Communication method, device and computer readable medium

Abstract

Embodiments of the present disclosure relate to methods, devices, and computer-readable storage media for communication. A network device sends a channel state information reference signal associated with a first number of ports at the network device to a terminal device; receives compressed channel state information generated based on the capabilities of the terminal device from the terminal device; and recovers channel state information associated with a second number of ports at the network device from the compressed channel state information, where the second number of ports is not less than the first number of ports. The terminal device receives a channel state information reference signal associated with a first number of ports at the network device from the network device; determines channel state information based on the channel state information reference signal; compresses the channel state information based on the capabilities of the terminal device; and sends compressed channel state information to the network device for recovering channel state information associated with the second number of ports. Therefore, CSI can be acquired more accurately while reducing complexity and overhead.

Description

Technical Field

[0001] Embodiments of the present disclosure relate generally to the field of telecommunications, and more particularly, to methods, devices, and computer-readable storage media for communications in massive multiple-input multiple-output (MIMO) systems. Background Art

[0002] In massive MIMO systems, especially those operating in frequency division duplexing (FDD) mode at millimeter wave (mmWave) frequencies, obtaining accurate channel state information (CSI) is crucial for beamforming or precoding design to achieve the throughput improvement brought by the large number of antennas. As the number of antenna ports increases, a key issue during CSI acquisition is the high overhead caused by the transmission of a large number of CSI reference signals (CSI-RS) and the large amount of CSI feedback.

[0003] Millimeter-wave MIMO channels with bidirectional modeling exhibit spatial sparsity. Given this, it is possible to compress the relevant CSI in massive MIMO systems, at least in the spatial domain, to reduce CSI feedback overhead. Currently, achieving good performance while reducing the overhead in the CSI acquisition process for massive MIMO is of great interest. Summary of the Invention

[0004] Generally speaking, example embodiments of the present disclosure provide a scheme for CSI acquisition.

[0005] In a first aspect, a communication method is provided. The method includes: transmitting, at a network device, a channel state information reference signal associated with a first number of ports at the network device to a terminal device; receiving, from the terminal device, compressed channel state information generated based on capabilities of the terminal device; and recovering, from the compressed channel state information, channel state information associated with a second number of ports at the network device, where the second number of ports is not less than the first number of ports.

[0006] In a second aspect, a communication method is provided. The method includes: receiving, at a terminal device, a channel state information reference signal associated with a first number of ports at the network device from a network device; determining channel state information based on the channel state information reference signal; compressing the channel state information based on a capability of the terminal device; and sending the compressed channel state information to the network device for use in recovering channel state information associated with a second number of ports at the network device, the second number of ports being not less than the first number of ports.

[0007] In a third aspect, a network device is provided. The network device comprises: at least one processor; and at least one memory including computer program code; the at least one memory and the computer program code are configured to, together with the at least one processor, cause the network device to perform the method according to the first aspect:

[0008] In a fourth aspect, a terminal device is provided. The terminal device includes: at least one processor; and at least one memory including computer program code; the at least one memory and the computer program code are configured to, together with the at least one processor, cause the terminal device to perform the method according to the second aspect:

[0009] In a fifth aspect, a non-transitory computer-readable medium is provided, wherein the non-transitory computer-readable medium includes program instructions, and the program instructions are configured to cause an apparatus to execute the method according to the first aspect.

[0010] In a sixth aspect, a non-transitory computer-readable medium is provided, wherein the non-transitory computer-readable medium includes program instructions, and the program instructions are configured to cause an apparatus to execute the method according to the second aspect.

[0011] It should be understood that the invention summary is not intended to identify the key or essential features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Some example embodiments will now be described with reference to the accompanying drawings, in which:

[0013] Figure 1 illustrates an example communication network in which example embodiments of the present disclosure may be implemented;

[0014] Figure 2 FIGURES illustrate a flow chart illustrating a communication process during CSI acquisition according to some embodiments of the present disclosure;

[0015] Figure 3 A flow chart illustrating a communication method implemented at a network device according to an example embodiment of the present disclosure is illustrated;

[0016] Figure 4 illustrates a flow chart of an example method of CSI recovery according to an example embodiment of the present disclosure;

[0017] Figure 5 A flowchart illustrating a communication method implemented at a terminal device according to an exemplary embodiment of the present disclosure is illustrated;

[0018] Figure 6 A flowchart illustrating an example method of CSI feedback according to an example embodiment of the present disclosure is illustrated;

[0019] Figure 7 An example comparison of the normalized mean square error (NMSE) performance between the proposed scheme and the existing schemes is shown in the figure using the cumulative distribution function (CDF);

[0020] Figure 8 An example comparison of throughput performance between the present solution and existing solutions using CDF is shown;

[0021] Figure 9 An example comparison of throughput performance and the number of terminal devices between the present solution and the existing solution is shown in the figure using CDF;

[0022] Figure 10 illustrates a simplified block diagram of an apparatus suitable for implementing an example embodiment of the present disclosure; and

[0023] Figure 11 A block diagram of an example computer-readable medium is illustrated according to an example embodiment of the present disclosure.

