Method and apparatus for operating in a broadband communication system

Abstract

A method implemented by a first device operating in a communication system includes: obtaining a channel representation of a set of channels between the first device and a second device, wherein the set of channels corresponds to a set of subcarriers, the first device has multiple antenna ports, and the second device has one or more antenna ports; the first device determines one or more communication filters based on at least the channel representation; the first device applies the one or more communication filters to communication on at least one antenna port of the multiple antenna ports of the first device, wherein the communication is performed on the set of subcarriers.

Description

[0001] This application claims the benefit of U.S. Provisional Application No. 62 / 862,495, filed on June 17, 2019, entitled “System and Method for Wideband MIMO Communications,” which is incorporated herein by reference in its entirety. Technical Field

[0002] The present invention relates generally to digital communication methods and apparatus, and in particular embodiments to methods and apparatus for operation in a broadband communication system. Background Art

[0003] Fifth-generation (5G) mobile systems rely on massive multiple-input multiple-output (MIMO) systems to provide the necessary high spectral efficiency (SE) to meet the ever-increasing user data rate demands. Various massive MIMO techniques have been investigated, most notably zero-forcing (ZF), eigen ZF (EZF), and weighted minimum mean square error (MMSE), which have demonstrated significant improvements in spectral efficiency. However, while these techniques effectively suppress intra-cell interference, they cannot control inter-cell interference, thus failing to fully leverage the potential of massive MIMO.

[0004] A major reason for not achieving the expected performance is the need to understand the Channel State Information (CSI) between all cooperating transmitting and receiving points. To address this issue, several distributed CSI acquisition methods have been proposed. Bi-directional training (BiT) is one such very promising technique, which was first used for adaptive antenna beamforming in the presence of simulated interference in time division duplex (TDD) communication systems. BiT has been shown to significantly suppress inter-cell and intra-cell interference in narrowband (NB) (i.e., frequency non-selective) multi-cell TDD communication systems.

[0005] However, most current and future communication systems are wideband (WB) systems, so there is a need to extend BiT to wideband communication systems. Summary of the Invention

[0006] According to a first aspect, a method implemented by a first device operating in a communication system is provided. The method includes: the first device obtaining a channel representation of a set of channels between the first device and a second device, wherein the set of channels corresponds to a set of subcarriers, the first device has multiple antenna ports, and the second device has one or more antenna ports; the first device determining one or more communication filters based on at least the channel representation; and the first device applying the one or more communication filters to communication on at least one of the multiple antenna ports of the first device, wherein the communication is performed on the set of subcarriers.

[0007] According to the first aspect, in a first implementation of the method, the channel representation comprises a channel matrix, a size of which is specified based on the multiple antenna ports of the first device and the one or more antenna ports of the second device.

[0008] According to the first aspect or any of the above implementations of the first aspect, in a second implementation of the method, determining one or more communication filters includes using a bi-directional training (BiT) process.

[0009] According to the first aspect or any of the above implementations of the first aspect, in a third implementation of the method, obtaining the channel representation of a group of channels includes receiving the channel representation of the group of channels from the second device.

[0010] According to the first aspect or any of the above implementations of the first aspect, in a fourth implementation of the method, obtaining the channel representation of a group of channels includes determining the channel representation of the group of channels based on the individual channel representations of the group of channels.

[0011] According to the first aspect or any of the above implementations of the first aspect, in a fifth implementation of the method, the channel representation is determined based on the phase domain components of the group of channels and the power / amplitude domain components of the group of channels.

[0012] According to the first aspect or any of the above implementations of the first aspect, in a sixth implementation of the method, the channel representation includes a phase domain component of the set of channels, and the phase domain component is constrained by a power / amplitude domain component of the set of channels.

[0013] According to the first aspect or any above-mentioned implementation manner of the first aspect, in the seventh implementation manner of the method, obtaining the channel representation of a group of channels includes: the first device receives a reference signal sent on the group of channels corresponding to the group of subcarriers; the first device estimates the respective channel representations based on the measurement of the received reference signal; the first device determines the channel representation based on the respective channel representations.

[0014] According to the first aspect or any above-mentioned implementation manner of the first aspect, in an eighth implementation manner of the method, obtaining the channel representation of a group of channels includes: the first device receiving an indication of a subcarrier index; and the first device selecting a channel estimation value associated with the subcarrier index as the channel representation of the group of channels.

[0015] According to the first aspect or any above-mentioned implementation manner of the first aspect, in a ninth implementation manner of the method, the one or more communication filters include at least one of a transmission precoder for a subset of the multiple antenna ports of the first device or a reception combiner for the multiple antenna ports of the first device.

[0016] According to the first aspect or any of the above implementations of the first aspect, in a tenth implementation of the method, the group of channels corresponds to a subset of the group of subcarriers.

[0017] According to the first aspect or any of the above implementations of the first aspect, in an eleventh implementation of the method, the communication includes at least one of sending a first message or receiving a second message.

[0018] According to a second aspect, a method implemented by a first device operating in a communication system is provided. The method comprises: receiving, by the first device, a signal transmitted on a set of channels corresponding to a set of subcarriers of the communication system, wherein the first device has multiple antenna ports; determining, by the first device, a channel representation of the set of channels based on the signal received on the set of channels corresponding to the set of subcarriers; and transmitting, by the first device, the channel representation.

[0019] According to the second aspect, in a first implementation manner of the method, the method further includes the first device receiving resource allocation information associated with resources for transmitting the received signal.

[0020] According to the second aspect or any of the above implementations of the second aspect, in a second implementation of the method, sending the channel representation includes the first device sending an indication of a subcarrier index associated with the channel representation.

[0021] According to the second aspect or any of the above implementations of the second aspect, in a third implementation of the method, sending the channel representation includes an indication that the first device sends the channel representation.

[0022] According to the second aspect or any of the above implementations of the second aspect, in a fourth implementation of the method, the method further includes: the first device determines one or more communication filters based on the channel representation of the set of channels, wherein the determining of one or more communication filters includes using a BiT process.

[0023] According to the second aspect or any of the above implementations of the second aspect, in a fifth implementation of the method, the size of the channel matrix is ​​specified based on the multiple antenna ports of the first device and the one or more antenna ports of the second device.

[0024] According to a third aspect, a device is provided. The device includes: a non-transitory memory including instructions; one or more processors in communication with the memory, wherein the one or more processors execute the instructions to: obtain a channel representation of a set of channels between a first device and a second device, wherein the set of channels corresponds to a set of subcarriers, the first device having multiple antenna ports, and the second device having one or more antenna ports; determine one or more communication filters based on at least the channel representation; and apply the one or more communication filters to communication on at least one of the multiple antenna ports of the first device, wherein the communication occurs on the set of subcarriers.

[0025] According to the third aspect, in a first implementation of the device, the one or more processors further execute the instructions to receive the channel representation of the set of channels from the second device.

[0026] According to the third aspect or any of the foregoing implementations of the third aspect, in a second implementation of the device, the one or more processors further execute the instructions to determine the channel representation of the group of channels based on the respective channel representations of the group of channels. According to the third aspect or any of the foregoing implementations of the third aspect, in a third implementation of the device, the channel representation is determined based on phase domain components of the group of channels and power / amplitude domain components of the group of channels.

[0027] According to the third aspect or any of the above implementations of the third aspect, in a fourth implementation of the device, the channel representation includes a phase domain component of the set of channels, and the phase domain component is constrained by a power / amplitude domain component of the set of channels.

[0028] According to the third aspect or any above-mentioned implementation manner of the third aspect, in a fifth implementation manner of the device, the one or more processors further execute the instructions to: receive a reference signal sent on the set of channels corresponding to the set of subcarriers; estimate the individual channel representations based on measurements of the received reference signals; and determine the channel representation based on the individual channel representations.

[0029] According to the third aspect or any above-mentioned implementation manner of the third aspect, in a sixth implementation manner of the device, the one or more processors further execute the instructions to: receive an indication of a subcarrier index; and select a channel estimation value associated with the subcarrier index as the channel representation of the set of channels.

[0030] According to the third aspect or any above-mentioned implementation manner of the third aspect, in a seventh implementation manner of the device, the channel representation includes a channel matrix, and the size of the channel matrix is ​​specified according to the multiple antenna ports of the first device and the one or more antenna ports of the second device.

[0031] According to the third aspect or any of the above implementations of the third aspect, in an eighth implementation of the device, the one or more communication filters are determined using a BiT process.

[0032] According to a fourth aspect, a device is provided, comprising: a non-transitory memory comprising instructions; one or more processors in communication with the memory, wherein the one or more processors execute the instructions to: receive a signal transmitted on a set of channels corresponding to a set of subcarriers of a communication system, wherein the first device has multiple antenna ports; determine a channel representation of the set of channels based on the signal received on the set of channels corresponding to the set of subcarriers; and transmit the channel representation.

[0033] According to the fourth aspect, in a first implementation of the device, the one or more processors further execute the instructions to receive resource allocation information associated with resources for transmitting the received signal.

[0034] According to the fourth aspect or any of the above implementations of the fourth aspect, in a second implementation of the device, the one or more processors further execute the instruction to send an indication of a subcarrier index associated with the channel representation.

[0035] According to the fourth aspect or any of the above implementations of the fourth aspect, in a third implementation of the device, the one or more processors further execute the instructions to send an indication of the channel representation.

[0036] According to the fourth aspect or any above-mentioned implementation manner of the fourth aspect, in the fourth implementation manner of the device, the one or more processors also execute the instructions to determine one or more communication filters based on the channel representation of the set of channels, wherein the determining of one or more communication filters includes using a BiT process.

[0037] An advantage of a preferred embodiment is that BiT is extended to WB communication systems without incurring much overhead compared to BiT for NB communication systems.

[0038] Another advantage of a preferred embodiment is that a distributed BiT implementation is provided, enabling communication devices to independently determine communication filters (e.g., transmit precoders or receive combiners), rather than using a centralized entity to determine communication filters. A centralized implementation may increase communication overhead and computational requirements of the centralized entity. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] For a more complete understanding of the present invention and its advantages, reference is now made to the following description taken in conjunction with the accompanying drawings, in which:

[0040] Figure 1 An exemplary wireless communication system is shown;

[0041] Figure 2 An exemplary communication system is shown, and a mathematical representation of signals transmitted in the communication system is provided;

[0042] Figure 3 A schematic diagram illustrating signaling and processing for multiple devices performing bi-directional training (BiT) and BiT transmission;

[0043] Figure 4A shows exemplary resources, highlighting channel measurement resources (CMR) and interference measurement resources (IMR);

[0044] Figure 4B An exemplary relationship between CMR and IMR and resource elements of an orthogonal frequency division multiplexing (OFDM) symbol or a physical resource block (PRB) is shown;

[0045] Figure 5is a flowchart of exemplary operations performed when a UE supports BiT training;

[0046] Figure 6 is a flow chart of an exemplary method for SRS communication;

[0047] Figure 7 and Figure 8 is a schematic diagram of RGB and shows an exemplary mapping relationship between SRS resources and ports;

[0048] Figure 9 An exemplary communication system is shown, highlighting the symbolic notation used in the discussion of the exemplary embodiments described herein;

[0049] Figure 10A Schematic diagram of a set of T matrices of channel coefficients and a narrowband representation H for exemplary embodiments described herein;

[0050] Figure 10B A schematic diagram of channel coefficients for a set of T matrices provided for exemplary embodiments described herein, highlighting narrowband representation;

[0051] Figure 11 A flowchart of exemplary operations performed by a device when performing a wideband BiT process is provided for exemplary embodiments described herein;

[0052] Figure 12A A high-level view of example operations performed by a device as it obtains a channel representation of a wideband channel is provided for example embodiments described herein;

[0053] Figure 12B A detailed view of exemplary operations performed by a device when acquiring a channel representation of a wideband channel is provided for exemplary embodiments described herein;

[0054] Figure 13 A schematic diagram illustrating a plurality of devices exchanging messages and performing processing in performing an iterative distributed broadband BiT process provided by the exemplary embodiments described herein;

[0055] Figure 14A A schematic diagram of multiple devices exchanging messages and performing processes for sharing subcarrier indices and using the subcarrier indices as narrowband representations of wideband channels, provided for exemplary embodiments described herein, wherein an access node determines the narrowband representation;

[0056] Figure 14B A schematic diagram of multiple devices exchanging messages and performing processes for sharing subcarrier indices and using the subcarrier indices as narrowband representations of wideband channels, wherein a UE determines the narrowband representation, according to exemplary embodiments described herein;

[0057] Figure 15AGraphs of cell spectral efficiency for BiT and ZF provided for exemplary embodiments described herein;

[0058] Figure 15B A data graph showing the SINR cumulative distribution function (CDF) of BiT and ZF provided by the exemplary embodiments described herein;

[0059] Figure 15C Data plots showing convergence of BiT and ZF for exemplary embodiments described herein;

[0060] Figure 16 An exemplary communication system provided by the exemplary embodiments described herein is shown.

[0061] Figure 17A and Figure 17B Exemplary devices are shown that can implement the methods and guidance provided by the present invention.

[0062] Figure 18 is a block diagram of a processing system that can be used to implement the devices and methods disclosed herein. DETAILED DESCRIPTION

[0063] The following describes in detail the structure and use of the embodiments disclosed herein. However, it should be understood that the present invention provides many applicable concepts that can be embodied in a variety of specific contexts. The specific embodiments discussed are merely illustrative of the specific structure and use of the embodiments and do not limit the scope of the present invention.

[0064] Figure 1 An exemplary wireless communication system 100 is shown. The communication system 100 includes an access node 110 having a coverage area 111. The access node 110 serves a plurality of user equipments (UEs), including UE 120 and UE 122. Transmissions from the access node 110 to the UEs are called downlink (DL) transmissions and occur on downlink channels (e.g., Figure 1 ), and the transmission from the UE to the access node 110 is called uplink (UL) transmission, which occurs on the uplink channel (such as Figure 1 ). Services may be provided to multiple UEs by a service provider connected to the access node 110 via a backhaul network 130 (eg, the Internet). The wireless communication system 100 may include multiple distributed access nodes 110.

[0065] There are several operating modes in a typical communication system. In a cellular operating mode, communication between multiple UEs passes through the access node 110, while in a device-to-device communication mode, such as a proximity service (ProSe) operating mode, UEs can communicate directly with each other. An access node may also be generally referred to as a NodeB, an evolved NodeB (eNB), a next generation (NG) NodeB (gNB), a master eNB (MeNB), a secondary eNB (SeNB), a master gNB (MgNB), a secondary gNB (SgNB), a network controller, a control node, a base station, an access point, a transmission point (TP), a transmission-reception point (TRP), a cell, a carrier, a macro cell, a femto cell, a pico cell, a relay station, a customer premises equipment (CPE), etc. A UE may also be generally referred to as a mobile station, mobile phone, terminal, user, subscriber, site, communication device, CPE, relay station, integrated access and backhaul (IAB) relay station, etc. It should be noted that when relay technology is used (depending on the relay station, picocell, CPE, etc.), especially when multi-hop relay technology is used, the boundary between the controller and the node controlled by the controller may become blurred, and dual nodes (controller or node controlled by the controller) may be deployed, where the first node providing configuration information or control information to the second node is the controller. Similarly, the concepts of uplink transmission and downlink transmission can also be expanded.

