block">S_{t}^{x} = e^{j2\pi f_{x} \frac{n_{3,l}^{(f^{*})}}{N} + j2\pi F_{x} t T_{OFDM}}, x = 0,1,...X - 1, \text{ wherein } n_{3,l}^{f^{**}} \text{ is one of } n_{3,l,b}^{f^{*}}, \max_{f} n_{3,l,b}^{f}, \max_{f} n_{3,l,a}^{f}, n_{3,l,a}^{f^{*}}, n_{ [Image disponible dans le document PDF, Image available in the PDF document] [Image disponible dans le document PDF, Image available in the PDF document] [Image disponible dans le document PDF, Image available in the PDF document] one of <com:Math><semantics>n3,l,bf<annotation encoding="application / x-tex">n_{3,l,b}^f< / annotation>< / semantics>< / com:Math> which is a value before remapping or <com:Math><semantics>n3,l,af<annotation encoding="application / x-tex">n_{3,l,a}^f< / annotation>< / semantics>< / com:Math> which is a value after remapping as done in Alt1. Alt5: <com:Math><semantics>stx=ej2πfxt(N+CP)N,x=0,1...,X−1<annotation encoding="application / x-tex">s_t^x = e^{\frac{j2\pi f_x t(N+CP)}{N}}, x = 0,1..., X-1< / annotation>< / semantics>< / com:Math> [Image disponible dans le document PDF, Image available in the PDF document] Alt6: [Image disponible dans le document PDF, Image available in the PDF document] Alt7: [Image disponible dans le document PDF, Image available in the PDF document] Alt8: Alt9: <com:Math><semantics>stx=ej2πfx(n3,l(f**)+t(N+CP))N,x=0,1...,X−1,wherein n3,lf** is one of n3,l,bf*,max⁡fn3,l,bf,max⁡fn3,l,af,<annotation encoding="application / x-tex">s_t^x = e^{\frac{j2\pi f_x(n_{3,l}^{(f^{**})} + t(N+CP))}{N}}, x = 0,1..., X-1, \text{wherein } n_{3,l}^{f^{**}} \text{ is one of } n_{3,l,b}^{f^*}, \max_f n_{3,l,b}^f, \max_f n_{3,l,a}^f,< / annotation>< / semantics>< / com:Math> [Image disponible dans le document PDF, Image available in the PDF document] [Image disponible dans le document PDF, Image available in the PDF document] [Image disponible dans le document PDF, Image available in the PDF document] [Image disponible dans le document PDF, Image available in the PDF document] <com:Math><semantics>fx=FxΛf<annotation encoding="application / x-tex">f_x = \frac{F_x}{\Lambda f}< / annotation>< / semantics>< / com:Math>. When the second type of vector set isn't reported for each layer, the subscript <com:Math><semantics>l<annotation encoding="application / x-tex">l< / annotation>< / semantics>< / com:Math> of <com:Math><semantics>n3,lf,n3,lf*,n3,lf**,n3,l,af,n3,l,bf<annotation encoding="application / x-tex">n_{3,l}^f, n_{3,l}^{f^*}, n_{3,l}^{f^{**}}, n_{3,l,a}^f, n_{3,l,b}^f< / annotation>< / semantics>< / com:Math> can be deleted. Note that <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> is a value associated with a third type of vector with index <com:Math><semantics>x<annotation encoding="application / x-tex">x< / annotation>< / semantics>< / com:Math>, which may be smaller than 3 or 2, for example, <com:Math><semantics>−432−μ≤fx≤432−μ<annotation encoding="application / x-tex">-\frac{4}{3}2^{-\mu} \le f_x \le \frac{4}{3}2^{-\mu}< / annotation>< / semantics>< / com:Math>, but <com:Math><semantics>f<annotation encoding="application / x-tex">f< / annotation>< / semantics>< / com:Math> is an index of second type of vector among M second type of vectors <com:Math><semantics>f∈{0,1,..M−1}<annotation encoding="application / x-tex">f \in \{0,1,..M-1\}< / annotation>< / semantics>< / com:Math>. The sub carrier space of the CSI-RS is <com:Math><semantics>Δf=2μ*15kHz<annotation encoding="application / x-tex">\Delta f = 2^{\mu} * 15kHz< / annotation>< / semantics>< / com:Math>. <com:Math><semantics>TOFDM<annotation encoding="application / x-tex">T_{OFDM}< / annotation>< / semantics>< / com:Math> is time duration of one OFDM symbol. For example [Image disponible dans le document PDF, Image available in the PDF document] [Image disponible dans le document PDF, Image available in the PDF document] wherein <com:Math><semantics>lOFDM<annotation encoding="application / x-tex">l_{OFDM}< / annotation>< / semantics>< / com:Math> is OFDM symbol index in a subframe. <com:Math><semantics>κ=64<annotation encoding="application / x-tex">\kappa = 64< / annotation>< / semantics>< / com:Math>. For different <com:Math><semantics>lOFDM<annotation encoding="application / x-tex">l_{OFDM}< / annotation>< / semantics>< / com:Math>, the <Formule mathématique disponible dans le document PDF, Math available in the PDF document>$T_{\it OFDM}$ may be different, then <Formule mathématique disponible dans le document PDF, Math available in the PDF document>$T_{\it OFDM}$ can be marked by <Formule mathématique disponible dans le document PDF, Math available in the PDF document>$T_{\it OFDM}$, <Formule mathématique disponible dans le document PDF, Math available in the PDF document>$t_{\it OFDM}$ <com:Math><semantics>N3<annotation encoding="application / x-tex">N_3< / annotation>< / semantics>< / com:Math> is number of frequency domain unit in one OFDM symbol or in one first type of time domain unit. <com:Math><semantics>N=Nuμ<annotation encoding="application / x-tex">N = N_u^{\mu}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>CP=Ncp,lOFDMμ<annotation encoding="application / x-tex">CP = N_{cp,l_{OFDM}}^{\mu}< / annotation>< / semantics>< / com:Math>, In another implementation, <com:Math><semantics>N=Nuμ / (κ2−μ)<annotation encoding="application / x-tex">N = N_u^{\mu} / (\kappa 2^{-\mu})< / annotation>< / semantics>< / com:Math> <com:Math><semantics>CP=Ncp,lOFDMμ / (κ2−μ)<annotation encoding="application / x-tex">CP = N^{\mu}_{cp,l_{OFDM}} / (\kappa 2^{-\mu})< / annotation>< / semantics>< / com:Math>. For example [Image disponible dans le document PDF, Image available in the PDF document] [Image disponible dans le document PDF, Image available in the PDF document] CP(cyclic prefix) [Image disponible dans le document PDF, Image available in the PDF document] We see that the CP length is different for different OFDM symbol set if normal CP is used. The first OFDM symbol set includes OFDM symbol with index <com:Math><semantics>lOFDM∈{0,7*2μ}<annotation encoding="application / x-tex">l_{OFDM} \in \{0,7*2^{\mu}\}< / annotation>< / semantics>< / com:Math>. The second OFDM symbol set includes OFDM symbol with index <com:Math><semantics>lOFDM∉{0,7*2μ}<annotation encoding="application / x-tex">l_{OFDM} \notin \{0,7*2^{\mu}\}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>lOFDM=0,1,...,14*2μ<annotation encoding="application / x-tex">l_{OFDM} = 0,1,...,14*2^{\mu}< / annotation>< / semantics>< / com:Math> <com:Math><semantics>t(N+CP)N=t*2192+z*16*2μ2048<annotation encoding="application / x-tex">\frac{t(N+CP)}{N} = \frac{t*2192 + z*16*2^{\mu}}{2048}< / annotation>< / semantics>< / com:Math>, wherein z is the number of OFDM symbols with longer CP from start of a first type of time domain unit with <com:Math><semantics>t=0<annotation encoding="application / x-tex">t = 0< / annotation>< / semantics>< / com:Math> to the end of the time unit with current t. If one first type of time domain unit includes one OFDM symbol, the t is an index of the OFDM symbol relative to start of <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math> which is not start of a slot then [Image disponible dans le document PDF, Image available in the PDF document] ~ ~ If the <com:Math><semantics>t<annotation encoding="application / x-tex">t< / annotation>< / semantics>< / com:Math> is an index of the OFDM symbol relative to start OFDM symbol of <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math> which is start of a slot then <com:Math><semantics>t(N+CP)N=t*2192+⌊t / (7*2μ)⌋*16*2μN<annotation encoding="application / x-tex">\frac{t(N+CP)}{N} = \frac{t*2192 + \left\lfloor t / (7*2^{\mu})\right\rfloor *16*2^{\mu}}{N}< / annotation>< / semantics>< / com:Math> If one of Alt 1,2,3,6,11,or 7 is adopted, then <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math> whose unit is Hz needs to be reported. We need to determine the quantify step for reporting <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math> should be determined. In some implementation, the quantification step may be 100Hz, or 50Hz, or other value. If one of Alt 4,5,8,9,11, or 10 is adopted, then <com:Math><semantics>fx=FxΔf<annotation encoding="application / x-tex">f_x = \frac{F_x}{\Delta f}< / annotation>< / semantics>< / com:Math>. The unit of <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> is 1, or we can say that the unit of <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> is <com:Math><semantics>Δf<annotation encoding="application / x-tex">\Delta f< / annotation>< / semantics>< / com:Math>. <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> needs to be reported. In some implementation, the quantify step for reporting <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> should be determined. For example, <com:Math><semantics>fx∈{0,0.1,0.2,...,2}<annotation encoding="application / x-tex">f_x \in \{0,0.1,0.2,...,2\}< / annotation>< / semantics>< / com:Math>. In some implementation, the quantify step depends on sub carrier spacing <com:Math><semantics>Δf<annotation encoding="application / x-tex">\Delta f< / annotation>< / semantics>< / com:Math>. For example, larger the reference sub carrier spacing <com:Math><semantics>Δf<annotation encoding="application / x-tex">\Delta f< / annotation>< / semantics>< / com:Math> is, the smaller the quantify step is. For example, if the reference sub carrier spacing <com:Math><semantics>Δf<annotation encoding="application / x-tex">\Delta f< / annotation>< / semantics>< / com:Math> is 15kHz, then the quantify step is 0.1, that is <com:Math><semantics>fx∈{0,0.05,0.1,...,2}<annotation encoding="application / x-tex">f_x \in \{0,0.05,0.1,...,2\}< / annotation>< / semantics>< / com:Math>. If the reference sub carrier spacing <com:Math><semantics>Δf<annotation encoding="application / x-tex">\Delta f< / annotation>< / semantics>< / com:Math> is 30kHz, then the quantify step is 0.05, that is <com:Math><semantics>fr∈{0,0.05,0.1,...,2}<annotation encoding="application / x-tex">f_r \in \{0,0.05,0.1,...,2\}< / annotation>< / semantics>< / com:Math>. The reference sub carrier spacing <com:Math><semantics>Δf<annotation encoding="application / x-tex">\Delta f< / annotation>< / semantics>< / com:Math> can be subcarrier space of the p CSI-RS ports, or a sub carrier space configured in one CSI reporting. In above description(we call it as a first method of for determining t of a third type of vector, or each of third type of vector), <com:Math><semantics>t=0,1...,C3−1<annotation encoding="application / x-tex">t = 0,1...,C_3-1< / annotation>< / semantics>< / com:Math>, that is for a frequency domain unit <com:Math><semantics>k<annotation encoding="application / x-tex">k< / annotation>< / semantics>< / com:Math>, the UE feedback <com:Math><semantics>C3<annotation encoding="application / x-tex">C_3< / annotation>< / semantics>< / com:Math> precoding matrices each of which corresponds to one domain unit of <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math>. The UE feedback a respective precoding matrix for each time unit with index t of <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math>, that is <com:Math><semantics>C4=C3<annotation encoding="application / x-tex">C_4 = C_3< / annotation>< / semantics>< / com:Math> The <com:Math><semantics>t<annotation encoding="application / x-tex">t< / annotation>< / semantics>< / com:Math> is relative to start of the <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math>. In a second method of for determining t of a third type of vector, the UE just feedback a precoding matrix for a set of a first type of time domain unit during <com:Math><semantics>T1<annotation encoding="application / x-tex">T_1< / annotation>< / semantics>< / com:Math>, each first type of time domain unit in the set includes one transmission occasion of the PCSI-RS ports, that is <com:Math><semantics>t=tCSI−RS,occasion−1,tCSI−RS,occasion−2,…,tCSI−RS,occasion−C1<annotation encoding="application / x-tex">t = t_{CSI-RS,occasion-1}, t_{CSI-RS,occasion-2}, \dots, t_{CSI-RS,occasion-C_1}< / annotation>< / semantics>< / com:Math> in formula (1-4), or other formula of third type of vector, <com:Math><semantics>tCSI−RS,occasion−i<annotation encoding="application / x-tex">t_{CSI-RS,occasion-i}< / annotation>< / semantics>< / com:Math> is the index of one first type domain time unit including the <com:Math><semantics>tCSI−RS<annotation encoding="application / x-tex">t_{CSI-RS}< / annotation>< / semantics>< / com:Math> transmission occasion during <com:Math><semantics>T1<annotation encoding="application / x-tex">T_1< / annotation>< / semantics>< / com:Math>. The first type of time domain unit index is relative to first OFDM symbol of a time unit with index t=0, wherein a time unit with index t=0 is before the CSI-RS transmission occasion with index <com:Math><semantics>i=0<annotation encoding="application / x-tex">i = 0< / annotation>< / semantics>< / com:Math>. For example, the CSI-RS 1 transmission occasion 1 is in OFDM 8 in slot 3 and one first type of time unit includes one OFDM symbol, and the t is relative to starting of slot 3, then <com:Math><semantics>tCSI−RS,occasion−1<annotation encoding="application / x-tex">t_{CSI-RS,occasion-1}< / annotation>< / semantics>< / com:Math> equals 9. <com:Math><semantics>C4=C1<annotation encoding="application / x-tex">C_4 = C_1< / annotation>< / semantics>< / com:Math> In a third method for determining t of a third type of vector, it is same with the second implementation except that [Image disponible dans le document PDF, Image available in the PDF document] That is the t is relative to a first type of time domain unit including the first transmission occasion of PCSI-RS ports. The first type of time domain unit with index 0 includes the first transmission occasion of PCSI-RS ports. <com:Math><semantics>C4=C1<annotation encoding="application / x-tex">C_4 = C_1< / annotation>< / semantics>< / com:Math> In a fourth method for determining t of a third type of vector, it is same with the first implementation except that [Image disponible dans le document PDF, Image available in the PDF document] wherein <com:Math><semantics>tCSI−RS,position<annotation encoding="application / x-tex">t_{CSI-RS, position}< / annotation>< / semantics>< / com:Math> is the i<com:Sup>th< / com:Sup> transmission occasion of the PCSI-RS ports during <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math>. t is relative to a OFDM symbol which is before the first OFDM symbol of a time unit including the first transmission occasion of the PCSI-RS ports in a burst. That is the first time unit of <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math> is before the first OFDM symbol of a first type of time unit including the first transmission occasion of the P CSI-RS ports in a burst. In some implementation, if all <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time units have same length in time such as in unit of second, then the t is a first type of time domain unit index. If at least two first type of time domain units have different length in time, the t is OFDM symbols index. In some implementation, if all <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time units have same number of OFDM symbols, then the t is a first type of time domain unit index. If at least two first type of time domain units have different numbers of OFDM symbols, the t is OFDM symbols index. For determining t of a third type of vector (or for each third type of vector in the second vector set), if the first or the fourth method is adopted, the PMI indicates <com:Math><semantics>N3<annotation encoding="application / x-tex">N_3< / annotation>< / semantics>< / com:Math> matrices for each first type of time domain unit during <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math> regardless whether the a first type of time domain unit includes a transmission occasion of the PCSI-RS ports, that is some first type of time domain units of the <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time domain units include one transmission occasion of the PCSI-RS ports and some first type of time domain units of the <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time domain units don't include any one transmission occasion of the PCSI-RS ports. The first type of time domain unit index is continuous. If the second or the third method is adopted, the PMI indicates <com:Math><semantics>N3<annotation encoding="application / x-tex">N_3< / annotation>< / semantics>< / com:Math> matrices for each first type of time domain unit which includes one transmission occasion of PCSI-RS ports. The first type of