[0024] Throughout the drawings, the same or similar reference numerals refer to the same or similar elements. DETAILED DESCRIPTION

[0025] The principles of the present disclosure will now be described with reference to some exemplary embodiments. It should be understood that these embodiments are described for illustrative purposes only and to help those skilled in the art understand and implement the present disclosure, and do not imply any limitation on the scope of the present disclosure. The disclosure described herein can be implemented in various ways other than those described below.

[0026] In the following description and claims, unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.

[0027] References in this disclosure to "one embodiment," "an embodiment," "an example embodiment," etc., indicate that the described embodiment may include a particular feature, structure, or characteristic, but not every embodiment necessarily includes the particular feature, structure, or characteristic. Furthermore, such phrases do not necessarily refer to the same embodiment. Furthermore, when a particular feature, structure, or characteristic is described in conjunction with an embodiment, it should be understood that it is within the knowledge of those skilled in the art to implement such feature, structure, or characteristic in conjunction with other embodiments, whether or not explicitly described.

[0028] It should be understood that although the terms "first" and "second" and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the example embodiments. As used herein, the term "and / or" includes any and all combinations of one or more of the listed terms.

[0029] The terms used herein are for the purpose of describing specific embodiments only and are not intended to limit the example embodiments. As used herein, unless the context clearly indicates otherwise, the singular forms "a," "an," and "the" are intended to include the plural forms as well. It will be further understood that the terms "including," "having," and "comprising" when used herein specify the presence of stated features, elements, and / or components, but do not preclude the presence or addition of one or more other features, elements, components, and / or combinations thereof.

[0030] As used in this application, the term "circuitry" may refer to one or more or all of the following:

[0031] (a) pure hardware circuit implementation (such as implementation in analog and / or digital circuitry only) and

[0032] (b) a combination of hardware circuitry and software such as (as applicable):

[0033] (i) a combination of analog and / or digital hardware circuitry and software / firmware, and

[0034] (ii) any portion of hardware processor(s) together with software (including digital signal processors), software and memory(s) that work together to enable a device (such as a mobile phone or server) to perform various functions), and

[0035] (c) Hardware circuit(s) and or processors, such as microprocessor(s) or portions of microprocessor(s), that require software (e.g., firmware) to operate, but which may not be present when not required for operation.

[0036] This definition of "circuitry" applies to all uses of this term in this application, including in any claims. As another example, as used in this application, the term "circuitry" also covers an implementation of merely a hardware circuit or processor (or multiple processors), or portions of a hardware circuit or processor, and their accompanying software and / or firmware. The term "circuitry" also covers (for example, and if applicable to a particular claim element) a baseband integrated circuit or processor integrated circuit for a mobile device, or a similar integrated circuit in a server, cellular network device, or other computing or networking equipment.

[0037] As used herein, the term "communication network" refers to a network that complies with any suitable communication standard, such as Long Term Evolution (LTE), LTE Advanced (LTE-A), Wideband Code Division Multiple Access (WCDMA), High Speed ​​Packet Access (HSPA), Narrowband Internet of Things (NB-IoT), etc. In addition, the communication between the terminal device and the network equipment in the communication network can be performed according to any suitable generation communication protocol, including but not limited to first generation (1G), second generation (2G), 2.5G, 2.75G, third generation (3G), fourth generation (4G), 4.5G, future fifth generation (5G) communication protocols and / or any other protocol currently known or developed in the future. The embodiments of the present disclosure can be applied to various communication systems. In view of the rapid development of communications, there will certainly be future types of communication technologies and systems that can be used to implement the present disclosure. It should not be considered that the scope of the present disclosure is limited to the aforementioned systems.

[0038] As used herein, the term "network device" refers to a node in a communication network through which a terminal device accesses the network and receives services from it. Depending on the terminology and technology used, a network device can refer to a base station (BS) or an access point (AP), such as a NodeB (NodeB or NB), an evolved NodeB (eNodeB or eNB), a NR NB (also known as a gNB), a remote radio unit (RRU), a radio head (RH), a remote radio head (RRH), a relay, or a low-power node such as a femto or pico.