[0066] A cell may include one or more uplink or downlink bandwidth parts (BWPs) allocated to a UE. Each BWP may have its own BWP-specific system parameters (numerology) and configuration. It should be noted that for a UE, not all BWPs need to be activated at the same time. A cell may correspond to one or more carriers. Typically, a cell (e.g., a primary cell (PCell) or a secondary cell (SCell)) is a component carrier (e.g., a primary component carrier (PCC) or a secondary CC (SCC)). For some cells, each cell may include multiple uplink carriers, one carrier with an associated downlink is called an uplink carrier or a non-supplementary uplink (UL, non-SUL) carrier, and other carriers without an associated downlink are called supplementary uplink (UL, SUL) carriers. A cell or carrier may be configured with a time slot or subframe format including downlink symbols and uplink symbols, and the cell or carrier may operate in time division duplex (TDD) mode. Generally speaking, for unpaired spectrum, the cell or carrier is in TDD mode, and for paired spectrum, the cell or carrier is in frequency division duplex (FDD) mode. According to one or more wireless communication protocols such as long term evolution (LTE), advanced LTE (LTE-A), 5G, 5G LTE, 5G NR, High Speed ​​Packet Access (HSPA), Wi-Fi 802.11a / b / g / n / ac, the access node can provide wireless access. For simplicity, only one access point and two UEs are shown in the figure, but it can be understood that the communication system can use multiple access nodes capable of communicating with multiple UEs.

[0067] In standard antenna-to-antenna channel estimation, the channel between two devices is estimated by having the first device send a known signal to the second device on one or more known time resources or frequency resources. The received signal on the second device can be expressed as:

[0068] y=Hx+n (1)

[0069] Where y is the received signal on the second device side, x is a known signal (which can be a reference signal, a pilot, or a pilot signal), H is the channel model or channel response, and n is the noise (and interference for some communication channels). Because x is known to the second device, the second device can determine or estimate H based on y.

[0070] The concepts of antenna, antenna element, and antenna port are often used interchangeably, but in some specific scenarios, they can represent different but related topics. For example, a transmit (Tx) antenna port can be composed of (or virtualized into) multiple antenna elements or antennas, and the receiver only sees one transmit antenna port and cannot see each of the multiple antenna elements or antennas. This virtualization can be achieved through techniques such as beamforming.

[0071] Figure 2 An exemplary communication system 200 is shown and a mathematical representation of signals transmitted in the communication system is provided. The communication system 200 includes an access node 205 communicating with a UE 210. Figure 2 As shown, the access node 205 uses a transmit filter v, while the UE 210 uses a receive filter w. Both the access node 205 and the UE 210 use linear precoding or a linear combination. Assume that H is the N of the multiple-input multiple-output (MIMO) system. rx ×N tx Matrix, that is, there is N tx transmit antennas and N rx receiving antennas. The size is N tx The transmit filter v of size N × Ns enables the transmitter to precode or beamform the transmitted signal, where Ns is the number of layers, ports, streams, symbols, pilots, messages, data, or known transmit sequences. The receive filter w in a multi-antenna system is of size N rx ×Ns, w means usually according to w H The above description is for the transmission from access node 205 to UE 210, i.e., downlink transmission. The transmission can also occur in the opposite direction (uplink transmission), in which case the channel matrix becomes H in the case of TDD. H (Among them, H H is the Hermitian matrix of the channel model H), w is the transmit filter, and v is the receive filter. The w used for transmission and the w used for reception can be the same or different, and the same is true for v.

[0072] The downlink (or forward) channel 215 between the access node 205 and the UE 210 has a channel model or channel response H, while the uplink (or backward or reverse) channel 220 between the UE 210 and the access node 205 has a channel model or channel response H. H (Another common practice is to express the uplink channel as H T , which is the transposed model of the channel model H. ) Although Figure 2 Only one access node and one UE are shown, but communication system 200 is not limited to this scenario. The access node can serve multiple UEs on different time-frequency resources (e.g., in a frequency division multiplexing-time division multiplexing (FDM-TDM) communication system, as in a typical cellular system) or on the same time-frequency resources (e.g., in a multi-user MIMO (MU-MIMO) communication system, where multiple UEs are paired together and transmissions to each UE are individually precoded). Intra-cell interference exists between paired UEs.

[0073] There may also be multiple access nodes in the network, some of which can collaboratively serve UE 210 through joint transmission (e.g., coherent joint transmission, incoherent joint transmission, coordinated multi-point transmission, etc.), dynamic point switching, etc. Other access nodes may not serve UE 210, and transmissions from these access points to their own UEs may cause inter-cell interference to UE 210. The scenario considered in this article is a scenario with multiple access nodes and multiple UEs, where these access points collaboratively serve the UEs using MU-MIMO technology.

[0074] Bi-directional training (BiT), also known as forward-backward training, is a general distributed training process with low computational complexity. It aims to train and update transmit precoders and receive combiners without exact CSI estimates. BiT is suitable for transmit beamformers (often also referred to as transmit precoders, transmit filters, spatial transmit filters, transmit filters, analog precoders, etc.) and receive combiners (often also referred to as receive filters, spatial receive filters, analog combiners, etc.) in TDD MIMO communication systems. In BiT, neither device participating in the BiT (transmitter or receiver) has prior information about the CSI, specifically detailed information about the channel, such as the channel matrix H or covariance matrix. The channel can be between a UE and one or more of its serving access nodes, or between the UE and one or more of its interfering access nodes. (Prior information typically requires information exchange between access nodes, such as RS information or channel information about the interfering link, so that the UE or access node can estimate the interfering link.) Iterative BiT includes forward training (e.g., in the downlink direction) and backward training (e.g., in the uplink direction) that are repeated until convergence is achieved. One-shot BiT includes one forward training step and one backward training step. BiT is able to adapt to unknown interference and can suppress interference without any channel estimation or CSI feedback, making BiT less sensitive to the orthogonality of the training sequence. BiT is discussed in more detail in commonly assigned patent application No. 15 / 983,692, filed on May 18, 2018, entitled "System and Method for Communications System Training," the entire contents of which are incorporated herein by reference.

[0075] Figure 3 Diagram 300 of exchanging signaling and performing processes for multiple devices performing BiT and BiT transmission. Diagram 300 shows an example of exchanging signaling and performing processes for an access node 305 and a UE 310 performing BiT and BiT transmission.

[0076] BiT begins with a BiT initialization phase 312. The BiT initialization phase 312 may include the access node 305 sending signaling configuration information to the UE 310 (step 315), and the UE 310 receiving the configuration information (step 317). The configuration information may include non-beamformed SRS information, which notifies the UE 310 of the time resources, frequency resources, or port resources allocated to transmit the non-beamformed SRS. The configuration information may also include other information. The configuration information may be sent to the UE 310 using an RRC message, a MAC message, or a DCI message. The configuration information may also include information regarding which type of SRS the UE 310 should transmit (e.g., non-beamformed SRS or beamformed SRS). The access node 305 also transmits a precoded RS as part of the signaling. The UE 310 transmits the SRS to the access node 305 (steps 323 and 325). For example, UE 310 transmits either a non-beamformed SRS or a beamformed SRS (event 327), as specified by access node 305. This phase is not specific to BiT. This phase can be used for non-BiT-oriented sounding. This phase can be used for purposes other than BiT and lasts for a period of time (e.g., a specified period of time or until a criterion is met), and the network determines when to initiate BiT based on this phase. Even after BiT completes initialization and performs iterations, sounding during this phase can still be used for other purposes, such as beam refinement.

[0077] The BiT initialization phase 312 may include BiT configuration and triggering of the BiT process. BiT configuration may be indicated using RRC signaling, MAC signaling, or DCI signaling. For example, one or more dedicated bits in the DCI are used to trigger the BiT process in a periodic mode, a semi-static mode, or an aperiodic mode using RRC high-layer configuration. For another example, a new RRC configured transmission mode is used to trigger the BiT process. For another example, a new MAC state is first configured via RRC and then activated via MAC signaling, thereby triggering the BiT process. For another example, RRC signaling is used to configure the BiT process in a periodic mode, a semi-static mode, or an aperiodic mode. For another example, the access node 305 may not explicitly trigger the BiT process, but may implicitly trigger the BiT process, for example, after completing the initial coarse beamforming phase. In this case, BiT is used to further optimize the beam found by the previous phase of BiT. For another example, the access node configures or specifies (via RRC signaling, MAC signaling, or DCI signaling) that the UE behavior is expected, and the UE performs this expected behavior. The expected UE behavior may be the BiT training operation to be performed by the UE, by which the BiT training process effectively begins. The expected UE behavior may include one or more steps corresponding to 339, 341, 343, 345, 347 and 349, which will be discussed further later.

[0078] BiT training 329 is a phase of the BiT process. In BiT training 329, precoded downlink signals and precoded uplink signals are sent by the access node 305 and the UE 310, respectively, so that the device receiving the precoded signals updates the corresponding receive filters and then updates the transmit precoding filters, or directly updates the transmit precoding filters. BiT training 329 can be an iterative process that continues until an end criterion is met or until signaling (RRC signaling, MAC signaling, or DCI signaling) is received to stop BiT or start a process different from BiT. BiT training 329 can occur in parallel with data transmission. For example, data transmission can begin after the transmit filter and receive filter have been determined, or even before the transmit filter and receive filter are optimized through continuous iterations of BiT training 329. Examples of termination criteria include: the signal plus interference to noise ratio (SINR) meets a specified standard (for example, the SINR change since the last iteration is less than a threshold (for example, 0.2dB)), the SINR has reached a certain threshold (for example, >10dB, which can already support a relatively high rate), the number of times the precoded downlink and uplink signals are sent reaches a certain threshold, etc. For example, BiT training 329 is not stopped, so that the transmit filter and the receive filter can track channel changes. For example, if BiT training 329 is interrupted, the tracking of channel changes is also interrupted.

[0079] The initial iteration of BiT training 329 includes: the access node 305 sends information about resources associated with signal reception (referred to herein as channel measurement resources (CMR)) or information about resources associated with interference transmission (referred to herein as interference measurement resources (IMR)) to the UE 310 (event 331). In other words, the access node configures resources for channel measurement (or signal reception) and interference measurement to the UE. The CMR can be a precoded downlink RS, such as an NZP CSI-RS. The UE 310 uses these signals to measure the signal sent by the access node 305 to the UE 310 or the channel between the access node 305 and the UE 310, and finds the best receive beam or the received signal properties in the general sense to update the UE's uplink precoder. The downlink signal is not limited to CSI-RS, but can also include data (in the physical downlink shared channel (PDSCH)) and DMRS for data PDSCH. Other examples of signals are mentioned below. In a sense, CMRs are not CMRs used for channel measurement or channel estimation in the traditional sense. Instead, CMRs are simply time or frequency resources on which the UE receives target signals for some or all layers, thereby using these target signals to perform the subsequent steps of the BiT process. Channel estimation may not be performed on these CMRs. The UE assumes that each port of the CMR is a target signal for a layer. The UE does not need to extract the target signal through sequence detection or other means (although it can do so). Instead, the UE uses one or more target signals received on the resources to calculate or obtain the receive combiner or transmit precoder.

[0080] IMR is typically used for signal transmission from an access node to other UEs, and UE 310 can use IMR to find the best receive beam, etc. In another embodiment, the UE processes one or more received transmissions on the IMR to optimize the receive beam found by considering the interference-free receive signal. IMR can also be used to measure interference. In a sense, IMR is not IMR used for interference measurement in the traditional sense. Instead, IMR is just a time resource or frequency resource that the UE uses to observe or receive interference signals of some or all layers on these resources, so as to use these interference signals to perform subsequent steps of the BiT process. The UE does not need to extract the interference signal (but can do so). Instead, the UE uses one or more interference signals received on the resources to calculate or obtain a receive combiner or transmit precoder.

[0081] Figure 4AAn exemplary resource 400 is shown, highlighting the CMR and IMR. Resource 400 includes downlink resources allocated for channel measurement, such as resources 405 and 407, and resources allocated for interference measurement, such as resources 410 and 412. Resources allocated for channel measurement are used to deliver NZP-CSI RS, resources allocated for interference measurement are used to deliver ZP-CSI RS, and so on. Figure 4B An exemplary relationship 450 between CMRs and IMRs and resource elements of an orthogonal frequency division multiplexed (OFDM) symbol or physical resource block (PRB) is shown. NZP CMRs 455 and 457 may occupy all resource elements of an OFDM symbol or PRB, or only a subset thereof, as shown in relationship 459. ZP IMRs 465 and 467 may occupy all resource elements of an OFDM symbol or PRB, or only a subset thereof, as shown in relationship 469. CMRs and IMRs may overlap.

[0082] For example, the UE associates a received signal (including a target signal, an interference signal, and noise) with a sequence representing a target signal of a layer to obtain a receive combiner for the layer, wherein the sequence may be generated from a scrambling ID associated with the transmit signal, and the scrambling ID has been configured for a CSI-RS resource or a DMRS resource associated with the UE, etc. For another example, the UE associates a received signal (including a target signal, an interference signal, and noise) with a sequence representing a target signal of a layer, and further determines an SINR indicator for the layer (the indicator represents the combined SINR of each layer, or implicitly represents the SINR indicator or directly calculates an indicator of a function of the SINR indicator of the layer, and the function is used to weight the precoder) to obtain a receive combiner or a transmit precoder for the layer, wherein the SINR indicator is obtained based on the following assumptions: assuming that the target signal is a signal on one or more CMRs and represented by a sequence, assuming that the one or more interference signals are interference signals on one or more IMRs; and the receive combiner or transmit precoder for the layer is weighted by a function of the SINR indicator for the layer. Exemplary CMR resources include NZP CSI-RS resources configured for channel measurement and interference measurement, NZP CSI-RS resources configured for channel measurement, DMRS resources, PDSCH resources, and the like. Exemplary IMR resources include NZP CSI-RS resources configured for channel measurement and interference measurement, NZP CSI-RS resources configured for channel measurement, NZP CSI-RS resources configured for interference measurement, zero-power (ZP) CSI-RS resources configured for interference measurement, CSI interference measurement resources, DMRS resources, a new downlink RS resource or a new RS resource, PDSCH resources, and the like. Access node 305 coordinates with other access nodes (particularly access nodes close to access node 305) to determine CMR and IMR. CMR information or IMR information provides UE 310 with information about the CMR or IMR sent by access node 305, such as time resource information, frequency resource information, or port resource information. The CMR information or IMR information may be indicated to UE 310 via an RRC message, a MAC message, or a DCI message. The CMR information or the IMR information may also be included within the beamformed SRS information sent by the access node 305 in event 319 .

[0083] Access node 305 applies transmit filter v to a downlink signal, such as a downlink RS (step 333), and transmits the precoded downlink signal to UE 310 (event 335). The transmission includes signals transmitted on one or more transmission ports and dedicated resources, where each transmission port corresponds to a transmission layer, and the number of transmission layers of the first transmission corresponds to the rank of the transmission.