time domain unit index can be non-continuous, or continuous. In some implementation, if X is larger than 1, the X <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>x=0,1,...,X−1<annotation encoding="application / x-tex">x = 0,1,...,X-1< / annotation>< / semantics>< / com:Math> or X <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>x=0,1,...,X−1<annotation encoding="application / x-tex">x = 0,1,...,X-1< / annotation>< / semantics>< / com:Math> is reported by differential method with a reference value. If the UE reports <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_{x}< / annotation>< / semantics>< / com:Math>, the reference value can be <com:Math><semantics>Freference<annotation encoding="application / x-tex">F_{reference}< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>F0<annotation encoding="application / x-tex">F_0< / annotation>< / semantics>< / com:Math>. The reference value is reported using absolute value or the reference value is a value determined by a rule or a fixed value which isn't reported by the UE. If the reference value is <com:Math><semantics>Freference<annotation encoding="application / x-tex">F_{reference}< / annotation>< / semantics>< / com:Math>, the UE reports <com:Math><semantics>Freference<annotation encoding="application / x-tex">F_{reference}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>Fx=Fx−Freference<annotation encoding="application / x-tex">F_x = F_x - F_{reference}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>x=0.1,2,...X−1<annotation encoding="application / x-tex">x = 0.1, 2, ... X - 1< / annotation>< / semantics>< / com:Math>. If the reference value is <com:Math><semantics>F0<annotation encoding="application / x-tex">F_0< / annotation>< / semantics>< / com:Math>, the UE reports <com:Math><semantics>F0<annotation encoding="application / x-tex">F_0< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>Fx=Fx−F0<annotation encoding="application / x-tex">F_x = F_x - F_0< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>x=1,2,...X−1<annotation encoding="application / x-tex">x = 1, 2, ... X - 1< / annotation>< / semantics>< / com:Math>. The period of reporting the reference value such as <com:Math><semantics>Freference<annotation encoding="application / x-tex">F_{reference}< / annotation>< / semantics>< / com:Math> (or <com:Math><semantics>F0<annotation encoding="application / x-tex">F_0< / annotation>< / semantics>< / com:Math> ) can be equal to period of reporting <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math> or be multiple of period of reporting <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math>. If the UE reports <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math>, the reference value can be <com:Math><semantics>freference<annotation encoding="application / x-tex">f_{reference}< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>f0<annotation encoding="application / x-tex">f_0< / annotation>< / semantics>< / com:Math>. The reference value is reported using absolute value or the reference value is a value determined by a rule or a fixed value which isn't reported by the UE. If the reference value is <com:Math><semantics>freference<annotation encoding="application / x-tex">f_{reference}< / annotation>< / semantics>< / com:Math>, the UE reports <com:Math><semantics>freference<annotation encoding="application / x-tex">f_{reference}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>fx′=fx−freference<annotation encoding="application / x-tex">f_x' = f_x - f_{reference}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>x=0,1,2,...X−1<annotation encoding="application / x-tex">x = 0,1,2,...X-1< / annotation>< / semantics>< / com:Math>. If the reference value is <com:Math><semantics>f0<annotation encoding="application / x-tex">f_0< / annotation>< / semantics>< / com:Math>, the UE reports <com:Math><semantics>f0<annotation encoding="application / x-tex">f_0< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>fx=fx−f0<annotation encoding="application / x-tex">f_x = f_x - f_0< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>x=1,2,...X−1<annotation encoding="application / x-tex">x = 1, 2, ... X - 1< / annotation>< / semantics>< / com:Math>. The period of reporting the reference value such as <com:Math><semantics>freference<annotation encoding="application / x-tex">f_{reference}< / annotation>< / semantics>< / com:Math> (or <com:Math><semantics>f0<annotation encoding="application / x-tex">f_0< / annotation>< / semantics>< / com:Math>) can be equal to period of reporting <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> or be multiple of period of reporting <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math>. In some implementation, the <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math>, or <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> can be directly used to generate the third type of vector. For example, the <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math> in above Alt 1-3,6,7,11 of format of the third type of vector with index <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> can be replaced with <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math>. The <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> in above Alt 4,5,8-10,12 of format of the third type of vector with index <com:Math><semantics>x<annotation encoding="application / x-tex">x< / annotation>< / semantics>< / com:Math> can be replaced with <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math>. If the reference value is <com:Math><semantics>x<annotation encoding="application / x-tex">x< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>x<annotation encoding="application / x-tex">x< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>x<annotation encoding="application / x-tex">x< / annotation>< / semantics>< / com:Math> the UE just reports X-1 <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> for <com:Math><semantics>x=1,2,...,X−1<annotation encoding="application / x-tex">x = 1, 2, ..., X-1< / annotation>< / semantics>< / com:Math>. <com:Math><semantics>F0=0<annotation encoding="application / x-tex">F_0 = 0< / annotation>< / semantics>< / com:Math>. The UE doesn't report <com:Math><semantics>F0<annotation encoding="application / x-tex">F_0< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>f0<annotation encoding="application / x-tex">f_0< / annotation>< / semantics>< / com:Math> to gNB. In another implementation, the X <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>x=0,1,...,X−1<annotation encoding="application / x-tex">x = 0,1,...,X-1< / annotation>< / semantics>< / com:Math> or X <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>x=0,1,...,X−1<annotation encoding="application / x-tex">x = 0,1,...,X-1< / annotation>< / semantics>< / com:Math> is reported by divided by the reference value. The UE reports <com:Math><semantics>F0<annotation encoding="application / x-tex">F_0< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>Fx′=Fx / F0<annotation encoding="application / x-tex">F_x' = F_x / F_0< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>x=1,2,...X−1<annotation encoding="application / x-tex">x = 1, 2, ... X - 1< / annotation>< / semantics>< / com:Math> if the reference value is <com:Math><semantics>F0<annotation encoding="application / x-tex">F_0< / annotation>< / semantics>< / com:Math>. The UE reports <com:Math><semantics>Freference<annotation encoding="application / x-tex">F_{reference}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>Fx′=Fx / Freference<annotation encoding="application / x-tex">F_x' = F_x / F_{reference}< / annotation>< / semantics>< / com:Math> <com:Math><semantics>x=0.1,2,...X−1<annotation encoding="application / x-tex">x = 0.1, 2, ... X - 1< / annotation>< / semantics>< / com:Math> if the reference value is <com:Math><semantics>Freference<annotation encoding="application / x-tex">F_{reference}< / annotation>< / semantics>< / com:Math> which isn't one of <com:Math><semantics>X<annotation encoding="application / x-tex">X< / annotation>< / semantics>< / com:Math> <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math> . The UE reports <com:Math><semantics>f0<annotation encoding="application / x-tex">f_0< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>fx=fx / f0<annotation encoding="application / x-tex">f_x = f_x / f_0< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>x=1,2,...X−1<annotation encoding="application / x-tex">x = 1,2,...X - 1< / annotation>< / semantics>< / com:Math> if the reference value is <com:Math><semantics>f0<annotation encoding="application / x-tex">f_0< / annotation>< / semantics>< / com:Math>. The UE reports <com:Math><semantics>freference<annotation encoding="application / x-tex">f_{reference}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>fx′=fx / freference<annotation encoding="application / x-tex">f'_x = f_x / f_{reference}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>x=0.1,2,...X−1<annotation encoding="application / x-tex">x = 0.1, 2, ... X - 1< / annotation>< / semantics>< / com:Math> if the reference value is <com:Math><semantics>freference<annotation encoding="application / x-tex">f_{reference}< / annotation>< / semantics>< / com:Math> which isn't one of <com:Math><semantics>X<annotation encoding="application / x-tex">X< / annotation>< / semantics>< / com:Math> <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math>. The quantify step of <com:Math><semantics>f0<annotation encoding="application / x-tex">f_0< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>freference<annotation encoding="application / x-tex">f_{reference}< / annotation>< / semantics>< / com:Math>, or <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> depends on sub carrier spacing <com:Math><semantics>Δf<annotation encoding="application / x-tex">\Delta f< / annotation>< / semantics>< / com:Math>. For example, larger the reference sub carrier spacing <com:Math><semantics>Δf<annotation encoding="application / x-tex">\Delta f< / annotation>< / semantics>< / com:Math> is, the smaller the quantify step is. For example, if the reference sub carrier spacing <com:Math><semantics>Δf<annotation encoding="application / x-tex">\Delta f< / annotation>< / semantics>< / com:Math> is 15kHz, then the quantify step <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> of the reference value is 0.1, that is <com:Math><semantics>freference<annotation encoding="application / x-tex">f_{reference}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>or<annotation encoding="application / x-tex">or< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>f0∈{0,0.1,0.2,...,2}<annotation encoding="application / x-tex">f_0 \in \{0,0.1,0.2,...,2\}< / annotation>< / semantics>< / com:Math> and the step of <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> is 0.01 that is <com:Math><semantics>fx∈{0,0.01,0.02,...,1}<annotation encoding="application / x-tex">f_x \in \{0,0.01,0.02,...,1\}< / annotation>< / semantics>< / com:Math>. If the reference sub carrier spacing <com:Math><semantics>Δf<annotation encoding="application / x-tex">\Delta f< / annotation>< / semantics>< / com:Math> is 30kHz, then the quantify step of is 0.05, that is <com:Math><semantics>freference<annotation encoding="application / x-tex">f_{reference}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>f0∈{0,0.05,0.1,...,2}<annotation encoding="application / x-tex">f_0 \in \{0,0.05,0.1,...,2\}< / annotation>< / semantics>< / com:Math> and the step of <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> is 0.005 that is <com:Math><semantics>fx∈{0,0.005,0.01,...,1}<annotation encoding="application / x-tex">f_x \in \{0,0.005,0.01,...,1\}< / annotation>< / semantics>< / com:Math>. In some implementation, the reference value such as <com:Math><semantics>freference,f0,Freference,F0<annotation encoding="application / x-tex">f_{reference}, f_0, F_{reference}, F_0< / annotation>< / semantics>< / com:Math> of multiple serving cell can be same / have relationship. The <com:Math><semantics>freference<annotation encoding="application / x-tex">f_{reference}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>f0<annotation encoding="application / x-tex">f_0< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Freference<annotation encoding="application / x-tex">F_{reference}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>F0<annotation encoding="application / x-tex">F_0< / annotation>< / semantics>< / com:Math> used to determine first precoding matrices corresponding to CSI-RS ports received in serving cell 1 and the <com:Math><semantics>freference,f0,Freference,F0<annotation encoding="application / x-tex">f_{reference}, f_0, F_{reference}, F_0< / annotation>< / semantics>< / com:Math> used to determine precoding matrices corresponding to CSI-RS ports received in serving cell 2 are same or have relationship. For example, [Image disponible dans le document PDF, Image available in the PDF document] [Image disponible dans le document PDF, Image available in the PDF document] [Image disponible dans le document PDF, Image available in the PDF document] [Image disponible dans le document PDF, Image available in the PDF document] wherein <com:Math><semantics>b=1<annotation encoding="application / x-tex">b=1< / annotation>< / semantics>< / com:Math>, or <com:Math><semantics>b<annotation encoding="application / x-tex">b< / annotation>< / semantics>< / com:Math> is based on at least one of: received signaling, a reported value, frequency carrier of serving cell 1, frequency carrier of serving cell 2, or ratio between the two frequency carriers of the two serving cells. For example, b equals a quotient result of frequency carrier of serving cell 1 divided by frequency carrier of serving cell 2. The granularity of the third type of vector set can be at least one of In some implementation, the third type of vector set is specific to a layer. The CSI reported by the UE includes information about v third type of vector sets each of which corresponds to one layer of <com:Math><semantics>v<annotation encoding="application / x-tex">v< / annotation>< / semantics>< / com:Math> layers. The precoding matrix <com:Math><semantics>Wtl<annotation encoding="application / x-tex">W_t^l< / annotation>< / semantics>< / com:Math> for each layer <com:Math><semantics>l=0,1,...,v−1<annotation encoding="application / x-tex">l = 0,1,...,v-1< / annotation>< / semantics>< / com:Math> is based on <com:Math><semantics>Sl,tx<annotation encoding="application / x-tex">S_{l,t}^x< / annotation>< / semantics>< / com:Math>, that is D equals <com:Math><semantics>v<annotation encoding="application / x-tex">v< / annotation>< / semantics>< / com:Math>, each of the <com:Math><semantics>v<annotation encoding="application / x-tex">v< / annotation>< / semantics>< / com:Math> third type of vector set is used to determine a column of each of the <com:Math><semantics>N3*C4<annotation encoding="application / x-tex">N_3 * C_4< / annotation>< / semantics>< / com:Math> precoding matrices and the column corresponds to the layer. The one set of PMI includes information about <com:Math><semantics>v<annotation encoding="application / x-tex">v< / annotation>< / semantics>< / com:Math> third type of vector sets. For example, the one set PMI includes <com:Math><semantics>v<annotation encoding="application / x-tex">v< / annotation>< / semantics>< / com:Math> set of <com:Math><semantics>Fx,Fx,fx,orfx<annotation encoding="application / x-tex">F_x, F_x, f_x, or f_x< / annotation>< / semantics>< / com:Math>. The <com:Math><semantics>Fx,Fx,fx,orfx<annotation encoding="application / x-tex">F_x, F_x, f_x, or f_x< / annotation>< / semantics>< / com:Math> in Alt 1- 12 can be replaced by <com:Math><semantics>Fl,x,Fl,x′,fl,x,orfl,x′<annotation encoding="application / x-tex">F_{l,x}, F'_{l,x}, f_{l,x}, or f'_{l,x}< / annotation>< / semantics>< / com:Math> respectively. In some implementation, the third type of vector set is specific to all layers. In some implementation, the third type of vector set is specific to each first type of vector with index <com:Math><semantics>i=0,1,...,L−1<annotation encoding="application / x-tex">i = 0,1,...,L-1< / annotation>< / semantics>< / com:Math>, the subscript of s has <com:Math><semantics>i<annotation encoding="application / x-tex">i< / annotation>< / semantics>< / com:Math>, such as <com:Math><semantics>Si,tx<annotation encoding="application / x-tex">S_{i,t}^x< / annotation>< / semantics>< / com:Math>. D equals <com:Math><semantics>L<annotation encoding="application / x-tex">L< / annotation>< / semantics>< / com:Math>. The one