[0039] The term "terminal device" refers to any terminal device that can perform wireless communication. By way of example and not limitation, a terminal device may also be referred to as a communication device, user equipment (UE), user station (SS), portable user station, mobile station (MS) or access terminal (AT). Terminal devices may include, but are not limited to, mobile phones, cellular phones, smart phones, voice over IP (VoIP) phones, wireless local loop phones, tablet computers, wearable terminal devices, personal digital assistants (PDAs), portable computers, desktop computers, image capture terminal devices such as digital cameras, game terminal devices, music storage and playback appliances, vehicle-mounted wireless terminal devices, wireless endpoints, mobile stations, laptop embedded equipment (LEE), laptop mounted equipment (LME), USB hardware dongles, smart devices, wireless client equipment (CPE), Internet of Things (IoT) devices, watches or other wearables, head-mounted displays (HMDs), vehicles, drones, medical devices and applications (e.g., remote surgery), industrial devices and applications (e.g., robots and / or other wireless devices operating in industrial and / or automated process chain scenarios), consumer electronic devices, devices operating in commercial and / or industrial wireless networks, etc. In the following description, the terms "terminal device," "communication device," "terminal," "user equipment," and "UE" may be used interchangeably.

[0040] Figure 1 1 illustrates an example communication network 100 in which embodiments of the present disclosure may be implemented. Figure 1 As shown in FIG, the network 100 includes a network device 110 and a terminal device 120 served by the network device 110. It should be understood that Figure 1 The number of network devices and terminal devices shown in FIG is for illustration purposes only and does not imply any limitation. Network 100 may include any suitable number of network devices and terminal devices suitable for implementing embodiments of the present disclosure.

[0041] like Figure 1 As shown in FIG, the network device 110 and the terminal device 120 can communicate with each other. The network device 110 can have multiple antennas for communicating with the terminal device 120. For example, the network device 110 can include four antennas 111, 112, 113, and 114. The terminal device 120 can also have multiple antennas for communicating with the network device 110. For example, the terminal device 120 can include four antennas 121, 122, 123, and 124. It should be understood that Figure 1 The number of antennas shown in FIG is for illustration purposes only and is not intended to be limiting. Each of the network device 110 and the terminal device 120 may provide any suitable number of antennas suitable for implementing embodiments of the present disclosure.

[0042] Communications in network 100 may conform to any suitable standard, including but not limited to LTE, LTE Evolution, LTE Advanced (LTE-A), Wideband Code Division Multiple Access (WCDMA), Code Division Multiple Access (CDMA), and Global System for Mobile Communications (GSM). Furthermore, communications may be performed according to any generation of communication protocols currently known or developed in the future. Examples of communication protocols include but are not limited to first generation (1G), second generation (2G), 2.5G, 2.75G, third generation (3G), fourth generation (4G), 4.5G, and fifth generation (5G) communication protocols.

[0043] Multiple channels or sub-channels can be established between the network device 110 and the terminal device 120 via multiple antennas 111, 112, 113, and 114 and multiple antennas 121, 122, 123, and 124 for data transmission. In order to achieve the throughput improvement brought by a large number of antennas, it is necessary to obtain accurate CSI associated with multiple channels or sub-channels for beamforming or precoding design in data transmission. During CSI acquisition, the network device 110 can send a CSI-RS to the terminal device 120 via each of the channels or sub-channels, and the terminal device 120 can measure the CSI on each of the channels or sub-channels based on the received CSI-RS and send the measured CSI to the network device 110.

[0044] For example, assume that the network device 110 is equipped with M T The network device 110 serves K terminal devices 120 simultaneously, where each terminal device 120 has receive antennas. The complete channel considered in the system is given by Indicates that is the total number of receiving antennas from all terminal devices 120. The mmWave channel is represented by the widely used cluster model, represented by H of the kth terminal device 120. k The network device 110 requires CSIH = [H (1) ,H (2) ,...,H (K) ] is used to design transmission processing such as precoding for the K terminal devices 120. In an FDD system, the downlink channel H can be obtained via feedback of the CSI estimated at the terminal device 120.

[0045] As the number of antennas increases, the CSI to be estimated at the terminal device 120 has a large dimension, thus requiring a large CSI-RS overhead in the downlink. On the other hand, once the terminal device 120 obtains the downlink CSI-RS, feeding back the entire explicit CSI will improve MIMO transmission performance but also incur a significantly large overhead.

[0046] In some conventional schemes, the network device 110 may send a non-precoded CSI-RS to the terminal device 120, and the terminal device 120 may determine the CSI by estimating the channel or subchannel based on the non-precoded CSI-RS. The terminal device 120 may compress the CSI to remove unnecessary redundancy by using, for example, principal component analysis (PCA) or orthogonal matching pursuit (OMP), and send the compressed CSI to the network device 110. The network device 110 may receive the compressed CSI and perform reconstruction of the compressed CSI. In this way, the CSI feedback overhead may be reduced. However, a disadvantage of this scheme is the consistent feedback of a large number of transmitted CSI-RSs and compressed information due to many antenna ports, such as the compression matrix or index of the corresponding codebook reconstructed at the network device 110. This will result in a large overhead for both transmitting the CSI-RS and the feedback.