[0084] The UE also receives interference on the interference resources. These interference resources may or may not overlap with the channel measurement resources. The UE finds the receive filter w based on the downlink signal transmission and the received interference. In other words, the UE's receive filter w is found based on the received transmission and the interference received on the resources associated with the signal reception and the interference transmission. For example, when the uplink precoder is an analog combiner, the analog beams in the downlink are selected based on the measured downlink signal and interference, and these analog beams can be different from or the same as the beams selected based only on the measured downlink signal.

[0085] The access node 305 sends quasi-co-located (QCL) information to the UE 310 (event 337). The QCL information indicates that there is a relationship between two or more reference signals or data signals (e.g., CMR and IMR) such that the two signals can be considered to have similar properties. For example, in a one-to-one QCL relationship, the receive beam used by the CMR is the same as the receive beam used by the IMR. Multiple signals can also be associated with one signal. The UE can assume that the antenna port corresponding to one signal is quasi-co-located with the antenna port corresponding to another signal in terms of Doppler frequency deviation, Doppler spread, average delay, delay spread, spatial Rx parameters, etc. (if applicable). For example, information related to the quasi-co-location type can be sent to the UE using the high-level parameter QCL-Type. Exemplary types of the parameter QCL-Type may include (it should be noted that the parameter may include a combination of the following exemplary types):

[0086] –QCL-TypeA': {Doppler frequency deviation, Doppler spread, average delay, delay spread},

[0087] –QCL-TypeB': {Doppler frequency deviation, Doppler spread},

[0088] –QCL-TypeC': {average delay, Doppler frequency shift},

[0089] –QCL-TypeD': {spatial Rx parameters}.

[0090] The QCL relationship can be indicated to the UE to inform the UE that the CMR and IMR can use the same receive beam, etc. For example, the QCL information can be indicated in a DCI message. The DCI message may include information about a CMR with a non-periodic RS (e.g., CSI-RS, etc.) and triggering, and an IMR with a non-periodic IMR triggering. Joint or separate DCI messages can be used to provide information on different resources. For example, the UE assumes that one or more CMR resources used for channel measurement (e.g., CSI-RS) and one or more IMR resources (e.g., CSI for interference measurement (CSI-IM) or NZP resources) have a QCL relationship, for example, one or more CMR resources have a spatial QCL relationship Type D (i.e., QCL-TypeD') with one or more IMRs. For another example, the UE uses the QCL relationship between the CMR and the IMR to obtain CQI. The presence of a QCL relationship between the CMR and the IMR indicates that the CMR reception and the IMR reception can use the same receive beam. For another example, no QCL relationship is used. For example, in a UE with full digital processing capabilities, the UE receives both CMR and IMR and uniformly determines the receive beams used for signal and interference reception. In the event that the CMR and IMR overlap, the UE may be configured with only the CMR for the BiT process but with a QCL-TypeD' relationship specified for the CMR. The UE then interprets the CMR and IMR as being configured to overlap.

[0091] UE 310 receives the precoded downlink signal using receive filter w (step 339). The UE does not need to know whether the signal is precoded. The UE processes the signal in a similar manner regardless of precoding. The precoded downlink signal includes CMR (e.g., downlink RS) and may include IMR. When UE 310 receives the precoded downlink signal, UE 310 expects to receive the downlink signal and associated interference. In this case, the QCL relationship includes not only the downlink signal but also the interference. When receiving a unicast downlink data channel according to BiT training (e.g., step 337), a spatial QCL relationship is supported between the joint downlink RS with interference and one or more DMRS antenna ports of the downlink data channel, where information about the one or more RS antenna ports is sent using a DCI message. For example, UE 310 separates the downlink signal (s) and the interference (i) from the received precoded downlink signal (step 341). In one embodiment, UE 310 removes the downlink signal s (the downlink signal is known to UE 310), leaving only the interference i. UE 310 can utilize interference cancellation receiver capabilities, etc. UE 310 uses optimization techniques to update receive filter w (steps 343 and 345). In other words, UE 310 obtains an updated receive filter w. UE 310 may also calculate or determine receive filter w. For example, UE 310 uses optimization techniques to maximize the downlink CQI, the communication system's SINR, channel capacity, error probability, etc. to update receive filter w. Some embodiments (e.g., those based on global utility) have been discussed above, and these embodiments may be combined here. For example, if UE 310 does not use interference cancellation receiver capabilities (or UE 310 does not have interference cancellation capabilities), for example, if UE 310 uses interference cancellation receiver capabilities for data reception, UE 310 uses similar assumptions to determine receive filter w. In this case, UE 310 receives and estimates the downlink signal s and interference i. UE 310 then obtains receive filter w, along with the resulting CQI, SINR, channel capacity, error probability, etc. Receive filter w (or its equivalent) also serves as a transmit filter, also denoted as w. As previously described, two filters are equivalent when they are conjugate transposed versions, conjugate versions, scaled versions, normalized versions, or identical to each other. In a multi-transmission layer deployment, the two filters can be weighted versions of each other, with the weights potentially being different for each transmission layer. For example, UE 310 does not acquire the receive filter w but instead directly acquires the transmit filter w. Directly acquiring the transmit filter w is applicable to BiT iterations that involve only training and not downlink data transmission. In other words, acquiring the receive filter is only required for downlink data transmission and is not required for BiT training.The expected UE behavior specified for the UE by the access node configuration or indication may be simply to derive the transmit filter w based on the associated CMR and optionally the IMR. Other UE behaviors, such as the intermediate steps of deriving a hypothetical receive filter w based on the associated CMR and optionally the IMR and then deriving the equivalent transmit filter, may be done by the UE as part of the UE implementation, but the network configuration or standard specification need not specify these behaviors.

[0092] UE 310 applies transmit filter w to an uplink signal, such as an uplink RS (step 347). The uplink RS can be an SRS, DMRS, or any other RS, among others. In one embodiment, the uplink RS can be uplink data that access node 305 can correctly decode. UE 310 transmits the precoded uplink signal to access node 305 (event 349). This step can be combined with the previous step of obtaining or updating transmit filter w. The expected UE behavior specified for the UE by the access node configuration or indication can include transmitting an uplink RS based on the associated CMR and, optionally, IMR. Access node 305 receives the precoded uplink signal using receive filter v (step 325). The access node does not need to know whether the signal is precoded. Access node 305 separates the uplink signal (s) and interference (i) from the received precoded uplink signal (step 351). Access node 305 uses optimization techniques to update receive filter v (steps 353 and 355). In other words, access node 305 obtains the updated receive filter v. Access node 305 may also calculate or determine the receive filter v. For example, access node 305 may use optimization techniques to optimize the overall downlink performance of the communication system. Other embodiments have been provided above, and these embodiments may be combined here. For example, access node 305 and UE 310 may use the same optimization technique. The second transmission is sent over one or more second transmission ports. For example, the number of first transmission layers in the first transmission and the number of second transmission layers in the second transmission are the same. As previously described, two corresponding filters are equivalent when they are conjugate transposed versions, conjugate versions, scaled versions, normalized versions, or identical to each other. For another example, the number of transmission layers in the second transmission is different from the number of transmission layers in the first transmission. The access node updates the UE using the number of transmission layers in the subsequent transmission. For another example, the access node may not obtain the receive filter v, but instead directly obtain the transmit filter v. This may be applicable to BiT iterations that only involve training and not BiT-based uplink data transmission. In other words, obtaining the receive filter is only required for uplink data transmission and not for BiT training.

[0093] Figure 5Flowchart of exemplary operations 550 performed when a UE supports BiT training. Multiple steps are described herein, each of which may have multiple embodiments. Furthermore, these embodiments may be combined. Operation 550 begins with the UE receiving a first transmission in a downlink (step 555). The first transmission in the downlink may be issued from one or more access nodes. The first transmission may include a first signal on one or more resources associated with signal reception previously configured or indicated to the UE. The first signal may be issued from one or more access nodes serving the UE, and the first signal may include one or more layers. The first signal may be an RS, such as an NZP CSI-RS or DMRS for downlink transmission, or a new or enhanced downlink RS for one or more ports, each port corresponding to a transmission layer. The total number of ports or layers is notified to the UE via control signaling (e.g., CSI-RS configuration signaling, MAC signaling, DCI, or a combination with RRC signaling), MAC signaling, and DCI. The total number of ports or layers is the rank of the first signal. Associated with the first signal may be first interference, i.e., the first downlink transmission including the first signal may be interfered with by other transmissions. The interference can be observed on resources associated with interference reception that were previously configured or indicated to the UE corresponding to the first transmission, and these resources can be the same as or different from the resources associated with signal reception. The association relationship between the first signal and the interference can be configured by RRC signaling, activated by MAC signaling, indicated by DCI signaling, or a default relationship when one or more interference reception resources are the same as one or more signal reception resources. The first interference can be an aggregation of multiple other transmissions, including expected inter-cell interference and intra-cell interference to other UEs. A portion of the interference can be indicated using port information, and the UE assumes that each port corresponds to an interfering transmission layer and takes into account the power factor (energy per resource element, EPRE) ratio configured or indicated to the UE. Other interference signals are not indicated using port information, and the UE assumes that these interference signals are present on all indicated interference resources, and so on.

[0094] The UE acquires a transmit precoder for uplink transmission (step 557). This acquisition is performed based on UE assumptions specified for this process, which may be defined in a standard specification or associated with signaling associated with the first downlink transmission. The UE assumptions include: the UE assumes that the rank of the downlink transmission of the target signal (e.g., a hypothetical PDSCH, or a PDSCH associated with a first transmission known to the UE to occur later) is equal to the first downlink signal, the interference corresponds to the first interference, and the receiver (which may be a hypothetical receiver, a receiver assumed by the UE to acquire the uplink precoder, a receiver to be received later by the associated PDSCH, a receiver to be measured later by the CSI for the associated PDSCH, etc.) has a receive combiner equivalent to the uplink precoder. In other words, the uplink precoder represents the downlink receive combiner (or vice versa), and therefore, the uplink precoded transmission based on the precoder represents the downlink signal or channel quality in the combining direction, the interference conditions in the combining direction, and the UE's ability to process the signal and avoid, mitigate, or cancel the interference (if any). The uplink transmission can then carry information to the access node, enabling the access node to adapt accordingly, such as link adaptation of the MCS, rank, or beamforming. The uplink rank is equal to the downlink rank, with a one-to-one mapping between the uplink and downlink layers. The uplink power allocation can be uneven, for example, associated with the combined downlink SINR of each downlink layer with a receive combiner.

[0095] The UE sends an uplink signal precoded using a precoder and an associated power allocation (step 559). The uplink signal can be an SRS on an SRS resource or one or more SRS resource sets, and can also be a DMRS. The UE can also send the combined downlink SINR of each downlink layer with a receive combiner. In some cases, when specified by the access node, the UE can also perform CSI measurements based on the first transmission and send CSI reporting information when the receiver hypothesis and the CSI hypothesis are the same as those of the uplink signal. Such CSI reporting information can be used later for link adaptation of the associated PDSCH. The UE process ends. However, the access node can instruct the UE to perform these steps continuously, occasionally with a specified period, or based on non-periodic signaling.

[0096] Apparatus and methods for dynamically or semi-statically indicating SRS transmission information are disclosed in commonly assigned U.S. patent application Ser. No. 62 / 800,336, filed Feb. 1, 2019, entitled “Device, Network, and Method for Sounding Reference Signal Transmission and Reception,” which is incorporated herein by reference in its entirety.

[0097] Instructing the UE to transmit SRS information includes instructing the UE to transmit SRS based on the indicated information rather than based on semi-dynamically configured SRS parameters. Thus, these apparatuses and methods adjust the SRS parameters to be used by the UE for SRS transmission. These apparatuses and methods enable flexible SRS transmission based on dynamically or semi-statically indicated SRS parameters. SRS resources transmitted based on these indicated SRS parameters can more accurately and timely reflect channel interference, channel conditions, and precoding, thereby improving channel estimation performance.

[0098] The dynamically indicated information may include information about one or more SRS parameters, such as SRS bandwidth, antenna port, frequency hopping, antennas or carriers switched within the indicated SRS bandwidth, SRS resources to be transmitted, precoding based on downlink signal measurement and interference measurement, SRS transmission power within the indicated SRS bandwidth, related RRC configuration, or any other information used for SRS transmission. In terms of resource mapping and multiplexing, the information about antenna ports may specify the number of antenna ports (number) or a subset of configured antenna ports. For example, the ports mapped to the SRS resources to be transmitted may be indicated or described. The mapping relationship between the dynamically indicated SRS ports and SRS resources (including the mapping relationship between REs and sequences, and the mapping relationship between CSs and comb teeth) may be predefined. The downlink RS port corresponding to the indicated port may also be indicated to the UE to perform measurements to obtain the precoding for transmitting the SRS on the indicated port. In the following description, the dynamically indicated information may also be generally referred to as SRS parameters. In the present invention, "antenna port" and "port" may be used interchangeably.

[0099] Figure 6Flowchart of an exemplary method 600 for SRS communication. In this example, an access node 602 semi-statically configures SRS transmission to a UE 604 and dynamically indicates SRS parameters to the UE 604, and the UE 604 transmits the SRS according to the configuration and indication. In step 612, the access node 602 semi-statically configures SRS transmission to the UE 604. For example, the access node 602 may semi-statically configure a plurality of SRS resource sets associated with different SRS parameter sets to the UE 604, such as each bandwidth part (BWP). These parameters may include SRS bandwidth, number of ports, cyclic shift, transmission comb, or other SRS parameters specified in 3GPP TS 38.214 V15.3.0 (2018-09), the entire contents of which are incorporated herein by reference. For each of the multiple SRS resource sets, the downlink CSI-RS resource (along with the corresponding bandwidth and number of ports) can be configured to have a spatial relationship with the corresponding SRS resource set in order to obtain appropriate precoding at the UE 604 side (e.g., Figure 2 The access node 602 may semi-statically configure one or more SRS resource sets to the UE 604. For example, the access node 602 may semi-statically indicate the SRS parameters associated with each SRS resource set to the UE 604 using RRC signaling. The access node 602 may select and trigger one of the SRS resource sets to be transmitted by the UE 604.

[0100] After step 612, the access node 602 may dynamically indicate one or more SRS parameters to the UE 604 in step 614. For simplicity of explanation, the SRS parameters configured in step 612 may generally be referred to as configured SRS parameters, while the SRS parameters indicated in step 614 may generally be referred to as indicated SRS parameters. Through step 614, the access node 602 further requests the UE 604 to transmit a triggered SRS resource set based on the SRS parameters specifically indicated in step 614. The one or more SRS parameters may include an SRS transmission bandwidth, which may be a subset of the configured SRS bandwidth within the triggered SRS resource set. For example, a set of resource blocks (RBs) or a subset of physical RBs (PRBs) within a set of RBs may have been semi-statically configured via higher layer signaling (e.g., via an SRS bandwidth configuration parameter C_SRS and an SRS bandwidth parameter B_SRS) (e.g., in step 612), and thus may be dynamically indicated to the UE 604 by the access node 602. While the description is primarily based on RBs, the same description can also be applied to subbands, RB groups (RBGs), subchannels, frequency domain bundles, or other types of frequency units. One or more SRS parameters may include antenna ports, which may be a subset of ports in a set of ports that have been semi-dynamically configured (e.g., in step 612) via higher layer signaling. For example, the semi-statically configured antenna ports include ports 1 to 4, and a subset of ports 1 to 4 (e.g., port 1 and port 3) may be dynamically indicated to UE 604, thereby using ports 1 and 3 to transmit SRS.