set of PMI includes information about <com:Math><semantics>L<annotation encoding="application / x-tex">L< / annotation>< / semantics>< / com:Math> third type of vector sets. For example, the one set PMI includes <com:Math><semantics>L<annotation encoding="application / x-tex">L< / annotation>< / semantics>< / com:Math> sets of <com:Math><semantics>Fx,Fx,fx,orfx<annotation encoding="application / x-tex">F_x, F_x, f_x, or f_x< / annotation>< / semantics>< / com:Math>. The <com:Math><semantics>Fx,Fx,fx,orfx<annotation encoding="application / x-tex">F_x, F_x, f_x, or f_x< / annotation>< / semantics>< / com:Math> in Alt 1- 12 can be replaced by <com:Math><semantics>Fi,x,Fi,x,orfi,x<annotation encoding="application / x-tex">F_{i,x}, F_{i,x}, or f_{i,x}< / annotation>< / semantics>< / com:Math> respectively. Each of the <com:Math><semantics>L<annotation encoding="application / x-tex">L< / annotation>< / semantics>< / com:Math> sets of third type of vectors is used to determine all columns of each of the <com:Math><semantics>N3*C4<annotation encoding="application / x-tex">N_3 * C_4< / annotation>< / semantics>< / com:Math> precoding matrices. In some implementation, the third type of vector set is specific to the first type of vector set; In some implementation, the third type of vector set is specific to each second type of vector with index <com:Math><semantics>f=0,1,...,M−1<annotation encoding="application / x-tex">f = 0,1,...,M-1< / annotation>< / semantics>< / com:Math>, the subscript of s has <com:Math><semantics>f<annotation encoding="application / x-tex">f< / annotation>< / semantics>< / com:Math>, such as <com:Math><semantics>Sf,tx<annotation encoding="application / x-tex">S_{f,t}^x< / annotation>< / semantics>< / com:Math> The one set of PMI includes information about M third type of vector sets. For example, the one set PMI includes M sets of <com:Math><semantics>Fx,Fx,fx,orfx<annotation encoding="application / x-tex">F_x, F_x, f_x, or f_x< / annotation>< / semantics>< / com:Math>. The <com:Math><semantics>Fx,Fx,fx,orfx<annotation encoding="application / x-tex">F_x, F_x, f_x, or f_x< / annotation>< / semantics>< / com:Math> in Alt 1- 12 can be replaced by <com:Math><semantics>Ff,x,Ff,x′,ff,x,orff,x′<annotation encoding="application / x-tex">F_{f,x}, F_{f,x}', f_{f,x}, or f_{f,x}'< / annotation>< / semantics>< / com:Math> respectively. Each of the M sets of third type of vectors is used to determine all columns of each of the <com:Math><semantics>N3*C4<annotation encoding="application / x-tex">N_3 * C_4< / annotation>< / semantics>< / com:Math> precoding matrices. The any Alt 1-12 can be used to determine the format of the third type of vector instead of only Alt 1,4,6,8 can be used. In some implementation, the third type of vector set is specific to each second type of vector with index <com:Math><semantics>f=0,1,...,M−1<annotation encoding="application / x-tex">f = 0,1,...,M-1< / annotation>< / semantics>< / com:Math>, the subscript of third type of vector s has <com:Math><semantics>f<annotation encoding="application / x-tex">f< / annotation>< / semantics>< / com:Math>, such as <com:Math><semantics>Sf,tx<annotation encoding="application / x-tex">S_{f,t}^{x}< / annotation>< / semantics>< / com:Math>. Different from above method that the UE reports M sets of <com:Math><semantics>Fx,Fx,fx,orfx<annotation encoding="application / x-tex">F_x, F_x, f_x, or f_x< / annotation>< / semantics>< / com:Math>, the UE reports only one set of M sets of <com:Math><semantics>Fx,Fx,fx,orfx<annotation encoding="application / x-tex">F_x, F_x, f_x, or f_x< / annotation>< / semantics>< / com:Math>, For different second parameter, the reported one set of <com:Math><semantics>Fx,Fx,fx,orfx<annotation encoding="application / x-tex">F_x, F_x, f_x, or f_x< / annotation>< / semantics>< / com:Math> can determine one set of third type of vectors such as <com:Math><semantics>Sf,tx<annotation encoding="application / x-tex">S_{f,t}^x< / annotation>< / semantics>< / com:Math> in Alt 1,.Alt4, Alt 6, or Alt 8. Then the UE can determine M set of third type of vectors. Each of the M sets of third type of vectors is used to determine all columns of each of the <com:Math><semantics>N3*C4<annotation encoding="application / x-tex">N_3 * C_4< / annotation>< / semantics>< / com:Math> precoding matrices and includes X third type of vectors. In some implementation, the <com:Math><semantics>X=M<annotation encoding="application / x-tex">X = M< / annotation>< / semantics>< / com:Math>. There is one to one mapping between the M second type of vectors and X third type of vectors. For example, Alt 11-12. In some implementation, the third type of vector set is specific to the second type of vector set, there is no subscript <com:Math><semantics>f<annotation encoding="application / x-tex">f< / annotation>< / semantics>< / com:Math> in the third type of vector. In a word, the subscript of one third type of vector can includes at least one of layer index <com:Math><semantics>l<annotation encoding="application / x-tex">l< / annotation>< / semantics>< / com:Math>, first type of vector index <com:Math><semantics>i<annotation encoding="application / x-tex">i< / annotation>< / semantics>< / com:Math>, or the second type of vector index <com:Math><semantics>f<annotation encoding="application / x-tex">f< / annotation>< / semantics>< / com:Math>. For example, <com:Math><semantics>six<annotation encoding="application / x-tex">s_i^x< / annotation>< / semantics>< / com:Math> can be marked by one of <com:Math><semantics>Sl,f,i,tx<annotation encoding="application / x-tex">S_{l,f,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,f,tx<annotation encoding="application / x-tex">S_{l,f,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,i,tx<annotation encoding="application / x-tex">S_{l,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,i,tx<annotation encoding="application / x-tex">S_{l,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,tx<annotation encoding="application / x-tex">S_{l,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,tx<annotation encoding="application / x-tex">S_{l,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,tx<annotation encoding="application / x-tex">S_{l,t}^x< / annotation>< / semantics>< / com:Math>, or <com:Math><semantics>Sl,tx<annotation encoding="application / x-tex">S_{l,t}^x< / annotation>< / semantics>< / com:Math>, is specific to one first type of vector with index <com:Math><semantics>i<annotation encoding="application / x-tex">i< / annotation>< / semantics>< / com:Math>, one second type of vector with index <com:Math><semantics>f<annotation encoding="application / x-tex">f< / annotation>< / semantics>< / com:Math>, and one layer index <com:Math><semantics>i<annotation encoding="application / x-tex">i< / annotation>< / semantics>< / com:Math>. The UE needs to report L*M* <com:Math><semantics>v<annotation encoding="application / x-tex">v< / annotation>< / semantics>< / com:Math> sets of third type of vectors, for example reports <com:Math><semantics>L*v*M<annotation encoding="application / x-tex">L * v * M< / annotation>< / semantics>< / com:Math> sets of <com:Math><semantics>Fx,Fx,fx,orfx<annotation encoding="application / x-tex">F_x, F_x, f_x, or f_x< / annotation>< / semantics>< / com:Math>, Or the UE reports <com:Math><semantics>L*v<annotation encoding="application / x-tex">L * v< / annotation>< / semantics>< / com:Math> sets of third type of vectors such as reports <com:Math><semantics>L*v<annotation encoding="application / x-tex">L^*v< / annotation>< / semantics>< / com:Math> sets of <com:Math><semantics>Fx,Fx′,fx,orfx′<annotation encoding="application / x-tex">F_x, F_x', f_x, or f_x'< / annotation>< / semantics>< / com:Math>. For M second type of vector <com:Math><semantics>f<annotation encoding="application / x-tex">f< / annotation>< / semantics>< / com:Math>, the reported <com:Math><semantics>L*v<annotation encoding="application / x-tex">L^*v< / annotation>< / semantics>< / com:Math> sets of <com:Math><semantics>Fx,Fx,fx,orfx<annotation encoding="application / x-tex">F_x, F_x, f_x, or f_x< / annotation>< / semantics>< / com:Math> can be extended to <com:Math><semantics>L*v*M<annotation encoding="application / x-tex">L^*v^*M< / annotation>< / semantics>< / com:Math>. Each set corresponds to one <com:Math><semantics>i<annotation encoding="application / x-tex">i< / annotation>< / semantics>< / com:Math>, one <com:Math><semantics>f<annotation encoding="application / x-tex">f< / annotation>< / semantics>< / com:Math> and one <com:Math><semantics>l<annotation encoding="application / x-tex">l< / annotation>< / semantics>< / com:Math>. There are <com:Math><semantics>X<annotation encoding="application / x-tex">X< / annotation>< / semantics>< / com:Math> third type of vectors in each set. The <com:Math><semantics>X<annotation encoding="application / x-tex">X< / annotation>< / semantics>< / com:Math> for different sets can be different or same. That is the number of third type of vectors in different of the D third type of vector sets can be different or same. <com:Math><semantics>Slx<annotation encoding="application / x-tex">S_l^x< / annotation>< / semantics>< / com:Math> in formula (1) and Alt 1-10 can be replaced with one of <com:Math><semantics>Sl,f,i,tx<annotation encoding="application / x-tex">S_{l,f,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>sl,f,tx<annotation encoding="application / x-tex">s_{l,f,t}^{x}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>sf,i,tx<annotation encoding="application / x-tex">s_{f,i,t}^{x}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>sl,i,tx<annotation encoding="application / x-tex">s_{l,i,t}^{x}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>sl,i,tx<annotation encoding="application / x-tex">s_{l,i,t}^{x}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>sf,tx<annotation encoding="application / x-tex">s_{f,t}^{x}< / annotation>< / semantics>< / com:Math>, or <com:Math><semantics>si,tx<annotation encoding="application / x-tex">s_{i,t}^{x}< / annotation>< / semantics>< / com:Math>, Any one of <com:Math><semantics>sl,f,i,tx<annotation encoding="application / x-tex">s_{l,f,i,t}^{x}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>sl,i,tx<annotation encoding="application / x-tex">s_{l,i,t}^{x}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>sl,i,tx<annotation encoding="application / x-tex">s_{l,i,t}^{x}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>sl,i,tx<annotation encoding="application / x-tex">s_{l,i,t}^{x}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>sl,i,tx<annotation encoding="application / x-tex">s_{l,i,t}^{x}< / annotation>< / semantics>< / com:Math>, or <com:Math><semantics>Si,tx<annotation encoding="application / x-tex">S_{i,t}^{x}< / annotation>< / semantics>< / com:Math> is got based on any of above Alt 1-12 of getting a third type of vector with index <com:Math><semantics>Si,tx<annotation encoding="application / x-tex">S_{i,t}^{x}< / annotation>< / semantics>< / com:Math>. <com:Math><semantics>Sl,f,tx<annotation encoding="application / x-tex">S_{l,f,t}^{x}< / annotation>< / semantics>< / com:Math> is specific to the first type of vector set and shared by all first type of vectors in the first type of vector set. <com:Math><semantics>Sf,i,tx<annotation encoding="application / x-tex">S_{f,i,t}^x< / annotation>< / semantics>< / com:Math> is specific to all layers and shared by all layers. <com:Math><semantics>Sl,i,tx<annotation encoding="application / x-tex">S_{l,i,t}^x< / annotation>< / semantics>< / com:Math> is specific to all the second type of vector sets and shared by all second type of vectors. <com:Math><semantics>Sl,tx<annotation encoding="application / x-tex">S_{l,t}^{x}< / annotation>< / semantics>< / com:Math> is specific to the first and second type of vector set and shared by all first and second type of vectors in the first and second type of vector set. <com:Math><semantics>Sf,tx<annotation encoding="application / x-tex">S_{f,t}^x< / annotation>< / semantics>< / com:Math> is only specific to each second type of vector. <com:Math><semantics>Si,tx<annotation encoding="application / x-tex">S_{i,t}^x< / annotation>< / semantics>< / com:Math> is only specific to each first type of vector. <com:Math><semantics>Stx<annotation encoding="application / x-tex">S_t^x< / annotation>< / semantics>< / com:Math> is shared by all layers, all first type of vector and all second type of vectors. The time domain index <com:Math><semantics>t<annotation encoding="application / x-tex">t< / annotation>< / semantics>< / com:Math> is increased in duration <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math>. In a word, the UE reports one set of PMI in one CSI reporting. The PMI indicates <com:Math><semantics>C4*N3<annotation encoding="application / x-tex">C_4 * N_3< / annotation>< / semantics>< / com:Math> precoding matrices. Each of <com:Math><semantics>C4*N3<annotation encoding="application / x-tex">C_4 * N_3< / annotation>< / semantics>< / com:Math> precoding matrices corresponds to one frequency domain unit and one first type of time domain unit. The information included in the one set PMI applies each of the <com:Math><semantics>C4*N3<annotation encoding="application / x-tex">C_4 * N_3< / annotation>< / semantics>< / com:Math> precoding matrices. wherein the <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> is larger than 1, and / or <com:Math><semantics>C4*N3<annotation encoding="application / x-tex">C_4 * N_3< / annotation>< / semantics>< / com:Math> is larger than 1. The one set of PMI includes information about the three type of vector sets and weighted coefficient <com:Math><semantics>al,i,f,x,t<annotation encoding="application / x-tex">a_{l,i,f,x,t}< / annotation>< / semantics>< / com:Math>. <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> is the number of elements included in one third type of vector. For example, in the first method of determining <com:Math><semantics>t<annotation encoding="application / x-tex">t< / annotation>< / semantics>< / com:Math> of a third type of vector, <com:Math><semantics>t=0,1,...,C3<annotation encoding="application / x-tex">t = 0,1,...,C_3< / annotation>< / semantics>< / com:Math>, then <com:Math><semantics>C4=C3<annotation encoding="application / x-tex">C_4 = C_3< / annotation>< / semantics>< / com:Math>. In the second method of determining <com:Math><semantics>t<annotation encoding="application / x-tex">t< / annotation>< / semantics>< / com:Math> of a third type of vector, <com:Math><semantics>t=tCSI−RS,occasion−1,tCSI−RS,occasion−2,...,tCSI−RS,occasion−C1,thenC4=C1<annotation encoding="application / x-tex">t = t_{CSI-RS,occasion-1}, t_{CSI-RS,occasion-2}, ..., t_{CSI-RS,occasion-C_1}, then C_4 = C_1< / annotation>< / semantics>< / com:Math>. In the third method of determining <com:Math><semantics>t<annotation encoding="application / x-tex">t< / annotation>< / semantics>< / com:Math> of a third type of vector, <com:Math><semantics>t=0<annotation encoding="application / x-tex">t = 0< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>tCSI−RS,position−2,...,tCSI−RS,position−C1<annotation encoding="application / x-tex">t_{CSI-RS,position-2},...,t_{CSI-RS,position-C_1}< / annotation>< / semantics>< / com:Math>, then <com:Math><semantics>C4=C1<annotation encoding="application / x-tex">C_4 = C_1< / annotation>< / semantics>< / com:Math>. In the fourth method of determining <com:Math><semantics>t<annotation encoding="application / x-tex">t< / annotation>< / semantics>< / com:Math> of a third type of vector, [Image disponible dans le document PDF, Image available in the PDF document] <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> is equal to or larger than <com:Math><semantics>C3<annotation encoding="application / x-tex">C_3< / annotation>< / semantics>< / com:Math>. The <com:Math><semantics>L*M*X*v<annotation encoding="application / x-tex">L^*M^*X^*v< / annotation>< / semantics>< / com:Math> weighted coefficients <com:Math><semantics>al,i,f,x,t<annotation encoding="application / x-tex">a_{l,i,f,x,t}< / annotation>< / semantics>< / com:Math> can reported by using at least one of following methods. Method 1. <com:Math><semantics>al,i,f,x,t=pl,⌊i / L⌋(1)pl,i,f,x(2)φl,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x,t} = p_{l,\lfloor i / L \rfloor}^{(1)} p_{l,i,f,x}^{(2)} \varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math>, the UE reports <com:Math><semantics>pl,⌊i / L⌋(1),pl,i,f,x(2),φl,i,f,x<annotation encoding="application / x-tex">p_{l,\lfloor i / L \rfloor}^{(1)}, p_{l,i,f,x}^{(2)}, \varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math>. Then formula (1) can be replaced as formula [Image disponible dans le document PDF, Image available in the PDF document] (5) If the <com:Math><semantics>stx<annotation encoding="application / x-tex">s_t^x< / annotation>< / semantics>< / com:Math> in formula (5) is replaced with <com:Math><semantics>sl,tx<annotation encoding="application / x-tex">s_{l,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>sl,f,tx<annotation encoding="application / x-tex">s_{l,f,t}^x< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>sl,i,f,tx<annotation encoding="application / x-tex">s_{l,i,f,t}^x< / annotation>< / semantics>< / com:Math> respectively, the formula (3-4) also can be updated by formula (5). Of course, the <com:Math><semantics>stx<annotation encoding="application / x-tex">s_t^x< / annotation>< / semantics>< / com:Math> in formula (5) can be replaced with one of <com:Math><semantics>Sl,f,i,tx<annotation encoding="application / x-tex">S_{l,f,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,f,tx<annotation encoding="application / x-tex">S_{l,f,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sf,i,tx<annotation encoding="application / x-tex">S_{f,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,i,tx<annotation encoding="application / x-tex">S_{l,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,tx<annotation encoding="application / x-tex">S_{l,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sf,tx<annotation encoding="application / x-tex">S_{f,t}^x< / annotation>< / semantics>< / com:Math>, or <com:Math><semantics>Si,tx<annotation encoding="application / x-tex">S_{i,t}^x< / annotation>< / semantics>< / com:Math> Method 2: <com:Math><semantics>al,i,f,x,t=pl,i / L(1)|pl,x,i / L(3)|pl,i,f,x(2)φl,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x,t} = p_{l,i / L}^{(1)} |p_{l,x,i / L}^{(3)}| p_{l,i,f,x}^{(2)} \varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math>, then formula (1) can be replaced as formula [Image disponible dans le document PDF, Image available in the PDF document] (6) If the <com:Math><semantics>stx<annotation encoding="application / x-tex">s_t^x< / annotation>< / semantics>< / com:Math> in formula (6) is replaced with <com:Math><semantics>sl,tx<annotation encoding="application / x-tex">s_{l,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>sl,f,tx<annotation encoding="application / x-tex">s_{l,f,t}^x< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>sl,i,f,tx<annotation encoding="application / x-tex">s_{l,i,f,t}^x< / annotation>< / semantics>< / com:Math> respectively, the formula (3-4) also can be updated by formula (6). Of course, the <com:Math><semantics>stx<annotation encoding="application / x-tex">s_t^x< / annotation>< / semantics>< / com:Math> in formula (6) can be replaced with one of <com:Math><semantics>Sl,f,i,tx<annotation encoding="application / x-tex">S_{l,f,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,f,tx<annotation encoding="application / x-tex">S_{l,f,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,i,tx<annotation encoding="application / x-tex">S_{l,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,i,tx<annotation encoding="application / x-tex">S_{l,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,tx<annotation encoding="application / x-tex">S_{l,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sf,tx<annotation encoding="application / x-tex">S_{f,t}^x< / annotation>< / semantics>< / com:Math>, or <com:Math><semantics>Si,tx<annotation encoding="application / x-tex">S_{i,t}^x< / annotation>< / semantics>< / com:Math>. The UE can reports and <com:Math><semantics>x**=max⁡xpl,x,q=⌊i* / L⌋(3)<annotation encoding="application / x-tex">x^{**} = \max_{x} p_{l,x,q=\lfloor i^* / L \rfloor}^{(3)}< / annotation>< / semantics>< / com:Math> for each layer <com:Math><semantics>l<annotation encoding="application / x-tex">l< / annotation>< / semantics>< / com:Math>. Alternative, <com:Math><semantics>x*=x**<annotation encoding="application / x-tex">x^* = x^{**}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>pl,x**,⌊i* / L⌋(3)<annotation encoding="application / x-tex">p_{l,x^{**},\lfloor i^* / L \rfloor}^{(3)}< / annotation>< / semantics>< / com:Math> doesn't needs to be reported by the UE because <com:Math><semantics>pl,x**,⌊i* / L⌋(3)=1<annotation encoding="application / x-tex">p_{l,x^{**},\lfloor i^* / L\rfloor}^{(3)} = 1< / annotation>< / semantics>< / com:Math>. Method 3: <com:Math><semantics>al,i,f,x,t=pl,|i / L|(1)pl,x,|i / L|(3)φl,x,|i / L|(3)pl,i,f,x(2)φl,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x,t} = p_{l,|i / L|}^{(1)} p_{l,x,|i / L|}^{(3)} \varphi_{l,x,|i / L|}^{(3)} p_{l,i,f,x}^{(2)} \varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math> then formula (1) can be replaced as formula [Image disponible dans le document PDF, Image available in the PDF document] (7) The UE can reports <com:Math><semantics>x**<annotation encoding="application / x-tex">x^{**}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>x**=max⁡xpl,x,q=⌊i* / L⌋(3)<annotation encoding="application / x-tex">x^{**} = \max_{x} p_{l,x,q=\lfloor i^* / L \rfloor}^{(3)}< / annotation>< / semantics>< / com:Math> for each layer <com:Math><semantics>l<annotation encoding="application / x-tex">l< / annotation>< / semantics>< / com:Math>. Alternative, <com:Math><semantics>x*=x**,pl,x**,⌊i* / L(3),φl,x**,⌊i* / L(3)<annotation encoding="application / x-tex">x^* = x^{**}, \quad p_{l,x^{**},\lfloor i^* / L}^{(3)}, \quad \varphi_{l,x^{**},\lfloor i^* / L}^{(3)}< / annotation>< / semantics>< / com:Math> doesn't needs to be reported by the UE because <com:Math><semantics>pl,x**,⌊i* / L(3)=1<annotation encoding="application / x-tex">p_{l,x^{**},\lfloor i^* / L}^{(3)} = 1< / annotation>< / semantics>< / com:Math>. <com:Math><semantics>𝝋l,x**,⌊i* / L⌋(3)=1<annotation encoding="application / x-tex">\boldsymbol{\varphi}_{l,x^{**},\lfloor i^* / L\rfloor}^{(3)} = 1< / annotation>< / semantics>< / com:Math> Method 4: <com:Math><semantics>al,i,f,x,t=pl,⌊i / L⌋(1)pl,x(3)φl,x(2)pl,i,f,x(2)φl,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x,t} = p_{l,\lfloor i / L \rfloor}^{(1)} p_{l,x}^{(3)} \varphi_{l,x}^{(2)} p_{l,i,f,x}^{(2)} \varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math> then formula (1) can be replaced as formula [Image disponible dans le document PDF, Image available in the PDF document] (8) The UE can report <com:Math><semantics>x**<annotation encoding="application / x-tex">x^{**}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>x**=max⁡xpl,x(3)<annotation encoding="application / x-tex">x^{**} = \max_{x} p_{l,x}^{(3)}< / annotation>< / semantics>< / com:Math> for each layer <com:Math><semantics>l<annotation encoding="application / x-tex">l< / annotation>< / semantics>< / com:Math>. Alternative, <com:Math><semantics>x*=x**<annotation encoding="application / x-tex">x^* = x^{**}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Pl,x**(3)<annotation encoding="application / x-tex">P_{l,x^{**}}^{(3)}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>φl,x**(3)<annotation encoding="application / x-tex">\varphi_{l,x^{**}}^{(3)}< / annotation>< / semantics>< / com:Math> doesn't need to be reported by the UE because <com:Math><semantics>Pl,x**(3)=1<annotation encoding="application / x-tex">P_{l,x^{**}}^{(3)} = 1< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>φl,x**(3)=1<annotation encoding="application / x-tex">\varphi_{l,x^{**}}^{(3)} = 1< / annotation>< / semantics>< / com:Math> Method 5: <com:Math><semantics>al,i,f,x,t=pl,⌊i / L⌋(1)pl,x(3),pl,i,f,x(2)φl,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x,t} = p_{l,\lfloor i / L \rfloor}^{(1)} p_{l,x}^{(3)}, p_{l,i,f,x}^{(2)} \varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math>, then formula (1) can be replaced as formula [Image disponible dans le document PDF, Image available in the PDF document] (9) The UE can report <com:Math><semantics>x**<annotation encoding="application / x-tex">x^{**}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>x**=max⁡xpl,x(3)<annotation encoding="application / x-tex">x^{**} = \max_{x} p_{l,x}^{(3)}< / annotation>< / semantics>< / com:Math> for each layer <com:Math><semantics>l<annotation encoding="application / x-tex">l< / annotation>< / semantics>< / com:Math>. Alternative, <com:Math><semantics>x*=x**<annotation encoding="application / x-tex">x^* = x^{**}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Pl,x**(3)<annotation encoding="application / x-tex">P_{l,x^{**}}^{(3)}< / annotation>< / semantics>< / com:Math> doesn't needs to be reported by the UE. Method 6: <com:Math><semantics>al,i,f,x,t=pl,⌊i / L⌋(1)pl,x(3)φl,x(3)pl,f(4)pl,i,f,x(2)φl,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x,t} = p_{l,\lfloor i / L \rfloor}^{(1)} p_{l,x}^{(3)} \varphi_{l,x}^{(3)} p_{l,f}^{(4)} p_{l,i,f,x}^{(2)} \varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math>, then formula (1) can be replaced as formula [Image disponible dans le document PDF, Image available in the PDF document] (10) The UE can reports <com:Math><semantics>x**<annotation encoding="application / x-tex">x^{**}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>x**=max⁡xpl,x(3)<annotation encoding="application / x-tex">x^{**} = \max_{x} p_{l,x}^{(3)}< / annotation>< / semantics>< / com:Math> for each layer <com:Math><semantics>l<annotation encoding="application / x-tex">l< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>pl,x(3)<annotation encoding="application / x-tex">p_{l,x}^{(3)}< / annotation>< / semantics>< / com:Math> doesn't needs to be reported by the UE . Alternatively, <com:Math><semantics>x*=x**<annotation encoding="application / x-tex">x^* = x^{**}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Pl,x**(3)<annotation encoding="application / x-tex">P_{l,x^{**}}^{(3)}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Pl,0(4)<annotation encoding="application / x-tex">P_{l,0}^{(4)}< / annotation>< / semantics>< / com:Math> doesn't needs to be reported by the UE because <com:Math><semantics>pl,x**(3)=1<annotation encoding="application / x-tex">p_{l,x^{**}}^{(3)} = 1< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>pl,0(4)=1<annotation encoding="application / x-tex">p_{l,0}^{(4)} = 1< / annotation>< / semantics>< / com:Math>. Method 7: <com:Math><semantics>al,i,f,x,t=pl,x,⌊i / L⌋(3),pl,i,f,x(2),φl,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x,t} = p_{l,x,\lfloor i / L \rfloor}^{(3)}, p_{l,i,f,x}^{(2)}, \varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math>, then formula (1) can be replaced as formula [Image disponible dans le document PDF, Image available in the PDF document] (11) If the <com:Math><semantics>Stx<annotation encoding="application / x-tex">S_t^x< / annotation>< / semantics>< / com:Math> in formula (11) is replaced with <com:Math><semantics>Sl,tx<annotation encoding="application / x-tex">S_{l,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,f,tx<annotation encoding="application / x-tex">S_{l,f,t}^x< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>Sl,i,f,tx<annotation encoding="application / x-tex">S_{l,i,f,t}^x< / annotation>< / semantics>< / com:Math> respectively, the formula (3-4) also can be updated by formula (11). Of course, the <com:Math><semantics>Stx<annotation encoding="application / x-tex">S_t^x< / annotation>< / semantics>< / com:Math> in formula (11) can be replaced with one of <com:Math><semantics>Sl,f,i,tx<annotation encoding="application / x-tex">S_{l,f,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,f,tx<annotation encoding="application / x-tex">S_{l,f,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sf,i,tx<annotation encoding="application / x-tex">S_{f,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,i,tx<annotation encoding="application / x-tex">S_{l,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,i,tx<annotation encoding="application / x-tex">S_{l,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sf,tx<annotation encoding="application / x-tex">S_{f,t}^x< / annotation>< / semantics>< / com:Math>, or <com:Math><semantics>Si,tx<annotation encoding="application / x-tex">S_{i,t}^x< / annotation>< / semantics>< / com:Math>. The UE can reports <com:Math><semantics>X**<annotation encoding="application / x-tex">X^{**}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>x**=max⁡xpl,x,q=⌊i* / L⌋(3)<annotation encoding="application / x-tex">x^{**} = \max_{x} p_{l,x,q=\lfloor i^* / L \rfloor}^{(3)}< / annotation>< / semantics>< / com:Math> for each layer <com:Math><semantics>l<annotation encoding="application / x-tex">l< / annotation>< / semantics>< / com:Math>. Alternative, <com:Math><semantics>x*=x**<annotation encoding="application / x-tex">x^* = x^{**}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>pl,x**,⌊i* / L⌋(3)<annotation encoding="application / x-tex">p_{l,x^{**},\lfloor i^* / L \rfloor}^{(3)}< / annotation>< / semantics>< / com:Math> doesn't needs to be reported by the UE because <com:Math><semantics>pl,x**,⌊i* / L⌋(3)=1<annotation encoding="application / x-tex">p_{l,x^{**},\lfloor i^* / L\rfloor}^{(3)} = 1< / annotation>< / semantics>< / com:Math>. Method 8: <com:Math><semantics>al,i,f,x,t=pl,x,|i / L(3)|φl,x,|i / L(3)|pl,i,f,x(2)|φl,i,f,x(2)|<annotation encoding="application / x-tex">a_{l,i,f,x,t} = p_{l,x,|i / L}^{(3)} |\varphi_{l,x,|i / L}^{(3)}| p_{l,i,f,x}^{(2)} |\varphi_{l,i,f,x}^{(2)}|< / annotation>< / semantics>< / com:Math> then formula (1) can be replaced as formula [Image disponible dans le document PDF, Image available in the PDF document] (12) The UE can reports <com:Math><semantics>x**<annotation encoding="application / x-tex">x^{**}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>x**=max⁡xpl,x,q=⌊l* / L⌋(3)<annotation encoding="application / x-tex">x^{**} = \max_{x} p_{l,x,q=\lfloor l^* / L \rfloor}^{(3)}< / annotation>< / semantics>< / com:Math> for each layer <com:Math><semantics>l<annotation encoding="application / x-tex">l< / annotation>< / semantics>< / com:Math>. Alternative, <com:Math><semantics>x*=x**,pl,x**,⌊i* / L⌋(3),φl,x**,⌊i* / L⌋(3)<annotation encoding="application / x-tex">x^* = x^{**}, \quad p_{l,x^{**},\lfloor i^* / L \rfloor}^{(3)}, \quad \varphi_{l,x^{**},\lfloor i^* / L \rfloor}^{(3)}< / annotation>< / semantics>< / com:Math> doesn't needs to be reported by the UE because <com:Math><semantics>pl,x**,⌊i* / L⌋(3)=1<annotation encoding="application / x-tex">p_{l,x^{**},\lfloor i^* / L \rfloor}^{(3)} = 1< / annotation>< / semantics>< / com:Math>. <com:Math><semantics>φl,x**,|i* / L|(3)=1<annotation encoding="application / x-tex">\varphi_{l,x^{**},|_{i^{*} / L}|}^{(3)} = 1< / annotation>< / semantics>< / com:Math> Method 9: <com:Math><semantics>al,i,f,x,t=pl,x,⌊i / L⌋(3)ej2πfxn3,lfNpl,i,f,x(2)φl,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x,t} = p_{l,x,\lfloor i / L \rfloor}^{(3)} e^{j2\pi f_x \frac{n_{3,l}^f}{N}} p_{l,i,f,x}^{(2)} \varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math>, then formula (1) can be replaced as formula [Image disponible dans le document PDF, Image available in the PDF document] (13) _ _ Method 10: <com:Math><semantics>al,i,f,x,t=pl,x(3)ej2πfxn3,lfNpl,i,f,x(2)φl,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x,t} = p_{l,x}^{(3)} e^{j2\pi f_x} \frac{n_{3,l}^f}{N} p_{l,i,f,x}^{(2)} \varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math>, then formula (1) can be replaced as formula [Image disponible dans le document PDF, Image available in the PDF document] (14) Method 11: <com:Math><semantics>al,i,f,x,t=pl,x(3)φl,f,x(3)pl,i,f,x(2)φl,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x,t} = p_{l,x}^{(3)} \varphi_{l,f,x}^{(3)} p_{l,i,f,x}^{(2)} \varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math>, then formula (1) can be replaced as formula [Image disponible dans le document PDF, Image available in the PDF document] (15) Method 12: <com:Math><semantics>al,i,f,x,t=pl,⌊i / L⌋(1)pl,x(3)φl,f,x(3)pl,i,f,x(2)φl,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x,t} = p_{l,\lfloor i / L \rfloor}^{(1)} p_{l,x}^{(3)} \varphi_{l,f,x}^{(3)} p_{l,i,f,x}^{(2)} \varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math>, then formula (1) can be replaced as formula [Image disponible dans le document PDF, Image available in the PDF document] (16) In above method 1-12, If the <com:Math><semantics>Stx<annotation encoding="application / x-tex">S_t^x< / annotation>< / semantics>< / com:Math> in any of formula (5-16) is replaced with <com:Math><semantics>Sl,tx<annotation encoding="application / x-tex">S_{l,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,f,tx<annotation