[0047] In some still conventional schemes, the network device 110 can use, for example, a beam grid (GoB) to send a beamformed CSI-RS to the terminal device 120, and the terminal device 120 can estimate the beamformed CSI-RS and feed back beamformed CSI with reduced dimensionality, as well as a CSI-RS resource indicator (CRI). This reduces the overhead of both CSI-RS transmission and feedback. However, the network device 110 needs to design a "second stage" precoding for the obtained beamformed CSI to map the layers to ports. In this way, performance will be limited by the beam selection of the beamformed CSI-RS because the precoding design is divided into two parts, one is the "second stage" precoding, and the other is port virtualization.

[0048] In some still conventional solutions, the two aforementioned solutions are combined to utilize beamformed CSI-RS at network device 110 and perform CSI compression at terminal device 120. This significantly reduces overhead, but network device 110 still needs to design "second-stage" precoding for the obtained beamformed CSI to map layers to ports. This is detrimental to improving MIMO transmission performance and complexity.

[0049] The inventors have noticed that millimeter wave MIMO channels with bidirectional modeling exhibit spatial sparsity. The relevant CSI in a massive MIMO system can be compressed, at least in the spatial domain, to reduce the CSI feedback overhead. A popular solution is to apply a compressed review (CS) formula at the terminal device 120, and the network device 110 can simply reconstruct the CSI from the compressed CSI feedback. In future mobile communications, due to the need for new functions, the terminal device 120 is required to have low power consumption and low complexity, while the network device 110 may be more powerful. Therefore, it is still very desirable to transfer the computational work from the terminal device 120 to the network device 110.

[0050] To solve the above and other potential problems, the embodiments of the present disclosure provide an improved CSI acquisition solution based on machine learning (ML), the mechanism of which is as follows: Figure 2 The high-level flow chart shown is illustrated.

[0051] Figure 2 A flowchart 200 illustrating a communication process during CSI acquisition according to some embodiments of the present disclosure is illustrated. A network device 110 transmits 210 CSI-RSs associated with a first number of ports at the network device 110 to a terminal device 120. Upon receiving the CSI-RSs, the terminal device 120 determines 220 relevant CSI based on channel measurements of the CSI-RSs. Upon determining the CSI, the terminal device 120 compresses 230 the CSI and transmits 240 the compressed CSI to the network device 110. Upon receiving the compressed CSI, the network device 110 recovers 250 CSI associated with a second number of ports at the network device 110 from the compressed CSI. The second number of ports is not less than the first number of ports, i.e., the second number of ports is equal to or greater than the first number of ports.

[0052] It should be understood that the ports herein may refer to virtual ports in port virtualization. In some embodiments, the second number of ports may be equal to the total number of ports on network device 110. In some alternative embodiments, the second number of ports may be less than the total number of ports on network device 110. Compared to conventional solutions, this solution can recover CSI for ports with more ports than the number of ports used for CSI feedback based on a predetermined ML model without requiring CRI-like feedback, thereby more accurately acquiring CSI while reducing complexity and overhead.

[0053] Some exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. However, those skilled in the art will readily appreciate that the detailed description given herein with respect to these drawings is for explanatory purposes only, as the present disclosure extends beyond these limited embodiments.

[0054] Figure 3 A flow chart of a communication method 300 implemented at a network device according to an example embodiment of the present disclosure is illustrated. The method 300 may be implemented at Figure 1 For the purpose of discussion, reference will now be made to the network device 110 shown in FIG. Figure 1 Method 300 is described. It should be understood that method 300 may also include additional blocks not shown and / or omit some of the blocks shown, and the scope of the present disclosure is not limited in this respect.

[0055] like Figure 3As shown in FIG, at block 310, network device 110 transmits a CSI-RS to be associated with a first number of ports at network device 110 to terminal device 120. In some embodiments, network device 110 may compress the CSI-RS and transmit the compressed CSI-RS to terminal device 120. In this case, the first number of ports may be less than the total number of ports at network device 110. In some embodiments, network device 110 may compress the CSI-RS based on antenna characteristics at network device 110. In some example embodiments, network device 110 may compress the CSI-RS based on antenna responses at network device 110.

[0056] For example, if the number of transmit antenna ports is M T , then apply the space compression matrix where N T <M T This is established by assuming that spatial redundancy exists at the transmitter of network device 110. Network device 110 only needs to send N T CSI-RS instead of M T , which reduces the overhead of CSI-RS.

[0057] An example of designing the transmission compression matrix F will now be described. Since the network device 110 does not know the arrival angle(s) from the terminal device 120 or the plurality of terminal devices, the large range is divided into N T intervals and apply Discrete Prolate Spherical Sequence (DPSS) beam spatial beamforming in each angular interval. For the n-th angular interval n=1,...,N T The semidefinite matrix of the DPSS method is defined as

[0058]

[0059] where Θ n and Φ n is the angular region of interest in two dimensions, It is the sampling point The array response at . By applying eigenvalue decomposition (EVD) to equation (2),

[0060]

[0061] Original DPSS beam space matrix F n It can be obtained by summing (equal combinations) the eigenvectors across the sampling region of interest and written as

[0062]

[0063] The matrix used to measure the static beam can be written as Because the beamforming matrix used for measurement depends only on the array response at network device 110, it is considered "static." It should be noted that the above examples are for illustration only, and other antenna characteristics besides array response can also be used to design the transmit compression matrix, and other forms of matrices can also be used. In this way, the quasi-static compression / manipulation of the CSI-RS at network device 110 presents a specific "beam" pattern, without requiring any beam selection on the end device 120 side, nor does it require CRI feedback from the end device 120.