[0101] In step 614, the access node 602 may dynamically indicate one or more SRS parameters via a DCI message, a medium access control (MAC) control element (CE), or other applicable control message. In one example, a new DCI format dedicated to dynamically or semi-statically indicating SRS parameters and triggering SRS transmission may be defined. The new DCI format may also include a field for CSI-RS triggering. In another example, the DCI format or group DCI format used for PDSCH scheduling or PUSCH scheduling may be modified and used to dynamically indicate SRS parameters. In this case, the resource allocation field of the DCI format may be used to dynamically indicate the SRS bandwidth to be used for SRS transmission. In another example, a new MAC CE may be defined to include one or more SRS parameters to be dynamically indicated.

[0102] In step 616, UE 604 may transmit a triggered SRS resource set based on the semi-dynamic SRS configuration (in step 612) and the dynamic indication (in step 614). One of the SRS resource sets configured in step 612 may be selected and triggered by access node 602 for transmission by UE 604. For example, a DCI message may be used to indicate one or more SRS parameters in step 614 and also trigger UE 604 to transmit the SRS resource set configured in step 612 (i.e., the triggered SRS resource set). In one illustrative example, the triggered SRS resource set is associated with an RB set (including a set of RBs) as the SRS bandwidth and a port set (including a set of ports) as the SRS antenna ports on which the SRS is to be transmitted, and the DCI message transmitted in step 614 includes an RB subset within the RB set and a port subset within the port set. In this case, UE 604 may transmit the triggered SRS resource set based on the RB subset and the port subset. Other SRS parameters (e.g., transmission comb, transmission comb offset, SRS sequence ID, etc.) associated with the triggered SRS resource set and configured in step 612 may still be used by the UE 604 to transmit the triggered SRS resource set if they are not changed or adjusted by the access node 602 through any signaling. In this case, dynamically indicating the SRS parameters can be regarded as adjusting or changing the values ​​of similar SRS parameters that have been semi-dynamically configured before, and the adjusted values ​​are used by the UE 604 for SRS transmission.

[0103] UE 604 can obtain a precoder for transmitting a triggered SRS resource set based on one or more downlink RS resources (i.e., one or more downlink RS signals) configured for channel measurement and interference measurement of an indicated SRS transmission bandwidth. The indicated SRS transmission bandwidth can be a subband of the transmission bandwidth between UE 604 and access node 602, or the entire transmission bandwidth. For example, when a subband precoder can be obtained based on downlink RS resources configured for channel measurement and interference measurement of a subband, subband precoding can be applied to SRS transmission in the subband. For another example, when a wideband precoder can be obtained based on downlink RS resources configured for channel measurement and interference measurement of the transmission bandwidth, wideband precoding can be applied to SRS transmission in the entire transmission bandwidth. The transmission bandwidth between UE 604 and access node 602 can be semi-statically configured by the base station, while the subband can be dynamically indicated to UE 604 by access node 602. UE 604 obtains a precoder for the indicated subband, precodes the SRS using the precoder, and transmits the precoded SRS. When access node 602 does not dynamically indicate a subband, UE 604 acquires a wideband precoder to precode the SRS and transmits the precoded SRS. In either case of using subband precoding or wideband precoding, the number of transmission layers and / or transmission ports may also be indicated by access node 602 to UE 604 via DCI or MAC CE, or configured via RRC. UE 604 may obtain SRS transmission power based on the dynamically indicated SRS transmission bandwidth rather than the semi-statically configured SRS transmission bandwidth.

[0104] The DCI message may be sent before step 614 and after step 612 to trigger the UE 604 to send the SRS resource set (i.e., the triggered SRS resource set). In this case, upon receiving the DCI message, the UE 604 may send the SRS resource set according to the SRS parameters (e.g., RB set and port set) associated with the triggered SRS resource set that the access node 602 has configured in step 612.

[0105] For example, after a UE receives a message that overrides certain SRS configurations configured via previous RRC signaling, the UE uses the new SRS configuration / parameters according to this message from now on until it receives another RRC signaling or another message that further overrides the SRS configuration / parameters. In other words, the configuration / parameters in the message remain applicable until further changes are made. For another example, after a UE receives a message that overrides certain SRS configurations configured via previous RRC signaling, the UE uses the new SRS configuration / parameters for one transmission according to this message, and then returns to the "normal state" of the SRS configuration / parameters according to the original RRC signaling. In other words, the configuration / parameters in the message are only used once. For another example, after a UE receives a message that overrides certain SRS configurations configured via previous RRC signaling, the UE uses the new SRS configuration / parameters for n transmissions or for m time slots according to this message, and then returns to the "normal state" of the SRS configuration / parameters according to the original RRC signaling. Here, the n transmissions or m time slots can be specified by the standard specification, in the RRC configuration signaling, in the MAC signaling, or generally via DCI or a message. For another example, the message includes a bit to indicate whether the new SRS configuration / parameters in the message are to be used only once and then restored, or to be used from now on.

[0106] For example, a new DCI format can be defined that can be specifically used to dynamically indicate SRS parameters and / or trigger SRS transmission. The new DCI format may include fields to indicate the SRS transmission bandwidth (which may include a bandwidth subset that has been configured through higher-layer signaling), the SRS transmission port (which may include a port subset that has been configured through higher-layer signaling), the SRS transmission power, or other SRS parameters that can be dynamically indicated to the UE for SRS transmission. Table 1 below shows an example of a new DCI format, including some of the fields included in the format. Fields for dynamically indicating other SRS parameters such as cyclic shift and SRS sequence ID may also be added.

[0107] Table 1

[0108]

[0109]

[0110] Table 1 shows 4 fields, including "SRS request", "frequency domain resource allocation", "antenna port" and "TPC command". The "SRS request" field is used to trigger the transmission of the SRS resource set. This field can have various lengths, for example, 1 bit or 2 bits. Therefore, different numbers of SRS resource sets can be selected and triggered through this field. The definition of the "frequency domain resource allocation" field can be similar to the definition of the "frequency domain resource allocation" field for PUSCH / PDSCH frequency domain resource allocation in the existing DCI format, as described in 3GPP TS 38.211V15.3.0 (2018-09), the entire content of which is incorporated by reference in this application. However, the frequency domain resources specified by this field need to be within the SRS bandwidth configured for one or more SRS resource sets through high-layer signaling. For example, this field can specify a subset of the SRS bandwidth that has been configured for the SRS resource set through high-layer signaling. The "antenna port" field can be used to specify the antenna port used for SRS transmission (and the downlink RS port for obtaining the SRS precoder). For each SRS port, there can be a downlink RS port configured to the UE to obtain the SRS port precoding. In this case, the downlink RS port is associated with the SRS port and can be indicated in the same field as the SRS port. The "Antenna Port" field can specify a subset of ports semi-statically configured via higher-layer signaling. The "TPC Command" field can specify the transmit power control command according to which the SRS transmission power can be adjusted.

[0111] To distinguish this new DCI format from other DCI formats, a new radio network temporary identifier (RNTI) can be configured for the new DCI format. The new DCI format can have a cyclic redundancy check (CRC) scrambled by the new RNTI. The UE can identify the new DCI format by the new RNTI.

[0112] In one example, this new DCI format can be used to dynamically trigger SRS transmission while indicating the SRS transmission bandwidth and port. In another example, the new DCI format can be used only to semi-statically adjust the SRS transmission bandwidth and port without triggering SRS transmission. In this case, a different DCI format can be used (for example, sent after semi-static adjustment) to dynamically trigger SRS transmission without further adjusting the SRS transmission bandwidth and port. In yet another example, the new DCI format can be used to semi-statically adjust the SRS transmission bandwidth and port while triggering SRS transmission. In this case, a different DCI format can also be used (for example, sent after sending a new DCI format message for semi-static adjustment) to dynamically trigger SRS transmission without adjusting the SRS transmission bandwidth and port, but using the adjusted transmission bandwidth and port.

[0113] For example, the existing DCI format 2_3, used to send a set of TPC commands for SRS transmission by one or more UEs, as described in 3GPP TS 38.212 V15.3.0 (2018-09), or a variant of DCI format 2_3, can be modified to dynamically indicate SRS parameters to the UE. The existing DCI format 2_3 includes an SRS request field and a TPC command field for each block, as described in 3GPP TS 38.212 V15.3.0 (2018-09).

[0114] For example, for each SRS request block in DCI format 2_3, one or more fields may be added to indicate the corresponding SRS transmission bandwidth (including a bandwidth subset configured through higher-layer signaling), SRS transmission port (including a port subset configured through higher-layer signaling), or other dynamically adjustable SRS parameters. Table 2 below shows an example of the modified DCI format 2_3, including a portion of the fields included in this format for an SRS request block.

[0115] Table 2

[0116]

[0117]

[0118] Table 2 shows four fields, including "SRS request", "frequency domain resource allocation", "antenna port" and "TPC command". Table 2 is similar to Table 1. However, the "frequency domain resource allocation" and "antenna port" fields are added to DCI format 2_3. The "SRS request" field is similar to the "SRS request" field in DCI format 2_3TypeB and has the same value, that is, 0 bits or 2 bits. This field is used to trigger the transmission of an SRS resource set. More than 2 bits can be used to trigger the transmission of an SRS resource set selected from a large number of SRS resource sets. The "frequency domain resource allocation" field is added, and the definition of this field can be similar to the definition of the "frequency domain resource allocation" field used for PUSCH / PDSCH frequency domain resource allocation in the existing DCI format, as described in 3GPP TS38.211V15.3.0 (2018-09). However, the frequency domain resource allocation specified by this field needs to be within the bandwidth configured for one or more SRS resource sets through high-layer signaling. The "antenna port" field has been added, which is used to specify the antenna port used for SRS transmission (and the downlink RS port for obtaining the SRS precoder). The "antenna port" field can specify a subset of ports that are semi-statically configured through high-layer signaling. The "TPC command" field is the same as the "TPC command" field in DCI format 2_3TypeB. This field includes a transmit power control command, according to which the transmit power of the SRS resource set used to send the trigger can be adjusted. Table 2 shows that "frequency domain resource allocation" and "antenna port" are two separate fields. In another embodiment, a field can be defined and added in DCI format 2_3 to uniformly specify "frequency domain resource allocation" and "antenna port".

[0119] Apparatus and methods for signaling control information are disclosed in commonly assigned patent application PCT / US19 / 46,898, filed on August 16, 2019, entitled “Methods and Apparatus for Signaling Control Information,” the entire contents of which are incorporated herein by reference. Figure 7 and Figure 8Schematic diagrams 700 and 800 of RGB 705, 707, 805, and 807 illustrate exemplary mapping relationships between SRS resources and ports. Schematic diagram 700 illustrates RGB 705 and 707 in a communication system with the following configuration: assuming DMRS type 1 (eight ports per RGB per cell for all paired UEs); another example may consider a 12-port DMRS, with eight ports associated with eight SRS port resources, which are selected from n available port resources (e.g., n=48 for comb 4 and n=16 for comb 2). SRS from neighboring cells may be multiplexed on n SRS port resources. In order to indicate to the UE whether one, two, or four of the n available SRS port resources are used, more bits are required than the available bits in the DCI message.

[0120] Schematic diagram 800 shows RGB 805 and 807, which are configured similar to Figure 7 For example, UE group CSI-RS or DMRS design is applied to SRS. For each cell, there are only 8 predefined SRS port resources (in Figure 8 The different shaded and patterned blocks in the SRS port resource area of ​​RGB are shown in the DCI message). Then, in the DCI message (e.g., a group DCI message), the UE is allocated a layer or port within the 8 predefined SRS port resources and indicated accordingly. For example, SRS resources are configured for all activated UEs in cell 1 (as shown by the unshaded blocks in the SRS port resource area of ​​RGB), and the SRS resources have the same 8 ports. The group DCI message indicates which of the 8 ports is allocated to a specific UE. For example, the rank [1, 2, 4, 1] is indicated to UE 1, 2, 3, and 4 without indicating a layer index. As another example, DMRS port mapping of resources is used. SRS resources are configured for all RGBs, but the scheduling or group DCI causes different UEs to be scheduled on different RGBs.

[0121] Other designs for group DCI indicating SRS control information are possible. In one embodiment, the UEs identified in the group DCI share a common resource allocation field (Type 0 or Type 1 indication). This may occur in a multi-user multiple input multiple output (MU-MIMO) setting, where UEs can share resource blocks or RBGs. In this case, a pre-configured UE that is not pre-scheduled has a field with the trigger condition set to 0 in the group DCI.

[0122] In another example, the UEs identified in the group DCI have separate fields to indicate the resource allocation fields for each UE. In this case, pre-configured UEs that are not pre-scheduled have a field with the trigger condition set to 0.

[0123] In any of the above examples, the UE identifier can be used to identify the pre-scheduled UE. In this way, only the pre-scheduled UE can decode the DCI. However, the UE will attempt to decode the DCI to check whether the DCI is triggered (pre-scheduled). For example, all UEs that detect the DCI can attempt to decode the DCI.

[0124] In any of the above examples, the DCI includes a combination of the listed fields or a subset of the above fields.

[0125] For another example, a modified DCI format, such as DCI format 0_1 ​​(uplink grant) or DCI format 1_1, can be used to dynamically configure the above SRS parameters and associated downlink PDSCH and / or CSI-RS parameters to a triggered UE (pre-scheduled UE).

[0126] Any 5G NR DCI design can be used to indicate that the above required information is added / modified to the DCI.

[0127] The above discussion is related to narrowband (NB) systems and operations, in which the entire bandwidth of the carrier is used for one BiT process. However, based on the narrowband representation of the subband channels, the systems and operations can be applied to any subband (or some other frequency unit including one or more physical resource blocks, etc.). In this case, one or more BiT processes can be operated on one or more subbands between the carrier bandwidth. Therefore, the discussion of narrowband systems and operations should not be interpreted as limiting the scope of the systems and operations previously presented.

[0128] BiT was developed for time division duplexing (TDD) communication systems to maximize the downlink weighted total rate. However, previous research on BiT has been limited to NB systems. Moreover, most fifth generation (5G) networks operate on carriers with large bandwidths, where the carriers consist of at least hundreds of subcarriers (tones), each with a different MIMO channel matrix and considerable frequency selectivity. Therefore, there is a need to extend BiT to 5G wideband (WB) communication systems.