encoding="application / x-tex">S_{l,f,t}^x< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>Sl,i,f,tx<annotation encoding="application / x-tex">S_{l,i,f,t}^x< / annotation>< / semantics>< / com:Math> respectively, the formula (3-4) also can be updated by any of formula (5-16). Of course, the <com:Math><semantics>Stx<annotation encoding="application / x-tex">S_t^x< / annotation>< / semantics>< / com:Math> in any formula (5-16) can be replaced with one of <com:Math><semantics>Sl,f,i,tx<annotation encoding="application / x-tex">S_{l,f,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,f,tx<annotation encoding="application / x-tex">S_{l,f,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sf,i,tx<annotation encoding="application / x-tex">S_{f,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,i,tx<annotation encoding="application / x-tex">S_{l,i,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sl,tx<annotation encoding="application / x-tex">S_{l,t}^x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Sf,tx<annotation encoding="application / x-tex">S_{f,t}^x< / annotation>< / semantics>< / com:Math>, or <com:Math><semantics>Si,tx<annotation encoding="application / x-tex">S_{i,t}^x< / annotation>< / semantics>< / com:Math> In above Method 1-12 , the UE needs to report index of a strongest coefficient for each layer <com:Math><semantics>l<annotation encoding="application / x-tex">l< / annotation>< / semantics>< / com:Math> such as <com:Math><semantics>pl,i*,f*,x*(2)=max⁡i,f,xpl,i,f,x(2)<annotation encoding="application / x-tex">p_{l,i^*,f^*,x^*}^{(2)} = \max_{i,f,x} p_{l,i,f,x}^{(2)}< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>al,i*,f*,x*=max⁡i,f,xal,i,f,x<annotation encoding="application / x-tex">a_{l,i^*,f^*,x^*} = \max_{i,f,x} a_{l,i,f,x}< / annotation>< / semantics>< / com:Math>, the UE reports <com:Math><semantics>i*,x*,f*<annotation encoding="application / x-tex">i^*, x^*, f^*< / annotation>< / semantics>< / com:Math> corresponding to the strongest coefficient for each layer <com:Math><semantics>l<annotation encoding="application / x-tex">l< / annotation>< / semantics>< / com:Math>. Alternative, the UE just reports information about <com:Math><semantics>i*<annotation encoding="application / x-tex">i^*< / annotation>< / semantics>< / com:Math>, the at least one of <com:Math><semantics>x*<annotation encoding="application / x-tex">x^*< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>f*<annotation encoding="application / x-tex">f^*< / annotation>< / semantics>< / com:Math> doesn't need to be reported by the UE and define that <com:Math><semantics>x*=0<annotation encoding="application / x-tex">x^* = 0< / annotation>< / semantics>< / com:Math>, and / or <com:Math><semantics>f*=0<annotation encoding="application / x-tex">f^* = 0< / annotation>< / semantics>< / com:Math>. In addition, <com:Math><semantics>pl,i*,f*,x*(2)=1<annotation encoding="application / x-tex">p_{l,i^*,f^*,x^*}^{(2)} = 1< / annotation>< / semantics>< / com:Math> <com:Math><semantics>pl,⌊i* / L⌋(1)=1<annotation encoding="application / x-tex">p_{l,\lfloor i^* / L \rfloor}^{(1)} = 1< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>φl,i*,f*,x*(2)=1<annotation encoding="application / x-tex">\varphi_{l,i^*,f^*,x^*}^{(2)} = 1< / annotation>< / semantics>< / com:Math>, then <com:Math><semantics>pl,i*,f*,x*(2)<annotation encoding="application / x-tex">p_{l,i^*,f^*,x^*}^{(2)}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>φl,i*,f*,x*(2)<annotation encoding="application / x-tex">\varphi_{l,i^*,f^*,x^*}^{(2)}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>pl,[i* / L](1)<annotation encoding="application / x-tex">p_{l, [i^* / L]}^{(1)}< / annotation>< / semantics>< / com:Math> don't needs to be feedback to gNB. Alternative, only <com:Math><semantics>pl,i*,f*,x*(2)=1<annotation encoding="application / x-tex">p_{l, i^*, f^*, x^*}^{(2)} = 1< / annotation>< / semantics>< / com:Math> <com:Math><semantics>pl,[i* / L](1)=1<annotation encoding="application / x-tex">p_{l, [i^* / L]}^{(1)} = 1< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>pl,i*,f*,x*(2)<annotation encoding="application / x-tex">p_{l, i^*, f^*, x^*}^{(2)}< / annotation>< / semantics>< / com:Math>. <com:Math><semantics>pl,l* / L(1)<annotation encoding="application / x-tex">p_{l,l^* / L}^{(1)}< / annotation>< / semantics>< / com:Math> don't needs to be feedback to gNB, but <com:Math><semantics>φl,l*,f*,x*<annotation encoding="application / x-tex">\varphi_{l,l^*,f^*,x^*}< / annotation>< / semantics>< / com:Math> needs to feedback to gNB. That is all remaining <com:Math><semantics>φl,i,f,x<annotation encoding="application / x-tex">\varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math> doesn't needed to normalized with reference <com:Math><semantics>φl,i*,f*,x*<annotation encoding="application / x-tex">\varphi_{l,i^*,f^*,x^*}< / annotation>< / semantics>< / com:Math>. In addition, <com:Math><semantics>pl,x*(3)=1<annotation encoding="application / x-tex">p_{l,x^*}^{(3)} = 1< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>φl,x*(3)=1<annotation encoding="application / x-tex">\varphi_{l,x^*}^{(3)} = 1< / annotation>< / semantics>< / com:Math>, and <com:Math><semantics>pl,x*(3)<annotation encoding="application / x-tex">p_{l,x^*}^{(3)}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>φl,x*(3)<annotation encoding="application / x-tex">\varphi_{l,x^*}^{(3)}< / annotation>< / semantics>< / com:Math> doesn't needs to feedback to gNB. In above Method 1-12 , <com:Math><semantics>pl,⌊i / L⌋(1)<annotation encoding="application / x-tex">p_{l,\lfloor i / L \rfloor}^{(1)}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>pl,i,f,x(2)<annotation encoding="application / x-tex">p_{l,i,f,x}^{(2)}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>pl,x(3)<annotation encoding="application / x-tex">p_{l,x}^{(3)}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>pl,x(4)<annotation encoding="application / x-tex">p_{l,x}^{(4)}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>pl,x,⌊i / L⌋(3)<annotation encoding="application / x-tex">p_{l,x,\lfloor i / L \rfloor}^{(3)}< / annotation>< / semantics>< / com:Math> are amplitude coefficients and their phase is 0 . <com:Math><semantics>pl,⌊i / L⌋(1)<annotation encoding="application / x-tex">p_{l,\lfloor i / L \rfloor}^{(1)}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>pl,i,f,x(2)<annotation encoding="application / x-tex">p_{l,i,f,x}^{(2)}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>pl,x(3)<annotation encoding="application / x-tex">p_{l,x}^{(3)}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>pl,x,⌊i / L⌋(4)<annotation encoding="application / x-tex">p_{l,x,\lfloor i / L \rfloor}^{(4)}< / annotation>< / semantics>< / com:Math> is equal to or larger than 0 and smaller than or equal to 1. <com:Math><semantics>φl,i,f,x<annotation encoding="application / x-tex">\varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>φl,x(3)<annotation encoding="application / x-tex">\varphi_{l,x}^{(3)}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>φl,f,x(3)<annotation encoding="application / x-tex">\varphi_{l,f,x}^{(3)}< / annotation>< / semantics>< / com:Math> are phase coefficient and their amplitudes are 1. <com:Math><semantics>l=0,1...,v−1,i=0,1,...,2*L−1,f=0,1,...M−1,x=0,1,...X−1<annotation encoding="application / x-tex">l = 0,1..., v - 1, i = 0,1,...,2 * L - 1, f = 0,1,...M - 1, x = 0,1,...X - 1< / annotation>< / semantics>< / com:Math>. For reporting the coefficient <com:Math><semantics>al,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x}< / annotation>< / semantics>< / com:Math>, the UE reports at least one of <com:Math><semantics>pl,⌊i / L⌋(1),pl,i,f,x(2),pl,x(3),pl,x(4),pl,x,⌊i / L⌋(3),φl,i,f,x(3),φl,x(3)<annotation encoding="application / x-tex">p_{l,\lfloor i / L \rfloor}^{(1)}, p_{l,i,f,x}^{(2)}, p_{l,x}^{(3)}, p_{l,x}^{(4)}, p_{l,x,\lfloor i / L \rfloor}^{(3)}, \varphi_{l,i,f,x}^{(3)}, \varphi_{l,x}^{(3)}< / annotation>< / semantics>< / com:Math>, or <com:Math><semantics>φl,f,x(3)<annotation encoding="application / x-tex">\varphi_{l,f,x}^{(3)}< / annotation>< / semantics>< / com:Math>. For each layer, to reports <com:Math><semantics>pl,i,f,x(2)<annotation encoding="application / x-tex">p_{l,i,f,x}^{(2)}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>φl,i,f,x<annotation encoding="application / x-tex">\varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math>, the UE reports a bitmap for each layer with <com:Math><semantics>L*M*X<annotation encoding="application / x-tex">L^*M^*X< / annotation>< / semantics>< / com:Math> bits. If one bit of the bitmap indicates value 1, the <com:Math><semantics>pl,i,f,x(2)<annotation encoding="application / x-tex">p_{l,i,f,x}^{(2)}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>φl,i,f,x<annotation encoding="application / x-tex">\varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math> corresponding the bit is in the CSI and reported to gNB, otherwise the <com:Math><semantics>pl,i,f,x(2)<annotation encoding="application / x-tex">p_{l,i,f,x}^{(2)}< / annotation>< / semantics>< / com:Math> and <com:Math><semantics>φl,i,f,x<annotation encoding="application / x-tex">\varphi_{l,i,f,x}< / annotation>< / semantics>< / com:Math> corresponding the bit isn't included in the CSI and define that its corresponding <com:Math><semantics>pl,i,f,x(2)=0<annotation encoding="application / x-tex">p_{l,i,f,x}^{(2)} = 0< / annotation>< / semantics>< / com:Math>. The relationship between bitmap of M and X, The total number of 1 in the bitmap for each <com:Math><semantics>x=0,1,2...,X−1<annotation encoding="application / x-tex">x = 0,1,2...,X-1< / annotation>< / semantics>< / com:Math> may be smaller than or equal to one threshold. The total number of 1 in the bitmap may be smaller than or equal to one threshold. In some implementation, the UE can report Q sets of CQIs in one CSI reporting, wherein <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> is larger than 1. Different sets of CQIs corresponds to different second type of time domain units. Different CQI in one set of CQI corresponds to different codewords or different frequency domain units. In some implementation, <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> is equal to or smaller than <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math>. In some implementation, one second type of time domain unit includes one or more the first type of time domain units. As shown in FIG. 3, one second type of time unit includes two first type of time domain units, then <com:Math><semantics>Q=⌈C42⌉<annotation encoding="application / x-tex">Q = \left\lceil \frac{C_4}{2} \right\rceil< / annotation>< / semantics>< / com:Math>. If <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> is an odd value, then the last or the starting second type of time domain units may include only one first type of time domain unit. If the CQI is wideband, the CQI is based on the 2 set of precoding matrices in two first type of time domain units included in one second type of time domain unit corresponding to the CQI. If the CQI is subband, the CQI is based on the 2 precoding matrices in two first type of time domain units included in one second type of time domain unit corresponding to the CQI, wherein the 2 precoding matrices corresponds to one subband of the CQI. For example, the UE gets <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> precoding matrix sets for <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time domain unit. Each of the <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> precoding matrix sets corresponds to one of <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> the first type of time domain units. The UE gets <com:Math><semantics>𝒬<annotation encoding="application / x-tex">\mathcal{Q}< / annotation>< / semantics>< / com:Math> of CQI sets for <com:Math><semantics>𝒬second type<annotation encoding="application / x-tex">\mathcal{Q}_{\text{second type}}< / annotation>< / semantics>< / com:Math> of time domain unit. In some implementation, the <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> set of CQIs can be reported using differential methods. The <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> set of CQIs corresponds to one reference CQI. Other CQI in the <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> set of CQIs are reported using differential value with reference to the reference CQI. The UE also needs to report the index of second time domain unit corresponding to the reference CQI. In another implementation, the reference CQI corresponds to a second time domain unit with a predefined index, such as with index 0. The UE doesn't reports the index of second time domain unit corresponding to the reference CQI In another implementation, each of the <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> set of CQI can be reported using differential methods. The Q set of CQIs corresponds to Q reference CQIs each of which corresponds to one of the Q set of CQIs. In some implementation, different sets of the <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> set of CQIs may include different number of CQI. For example, for a second type of time domain unit which includes at least one of transmission occasion of P CSI-RS ports corresponding to the CQI, the set of CQIs of the second type of time domain unit includes subband CQI. But for a second type of time domain unit which includes no transmission occasion of P CSI-RS ports corresponding to the CQI, the set of CQIs of the second type of time domain unit doesn't include subband CQI and only include wideband CQI. In some implementation, the <com:Math><semantics>Q=⌈C1q⌉<annotation encoding="application / x-tex">Q = \left\lceil \frac{C_1}{q} \right\rceil< / annotation>< / semantics>< / com:Math>. <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> is the number of CSI-RS transmission occasions in a burst corresponding to the CSI reporting including at least one of PMI, CQI. The <com:Math><semantics>q<annotation encoding="application / x-tex">q< / annotation>< / semantics>< / com:Math> is number of CSI-RS transmission occasions in one second type of time domain unit. <com:Math><semantics>q<annotation encoding="application / x-tex">q< / annotation>< / semantics>< / com:Math> is an integer which is larger than 0. The UE only reports CQI for the second type of time domain unit including <com:Math><semantics>q<annotation encoding="application / x-tex">q< / annotation>< / semantics>< / com:Math> CSI-RS transmission occasions except one or two second type of time domain units which includes at least one and less than <com:Math><semantics>q<annotation encoding="application / x-tex">q< / annotation>< / semantics>< / com:Math> CSI-RS transmission occasions. Similarly, the UE can reports R values of RI(rank indicator) in one CSI reporting, wherein R is larger than 1. Different values of the R values RIs corresponds to different third type of time domain units. The total layers(or total column) of each of <com:Math><semantics>C4*N3<annotation encoding="application / x-tex">C_4 * N_3< / annotation>< / semantics>< / com:Math> precoding matrices is determined by one of the R values of RI. For example, the one RI is the first RI of a first type of time domain unit with index 0, or with first CSI-RS transmission occasion. In some implementations, the one RI can be maximal of the R values of RI. For a third type of time domain unit, the columns of each of <com:Math><semantics>C5*N3<annotation encoding="application / x-tex">C_5 * N_3< / annotation>< / semantics>< / com:Math> precoding matrices indicated by the PMI may be different from the RI of the third type of time domain units, then the CQI of the third type of time domain unit is based on RI columns of each of precoding matrices of the third type of time domain unit indicated by the PMI and indicated RI of the third type of time domain unit. <com:Math><semantics>C5<annotation encoding="application / x-tex">C_5< / annotation>< / semantics>< / com:Math> is the number of the <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time domain units in the one third type of time units. For example, the PMI indicates <com:Math><semantics>C4*N3<annotation encoding="application / x-tex">C_4 * N_3< / annotation>< / semantics>< / com:Math> precoding matrices for <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type domain units. Each of the <com:Math><semantics>C4*N3<annotation encoding="application / x-tex">C_4 * N_3< / annotation>< / semantics>< / com:Math> precoding matrices