[0064] In some alternative embodiments, network device 110 may transmit original or uncompressed CSI-RS. In this case, the first number of ports may be equal to the total number of ports at network device 110. In these embodiments, network device 110 may also transmit information about transmission compression to terminal device 120 for compressing CSI at terminal device 120, which will be discussed later in conjunction with FIG. Figure 5 and Figure 6 For example, the terminal device 120 may transmit information about the transmission compression matrix F. It should be noted that any other suitable form of information about the transmission compression may also be used.

[0065] At block 320, the network device 110 receives compressed CSI (eg, Y (k) , where k represents the kth terminal device). According to an embodiment of the present disclosure, compressed CSI is generated based on the capabilities of the terminal device 120. In some embodiments, the compression mode of compressed CSI can be divided into a first compression mode that is friendly to the common codebook ML and a second compression mode that is not friendly to the common codebook ML.

[0066] In some embodiments where the terminal device 120 enables the first compression mode, a common codebook may be used to generate compressed CSI based on the antenna characteristics at the terminal device 120. The codebook should be accurately defined to meet the compression requirements. The compression does not vary with dynamic channel propagation characteristics, but may vary with different antenna patterns of different terminal devices. This is simple for ML-enabled CSI recovery with good performance. In some embodiments where the terminal device 120 enables the second compression mode, a unique codebook may be used to generate compressed CSI based on the channel propagation characteristics at the terminal device 120. Additional details regarding compression associated with compressed CSI will be provided later in conjunction with Figure 5 and Figure 6 Provide a description.

[0067] After receiving the compressed CSI, at block 330, network device 110 recovers CSI associated with a second number of ports at network device 110 (eg, H (k), where k represents the kth terminal device). According to an embodiment of the present disclosure, the second number of ports is not less than the first number of ports. In some embodiments, the second number of ports may be equal to the total number of ports at network device 110. In some alternative embodiments, the second number of ports may be less than the total number of ports at network device 110.

[0068] According to an embodiment of the present disclosure, the network device 110 may recover the CSI associated with the second number of ports from the compressed CSI based on a predetermined ML model. Different models may be used for different compression modes associated with the compressed CSI. Figure 4 Provide a description. Figure 4 A flow chart illustrating an example method 400 for CSI recovery according to an example embodiment of the present disclosure is shown. The method 400 may be performed in Figure 1 For the purpose of discussion, reference will be made to the network device 110 shown in FIG. Figure 1 Method 400 is described. It should be understood that method 400 may also include additional blocks not shown and / or omit some of the blocks shown, and the scope of the present disclosure is not limited in this respect.

[0069] At block 410, the network device 110 may determine whether the mode is the first compression mode or the second compression mode. Upon receiving the compressed CSI, the network device 110 may determine the compression mode of the compressed CSI. In some embodiments, the mode may be determined or triggered by the terminal device 120. In this case, the network device 110 may receive an indicator representing the mode from the terminal device 120 and determine the mode based on the indicator. For example, in some embodiments, the network device 110 may receive the indicator in a physical uplink control channel (PUCCH). In some embodiments, the network device 110 may receive the indicator together with the compressed CSI. It should be noted that sending the indicator from the terminal device 120 may be performed in any other suitable manner and is not limited to the above examples.

[0070] In some alternative embodiments, the mode may be determined or triggered by network device 110. In some embodiments, network device 110 may receive information about the capabilities of terminal device 120 from terminal device 120 and determine the mode based on the capabilities of terminal device 120. For example, in some embodiments where terminal device 120 enables the first compression mode, network device 110 may determine the mode as the first compression mode. In some embodiments where terminal device 120 does not enable the first compression mode, network device 110 may determine the mode as the second compression mode.

[0071] In these embodiments where the mode is determined by network device 110, network device 110 may further transmit an indicator indicating the mode to terminal device 120 for CSI compression. For example, in some embodiments, network device 110 may transmit the indicator in a physical downlink control channel (PDCCH). In some embodiments, network device 110 may transmit the indicator together with the CSI-RS. It should be noted that the transmission of the indicator from network device 110 may be performed in any other suitable manner and is not limited to the above examples.