[0129] In a WB system, multiple subcarriers can be used to form a resource block (RB) or a physical RB (PRB). For LTE and NR, the number of subcarriers is 12. In this case, multiple PRBs can be used to form subbands, RB groups (RBGs), precoding resource block groups (PRGs), frequency blocks, frequency domain units, bandwidth units, etc. for different purposes. One or more of them (for example, subbands, RBGs, PRGs, frequency blocks, frequency domain units, bandwidth units, etc.) form a bandwidth part (BWP) or a carrier, and the bandwidth part or carrier can be associated with a serving cell.

[0130] Aspects of the present invention provide BiT for 5G broadband communication systems. First, a global centralized optimization problem is proposed for the broadband communication system. Then, the (suboptimal) solution is distributed among the access nodes and the UEs, resulting in a broadband BiT solution that iteratively adjusts the communication filters (e.g., transmit and receive filters) of each access node and each UE using only local information. The broadband BiT solution can be viewed as a set of narrowband BiT solutions, each of which is performed based on a suitable narrowband representation of a set of subcarriers with different channels while maintaining the first and second moments of the channel. Simulation study results are provided to evaluate the performance of the BiT solution in a broadband communication system.

[0131] A simple approach to extending BiT to wideband operation is to implement a BiT process for each subcarrier in a wideband communication system. However, because a single wireless carrier (channel) may contain at least hundreds of subcarriers, and an access node or UE may need to support one or more carriers, operating a wideband communication system using a BiT process performed separately for each subcarrier is impractical. Another simple approach to extending BiT to wideband operation is to select a subcarrier within a subband, such as the center subcarrier within the subband, to represent the subband, and then apply narrowband BiT based on the selected subcarrier. Alternatively, one could consider simply averaging the per-subcarrier MIMO channel matrix within the subband. However, analysis and evaluation have shown that these approaches, as well as several other simple extensions to wideband communication systems, can lead to significant performance degradation. A fundamental reason for the performance degradation may be the rapid rotation of multipath phases between subcarriers in the wideband channel. Furthermore, in some extreme cases, the mean of the channel matrix within a subband can be close to zero, making this simple narrowband representation of the subband insignificant. Issues associated with this simple narrowband representation of the wideband channel are discussed in detail below.

[0132] According to an exemplary embodiment, a global centralized optimization problem for a broadband communication system is provided. In one embodiment, the global centralized optimization problem assumes that ideal global channel state information (CSI) is available at a centralized location. Suboptimal approximate solutions are obtained and analyzed. The centralized solution is also distributed between access nodes and UEs, resulting in a distributed solution that relies only on local CSI and local decision making. In one embodiment, the distributed solution exploits TDD channel reciprocity, allowing access nodes to obtain information about the downlink channel and inter-cell / intra-cell interference.

[0133] The following wideband constraints are considered: (1) the subcarriers have distinct channels; and (2) in order for the receiver to receive and process the transmitted signal sufficiently accurately, for example to perform channel estimation for computing receive and transmit filters or for demodulation, the receiver needs to process multiple subcarriers uniformly, combined with certain common channel assumptions (discussed below). To this end, a narrowband representation of the wideband channel is developed.

[0134] Use the following notation: X n×m Used to represent a complex matrix of size n×m, X′ is the Hermitian transpose of X, tr(X) is the trace of X, and if X is orthogonal, then X (i) is the (i,i)th (diagonal) element of X, X (1:k,1:l) is the upper left k×l block. In addition, ‖X‖ F is the Frobeius norm of the matrix X, and |x|2 is the vector 2-norm of the vector x. For vector x, diag(x) n×m is an n×m "diagonal" matrix where the diagonal elements are equal to the diagonal elements in x and all other elements are 0. The general concept of a matrix diagonal or diagonal matrix for square matrices also extends to non-square matrices.

[0135] Figure 9 900 is an exemplary communication system 900, highlighting the symbolic notation used in the discussion of the exemplary embodiments described herein. The communication system 900 includes L access nodes (including AN1 905, AN2 906, ..., AN L 907) and K UEs (including UE1 910, UE2 911...UE K 912), these access points and UEs are paired and carried on a group of T subcarriers (which can be called subbands). Each access point has N antennas and each UE has M antennas. For simplicity, assume that the association relationship between the access node and the UE (BS l k Serving UE k), UE pairing, layer selection (UE k's d k layer) and power allocation (p of UE k's dth layer)k,d ) is predetermined and fixed in the present invention. Only the access node precoder {v k,d} (common to all T subcarriers in the subband and the dth layer of UE k) and UE combiner {w k,d,t} (different for different subcarriers within the subband and the dth layer of UEk) needs to be optimized.

[0136] For a subband, the centralized global optimization problem is formulated as follows, denoted as Problem F1:

[0137]

[0138]

[0139] Among them, α k is the fairness weight of each UE. The SINR of each subcarrier per layer can be expressed as:

[0140]

[0141] Among them, H l,k,t is the downlink MIMO channel of subcarrier t between access node l and UE k, C k,t is the received covariance matrix of subcarrier t at UE k, given by:

[0142]

[0143] In addition, assuming

[0144]

[0145] So

[0146]

[0147] Therefore, the optimized UE receiving combiner corresponding to each subcarrier and each layer obtained according to the solution of the Rayleigh Quotient and / or the generalized eigenvalue problem can be:

[0148]

[0149] Using fractional programming techniques, the terms in the objective function of problem F1 become:

[0150] logβ k,d,t -β k,d,t e k,d,t +1,

[0151] in,

[0152] Lagrange multipliers can be applied to transform problem F1 into the following unconstrained optimization problem, expressed as problem F2:

[0153]

[0154] Among them, μ k,d is the Lagrange multiplier of the access node l to be searched to satisfy the power constraint.

[0155] The following optimal equations can be obtained for all k, d, and t, and can be solved using methods such as coordinate block descent. First, β k,d,t :

[0156]

[0157] In order to determine v k,d , collect v involving k and d fixed k,d Item (note that v k,d In each C κ,t ), the derivation is as follows:

[0158]

[0159] In each In, only is related. Therefore, we only need to take the derivative of the following function:

[0160]

[0161]

[0162] therefore,

[0163]

[0164] in,

[0165]

[0166] The following identity connects the access node-side precoding and the UE-side combining, indicating that for the resulting solution, there is a strong coupling between the access node and the UE:

[0167]

[0168] The global centralized solution can be expressed as:

[0169]

[0170]

[0171]

[0172]

[0173]

[0174] The global centralized solution can be executed iteratively. Although there is no theoretical guarantee of convergence or achieving the global optimum, it is generally observed that the global centralized solution achieves convergence and improves performance compared to other methods.

[0175] According to an exemplary embodiment, the global centralized solution is distributed. If UE k uses precoder w on subcarrier t of layer d k,d,t The access node sends a sounding signal (eg, a sounding reference signal or SRS) and the access node receives the sounding signal and sends a sounding reference signal (eg, a sounding reference signal or SRS) ... k,d,t The associated sounding signal and the covariance matrix from all sounding signals (including serving signal and interference) and all subcarriers are used to estimate v k,d , then the global centralized solution can be distributed and approximated by the communication devices (e.g., access nodes and UEs) in the TDD system. However, in order to satisfy the above broadband constraint (2), the precoder w k,d,t Need to be common to all t (so it is expressed as ) to enable the access point to filter the wideband channel in the time domain (e.g., based on the channel power delay profile or the channel impulse response) while remaining close enough to the combiner w k,d,t , or according to a Determined v k,d and according to all w k,d,t Definitely close enough. In addition, even if Shared, each subcarrier corresponds to a channel are also different. In order to Get the shared v k,d ,One method is to obtain a common channel matrix for all ,subcarriers, and the common channel matrix should be close to the channel matrix of ,the subcarriers in a certain sense.

[0176] Therefore, the following problem needs to be solved: Assume a set of T matrices H1,…,H t ,…,H T ∈C n×m , solve H∈C n×m To best approximate this set of matrices. This problem is called solving a narrowband representation of a set of wideband channels.

[0177] The channel coefficients can be simply averaged. A simple way to determine the narrowband representation H of a set of wideband channels is to know a set of T matrices H1,…,H t ,…,H T ∈C n×m , where n≥m, then H∈Cn×m It can be expressed as:

[0178]

[0179] Simple calculations show that the optimal solution is However, in some cases, the simple average of the channel coefficients does not represent H well. t For example, if the channel coefficients are uniformly distributed on a circle around the origin, the simple mean becomes 0. An example of this situation is Figure 10A As shown, Figure 10A 1000 is a schematic diagram of a set of T matrix channel coefficients and a narrowband representation H. Figure 10A As shown, the channel coefficients of this set of T matrices (e.g., coefficients 1005 and 1006) are uniformly distributed on a circle centered at the origin. The simple average of the channel coefficients produces a magnitude equal to 0. Figure 10A , which is also located at the origin. Therefore, a narrowband representation of a channel with a value of 0, or a size / power / energy equal to 0, cannot be effectively used for BiT operation.

[0180] However, since each w k,d,t According to the corresponding is set, so the effective channels on the subcarriers are constructively added, so is not an ill-posed problem. However, is an ill-posed problem. Therefore, we need to find a different representation of the sum of the channel matrices.

[0181] According to one embodiment, one way to prevent the problem of size equal to 0 due to summation is to select one subcarrier in the subband as the narrowband representation of the wideband channel and perform the narrowband BiT algorithm on the representative subcarrier. The narrowband representation of the wideband channel is called H NB . For example, a simple technique for selecting subcarriers is to select the center subcarrier as the narrowband representation. This simple technique does not require any additional processing or any overhead. However, this approach can cause problems if the representative subcarrier happens to encounter deep fading, unusually strong or weak interference, etc., which can cause the representative subcarrier to be substantially different from the other subcarriers in the subband. One can try to select the best subcarrier as the representative subcarrier, but the search is usually an exhaustive search, etc. In general, relying on only one subcarrier instead of multiple subcarriers will lack robustness. Therefore, it is necessary to find the best representation based on the channels of all available subcarriers in the subband.

[0182] According to one embodiment, in order to avoid the problem of the size being equal to 0 discussed above, it is beneficial to maintain the power / energy of the wideband channel in a narrowband representation. That is, the second-order moment of the narrowband representation can be close to the second-order moment of the wideband channel, and "close" can be measured according to certain matrix norms of the difference between the narrowband representation and the wideband channel. In other words, the difference between the second-order moments should be less than the power threshold, or in a certain sense, the second-order moment difference should be minimized (optimized). For the matrix H, the second-order moment can be expressed as HH' or H'H. Therefore, it is necessary to calculate the second-order moment error of the wideband channel in any way.

[0183] Of course, restricting only the second-order moments does not guarantee that the first-order moments are uniquely determined, nor does it guarantee that the first-order moments are close to the first-order moments of the wideband channel. Therefore, the multi-step process involves first solving for a set of H that satisfies the second-order moment optimality, and then applying the first-order moment conditions to find an acceptable narrowband representation that is optimal in terms of both the first-order and second-order moments. In other words, the difference between the phase of the narrowband representation and the phase of the wideband channel must be less than a phase threshold or, in some sense, minimized (optimized). The first-order moments are related to the phase of the wideband and narrowband channels and can be referred to as phase domain components, phase domain quantities, first-order moments of the channels, or phase domain approximations. The second-order moments are related to the power (or energy, magnitude, or amplitude) of the wideband and narrowband channels and can be referred to as power (or energy, magnitude, or amplitude) domain components, power (or energy, magnitude, or amplitude) domain quantities, first-order moments of the channels, or power (or energy, magnitude, or amplitude) domain approximations.

[0184] The power threshold / optimization and / or phase threshold / optimization may be specified by a technical standard or an operator of a broadband communication system. In addition, historical performance information may be used to adjust the power threshold and / or phase threshold to help the communication system meet performance standards, such as data rate, quality of service (QoS) limits, etc.

[0185] Finding a narrowband representation that preserves the power / energy of the wideband channel is called Problem F3. Problem F3 is also known as the second-order moment optimality condition, which is related to the power of the wideband channel and the narrowband channel. Problem F3 can be expressed as: Given a set of T matrices H1,…,H t ,…,H T ∈C n×m , where n≥m, solve H∈C according to the following expression n×m :

[0186]

[0187] in,

[0188]

[0189] Assume that without loss of generality, the eigenvalue decomposition (EVD) of L and R is possible: L = US L U′, R=VS R V′, where U is an n×n unitary matrix, V is an m×m unitary matrix, and S L and S R The eigenvalues ​​in are sorted in descending order. Assume that S s =[S,0 m×(n-m) ]′, where S is a diagonal line, such that

[0190] Proposition 1: The solution to problem F3 can be expressed as:

[0191]

[0192] Proof of Proposition 1:

[0193] Lemma 1 (Van Newman's trace inequality): Given A and B in, assuming a1,…,a n and b1,…,b n are the singular values ​​of A and B, respectively, and are sorted in descending order.

[0194]

[0195] If and only if there exists a unitary matrix U AB and V AB When A=U AB diag([a1,…,a n ]) n×m V′ AB and B=U AB diag([b1,…,b n ]) n×m V′ AB , the inequality holds, where diag([a1,…,a n ]) n×m Indicates setting the diagonal matrix to n×m.

[0196] Lemma 2: Given A and B in, assuming a1,…,a n and b1,…,b n are the singular values ​​of A and B, respectively, and are sorted in descending order.

[0197]

[0198] If and only if there exists a unitary matrix U AB and V ABWhen A=U AB diag([a1,…,a n ]) n×m V′ AB and B=U AB diag([b1,…,b n ]) n×m V′ AB , the inequality holds.

[0199] Assume that the following singular value decomposition (SVD) of H is possible: H = U H S H V′ H , where U H is an n×n unitary matrix, V H is an m×m unitary matrix, S H The singular values ​​in are sorted from large to small. Because n≥m, At this time,

[0200]

[0201]

[0202] also,

[0203]

[0204]

[0205] In the above, the first inequality follows from Lemma 2 and holds if and only if for some diagonal matrix D with only phase rotations, U H =U and DV H =V(Note that U H , U, D, V H and V are not necessarily unique), the inequality holds; the second equality is generated by the parallelogram identity; in the last inequality, if and only if for all i there exists This inequality holds. Therefore, is the solution set of problem F3.

[0206] That is, a set of optimized H is determined by the eigenspaces of L and R, the means of the eigenvalues ​​of L and R, and any phase rotation captured in the diagonal matrix D only includes the complex phase.

[0207] Problem F3 is equivalent to Problem F3' and can be expressed as:

[0208]

[0209] This equivalence is interesting because it indicates that the same narrowband representation can be obtained using two different techniques. For Problem F3, the technique involves minimizing the squared error with respect to the mean of the per-subcarrier second-order moments. For Problem F3', the technique minimizes the mean of the squared error with respect to the per-subcarrier second-order moments. It turns out that the two techniques are equivalent.

[0210] To prove this, rewrite Problem F3' as

[0211]

[0212] Expand and remove constant items and The problem becomes minimizing

[0213] Ttr(-LHH′-HH′L+HH′HH′)+Ttr(-RH′HH′HR+H′HH′H)

[0214] =Ttr(-2H'LH+H'HH'H)+Ttr(-2HRH'+HH'HH').