includes 4 columns each of which corresponds to one layers. The UE reports RI for 8 third type of time domain units as [4,3,4,2,1,4,4,3,1,1]. For a third type of time domain unit with RI value doesn't equal to 4, the UE gets the CQI for each second type of time domain unit of the third type of time domain unit, according to RI values of columns of each of the <com:Math><semantics>C5*N3<annotation encoding="application / x-tex">C_5 * N_3< / annotation>< / semantics>< / com:Math> precoding matrices of the third type of time domain unit. For example, for the third type of time domain unit with index 1, the indicated RI is 3, the UE gets the CQI for each second type of time domain unit of the third type of time domain unit with index 1, according to 3 columns from 4 columns of each of the precoding matrices of the third type of time domain unit. The 3 columns of the 4 columns can be first 3 columns of the 4 columns, or indicated by information included in the CSI. In some implementation, one third type of time domain unit includes one or more second type of time domain units. If one third type of time domain unit includes one second type of time domain unit, the third type of time domain unit equals to the second type of time domain unit, that is the second and third type of time domain units are same type of time unit. In some implementation, the third type of time units equals time duration <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math>, then R=1. In some implementation, the <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time unit, <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> second type of time domain unit and R second type of time domain unit are all in duration <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math>. In some implementation, the second type of time units equals time duration <com:Math><semantics>T3<annotation encoding="application / x-tex">T_3< / annotation>< / semantics>< / com:Math>, then <com:Math><semantics>Q=1<annotation encoding="application / x-tex">Q=1< / annotation>< / semantics>< / com:Math>. In a word, the UE can report at least one of PMI, Q sets of CQIs, or R value of RI in one CSI reporting. Each of the Q sets of CQIs corresponds to one second type of time domain unit. The <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> sets of CQIs are for <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> second type of time domain units. The <com:Math><semantics>R<annotation encoding="application / x-tex">R< / annotation>< / semantics>< / com:Math> value of RI corresponds to R third type of time domain units. Each of the R value of RI corresponds to one of <com:Math><semantics>R<annotation encoding="application / x-tex">R< / annotation>< / semantics>< / com:Math> third type of time domain units. The PMI indicates <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> sets of precoding matrices for <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time domain units. Each of the <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> sets includes <com:Math><semantics>N3<annotation encoding="application / x-tex">N_3< / annotation>< / semantics>< / com:Math> precoding matrices. wherein <com:Math><semantics>Q,R,C4<annotation encoding="application / x-tex">Q, R, C_4< / annotation>< / semantics>< / com:Math> is integer which is larger than 0. And / or at least one of <com:Math><semantics>Q,R,C4<annotation encoding="application / x-tex">Q, R, C_4< / annotation>< / semantics>< / com:Math> is larger than 1. In some implementation, <com:Math><semantics>C4≥Q≥R<annotation encoding="application / x-tex">C_4 \ge Q \ge R< / annotation>< / semantics>< / com:Math>. In some implementation, The CSI reporting should be configured with CMR (channel measurement reference such as the P CSI-RS ports)( or CMR+IMR (interference measurement reference)) and TRS or SSB. The UE gets the CSI based the CMR (or CMR+IMR) and at least one of TRS, or SSB (synchronization signal block). In the following we give the example of the TRS, it also applied to SSB. To get the estimation of doppler shift or doppler spread such as <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math> or, <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math>, it needs dense time domain TRS which is at least in coherent time. But more time occasions of CSI-RS is needed to get coefficient(such as <com:Math><semantics>al,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x}< / annotation>< / semantics>< / com:Math>) and spatial domain vector set (such as <com:Math><semantics>vi<annotation encoding="application / x-tex">v_i< / annotation>< / semantics>< / com:Math>) and frequency domain vector selection(such as <com:Math><semantics>ylkf<annotation encoding="application / x-tex">y_{lk}^f< / annotation>< / semantics>< / com:Math>), then the reported CSI may be more accurate. The measurement metric for TRS and CSI-RS are different. For example, the measurement metric for TRS is <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>Fx<annotation encoding="application / x-tex">F_x< / annotation>< / semantics>< / com:Math> or, <com:Math><semantics>fx<annotation encoding="application / x-tex">f_x< / annotation>< / semantics>< / com:Math> it needn't denser time domain sample. It can be got as long as at least one path isn't blocked. The phase change between two TRS symbol shouldn't be larger than <com:Math><semantics>π<annotation encoding="application / x-tex">\pi< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>2π<annotation encoding="application / x-tex">2\pi< / annotation>< / semantics>< / com:Math>. For example, <com:Math><semantics>FxΔt≤π<annotation encoding="application / x-tex">F_x \Delta t \leq \pi< / annotation>< / semantics>< / com:Math>, or <com:Math><semantics>FxΔt≤2π<annotation encoding="application / x-tex">F_{x}\Delta t \leq 2\pi< / annotation>< / semantics>< / com:Math>, wherein <com:Math><semantics>Δt<annotation encoding="application / x-tex">\Delta t< / annotation>< / semantics>< / com:Math> is time gap between two TRS symbols. The measurement metric of CSI-RS is for spatial domain vector (e.g., the first type of vector) and frequency domain vector (e.g., the second type of vector) selection and weighted coefficient <com:Math><semantics>al,i,f,x<annotation encoding="application / x-tex">a_{l,i,f,x}< / annotation>< / semantics>< / com:Math> determination as shown in formula 1-12. It needs more time occasion considering some path is blocked at some instance or considering interference burst. Alt1: TRS and CSI-RS have same period and TRS is denser than CSI-RS in one slot. One CSI-RS reporting period includes one period of CSI-RS and one period TRS. Alt2: The period of CSI-RS and TRS can be same. One CSI-RS period includes <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> CSI-RS transmission occasions of the P CSI-RS ports and <com:Math><semantics>CR<annotation encoding="application / x-tex">C_R< / annotation>< / semantics>< / com:Math> TRS transmission occasions. TRS is denser than CSI-RS in one slot and there are one than one CSI-RS transmission occasions in one CSI-RS period. The <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> CSI-RS transmission occasions are in one burst of the P CSI-RS. The <com:Math><semantics>CR<annotation encoding="application / x-tex">C_R< / annotation>< / semantics>< / com:Math> TRS transmission occasions are in one burst. The gap between continuous TRS transmission occasions of the <com:Math><semantics>CR<annotation encoding="application / x-tex">C_R< / annotation>< / semantics>< / com:Math> transmission occasions are smaller than the gap between continuous CSI-RS transmission occasions of the <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> transmission occasions. In some implementation, <com:Math><semantics>C1≥CR<annotation encoding="application / x-tex">C_1 \ge C_R< / annotation>< / semantics>< / com:Math>. In some implementation, the burst including <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> CSI-RS transmission occasions of the P CSI-RS ports is longer than the burst including the <com:Math><semantics>CR<annotation encoding="application / x-tex">C_R< / annotation>< / semantics>< / com:Math> TRS transmission occasions. Alt 3: One CSI-RS period includes <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> CSI-RS resources and <com:Math><semantics>CR<annotation encoding="application / x-tex">C_R< / annotation>< / semantics>< / com:Math> TRS transmission resources. The <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> CSI-RS resource has relationship such as same number of CSI-RS ports, one to one mapping between CSI-RS port, same period. But <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> CSI-RS resources may need more description to clarify that the CSI is based on <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> CSI-RS resource. The N TRS resources have relationship. One CSI-RS period corresponds to one CSI reporting. The M CSI-RS transmission occasion can be equally / unequally distributed. Alt 4: CSI-RS and TRS are same signal. Alt 5: One of CSI-RS port are TRS. In above description, the CSI is determined according to P CSI-RS ports received on multiple transmission occasions in one period of P CSI-RS ports. The gNB should configure information about the multiple transmission occasions of the P CSI-RS ports. In a first implementation, the multiple transmission occasions corresponding to one CSI-RS resource which includes the P CSI-RS ports. The gNB should configure information about the multiple transmission occasions of the one CSI-RS resource by configuring at least one of a OFDM symbol index of the starting transmission occasion, the number of transmission occasions, OFDM symbol gap between two continuous transmission occasions, slot gap between two continuous transmission occasions, a set of slot offset, an OFDM symbol index for each transmission occasions, a set of OFDM symbol gap. For example, the UE can configure the multiple transmission occasion using one or more following Method 1-6. Method 1: the gNB configures OFDM symbols of starting transmission occasion, the number of transmission occasions and OFDM symbol gap between two continuous transmission occasions, then the UE can get the multiple transmission occasions. One transmission occasion can includes one or more OFDM symbol, wherein different OFDM symbols of one transmission occasion includes different CSI-RS ports of the one CSI-RS resource. Method 2: The gNB configure a OFDM symbol index of starting transmission occasion and a set of slot offset for the one CSI-RS resource. The multiple transmission occasions of the one CSI-RS resource are in different slot and in OFDM symbol with same index. For example the first transmission occasion are in {OFDM symbol 1 and OFDM symbol 4} in slot n1 which is indicated by the first entry of the slot offset set, the second transmission occasion are in {OFDM symbol 1 and OFDM symbol 4} in slot n2 which is indicated by the first entry of the slot offset set. Method 3: The gNB configure a OFDM symbol index of starting transmission occasion and the number of transmission occasions of the one CSI-RS resource. The multiple transmission occasions are in continuous slots and with same OFDM symbol index. Alternatively, the multiple transmission occasions are in continuous available slots and with same OFDM symbol index. Symbols of the one CSI-RS resource in one available slot are available symbols. For example, it isn't an uplink symbol, a symbol for PRACH (physical random access channel), an symbol of SSB. Method 4: The gNB configures a OFDM symbol index of starting transmission occasion, slot gap, the number of transmission occasions of the one CSI-RS resource. The multiple transmission occasions are in different slots with same OFDM symbol index. The slot gap between two continuous transmission occasions is the indicated slot gap. Method 5: The gNB configures OFDM symbol index for each transmission occasions, the gap between two transmission occasion can be different for different two continuous transmission occasions. For example, the gap between transmission occasion 1 and 2 is 3 OFDM symbols, but the gap between transmission occasion 2 and 3 is 2 OFDM symbols. The offset of two OFDM symbols can be different for different transmission occasions. For example, the first transmission occasion are in {OFDM 1 and OFDM 4} in one slot, the second transmission occasion can be {OFDM 5 and OFDM 6} in the same one slot. Method 6: The gNB configures OFDM symbol index for starting transmission occasion and a set of OFDM symbol gap. Then the gap between two continuous transmission occasion can be different for different two continuous transmission occasions but the offset between two OFDM symbols of one OFDM symbol is same. The OFDM symbol gap between the first transmission occasion and the second transmission occasion is the first entry of the set of OFDM symbol gap. The OFDM symbol gap between the second transmission occasion and the third transmission occasion is the second entry of the set of OFDM symbol gap, and so on The gNB can configure the one CSI-RS resource to UE and the UE gets CSI based on the configured CSI-RS resource. The CSI does not include CRI. Alternatively, the gNB configures a set of CSI-RS resource with multiple CSI-RS resources each of which with multiple transmission occasions configured by one Method 1 to method 6. The UE selects one CSI-RS resource and gets CSI based on the selected CSI-RS resource, the CSI includes CRI indicating the selected CSI-RS resources. In a second implementation, the gNB configures the multiple occasions of the P CSI-RS ports using multiple CSI-RS resource, each transmission occasion corresponds to a CSI-RS resource. Different transmission occasion corresponds to different transmission occasions. The multiple transmission occasion can be in one or more CSI-RS resource sets. If they are in more than one CSI-RS resource sets, each set corresponds to one period and slot offset And the CSI is based one setting of CSI-RS resource sets, wherein one CSI-RS resource setting includes one or more CSI-RS resource sets. The gNB can directly configure one CSI-RS setting to let UE get the CSI. Alternatively, the gNB can configures multiple CSI-RS setting, If they are all in one CSI-RS resource set, the one set can corresponding to one period and more than one slot offset. Each CSI-RS resource can be configured with on slot offset. The multiple CSI-RS resources satisfies some restriction. The restriction includes that some parameter needs to be same for these CSI-RS resources in the CSI-RS resource set. The parameter includes at least one of: the number of CSI-RS ports, RB location, bandwidth, or period, In addition, these CSI-RS resources in the CSI-RS resource set are quasi-co-located QCL-ed with respective to QCL type A and QCL-type D, where applicable. Different CSI-RS resources in the CSI-RS resource set are in different time domain symbols in one slot or in different slots. If one transmission occasion of the P CSI-RS ports includes more than one OFDM symbols, wherein different OFDM symbols of one transmission occasion includes different CSI-RS ports of the P CSI-RS ports, the gap between two transmission occasions is gap between starting OFDM symbols of two continuous transmission. One transmission occasion can be in one slot or multiple slot. Alternatively, it limits that one transmission occasions can only be one slot. In some implementation, the starting symbols of later transmission occasions is after the end of the former transmission occasion. For example, the first transmission occasion includes {OFDM symbol 1 including CSI-RS port {0-7}, OFDM symbol 4 including CSI-RS ports {8-15}} of slot n, the second transmission should start from OFDM 9 of slot n, such as the gap is 5, then the second transmission occasion includes {OFDM 6, OFDM 9 }. In another implementation, the starting symbols of later transmission occasions is after the starting of the former transmission occasion and the two transmission occasion shouldn't be in one same symbols, that is the gap doesn't equal to offset of two OFDM symbols of one transmission occasion, such as the gap is 2, the second transmission occasion includes {OFDM 3, OFDM 6}. In a third implementation, the starting symbols of later transmission occasions is after the starting of the former transmission occasion and the two transmission occasion can be in one same symbols, that is the gap can equal offset of two OFDM symbols of one transmission occasion, such as the