[0072] refer to Figure 4 If the mode is determined to be the first compression mode at block 410, then at block 420, network device 110 may restore the CSI associated with the second number of ports based on a first predetermined model that maps the compressed CSI to the CSI associated with the second number of ports. In some embodiments, due to As the input of the first predetermined model, H (k) As output. By concatenating its real and imaginary parts in the third dimension, is the matrix Y (k) The tensor symbol for .

[0073] According to embodiments of the present disclosure, a first predetermined model can be pre-generated by training a neural network using a large amount of compressed CSI as input and the corresponding known precise CSI as output. During the training process, all possible codebook cases for CSI compression are considered as input, so the first predetermined model for CSI recovery is not limited by a specific terminal device. It should be understood that the first predetermined model can be obtained through any suitable neural network training method, and this disclosure is not limited in this regard.

[0074] If it is determined at block 410 that the mode is the second compression mode, then at block 430, the network device 110 may receive an indication of a codebook for compressing the CSI from the terminal device 120 (eg, W (k) , where k represents the kth terminal device). The indication can be made in any suitable manner. Feedback of the codebook indication does not incur significant overhead because the codebook size is small, limited only by the number of antennas at the terminal device 120, and is related to receive compression.

[0075] At block 440, the network device 110 may generate processed CSI based on the codebook (e.g., ). For example, in some embodiments, compressed CSIY (k) The compressed CSIY (k) Vectorized processing can be written as:

[0076]

[0077] where Y (k) represents the compressed CSI associated with the kth terminal device, y (k) represents the vectored compressed CSI associated with the kth terminal device, W (k) represents the receiving compression matrix associated with the kth terminal device, H (k) Denotes the exact channel matrix associated with the kth terminal device, and F denotes the transmit compression matrix. The processed CSI can then be obtained by the least squares (LS) method.

[0078]

[0079] in Indicates the processed CSI, Y (k) is the compressed CSI associated with the kth terminal device, W (k) represents the receiving compression matrix associated with the kth terminal device, H (k) represents the exact channel matrix associated with the kth terminal device, F represents the transmit compression matrix, represents the number of receiving antenna ports at the terminal device 120, M T represents the number of transmit antenna ports at the network device 110 .

[0080] It should be noted that the above equation is for illustration only and is not intended to limit the present application, and any other suitable implementation may be adopted.

[0081] At block 450, network device 110 may recover CSI associated with the second number of ports (eg, H ports) based on a second predetermined model that maps the processed CSI to CSI associated with the second number of ports. (k) ). In some embodiments, it is possible to The real and imaginary parts of the tensor are reformed into The processed CSI is further processed to serve as input to a second predetermined model. Then, CSI associated with the second number of ports may be obtained as output of the second predetermined model.

[0082] According to embodiments of the present disclosure, a second predetermined model can be pre-generated by training a neural network using a large amount of processed CSI as input and corresponding known precise CSI as output. The processed CSI is generated from compressed CSI based on codebook instructions. It should be understood that the second predetermined model can be obtained through any suitable neural network training method, and this disclosure is not limited thereto.

[0083] Referenced above Figure 3 and Figure 4An example implementation at the network device 110 is described. In this way, CSI in uncompressed dimensions can be recovered from CSI feedback with a limited number of ports, which is enabled by a neural network trained for two modes. Accordingly, reference will now be made to Figures 5 and 6 An example implementation at the terminal device 120 is described.

[0084] Figure 5 The flowchart of the communication method 500 implemented at the terminal device according to an exemplary embodiment of the present disclosure is shown. The method 500 may be implemented at Figure 1 For the purpose of discussion, reference will be made to the terminal device 120 shown in FIG. Figure 1 Method 500 is described. It should be understood that method 500 may also include additional blocks not shown and / or omit some of the blocks shown, and the scope of the present disclosure is not limited in this respect.

[0085] like Figure 5 As shown in , at block 510, the terminal device 120 receives a CSI-RS associated with a first number of ports at the network device 110 from the network device 110. In some embodiments, the terminal device 120 may receive a compressed CSI-RS. In this case, the first number of ports may be less than the total number of ports at the network device 110. In some embodiments, the terminal device 120 may receive a compressed CSI-RS generated based on antenna characteristics at the network device 110. For example, the compressed CSI-RS may be generated according to equations (1)-(3) above. For more details, please refer to the combined Figure 3 A similar description is given for block 310 of FIG.

[0086] In some alternative embodiments, terminal device 120 may receive raw or unprocessed CSI-RS. In this case, the first number of ports may be equal to the total number of ports at network device 110. In these embodiments, terminal device 120 may also receive information about transmit compression from network device 110 for later use in CSI compression. For example, terminal device 120 may receive information about a transmit compression matrix F. It should be noted that any other suitable form of information about transmit compression may also be used.

[0087] At block 520, the terminal device 120 determines CSI based on the CSI-RS. According to an embodiment of the present disclosure, the terminal device 120 may determine the CSI associated with the first number of ports by channel estimation based on the CSI-RS. It should be noted that the determination of CSI may be performed in any suitable manner, and the present disclosure is not limited thereto.