[0215] On the other hand, the goal in problem F3 can be rewritten as

[0216] tr(L-HH′)′(L-HH′)+tr(RH′H)′(RH′H).

[0217] Expanding the above expression for Problem F3 and removing the constant term, the quantity to be minimized is the same as for Problem F3', except for the constant scaling factor T.

[0218] In addition, D should be solved while satisfying the first-order moment optimality criterion, which is called problem F4 and can be expressed as:

[0219]

[0220] Proposition 2: If and the (i,i)th element of X has phase θ i , then the solution to problem F4 is given by given.

[0221] The proof of Proposition 2 is as follows: The goal is equivalent to:

[0222]

[0223] Because tr(H′ t H t +S 2 ) is a constant, so the optimization problem reduces to

[0224]

[0225] Since D is diagonal, it is easy to verify that the optimal solution is to compensate the phase of the diagonal elements of X so that all elements add constructively. That is, if the (i,i)th element of X is Among them, r i >0, then the (i,i)th element of D is

[0226] Numerical experiments show that the above technique can solve the problem of size equal to 0 because the second-order moments are preserved (omitted for brevity). An example of a wideband channel and the optimal narrowband representation is shown in Figure 10B As shown in Figure 2. Both the optimal second-order moment condition and the optimal first-order moment approximation under the second-order moment constraint discussed above require the generation of a suitable representation. If only the phase is chosen correctly, the obtained channel coefficients will not be a good representation of the wideband channel.

[0227] Figure 10B 1050 is a diagram of a set of T matrices of channel coefficients, highlighting the narrowband representation. Figure 10B As shown, the channel coefficients for the set of T matrices may include coefficients 1055 and 1056. Point 1060 represents a narrowband representation based solely on first-order moment optimality, an example of which is a simple mean. Point 1065 represents a narrowband representation based solely on second-order moment optimality. Point 1070 represents a narrowband representation based on both first-order moment optimality and second-order moment optimality. Point 1070 may be the optimal narrowband representation for the set of T matrices.

[0228] According to an exemplary embodiment, a narrowband representation of a wideband channel is provided to a BiT process that adapts to the communication filters (i.e., transmit filters (transmit precoders, transmit beamformers, etc.) and receive filters (receive combiners, receive beamformers, etc.)) in the communication system. In one embodiment, the BiT process can be a one-time process in which an uplink training step and a downlink training step occur, or it can be an iterative process in which uplink training and downlink training occur alternately. In one embodiment, on each subband, each transmitter (e.g., UE or access node) uses the same precoder on all subcarriers in the subband, so that the receiver can perform joint channel estimation on the subcarriers. The transmit precoder is obtained from the subcarriers of the subband in the previous training step. However, the receiver can use different receive combiners on different subcarriers.

[0229] The wideband BiT process can be viewed as a set of narrowband BiT processes, each of which is performed on a suitable narrowband representation of a set of subcarriers. Obviously, finding a suitable representation is the key to solving the wideband problem.

[0230] In one embodiment, a distributed broadband BiT process is provided that leverages network reciprocity to enable access nodes to obtain information about serving signals / channels, inter-cell interference, and intra-cell interference. The distributed broadband BiT process uses cooperative uplink sounding to mitigate inter-cell and intra-cell interference while improving downlink SINR and overall rate.

[0231] Although the discussion of the exemplary embodiments provided herein focuses on narrowband representations of wideband channels, the exemplary embodiments can be used for any representation of wideband channels. Therefore, the discussion of narrowband representations should not be interpreted as limiting the scope of the exemplary embodiments.

[0232] Figure 11 1 is a flow chart of example operations 1100 performed by a UE when performing a wideband BiT procedure. Operations 1100 may represent operations performed on a device when performing a wideband BiT procedure to determine a communication filter for a wideband communication system. The device may be a network entity, such as an access node, or some other device in a wideband communication system, such as a dedicated entity for performing a wideband BiT procedure.

[0233] The device can communicate with another device. For example, the device can be an access node that receives information from a UE. In another example, the device can be a UE that receives information from an access node. The device can be referred to as a receiving device, and the device that receives information from the receiving device can be referred to as a transmitting device. The device (receiving device) can have multiple receiving antennas (also referred to as antenna ports). The transmitting device can have one or more transmitting antennas, but the number of transmitting antennas in the transmitting device is less than or equal to the number of receiving antennas in the transmitting device.

[0234] A transmitting device may have one or more antenna elements, but the number of ports virtualized and seen by a receiving device may be small. For example, an access node may have a panel of tens to hundreds of antenna elements, but the number of ports used for RS transmission is typically limited to 16 or 32 for CSI-RS or 8 or 12 for downlink DMRS. These ports can be associated with a transmission layer in a MIMO system. For example, for downlink DMRS and a specific downlink CSI-RS used for pre-scheduling / sounding, each port is associated with a transmission layer for the associated PDSCH. By transmitting RS on 8 ports in a transmitting device (e.g., an access node), a receiving device (e.g., a UE) can receive using 4 ports, and the associated channel matrix size is 8×4. The receiving device can perform a channel estimation on each RS subcarrier and obtain a (separate) channel representation, namely an 8×4 channel matrix. The receiving device can also obtain channel estimates based on certain filtering methods and obtain (individual) channel representations for other non-RS subcarriers. The receiving device can then derive a single channel representation based on multiple channel representations. The receiving device can then begin transmitting and the transmitting device can begin receiving. The above description mostly applies.

[0235] In one embodiment, the UE transmits SRS using 4 ports equal to its receive ports, while the access node receives using all 8 of its ports, or even more ports when ports are virtualized (e.g., 32 ports), in which case the access node obtains a (respective) channel representation of size 4×32. In another embodiment, due to limited UE uplink MIMO capabilities, the UE transmits SRS using 1 port less than its receive ports, while the access node receives using all 32 of its ports, in which case the access node obtains a channel representation of size 1×32. If the number of ports (layers) is 2, the next transmission can be a transmission from the access node to the UE via MIMO of size 2×4, and so on. For each reception, the receiving device generates a channel representation of the corresponding size from the (respective) representation and determines one or more communication filters based on the channel representation (and possibly other quantities, such as interference measurements, the total receive covariance matrix, based on the BiT algorithm, etc.). The communication filters can be receive filters (combiners, receive beamformers) so that the receiving device can use the filters to receive other (non-RS) transmissions. Even though each subcarrier can have a receive filter, it may be more practical to have each receive filter cover several subcarriers. The receive filter size is consistent with the receive port size. However, for the next transmission, a transmit filter (precoder, transmit beamformer) is typically derived based on the channel representation (and possibly other quantities). The transmit filter size is consistent with the transmit port size, which may be different from the receive filter size.

[0236] In one embodiment, the broadband BiT process can be a centralized broadband BiT process or a distributed broadband BiT process. In a centralized broadband BiT process, one device performs the broadband BiT process, including determining the communication filter. The centralized broadband BiT process can occur within the device, as described below. In a distributed broadband BiT process, multiple devices perform the broadband BiT process, where one or more devices initiate the broadband BiT process and share initial communication filter values. The multiple devices then perform localized BiT processing to optimize the communication filter. In either the centralized or distributed broadband BiT process, the BiT process can be performed once or iteratively.

[0237] Operation 1100 begins by the device obtaining a channel representation of a wideband channel (step 1105). The channel representation can be used for a set of channels between the device and another device. The device can obtain the channel representation of the wideband channel of the wideband communication system by maintaining the power and energy of the wideband channel. For example, the channel representation of the wideband channel is selected so that the difference between the second-order moment (power) of the channel representation and the second-order moment of the wideband representation is less than a power threshold or as small as possible. In addition, the channel representation of the wideband channel is selected so that the difference between the first-order moment (phase) of the channel representation and the first-order moment of the wideband representation is less than a phase threshold or as small as possible. For example, the channel representation can be a narrowband representation.

[0238] In an implementation of a distributed BiT process, devices share (or indicate, send, etc.) information about a channel representation of a wideband channel (step 1107). The channel representation of the wideband channel can be sent in one or more messages to the communication device executing the distributed BiT process. For example, information about the channel representation of the wideband channel is provided to an access node, and the access node sends a broadcast message including the channel representation of the wideband channel. For another example, the information about the channel representation of the wideband channel is sent in a multicast message. For another example, the information about the channel representation of the wideband channel is sent in multiple unicast messages. For another example, the device that obtains the channel representation of the wideband channel sends one or more messages including information about the channel representation of the wideband channel.

[0239] The device determines the communication filter (step 1109). For example, the communication filter is determined by BiT. The device determines the communication filter in the following manner: starting from the channel representation of the wideband channel, and determining / calculating / optimizing the communication filter based on the signals received from other communication devices. For example, if the device is an access node, the access node uses the sounding signal (e.g., sounding reference signal) sent by the UE to optimize the communication filter. For another example, if the device is a UE, the UE uses the downlink signal (e.g., reference signal) sent by the access node to optimize the communication filter. If the BiT process is an iterative process, multiple optimization iterations can be performed. For example, the optimization of the communication filter can continue until a performance threshold is met. Alternatively, the optimization of the communication filter can continue until a specified number of iterations are performed.

[0240] In an implementation of a centralized BiT process, the device may use signals received from other devices in the communication system to determine and / or optimize the communication filter. After determining the communication filter, the device may share (or indicate, transmit, etc.) information about the communication filter with other devices in the communication system. For example, the communication filter may be transmitted in one or more messages. For example, these messages may be broadcast messages, multicast messages, or unicast messages. Information about the communication filter may include vectors, matrices, quantized forms of vectors / matrices, codewords / indexes in a codebook, differential vectors / matrices / indexes related to a previous quantity, other forms of information sufficient for a receiver to obtain the communication filter, and the like.

[0241] The device uses a communication filter to communicate (step 1111). For example, the device applies a receive filter to an incoming message to receive the message. For example, applying a receive filter to an incoming message includes using the receive filter as a receive combiner to receive the message. To illustrate the use of a receive filter to receive a message, consider the following case: the size of each channel representation and the obtained channel is n×p, where n is the number of receive ports on the receiving device and p is the number of transmit ports / layers / streams of the RS / message to be received. The receive combiner can be a p×n matrix on each receive subcarrier and multiplied by the received signal on each subcarrier. In other words, each transmission layer is multiplied by a vector of length n (combination vector) on n receive ports. By doing so, p transmission layers of the message can be received. For another example, the device applies a transmission filter to an outgoing message before sending the message. For example, applying a transmission filter to an outgoing message includes applying the transmission filter as a precoder (beamformer) to the message. To illustrate the use of transmit filters to transmit messages, consider the following case: the size of each channel representation and the resulting channel representation is n×p, where n is the number of receive ports on the receiving device, p is the number of transmit ports / layers / streams receiving the RS / message, and the number of physical transmit ports in the device is m, which can be equal to or less than n, and the number of virtual transmit ports (or layers, streams, etc.), which can be equal to or less than m. The precoder can be an m×q matrix across the set of transmit subcarriers in the subband, multiplied across the q layers / ports of the message. In other words, each transmission layer is multiplied by a vector of length m (the precoding vector) across the m physical transmit ports, resulting in a q-layer (or q virtual port) message for transmission.

[0242] Figure 12A FIG1 is a schematic diagram of example operations 1200 performed by a device when acquiring a channel representation of a wideband channel. Operations 1200 may be representative of operations performed by a device when acquiring a channel representation of a wideband channel.

[0243] Operation 1200 begins with the device estimating a wideband channel (step 1205). The wideband channel may be estimated based on a signal received on the wideband channel. In one embodiment, an estimated value is determined for each wideband channel. The device determines a channel representation for the wideband channel (step 1207). For example, the channel representation for the wideband channel may be a narrowband representation. For example, the channel representation for the wideband channel may be determined based on the estimated value for the wideband channel. The channel representation for the wideband channel may be determined using the (centralized or distributed) wideband BiT process described above.

[0244] Figure 12BDetailed view of example operation 1250 performed when a device acquires a channel representation of a wideband channel. Operation 1250 may represent operations performed on the device side when the device acquires a channel representation of a wideband channel. For example, a narrowband representation is an example of a channel representation of a wideband channel.

[0245] Operation 1250 begins with the device estimating a wideband channel (step 1255). The wideband channel may be estimated based on a signal received on the wideband channel. In one embodiment, an estimate is determined for each wideband channel. The wideband channel may be estimated based on a reference signal sent from an access node to a UE, in which case the reference signal may be a channel state information reference signal (CSI-RS). The wideband channel may be estimated based on a reference signal sent from a UE to an access node, in which case the reference signal may be a sounding reference signal (SRS). The reference signal may not appear in every subcarrier in a subband, for example, the reference signal may be a comb structure and appear once every 4 (or some other number) subcarriers. On subcarriers where there is no reference signal, the receiver may perform interpolation / extrapolation of the estimate. H1 to H T It can be a channel estimate value on only the subcarrier carrying the pilot or on all subcarriers in the subband. For example, if a subband includes 4 PRBs and each PRB includes 12 subcarriers, the subband includes a group of 48 subcarriers; a 2-port CSI-RS (or SRS, DMRS, etc.) can be sent once every 4 subcarriers, so 12 channel matrices can be obtained from the channel estimate without additional interpolation, and then a channel representation of all 48 subcarriers of the subband can be obtained from the 12 channel matrices, or alternatively, more than 12 (for example, 48) channel matrices can be obtained from the channel estimate with additional interpolation, and then a channel representation of all 48 subcarriers of the subband can be obtained from the more than 12 channel matrices. The reference signal can be beamformed using one or more ports, each port being associated with a transmission layer.

[0246] The device determines the left and right second-order moments L and R (step 1257). As previously mentioned, the left and right second-order moments L and R can be expressed as:

[0247]

[0248] The device determines the EVD of L and R (step 1259). The EVD of L and R can be expressed as:

[0249] L=US L U′ and R = VS R V′,

[0250] Where U is an n×n unitary matrix and V is an m×m unitary matrix. The device determines S (step 1261). The device determines S so that the relationship The device initializes S S , X and D (step 1263). For example, the device initializes

[0251] S S =[S,0 m×(n-m) ]′,

[0252]

[0253] D is diagonal, where the elements

[0254] The device generates a channel representation of the wideband channel (step 1265). The channel representation of the wideband channel is generated based on U, S S , D and V', etc., and can be expressed as:

[0255] H NB =US S DV′.

[0256] Steps 1257 to 1265 may be collectively referred to as determining a channel representation of the wideband channel (as shown in step 1267 ).

[0257] Figure 13 Diagram 1300 illustrating exchanging messages and performing processes for multiple devices performing an iterative distributed broadband BiT procedure. Diagram 1300 shows an access node 1305 and a UE 1310 exchanging messages and performing processes when the devices perform an iterative distributed broadband BiT procedure.