gap is 3, then the second transmission occasion includes {OFDM 4, OFDM 7}. One CRI corresponds to one CSI-RS resource, one CSI-RS resource set, or one CSI-RS resource setting. The UE reports a capability about at least one of the number of CSI-RS transmission occasions in a burst of the P CSI-RS ports used to generate the PMI, such as <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math>; the number of elements in one third type of vector such as the value of <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math>: the number of elements in one second type of vector such as the value of <com:Math><semantics>N3<annotation encoding="application / x-tex">N_3< / annotation>< / semantics>< / com:Math>, the <com:Math><semantics>L,M<annotation encoding="application / x-tex">L,M< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>X<annotation encoding="application / x-tex">X< / annotation>< / semantics>< / com:Math>. That is the UE will report its supported maximal value of <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math>. Because the UE needs to gets an actually precoding matrix for each transmission occasion, and get the PMI for the <com:Math><semantics>N4<annotation encoding="application / x-tex">N_4< / annotation>< / semantics>< / com:Math> first type of time domain units. It requests UE capability reporting. In some implementation, the UE reports its capability about supported value sets about at least one of <com:Math><semantics>C1,C4,N3,L,M<annotation encoding="application / x-tex">C_1, C_4, N_3, L, M< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>X<annotation encoding="application / x-tex">X< / annotation>< / semantics>< / com:Math>. For example, the UE reports its supported value set of <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> is <com:Math><semantics>{4,8,16}<annotation encoding="application / x-tex">\{4,8,16\}< / annotation>< / semantics>< / com:Math>, In some implementation, the UE reports its capability about supported maximal value about at least one of <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>N3<annotation encoding="application / x-tex">N_3< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>L<annotation encoding="application / x-tex">L< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>M<annotation encoding="application / x-tex">M< / annotation>< / semantics>< / com:Math> or <com:Math><semantics>X<annotation encoding="application / x-tex">X< / annotation>< / semantics>< / com:Math>. In some implementation, the UE reports a capability <com:Math><semantics>C1<annotation encoding="application / x-tex">C_1< / annotation>< / semantics>< / com:Math> Example 2 For power allocation between panels and panel selection, the UE needs to feedback RSRP (reference signal received power) or RSRP offset to gNB. There are following method to report this information: In some implementation, the gNB configures a set of CSI-RS / SSB resource set, the UE reported it selected CRI and corresponding RSRP / SINR (signal to interference and noise ratio). The maximal number of the reported pair of (CRI, RSRP) need to be extended, for example larger than 4; In some implementation, the UE reports multiple pairs of (CRI, RSRP), the multiple RSRP is reported by quotient method such as the reported value is relative value: <com:Math><semantics>RSRPi,relative=RSRPiRSRPmax∈[0,1],i=0,1...E−1<annotation encoding="application / x-tex">RSRP_{i,relative} = \frac{RSRP_i}{RSRP_{max}} \in [0,1], i = 0,1...E-1< / annotation>< / semantics>< / com:Math> <com:Math><semantics>RSRPmax<annotation encoding="application / x-tex">RSRP_{max}< / annotation>< / semantics>< / com:Math> corresponds to the first reported <com:Math><semantics>RSRP0,relative=1<annotation encoding="application / x-tex">RSRP_{0,relative} = 1< / annotation>< / semantics>< / com:Math> CRI / SSBRI, isn't reported to gNB. In another implementation, the UE reports <com:Math><semantics>RSRPmax<annotation encoding="application / x-tex">RSRP_{\max}< / annotation>< / semantics>< / com:Math> absolute value of for the first CRI and relative value of other reported CRI / SSBRI <com:Math><semantics>RSRPi,relative,i=1,2,...,E−1<annotation encoding="application / x-tex">RSRP_{i,relative}, i = 1, 2, ..., E - 1< / annotation>< / semantics>< / com:Math> . E is the quantity of the reported CRI / SSBRI. In some implementation, the UE reports RSRP or relative value of each CRI, doesn't report CRI, just RSRP / relative RSRP. For example, the UE will reports absolute value of RSRP / relative RSRP for each CSI-RS resource / SSB resource in a candidate set. In some implementation, the UE reported E CRI / SSBRI and one combination RSRP / SINR. For example, the UE reports E CRI / SSBRI and one RSRP(or SINR), wherein the one RSRP / SINR correspond to the E reported CRI / SSBRI, combination RSRP. Various preferred solutions adopted by embodiments are listed as below. One solution includes a method of wireless communication (e.g., method 400 depicted in FIG. 4), comprising: determining (402), by a first communication device, channel state information CSI corresponding to P ports, where P is a positive integer; and transmitting (404), by the first communication device to a second communication device, the CSI; wherein the CSI includes at least one set of a precoding matrix indicator PMI, <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> sets of channel quality indicators CQIs, or <com:Math><semantics>R<annotation encoding="application / x-tex">R< / annotation>< / semantics>< / com:Math> value of rank indicator RI, wherein the one set PMI indicates <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> sets of precoding matrices corresponding to <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time domain units, the <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> sets of CQIs corresponds to <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> second type of time domain units, the <com:Math><semantics>R<annotation encoding="application / x-tex">R< / annotation>< / semantics>< / com:Math> values of RI corresponds to <com:Math><semantics>R<annotation encoding="application / x-tex">R< / annotation>< / semantics>< / com:Math> third type of time domain units, and wherein <com:Math><semantics>QR<annotation encoding="application / x-tex">Q_{R}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> are positive integers. The first communication device can be a first base station, or a first UE; The second communication device can be a second base station, or a second UE Another solution includes a method of wireless communications (e.g., method 800 depicted in FIG. 8) includes: receiving (802), from a first communication device by a second communication device, a channel state information CSI corresponding to P ports, where P is a positive integer; wherein the CSI includes at least one set of a precoding matrix indicator PMI, <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> sets of channel quality indicators CQIs, or R value of rank indicator RI, wherein the one set PMI indicates <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> sets of precoding matrices corresponding to <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time domain units, the <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> sets of CQIs corresponds to <com:Math><semantics>Q<annotation encoding="application / x-tex">Q< / annotation>< / semantics>< / com:Math> second type of time domain units, the <com:Math><semantics>R<annotation encoding="application / x-tex">R< / annotation>< / semantics>< / com:Math> values of RI corresponds to R third type of time domain units, and wherein <com:Math><semantics>QR<annotation encoding="application / x-tex">Q_{R}< / annotation>< / semantics>< / com:Math>, <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> are positive integers. Another solution includes a method of wireless communication (e.g., method 500 depicted in FIG. 5), comprising: receiving (502), by a first communication device from a second communication device, signal transmissions comprising transmissions of a first reference signal and a second reference signal according to a pattern; and transmitting (504), by the first communication device to the second communication device, channel state information according to the first reference signal and the second reference signal. The wireless communication device feedback one set of PMI (precoding matrix indicator). The one set of PMI indicate <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> set of precoding matrices corresponding to <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> first type of time domain units. The one set of PMI applies for each precoding matrix of <com:Math><semantics>C4<annotation encoding="application / x-tex">C_4< / annotation>< / semantics>< / com:Math> set of precoding matrices. It captures the frequency domain feature and time domain feature of the channel. All precoding matrices share same information indicated by the one set of PMI. It reduces overhead of CSI and the second communication device can get more accurate CSI because the PMI adopts the frequency domain feature and time domain feature of the channel and we propose a model and the PMI just includes some parameters of the model. The present document also reports RI / CQI for multiple time domain units in one CSI-reporting. It allows the second communication device to schedule with high flexibility based on the reported CSI Another solution includes a method of wireless communication (e.g., method 900 depicted in FIG. 9), comprising: receiving (902), by a first communication device from a second communication device, signal transmissions comprising transmissions of a first reference signal and a second reference signal according to a pattern; and transmitting (904), by the first communication device to the second communication device, channel state information according to the first reference signal and the second reference signal. FIG. 6 shows an exemplary block diagram of a hardware platform 700 that may be a part of a network device (e.g., base station) or a communication device (e.g., a network node or a user equipment (UE)). The hardware platform 700 includes at least one processor 710 and a memory 705 having instructions stored thereupon. In some embodiments, the memory 705 may be omitted or may be internal to the processor 710. The instructions upon execution by the processor 710 configure the hardware platform 700 to perform the operations described in FIGS. 1 to 8 in the various embodiments described in this patent document. The transmitter 715 transmits or sends information or data to another device. For example, a network device transmitter can send a message to a user equipment. The receiver 720 receives information or data transmitted or sent by another device. For example, a user equipment can receive a message from a network device. The implementations as discussed above will apply to a wireless communication. FIG. 7 shows an example of a wireless communication system (e.g., a 5G or NR cellular network) that includes a base station 820 and one or more user equipment (UE) 811, 812 and 813. In some embodiments, the UEs access the BS (e.g., the network) using a communication link to the network (sometimes called uplink direction, as depicted by dashed arrows 831, 832, 833), which then enables subsequent communication (e.g., shown in the direction from the network to the UEs, sometimes called downlink direction, shown by arrows 841, 842, 843) from the BS to the UEs. In some embodiments, the BS send information to the UEs (sometimes called downlink direction, as depicted by arrows 841, 842, 843), which then enables subsequent communication (e.g., shown in the direction from the UEs to the BS, sometimes called uplink direction, shown by dashed arrows 831, 832, 833) from the UEs to the BS. The UE may be, for example, a smartphone, a tablet, a mobile computer, a machine to machine (M2M) device, an Internet of Things (IoT) device, and so on. From the foregoing, it will be appreciated that specific embodiments of the presently disclosed technology have been described herein for purposes of illustration, but that various modifications may be made without deviating from the scope of the invention. Accordingly, the presently disclosed technology is not limited except as by the appended claims. The disclosed and other embodiments, modules and the functional operations described in this document can be implemented in digital electronic circuitry, or in computer software, firmware, or hardware, including the structures disclosed in this document and their structural equivalents, or in combinations of one or more of them. The disclosed and other embodiments can be implemented as one or more computer program products, i.e., one or more modules of computer program instructions encoded on a computer readable medium for execution by, or to control the operation of, data processing apparatus. The computer readable medium can be a machine-readable storage device, a machine-readable storage substrate, a memory device, a composition of matter effecting a machine-readable propagated signal, or a combination of one or more them. The term "data processing apparatus" encompasses all apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, or multiple processors or computers. The apparatus can include, in addition to hardware, code that creates an execution environment for the computer program in question, e.g., code that constitutes processor firmware, a protocol stack, a database management system, an operating system, or a combination of one or more of them. A propagated signal is an artificially generated signal, e.g., a machine-generated electrical, optical, or electromagnetic signal, that is generated to encode information for transmission to suitable receiver apparatus. A computer program (also known as a program, software, software application, script, or code) can be written in any form of programming language, including compiled or interpreted languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program does not necessarily correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network. The processes and logic flows described in this document can be performed by one or more programmable processors executing one or more computer programs to perform functions by operating on input data and generating output. The processes and logic flows can also be performed by, and apparatus can also be implemented as, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application specific integrated circuit). Processors suitable for the execution of a computer program include, by way of example, both general and special purpose microprocessors, and any one or more processors of any kind of digital computer. Generally, a processor will receive instructions and data from a read only memory or a random-access memory or both. The essential elements of a computer are a processor for performing instructions and one or more memory devices for storing instructions and data. Generally, a computer will also include, or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices for storing data, e.g., magnetic, magneto optical disks, or optical disks. However, a computer need not have such devices. Computer readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media and memory devices, including by way of example semiconductor memory devices, e.g., EPROM, EEPROM, and flash memory devices; magnetic disks, e.g., internal hard disks or removable disks; magneto optical disks; and CD ROM and DVD-ROM disks. The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry. While this patent document contains many specifics, these should not be construed as limitations on the scope of any invention or of what may be claimed, but rather as descriptions of features that may be specific to particular embodiments of particular inventions. Certain features that are described in this patent document in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination. Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a subcombination or variation of a subcombination. Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. Moreover, the separation of various system components in the embodiments described in this patent document should not be understood as requiring such separation in all embodiments. Only a few implementations and examples are described, and other implementations, enhancements and variations can be made based on what is described and illustrated in this patent document.