[0088] At block 530, the terminal device 120 compresses the CSI based on the capabilities of the terminal device 120. In some embodiments, the terminal device may determine a compression mode and compress the CSI based on the mode. As described above, in some embodiments, the compression mode for compressing the CSI may be divided into a first compression mode that is friendly to the common codebook ML and a second compression mode that is not friendly to the common codebook ML. Different compression modes are used for different modes to facilitate recovery of the CSI at the network device 110. Figure 6 Provide a description.

[0089] Figure 6 A flow chart illustrating an example method 600 for CSI feedback according to an example embodiment of the present disclosure is shown. The method 600 may be Figure 1 For the purpose of discussion, reference will be made to the terminal device 120 shown in FIG. Figure 1 Method 600 is described. It should be understood that method 600 may also include additional blocks not shown and / or omit some of the blocks shown, and the scope of the present disclosure is not limited in this respect.

[0090] At block 610, the terminal device 120 may determine whether to enable the first compression mode or the second compression mode. In some embodiments, the mode may be determined or triggered by the network device 110. In this case, the terminal device 120 may send information about the capabilities of the terminal device 120 to the network device 110 to determine the mode. Accordingly, the terminal device 120 may receive an indicator representing the mode from the network device 110 and determine the mode based on the indicator. For example, in some embodiments, the terminal device 120 may receive the indicator in the PDCCH. In some embodiments, the terminal device 120 may receive the indicator together with the CSI-RS. It should be noted that receiving the indicator from the network device 110 may be performed in any other suitable manner and is not limited to the above examples.

[0091] In some alternative embodiments, the mode may be determined or triggered by the terminal device 120. In some embodiments, the terminal device 120 may determine the mode based on the capabilities of the terminal device 120. For example, in some embodiments where the terminal device 120 enables the first compression mode, the terminal device 120 may determine the mode as the first compression mode. In some embodiments where the terminal device 120 does not enable the first compression mode, the terminal device 120 may determine the mode as the second compression mode.

[0092] In these embodiments where the mode is determined by terminal device 120, terminal device 120 may further transmit an indicator indicating the mode to network device 110 for use in CSI recovery. For example, in some embodiments, terminal device 120 may transmit the indicator in a PUCCH. In some embodiments, terminal device 120 may transmit the indicator along with CSI feedback. It should be noted that transmitting the indicator from terminal device 120 may be performed in any other suitable manner and is not limited to the above examples.

[0093] refer to Figure 6 If it is determined at block 610 that the first compression mode is enabled, then at block 620 the terminal device 120 may compress the CSI based on the antenna characteristics at the terminal device 120. For example, in some embodiments, the UE applies a receive compression To further reduce the CSI feedback overhead. The resulting compressed CSI can be expressed as

[0094]

[0095] where Y (k) represents the compressed CSI associated with the kth terminal device, W (k) represents the receiving compression matrix associated with the kth terminal device, H (k) represents the exact channel matrix associated with the kth terminal device, F represents the transmit compression matrix, represents the number of receiving antenna ports at the terminal device 120, and N T Indicates the number of CSI-RS.

[0096] In some embodiments where the first compression mode is enabled, the terminal device 120 may apply a "static" beamspace compression that matches the array response within its range of interest. Similar to equation (1), based on the semidefinite matrix of the DPSS defined by

[0097]

[0098] in represents the sampling point for each terminal device 120 EVD is applied to this matrix J UE =U∑U H And the compressed matrix W can be obtained by taking its main features as follows (k)

[0099]

[0100] This ensures good array gain in the region of interest and is independent of the different channels.

[0101] In some embodiments, a common codebook may be defined, W (k) can be mapped to the public codebook. (k) The design of follows the criterion of “maximizing the effective array gain in the angular region of interest” and is “static” (not dependent on the channel / time variation), so W can be obtained for each terminal device’s antenna configuration. (k) . A codebook is also needed to represent W (k) In some embodiments, different terminal devices may have different W (k) , but for each terminal device, W (k) It is fixed according to its own pattern.

[0102] In some embodiments, the public codebook can be designed to accommodate different numbers of antenna elements. Since the terminal device may not have many antennas, the size of the public codebook will not be large. (k) The choice of matching codebook will also be small.

[0103] In some embodiments where an uncompressed CSI-RS is received, the terminal device 120 may determine the receive compression matrix W based on the received information about transmit compression (eg, the transmit compression matrix F). (k) , for example according to equations (7) and (8). The resulting compressed CSI can be written as:

[0104] Y (k) =w (k) HF (9)

[0105] where Y (k) represents compressed CSI, W (k) represents the receiving compression matrix, H represents the precise channel matrix, and F represents the transmitting compression matrix.

[0106] It should be noted that the above equation is provided for illustration only and is not intended to limit the present application, and any other suitable implementations may be adopted.