[0258] Access node 1305 configures measurement resources (event 1315). The measurement resources may be signal measurement resources or channel measurement resources (CMR) and are configured by access node 1305. Examples of CMRs may include non-zero power (NZP) CSI-RS resources, NZP CSI-RS resource sets, NZP CSI-RS resource settings / configurations, synchronization signal blocks (SSBs), etc. Access node 1305 sends information about the measurement resources to UE 1310. Access node 1305 may configure interference measurement resources (IMR) (event 1317). The interference measurement resources may be used by UE 1310 to measure interference caused by transmissions from other devices. Examples of IMR may include zero power (ZP) CSI-RS resources, CSI-interference-measurement (CSI-IM) resources, CSI-IM resource sets, CSI-IM resource settings, NZP CSI-RS resources, NZP CSI-RS resource sets, NZP CSI-RS resource settings / configurations, SSBs, etc. IMR and CMR have a quasi-co-location (QCL) relationship. Access node 1305 sends information about interference measurement resources to UE 1310. Configuration of interference measurement resources may be optional. Access node 1305 configures one or more SRS resources or one or more SRS resource sets (event 1319). Access node 1305 also configures resources that UE 1310 may use to send SRS to enable access node 1305 (and possibly other devices) to perform channel measurements. Access node 1305 sends information about SRS resources to UE 1310. Access node 1305 transmits on signal measurement resources or channel measurement resources and interference measurement resources (if configured) (event 1321). Access node 1305 may transmit a reference signal or some other signal known to UE 1310 on the signal measurement resources, channel measurement resources, or interference measurement resources.

[0259] UE 1310 estimates the wideband channel based on the signal received on the signal measurement resource or the channel measurement resource (step 1323). If the interference measurement resource is configured, the UE 1310 may estimate the interference based on the signal received on the interference measurement resource. UE 1310 determines the channel representation of the wideband channel (step 1325). For example, the UE 1310 may determine the narrowband representation of the wideband channel. The UE 1310 may determine the channel representation of the wideband channel using the techniques discussed above. UE 1310 determines the SRS precoder (step 1327). The SRS precoder (communication filter) may be determined based on the channel (e.g., narrowband) representation of the wideband channel. The SRS may have one or more ports (i.e., rank 1 or higher), each port being associated with a transmission layer. For example, the SRS precoder may be associated with an H NB Alignment. For example, the SRS precoder can be H NB One or more main singular vectors, where the number of vectors is equal to the number of SRS ports. For another example, the SRS precoder can be R –1 H NB , where R is the covariance matrix obtained from the IMR including interference and noise, or R is the covariance matrix obtained from the CMR and IMR including signal, interference and noise. The SRS precoder can be R –1 H NB One or more primary singular vectors of . For example, UE 1310 may use a distributed BiT process. UE 1310 transmits an SRS (event 1329). For example, UE 1310 precodes the SRS using an SRS precoder. The SRS is transmitted in the SRS resource (as configured in event 1319).

[0260] The access node 1305 estimates the wideband channel based on the SRS received on the SRS resource (step 1331). The access node 1305 determines the channel representation of the wideband channel (step 1333). The access node 1305 can determine the channel (e.g., narrowband) representation of the wideband channel by the techniques discussed above. The access node 1305 determines the CSI-RS precoder (step 1335). The CSI-RS precoder (communication filter) can be determined based on the channel representation of the wideband channel. The CSI-RS can have one or more ports (i.e., rank 1 or higher), each port being associated with a transmission layer. For example, the CSI-RS precoder can be associated with H' NB Alignment. For another example, the SRS precoder can be H' NB One or more main singular vectors, where the number of vectors is equal to the number of CSI-RS ports. For another example, the CSI-RS precoder can be R –1 H' NB, where R is the uplink covariance matrix obtained on the sounding resources including interference and noise, or R is the covariance matrix obtained on the sounding resources including signal, interference and noise. The CSI-RS precoder can be R –1 H' NB

[00106] One or more primary singular vectors of the signal. For example, access node 1305 may use a distributed BiT process. Access node 1305 transmits a CSI-RS (event 1329). The CSI-RS is transmitted on a signal or channel measurement resource. For example, access node 1305 precodes the CSI-RS using a CSI-RS precoder.

[0261] Events and steps 1321 to 1337 may be iterated multiple times (illustrated simply as event 1339) until a convergence criterion is met or the number of iterations is met. If the access node 1305 and the UE 1310 use a one-time distributed BiT process instead of an iterative distributed BiT process, the access node 1305 and the UE 1310 do not iterate (as indicated by event 1339). In either case, a physical downlink shared channel (PDSCH) transmission may be performed using a precoder (e.g., a precoder acquired by the access node 1305), and the UE may receive the PDSCH.

[0262] In one embodiment, the subcarriers shared by the access node and the UE are used as channel representations of the wideband channel. After determining the channel representation of the wideband channel, the subcarrier closest to the channel representation is selected and sent to indicate the channel representation. "Closest" can mean that the channel matrix of the subcarrier and the matrix H are the same in terms of a certain matrix norm (e.g., Frobeius norm). NB The difference between the two quantities is minimal across all subcarriers. For example, the distance (or a metric derived from a norm, etc.) between a first quantity A and a second quantity B can be defined as ||AB||, i.e., the norm difference between the two quantities. Based on this definition of distance, the proximity of two quantities can be described (e.g., A is closest to B, and so on). Both the access node and the UE use this subcarrier as a channel representation. Since both the access node and the UE know the subcarrier, a simple indicator of the subcarrier is the subcarrier index.

[0263] Figure 14A Schematic diagram 1400 of exchanging messages and performing processes for multiple devices that share a subcarrier index and use the subcarrier index as a channel representation for a wideband channel, wherein an access node determines the channel representation. Schematic diagram 1400 shows an access node 1405 and a UE 1410 exchanging messages and performing processes when multiple devices share a subcarrier index and use the subcarrier index as a channel representation for a wideband channel, wherein the access node 1405 determines the channel representation.

[0264] UE 1410 transmits an SRS (event 1415). The SRS may be precoded using an SRS precoder and transmitted on the SRS resources configured by access node 1405. The configuration of measurement resources, SRS resources, etc. is not performed in Figure 14A However, resource allocation can continue as Figure 13 As shown in events 1315 to 1319.

[0265] The access node 1405 estimates the wideband channel based on the SRS received on the SRS resource (step 1417). The access node 1405 determines a channel representation of the wideband channel (step 1419). The access node 1405 can determine the channel representation of the wideband channel using the techniques discussed above. The access node 1405 determines subcarrier J such that the channel H J closest to the channel representation (step 1421). This determination can be performed by solving the following optimization problem: J = argmin || H t –H NB ||, where t=1, 2, ... T. Access node 1405 sends the index corresponding to subcarrier J (event 1423). UE 1410 estimates the channel H on subcarrier J J , and use H J As a channel representation (step 1425).

[0266] The BiT process (e.g., iterative distributed BiT process, one-shot distributed BiT process, etc.) may occur after determining the channel representation of the wideband channel, but this is not discussed in detail. Figure 14A Shown in.

[0267] Figure 14B Diagram 1450 illustrates exchanging messages and performing processing for multiple devices that share a subcarrier index and use the subcarrier index as a channel representation for a wideband channel, wherein the UE determines the channel representation. Diagram 1450 shows an access node 1455 and a UE 1460 exchanging messages and performing processing when multiple devices share a subcarrier index and use the subcarrier index as a channel representation for a wideband channel, wherein the UE 1460 determines the channel representation.

[0268] Access node 1455 transmits CSI-RS (event 1465). CSI-RS may be precoded using a CSI-RS precoder and transmitted on measurement resources configured by access node 1455. Configuration of measurement resources, CSI-RS resources, etc. is not performed in Figure 14B However, resource allocation can continue as Figure 13 As shown in events 1315 to 1319.

[0269] UE 1460 estimates the wideband channel based on the CSI-RS received on the measurement resource (step 1467). UE 1460 determines a channel representation of the wideband channel (step 1469). UE 1460 may determine the channel representation of the wideband channel using the techniques discussed above. UE 1460 determines subcarrier J such that channel H J closest to the channel representation (step 1471). This determination can be performed by solving the following optimization problem: J = argmin || H t –H NB ||, where t=1, 2, ... T. UE 1460 sends the index corresponding to subcarrier J (event 1473). Access node 1455 estimates the channel H on subcarrier J. J , and use H J As a channel representation (step 1475).

[0270] The BiT process (e.g., iterative distributed BiT process, one-shot distributed BiT process, etc.) may occur after determining the channel representation of the wideband channel, but this is not discussed in detail. Figure 14B Shown in.

[0271] In an example embodiment, in an iterative distributed BiT process (e.g., in wireless-to-everything deployments, fixed wireless access deployments, etc.), the access node and the UE must have the same channel representation. Therefore, the access node and the UE can use the same standardized BiT process algorithm to determine the channel representation to optimize performance.

[0272] In an example embodiment, during an iterative distributed BiT process (e.g., in wireless-to-everything deployments, fixed wireless access deployments, etc.), the access node and the UE must have the same channel representation. Therefore, the access node and the UE can use the same standardized BiT processing algorithm to determine the common subcarriers that are closest to the narrowband representation to optimize performance.

[0273] In an example embodiment, in a non-iterative distributed BiT procedure, such as a one-shot BiT procedure, the UE determines the channel representation using a standardized BiT procedure algorithm.

[0274] In an example embodiment, in a non-iterative distributed BiT process, such as a one-shot BiT process, there is no standardized BiT process algorithm, but there are standardized UE assumptions regarding transmit precoding over wideband channels. For example, the UE derives the uplink transmit precoder for each layer based on a representation of the downlink wideband channel matrix for each layer, where the representation preserves the first and second moments of the wideband channel matrix (or the representation is derived based on the channel matrix while preserving the second moments of the channel matrix).

[0275] In an example embodiment, a channel representation of the downlink wideband channel matrix (Nt×Nr×Ntones) is provided in a compressed feedback report (Nt×Nr) and is obtained based on standardized UE assumptions. For example, in an FDD system, the access node sends CSI-RS on the N subcarriers of a subband, and on each subcarrier, the size of the MIMO channel matrix is ​​Nt×Nr, and the channel matrix for each subcarrier may be different. The UE performs channel estimation on these subcarriers and generates a CSI report. The CSI report typically includes a channel quality indicator (CQI), a precoding matrix indication (PMI), a singular vector, etc. The information included in the CSI report is quite limited. However, in order for the access node to provide higher spectral efficiency through MU pairing and other means, it is desired that the access node have complete downlink channel information, such as channel matrix information. Since each subcarrier has a different channel matrix, this is generally difficult to achieve in an FDD WB system. In this case, the UE can feedback a channel matrix H NB , which is the best representation of the channel matrix for all N subcarriers and can preserve power / phase information. The access node can then operate based on the complete downlink channel information. In this way, the spectral efficiency performance in MU-MIMO / massive MIMO is much higher than when only CQI / PMI / other limited channel information is reported. To this end, the access node can instruct the UE to generate a CSI report based on the WB channel in the subband. The CSI report is a matrix of size Nt×Nr, whose second-order moment is closest to the second-order moment of the WB channel and whose first-order moment is also closest to the first-order moment of the WB channel under the known second-order moment constraints. The CSI report, including the matrix, can be quantized to reduce communication overhead.

[0276] Simulation results are used to evaluate the performance of the wideband BiT algorithm in a multi-cell broadband system. The deployment scenario is based on the 3GPP model. A benchmark using ZF-based massive MIMO shows significantly worse spectral efficiency (SE) performance than BiT. Figure 15A 1500 is a graph of cell spectrum efficiency for BiT and ZF. By using the wideband BiT technique proposed in the exemplary embodiments herein, SE gains exceeding 50% are achieved compared to ZF in many deployment scenarios (some results are omitted for brevity). A simple approach to extend BiT to wideband by using the center subcarrier as a narrowband representation is also compared. Compared to the approach with the optimal narrowband representation, the SE gain from Figure 15A It is easy to see that there is a performance degradation. Figure 15BFigure 1530 shows the SINR cumulative distribution function (CDF) of BiT and ZF. The first curve 1535 shows the SINR CDF of ZF, and the second curve 1540 shows the SINR CDF of BiT. The results confirm that the SE performance of wideband BiT is improved, and the performance improvement is due to the improvement of SINR by avoiding inter-cell interference, such as Figure 15B shown. Figure 15C 1560 is a graph showing the convergence of BiT and ZF. The first curve 1565 shows the convergence of ZF, and the second curve 1570 shows the convergence of BiT. The number of BiT training iterations can be very small, typically achieving about 90% of steady-state performance in two rounds of training (e.g., Figure 15C shown).

[0277] Figure 16 An exemplary communication system 1600 is shown. Generally, system 1600 enables multiple wireless or wired users to transmit and receive data and other content. System 1600 can implement one or more channel access methods, such as code division multiple access (CDMA), time division multiple access (TDMA), frequency division multiple access (FDMA), orthogonal FDMA (OFDMA), single-carrier FDMA (SC-FDMA), or non-orthogonal multiple access (NOMA).

[0278] In this example, the communication system 1600 includes electronic devices (EDs) 1610a to 1610c, radio access networks (RANs) 1620a and 1620b, a core network 1630, a public switched telephone network (PSTN) 1640, the Internet 1650, and other networks 1660. Figure 16 A certain number of these components or elements are shown, but system 1600 may include any number of these components or elements.

[0279] EDs 1610a through 1610c are configured to operate and / or communicate within system 1600. For example, EDs 1610a through 1610c are configured to transmit or receive via a wireless communication channel or a wired communication channel. EDs 1610a through 1610c represent any suitable end-user device and may include (or may be referred to as) user equipment (UE), a wireless transmit / receive unit (WTRU), a mobile station, a fixed or mobile subscriber unit, a cellular phone, a personal digital assistant (PDA), a smartphone, a laptop, a computer, a touchpad, a wireless sensor, or a consumer electronic device.

[0280] RAN 1620a includes a base station 1670a, and RAN 1620b includes a base station 1670b. Both base stations 1670a and 1670b are configured to wirelessly connect to one or more of EDs 1610a to 1610c to enable access to a core network 1630, PSTN 1640, the Internet 1650, and / or other networks 1660. For example, base stations 1670a and 1670b may include (or may be) one or more of several well-known devices, such as a base transceiver station (BTS), a Node-B (NodeB), an evolved NodeB (eNodeB), a next generation (NG) NodeB (gNB), a Home NodeB, a Home eNodeB, a site controller, an access point (AP), or a wireless router. EDs 1610 a to 1610 c are used to connect to and communicate with the Internet 1650 , and can access a core network 1630 , a PSTN 1640 , or other networks 1660 .

[0281] exist Figure 16 In the illustrated embodiment, base station 1670a is part of RAN 1620a, which may include other base stations, components, and / or devices. Similarly, base station 1670b is part of RAN 1620b, which may include other base stations, components, or devices. Base stations 1670a and 1670b each transmit and / or receive wireless signals within a specific geographic region (region / area), sometimes referred to as a "cell." In some embodiments, multiple transceivers may be used in each cell using multiple-input multiple-output (MIMO) technology.

[0282] Base stations 1670a and 1670b communicate with one or more of EDs 1610a through 1610c using wireless communication links over one or more air interfaces 1690. The air interface 1690 may utilize any suitable radio access technology.