Claims

1. A method of wireless communication, comprising: determining, by a first communication device, channel state information CSI corresponding to P Channel State Information Reference Signal (CSI-RS) ports, where P is a positive integer; and transmitting, by the first communication device to a second communication device, the CSI in one CSI reporting; wherein the CSI in the one CSI reporting includes one set of a precoding matrix indicator PMI, Q sets of channel quality indicators CQIs, and R values of rank indicator RI, wherein the one set of PMI indicates C 4 C_4 sets of precoding matrices corresponding to C 4 C_4 first type of time domain units, the Q Q sets of CQIs have a one to one correspondence to Q Q second type of time domain units, where Q is greater than 1, the R values of RI corresponds to R third type of time domain units, wherein Q Q , R R , C 4 C_4 are positive integers, and wherein at least one of Q R Q_{R} , C 4 C_4 is greater than 1, C 4 ≥ Q ≥ R C_4 \ge Q \ge R , one of the R R third type of time domain units includes one or more second type of time domain units, one of the R third type of time domain unit includes one or more first type of time domain units, one of the Q Q second type of time domain unit includes one or more of the C 4 C_4 first type of time domain units, each of the C 4 C_4 first type of time units corresponds to one or more slots.

2. The method of claim 1, wherein the one set of PMI includes information about one first type of vector set, information about v second type of vector sets, information about D third type of vector sets, index of a strongest coefficient and E coefficients, wherein v is a total number of layers, D and E are positive integers, wherein each first type of vector from the one first type of vector set includes P / 2 P / 2 elements, wherein the P / 2 P / 2 elements of one first type of vector have a same amplitude and each of the P / 2 P / 2 elements has a respective phase, wherein each second type of vector from one of the v v second type of vector sets includes N 3 N_3 elements each of which corresponds to one frequency domain unit k k , wherein k t h k^{th} elements of each second type of vector have a format: [Image disponible dans le document PDF, Image available in the PDF document] wherein f is an index of a corresponding second type of vector that corresponds to one value of n 3 f n_3^f , and wherein each third type of vector from one of the D third type of vector sets includes C 4 C_4 elements each of which corresponds to one of the C 4 C_4 first type of time domain units.

3. The method of claim 1, wherein each precoding matrix of the C 4 C_4 sets of precoding matrices is based on three types of vector sets including a first type of vector set, a second type of vector set and a third type of vector set, wherein each first type of vector from the first type of vector set includes P / 2 P / 2 elements, wherein the P / 2 P / 2 elements of one first type of vector have a same amplitude and each of the P / 2 P / 2 elements has a respective phase, wherein each second type of vector from the second type of vector set includes N 3 N_3 elements each of which corresponds to one frequency domain unit k k , wherein k t h k^{th} elements of each second type of vector have a format: y k ( f ) = e j 2 π n 3 f k N 3 y_k^{(f)} = e^{j\frac{2\pi n_3^f k}{N_3}} , n 3 f ∈ { 0 , 1 , . . . , N 3 − 1 } n_3^f \in \{0,1,...,N_3-1\} wherein f f is an index of a corresponding second type of vector that corresponds to one value of n 3 f n_3^f , and wherein each third type of vector from the third type of vector set includes C 4 C_4 elements each of which corresponds to one of the C 4 C_4 first type of time domain units.

4. The method of claim 1, wherein an l t h l^{th} column W k , t l W_{k,t}^l of a precoding matrix W k , t W_{k,t} of the C 4 C_4 sets of precoding matrix is specified as [Image disponible dans le document PDF, Image available in the PDF document] wherein W k , t l W_{k,t}^l is the l t h l^{th} column of the precoding matrix W k , t W_{k,t} of a frequency domain unit k k , a first type of time domain unit corresponding to t t , and a layer l l ; [Image disponible dans le document PDF, Image available in the PDF document] wherein v i , i ∈ { 0 , 1 , . . . , L − 1 } v_i, i \in \{0,1,..., L-1\} is a first type of vector in a first type of vector set, y k , l ( f ) y_{k,l}^{(f)} is k t h k^{th} element of a second type of vector with index f f in a second type of vector set, s l , t x s_{l,t}^{x} is an element corresponding to t t of a third type of vector with index x x in a third type of vector set specific to layer l l ; a l , i , f , j a_{l,i,f,j} is a value with phase and amplitude, the amplitude of a l , i , f , j a_{l,i,f,j} is smaller than or equal to 1; t t is a time domain index corresponding to one of the C 4 C_4 first type of time domain units; wherein S l , t x S_{l,t}^{x} is specific to the first type of vector set and the second type of vector set and shared by all first type of vectors in the first type of vector set and all second type of vectors in the second type of vector set. The method of claim 4, wherein a l , i , f , x , t = p l , ⌊ i / L ⌋ ( 1 ) p l , i , f , x ( 2 ) φ l , i , f , x a_{l,i,f,x,t} = p_{l,\lfloor i / L \rfloor}^{(1)} p_{l,i,f,x}^{(2)} \varphi_{l,i,f,x}5. wherein p l , ⌊ i / L ⌋ ( 1 ) p_{l,\lfloor i / L \rfloor}^{(1)} and p l , i , f , x ( 2 ) p_{l,i,f,x}^{(2)} are amplitude coefficients and their phase is 0; p l , ⌊ i / L ⌋ ( 1 ) p_{l,\lfloor i / L \rfloor}^{(1)} and p l , i , f , x ( 2 ) p_{l,i,f,x}^{(2)} being equal to or larger than 0 and smaller than or equal to 1. φ l , i , f , x \varphi_{l,i,f,x} are phase coefficient and their amplitudes are 1 for l = 0 , 1 . . . , v − 1 , i = 0 , 1 , . . . , 2 * L − 1 , f = 0 , 1 , . . . M − 1 , x = 0 , 1 , . . . X − 1 l = 0,1..., v-1, i = 0,1,...,2 * L-1, f = 0,1,...M-1, x = 0,1,...X-1 , and the one set of PMI includes v sets of indexes each of which corresponds to a strongest coefficient a l , i , f , j a_{l,i,f,j} of the corresponding layer, wherein each of the v v sets of indexes includes i * ∈ { 0 , 1 , . . . , L − 1 } i^* \in \{0,1,...,L-1\} and x * ∈ { 0 , 1 , . . . , X − 1 } x^* \in \{0,1,...,X-1\} .

6. The method of claim 2, wherein one of the D third type of vector sets is specific to a layer and D equals ν \nu , wherein each of the D third type vector sets is used to determine a column of each precoding matrix of the C 4 C_4 sets of precoding matrix; and one third type vector with index x x is determined by f x f_x and a time domain index corresponding to one of the C 4 C_4 first type of time domain units, wherein a unit of f x f_x is 1.

7. A method of wireless communications, comprising: receiving, from a first communication device by a second communication device, channel state information, CSI corresponding to P Channel State Information Reference Signal (CSI-RS) ports in one CSI reporting, where P is a positive integer; and performing, by the second communication device, a communication with the first communication device according to the CSI, wherein the CSI in the one CSI reporting includes at least one set of a precoding matrix indicator PMI, Q Q sets of channel quality indicators CQIs, and R R value of rank indicator RI, wherein the one set of PMI indicates C 4 ^{C_4} sets of precoding matrices corresponding to C 4 ^{C_4} first type of time domain units, the 𝒬 \mathcal{Q} sets of CQIs have one to one correspondence to 𝒬 \mathcal{Q} second type of time domain units, Q is larger than 1, the R R values of RI corresponds to R R third type of time domain units, wherein Q R Q_{R} , C 4 C_4 are positive integers, and wherein at least one of Q R Q_{R} , C 4 C_4 is greater than 1, C 4 ≥ Q ≥ R C_4 \ge Q \ge R , one of the R third type of time domain units includes one or more second type of time domain units, one of the R R third type of time domain unit includes one or more first type of time domain units, and one of the Q Q second type of time domain unit includes one or more of the C 4 C_4 first type of time domain units, each of the C 4 C_4 first type of time units corresponds to one or more slots.

8. The method of claim 7, wherein the one set of PMI includes information about one first type of vector set, information about v second type of vector sets, information about D third type of vector sets, index of a strongest coefficient and E coefficients, wherein v is a total number of layers, D and E are positive integers, wherein each first type of vector from the one first type of vector set includes P / 2 P / 2 elements, wherein the P / 2 P / 2 elements of one first type of vector have a same amplitude and each of the P / 2 P / 2 elements has a respective phase, wherein each second type of vector from one of the v v second type of vector sets includes N 3 N_3 elements each of which corresponds to one frequency domain unit k k , wherein k t h k^{th} elements of each second type of vector have a format: [Image disponible dans le document PDF, Image available in the PDF document] wherein f is an index of a corresponding second type of vector that corresponds to one value of n 3 f n_3^f , and wherein each third type of vector from one of the D third type of vector sets includes C 4 C_4 elements each of which corresponds to one of the C 4 C_4 first type of time domain units.

9. The method of claim 7, wherein each precoding matrix of the C 4 C_4 sets of precoding matrices is based on three types of vector sets including a first type of vector set, a second type of vector set and a third type of vector set, wherein each first type of vector from the first type of vector set includes P / 2 P / 2 elements, wherein the P / 2 P / 2 elements of one first type of vector have a same amplitude and each of the P / 2 P / 2 elements has a respective phase, wherein each second type of vector from the second type of vector set includes N 3 N_3 elements each of which corresponds to one frequency domain unit k k , wherein k t h k^{th} elements of each second type of vector have a format: [Image disponible dans le document PDF, Image available in the PDF document] wherein f f is an index of a corresponding second type of vector that corresponds to one value of n 3 f n_3^f , and wherein each third type of vector from the third type of vector set includes C 4 C_4 elements each of which corresponds to one of the C 4 C_4 first type of time domain units.

10. The method of claim 7, wherein an l t h l^{th} column W k , t l W_{k,t}^l of a precoding matrix W k , t W_{k,t} of the C Δ C_{\Delta} sets of precoding matrix is specified as [Image disponible dans le document PDF, Image available in the PDF document] wherein W k , t l W_{k,t}^l is the l t h l^{th} column of the precoding matrix W k , t W_{k,t} of a frequency domain unit k k , a first type of time domain unit corresponding to t t , and a layer l l ; wherein k = 0 , 1 , . . . N 3 − 1 k = 0,1,...N_3 - 1 , t = t 0 , t 1 , . . . , t C 4 − 1 t = t_0, t_1,..., t_{C_4-1} , and l = 0 , 1 , . . . , v − 1 l = 0,1,...,v-1 ; wherein v i , i ∈ { 0 , 1 , . . . , L − 1 } v_i, i \in \{0,1,...,L-1\} is a first type of vector in a first type of vector set, 𝒴 k , l ( j ) \mathcal{Y}_{k,l}^{(j)} is k t h k^{th} element of a second type of vector with index f f in a second type of vector set, S l , t x S_{l,t}^{x} , is an element corresponding to t t of a third type of vector with index x x in a third type of vector set; a l , i , f , j a_{l,i,f,j} is a value with phase and amplitude, the amplitude of a l , i , f , j a_{l,i,f,j} is smaller than or equal to 1; t is a time domain index corresponding to one of the C 4 C_4 first type of time domain units; wherein S l , t λ S_{l,t}^{\lambda} is specific to the first type of vector set and the second type of vector set and shared by all first type of vectors in the first type of vector set and all second type of vectors in the second type of vector set.

11. The method of claim 10, wherein a l , i , f , x , t = p l , ⌊ i / L ⌋ ( 1 ) p l , i , f , x ( 2 ) φ l , i , f , x a_{l,i,f,x,t} = p_{l,\lfloor i / L \rfloor}^{(1)} p_{l,i,f,x}^{(2)} \varphi_{l,i,f,x} wherein p l , ⌊ i / L ⌋ ( 1 ) p_{l,\lfloor i / L\rfloor}^{(1)} , p l , i , f , x ( 2 ) p_{l,i,f,x}^{(2)} , are amplitude coefficients and their phase is 0 0 ; p l , ⌊ i / L ⌋ ( 1 ) p_{l,\lfloor i / L\rfloor}^{(1)} , p l , i , f , x ( 2 ) p_{l,i,f,x}^{(2)} , is equal to or larger than 0 and smaller than or equal to 1. φ l , i , f , x \varphi_{l,i,f,x} are phase coefficient and their amplitudes are 1 for l = 0 , 1 . . . , v − 1 , i = 0 , 1 , . . . , 2 * L − 1 , f = 0 , 1 , . . . M − 1 , x = 0 , 1 , . . . X − 1 l = 0,1..., v-1, i = 0,1,...,2 * L-1, f = 0,1,...M-1, x = 0,1,...X-1 . and the one set of PMI includes v sets of indexes each of which corresponds to a strongest coefficient a l , i , f , j a_{l,i,f,j} of the corresponding layer, wherein each of the v v sets of indexes includes i * ∈ { 0 , 1 , . . . , L − 1 } i^* \in \{0,1,...,L-1\} and x * ∈ { 0 , 1 , . . . , X − 1 } x^* \in \{0,1,...,X-1\}12. The method of claim 8, wherein: one third type of vector set is specific to a layer and D equals v v , wherein each of the D third type vector sets is used to determine a column of each precoding matrix of the C 4 C_4 sets of precoding matrix; or one third type vector with index x x is determined by f x f_x and a time domain index wherein a unit of the f x f_x is 1.

13. A communication device comprising at least one processor that is configured to cause the communication device to implement a method recited in any one of claims 1 to 6.

14. A communication device comprising at least one processor that is configured to cause the communication device to implement a method recited in any one of claims 7 to 12.