[0283] It is contemplated that system 1600 may utilize multi-channel access capabilities, including the aforementioned schemes. In specific embodiments, the base station and ED implement 5G New Radio (NR), LTE, LTE-A, or LTE-B. Of course, other multiple access schemes and wireless protocols may also be used.

[0284] RANs 1620a and 1620b communicate with core network 1630 to provide voice, data, application, Voice over Internet Protocol (VoIP), or other services to EDs 1610a through 1610c. It will be appreciated that RANs 1620a and 1620b or core network 1630 may communicate directly or indirectly with one or more other RANs (not shown). Core network 1630 may also serve as a gateway to other networks, such as PSTN 1640, Internet 1650, and other networks 1660. Furthermore, some or all of EDs 1610a through 1610c may include functionality to communicate with different wireless networks over different wireless links using different wireless technologies or protocols. EDs may communicate with a service provider or switch (not shown) and with Internet 1650 via wired communication channels, rather than wirelessly (or in addition to wirelessly).

[0285] Although Figure 16 An example of a communication system is shown, but Figure 16 For example, communication system 1600 may include any number of EDs, base stations, networks, or other components in any suitable configuration.

[0286] Figure 17A and Figure 17B The present invention provides exemplary devices that can implement the various methods and instructions provided by the present invention. Figure 17A An exemplary ED 1710 is shown, Figure 17B Shown is an exemplary base station 1770. These components may be used in system 1600 or any other suitable system.

[0287] like Figure 17AAs shown, ED 1710 includes at least one processing unit 1700. Processing unit 1700 implements various processing operations of ED 1710. For example, processing unit 1700 may perform signal encoding, data processing, power control, input / output processing, or any other function that enables ED 1710 to operate in system 1600. Processing unit 1700 also supports the methods and instructions described in detail above. Each processing unit 1700 includes any suitable processing device or computing device for performing one or more operations. Each processing unit 1700 may include a microprocessor, a microcontroller, a digital signal processor, a field programmable gate array, an application-specific integrated circuit, or the like.

[0288] ED 1710 also includes at least one transceiver 1702. Transceiver 1702 is used to modulate data or other content for transmission via at least one antenna or a network interface controller (NIC) 1704. Transceiver 1702 is also used to demodulate data or other content received via at least one antenna 1704. Each transceiver 1702 includes any suitable structure for generating signals for wireless or wired transmission or for processing signals received via wireless or wired means. Each antenna 1704 includes any suitable structure for transmitting or receiving wireless or wired signals. One or more transceivers 1702 can be used in ED 1710, and one or more antennas 1704 can be used in ED 1710. Although transceiver 1702 is shown as a separate functional unit, transceiver 1702 can also be implemented using at least one transmitter and at least one separate receiver.

[0289] ED 1710 also includes one or more input / output devices 1706 or interfaces (e.g., a wired interface to the Internet 1650). Input / output devices 1706 facilitate interaction with users or other devices on a network (network communications). Each input / output device 1706 includes any suitable structure for providing information to or receiving information from a user, such as a speaker, microphone, numeric keypad, keyboard, display, or touch screen, including network interface communications.

[0290] In addition, ED 1710 includes at least one memory 1708. Memory 1708 stores instructions and data used, generated, or collected by ED 1710. For example, memory 1708 may store software or firmware instructions executed by one or more processing units 1700 and store data used to reduce or eliminate interference in incoming signals. Each memory 1708 includes any suitable one or more volatile or non-volatile storage and retrieval devices. Any suitable type of memory may be used, such as random access memory (RAM), read-only memory (ROM), a hard disk, an optical disk, a subscriber identity module (SIM) card, a memory stick, a secure digital (SD) memory card, etc.

[0291] like Figure 17B As shown, base station 1770 includes at least one processing unit 1750, at least one transceiver 1752 (including transmitter and receiver functionality), one or more antennas 1756, at least one memory 1758, and one or more input / output devices or interfaces 1766. A scheduler, as will be appreciated by those skilled in the art, is coupled to processing unit 1750. The scheduler may be included within base station 1770 or may operate separately from base station 1770. Processing unit 1750 implements various processing operations for base station 1770, such as signal encoding, data processing, power control, input / output processing, or any other functions. Processing unit 1750 may also support the methods and instructions described in detail above. Each processing unit 1750 comprises any suitable processing device or computing device for performing one or more operations. Each processing unit 1750 may comprise a microprocessor, a microcontroller, a digital signal processor, a field programmable gate array, or an application-specific integrated circuit, among others.

[0292] Each transceiver 1752 includes any suitable structure for generating signals for wireless or wired transmission to or from one or more EDs or other devices. Each transceiver 1752 also includes any suitable structure for processing signals received wirelessly or wired from one or more EDs or other devices. Although the transmitter and receiver are shown combined as a transceiver 1752, the transmitter and receiver can be separate components. Each antenna 1756 includes any suitable structure for transmitting or receiving wireless or wired signals. Although a shared antenna 1756 is shown coupled to the transceiver 1752, one or more antennas 1756 can be coupled to the transceiver 1752, such that separate antennas 1756 are coupled to the transmitter and receiver (if configured as separate components). Each memory 1758 includes any suitable one or more volatile or non-volatile storage and retrieval devices. Each input / output device 1766 facilitates interaction with users or other devices on a network (network communications). Each input / output device 1766 includes any suitable structure for providing information to or receiving information from a user, including network interface communications.

[0293] Figure 18 18 is a block diagram of a computing system 1800 that can be used to implement the devices and methods disclosed herein. For example, the computing system can be any entity of a UE, an access network (AN), mobility management (MM), session management (SM), a user plane gateway (UPGW), or an access stratum (AS). A specific device may use all or only a subset of the components shown, and the degree of integration of the device may vary. In addition, a device may include multiple instances of a component, such as multiple processing units, multiple processors, multiple memories, multiple transmitters, multiple receivers, etc. Computing system 1800 includes a processing unit 1802. The processing unit includes a central processing unit (CPU) 1814, a memory 1808, and may also include a mass storage 1804 connected to a bus 1820, a video adapter 1810, and an I / O interface 1812.

[0294] Bus 1820 can be one or more of any type of bus architecture, including a memory bus or memory controller, a peripheral bus, or a video bus. CPU 1814 can include any type of electronic data processor. Memory 1808 can include any type of non-transient system memory, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), read-only memory (ROM), or a combination thereof. In one embodiment, memory 1808 can include ROM for use at boot time and DRAM for storing programs and data for use during program execution.

[0295] The mass storage 1804 may include any type of non-transitory storage device for storing data, programs, and other information and making such data, programs, and other information accessible via the bus 1820. The mass storage 1804 may include one or more of a solid-state drive, a hard disk drive, a magnetic disk drive, an optical disk drive, and the like.

[0296] A video adapter 1810 and an I / O interface 1812 are provided to couple external input and output devices to the processing unit 1802. As shown, examples of input and output devices include a display 1818 coupled to the video adapter 1810 and a mouse, keyboard, or printer 1816 coupled to the I / O interface 1812. Other devices may be coupled to the processing unit 1802, and other or fewer interface cards may be used. For example, a serial interface such as a Universal Serial Bus (USB) (not shown) may be used to provide an interface for external devices.

[0297] Processing unit 1802 also includes one or more network interfaces 1806, which can include wired links such as Ethernet cables or wireless links to access nodes or different networks. Network interface 1806 enables processing unit 1802 to communicate with remote units via a network. For example, network interface 1806 can provide wireless communication via one or more transmitters / transmit antennas and one or more receivers / receive antennas. In one embodiment, processing unit 1802 is coupled to a local area network 1822 or a wide area network to process and communicate data with remote devices such as other processing units, the Internet, or remote storage facilities.

[0298] It should be understood that one or more steps in the embodiment method provided herein can be performed by corresponding units or modules. For example, a signal can be sent by a transmitting unit or a transmitting module. A signal can be received by a receiving unit or a receiving module. A signal can be processed by a processing unit or a processing module. Other steps can be performed by an acquisition unit or a module, a determination unit or a module, an estimation unit or a module, a selection unit or a module, a sharing unit or a module, an adaptation unit or a module. The corresponding units or modules can be hardware, software or a combination thereof. For example, one or more units or modules can be integrated circuits, such as field programmable gate arrays (FPGAs) or application-specific integrated circuits (ASICs).

[0299] Although the present invention and its advantages have been described in detail, it should be understood that various changes, substitutions and alterations can be made herein without departing from the scope of the invention as defined by the appended claims.

Claims

1. A method implemented by a first device operating in a communication system, characterized in that: The method comprises: The first device obtains a channel representation of a set of channels between the first device and the second device, wherein the set of channels corresponds to a set of subcarriers, the first device has multiple antenna ports, and the second device has one or more antenna ports; The channel representation is such that a difference between a second-order moment of the channel representation and a second-order moment of the broadband representation is less than a power threshold; The first device determines one or more communication filters based at least on the channel representation; The first device applies the one or more communication filters to communications on at least one of the plurality of antenna ports of the first device, wherein the communications occur over the set of subcarriers.

2. The method according to claim 1, characterized in that The channel representation includes a channel matrix whose size is specified based on the plurality of antenna ports of the first device and the one or more antenna ports of the second device.

3. The method according to claim 1, characterized in that Determining the one or more communication filters includes using a Bidirectional Training (BiT) process.

4. The method according to claim 1, wherein The obtaining of a channel representation of a set of channels comprises receiving the channel representation of the set of channels from the second device.

5. The method according to claim 1, wherein The obtaining of a channel representation of a group of channels comprises determining the channel representation of the group of channels based on the individual channel representations of the group of channels.

6. The method according to claim 5, characterized in that The channel representation is determined based on a phase domain component of the set of channels and a power / amplitude domain component of the set of channels.

7. The method according to claim 6, characterized in that The channel representation includes a phase domain component of the set of channels, the phase domain component being constrained by a power / amplitude domain component of the set of channels.

8. The method according to any one of claims 5 to 7, characterized in that The obtaining of a channel representation of a group of channels comprises: The first device receives a reference signal sent on the set of channels corresponding to the set of subcarriers; The first device estimates the respective channel representations based on measurements of the received reference signal; The first device determines the channel representation based on the respective channel representations.

9. The method according to claim 1, characterized in that The obtaining of a channel representation of a group of channels comprises: The first device receives an indication of a subcarrier index; The first device selects a channel estimate value associated with the subcarrier index as the channel representation of the set of channels.

10. The method according to any one of claims 1 to 9, characterized in that The one or more communication filters include at least one of a transmit precoder for a subset of the plurality of antenna ports of the first device or a receive combiner for the plurality of antenna ports of the first device.

11. The method according to any one of claims 1 to 10, characterized in that The set of channels corresponds to a subset of the set of subcarriers.

12. The method according to any one of claims 1 to 11, characterized in that The communicating includes at least one of sending a first message or receiving a second message.

13. A method implemented by a first device operating in a communication system, characterized in that: The method comprises: The first device receives a signal transmitted on a set of channels corresponding to a set of subcarriers of the communication system, wherein the first device has a plurality of antenna ports; determining, by the first device, a channel representation of the set of channels based on the signals received on the set of channels corresponding to the set of subcarriers; The channel representation is such that a difference between a second-order moment of the channel representation and a second-order moment of the broadband representation is less than a power threshold; The first device sends the channel representation.

14. The method according to claim 13, wherein: The method also includes receiving, by the first device, resource allocation information associated with resources for communicating the received signal.

15. The method according to claim 13 or 14, characterized in that The sending the channel representation includes the first device sending an indication of a subcarrier index associated with the channel representation.

16. The method according to any one of claims 13 to 15, characterized in that The sending the channel representation comprises an indication by the first device to send the channel representation.

17. The method according to any one of claims 13 to 16, characterized in that The method further includes determining, by the first device, one or more communication filters based on the channel representations of the set of channels, wherein determining the one or more communication filters includes using a Bidirectional Training (BiT) process.

18. The method according to claim 13, characterized in that The channel representation includes a channel matrix whose size is specified based on the plurality of antenna ports of the first device and one or more antenna ports of the second device.

19. A device, characterized in that The device comprises: Non-transitory memory including instructions; one or more processors in communication with the memory, wherein the one or more processors execute the instructions to: Obtaining a channel representation of a set of channels between a first device and a second device, wherein the set of channels corresponds to a set of subcarriers, the first device has a plurality of antenna ports, and the second device has one or more antenna ports; The channel representation is such that a difference between a second-order moment of the channel representation and a second-order moment of the broadband representation is less than a power threshold; determining one or more communication filters based at least on the channel representation; The one or more communication filters are applied to communications on at least one of the plurality of antenna ports of the first device, wherein the communications occur over the set of subcarriers.

20. The device according to claim 19, characterized in that The one or more processors also execute the instructions to receive the channel representation of the set of channels from the second device.

21. The device according to claim 19, characterized in that The one or more processors further execute the instructions to determine the channel representation of the set of channels based on the individual channel representations of the set of channels.

22. The device according to claim 21, characterized in that The channel representation is determined based on a phase domain component of the set of channels and a power / amplitude domain component of the set of channels.

23. The device according to claim 22, characterized in that The channel representation includes a phase domain component of the set of channels, the phase domain component being constrained by a power / amplitude domain component of the set of channels.

24. The apparatus according to any one of claims 21 to 23, characterized in that The one or more processors further execute the instructions to: receiving a reference signal sent on the set of channels corresponding to the set of subcarriers; estimating the respective channel representations based on measurements of the received reference signals; The channel representation is determined based on the respective channel representations.

25. The apparatus according to claim 19, wherein The one or more processors further execute the instructions to: receiving an indication of a subcarrier index; A channel estimate value associated with the subcarrier index is selected as the channel representation of the set of channels.

26. The apparatus according to any one of claims 19 to 25, characterized in that The channel representation includes a channel matrix whose size is specified based on the plurality of antenna ports of the first device and the one or more antenna ports of the second device.

27. The apparatus according to any one of claims 19 to 26, characterized in that The one or more communication filters are determined using a Bidirectional Training (BiT) process.

28. A first device, characterized in that: The first device includes: Non-transitory memory including instructions; one or more processors in communication with the memory, wherein the one or more processors execute the instructions to: receiving a signal transmitted on a set of channels corresponding to a set of subcarriers of a communication system, wherein the first device has a plurality of antenna ports; determining a channel representation of the set of channels based on the signals received on the set of channels corresponding to the set of subcarriers; The channel representation is such that a difference between a second-order moment of the channel representation and a second-order moment of the broadband representation is less than a power threshold; The channel representation is transmitted.

29. The first device according to claim 28, characterized in that The one or more processors further execute the instructions to receive resource allocation information associated with resources for communicating the received signal.

30. The first device according to claim 28 or 29, characterized in that The one or more processors further execute the instructions to send an indication of a subcarrier index associated with the channel representation.

31. The first device according to any one of claims 28 to 30, characterized in that The one or more processors further execute the instructions to send an indication of the channel representation.

32. The first device according to any one of claims 28 to 31, characterized in that The one or more processors further execute the instructions to determine one or more communication filters based on the channel representations of the set of channels, wherein determining one or more communication filters includes using a Bidirectional Training (BiT) process.

Images