[0107] refer to Figure 6 If, at block 610, it is determined that the second compression mode is enabled, then at block 630, terminal device 120 may compress the CSI based on the channel propagation characteristics at terminal device 120. In some embodiments, terminal device 120 may compress the CSI in the spatial domain by applying a dynamic channel correlation scheme. For example, based on the estimated channel, terminal device 120 may find important components and align them with a predefined codebook, such as a DFT codebook. This scheme ensures that terminal device 120 selects a compression matrix from the codebook to maximize the effective array gain during reception, which is dynamic and depends on the CSI.

[0108] In some embodiments where the second compression mode is enabled, at block 640, terminal device 120 may send an indication of a codebook used to compress the CSI to network device 110 for use by network device 110 in CSI recovery. In some embodiments, terminal device 120 may send an indication of the codebook along with the compressed CSI to network device 110. In some alternative embodiments, terminal device 120 may send an indication of the codebook separately to network device 110. Compared to the second compression mode, in the first compression mode, it is not necessary to estimate the codebook used to obtain the received compression matrix W. (k) channel information, thereby reducing complexity on the terminal device.

[0109] return Figure 5 At block 540, terminal device 120 sends the compressed CSI to network device 110 for use in recovering CSI associated with a second number of ports at network device 110, the second number of ports being no less than the first number of ports. In this manner, CSI compression based on the capabilities of terminal device 120 may facilitate recovery of uncompressed dimension CSI from compressed CSI associated with a limited number of ports.

[0110] The disclosed solution utilizes deep neural networks, which inherently exploit the spatial sparsity of mmWave massive MIMO channels, enabling enhanced CSI recovery for more ports than the number of ports fed back. Furthermore, CSI compression is split between the network equipment and the end device. This is simple to implement because transmit compression is static, requiring no extensive CSI knowledge, and the end device does not need to perform complex compression as with OMP-based solutions.

[0111] Furthermore, compared to the non-precoded case, this solution can reduce CSI-RS overhead. Compared to the beamforming case, this solution offers greater flexibility in designing transmit precoding, as it does not separate beams and "second-stage" precoding. Therefore, it can also support a larger number of transmit layers. Furthermore, this solution does not require CSI feedback. The proposed solution flexibly supports different modes of CSI compression based on the capabilities of different terminal devices.

[0112] The inventors have summarized the differences between this solution and the existing solutions, as shown in Table 1.

[0113] Table 1: Comparison with existing solutions

[0114]

[0115] The inventors performed simulations by considering a millimeter-wave multi-user MIMO system. The simulation settings are shown in Table 2.

[0116] Table 2: Simulation settings

[0117]

[0118] In the simulation, a deep convolutional neural network (CNN) was implemented by MATLAB 2018a. Four CNN units were considered, and L with consecutive units was applied in the convolutional layers. f = [8, 16, 32, 64] filters. The fully connected layer following the connected CNN unit has the same number of neurons as the output size. The total dataset consists of 50,000 samples, 85% of which are used for training and the rest for validation. The training rate is 0.0002 and the gradient descent rate is 0.9. Gradient descent is performed using a mini-batch size of 2048.

[0119] The inventors evaluated the proposed scheme for both the first compression mode (Mode 0) and the second compression mode (Mode 1) using different compression values ​​and compared them with an OMP-based algorithm and a perfect case. The proposed scheme achieved compression values ​​of 8 (8 out of 32, the number of transmit antenna ports) and 1 or 2 (1 or 2 out of 4, the number of receive antennas) for the transmitter, i.e., 8 or 16 out of 128. For the OMP method, the compression value varies with different channel implementations due to the iterative algorithm. The stopping criteria were adjusted to achieve average compression values ​​comparable to those of our proposed method, which achieved 9 and 17 out of 128 cases. Lower compression values ​​indicate higher compression ratios.

[0120] Figure 7 An example comparison 700 of the NMSE performance between the present solution and the existing solution is shown in FIG. Figure 7 As shown in , 710 represents the mode 0 case with a compression value of 16 in this scheme, 720 represents the OMP case with a compression value of 17 in the existing scheme, 730 represents the mode 0 case with a compression value of 8 in this scheme, 740 represents the mode 1 case with a compression value of 8 in this scheme, 750 represents the mode 0 case with a compression value of 16 in this scheme, and 760 represents the OMP case with a compression value of 9 in the existing scheme.

[0121] It can be observed that Mode 0, with a compression value of 16, achieves the best performance, outperforming Mode 1. Although Mode 0, with a value of 8, does not outperform Mode 1, its static beamspace compression with a smaller compression value may lose its adaptability, and a higher compression value may be sufficient. In Mode 1, the compression at the end device varies dynamically with the channel. Therefore, Mode 1's performance is more stable. The inventors also evaluated OMP-based methods with different compression values. The performance was slightly better than Mode 1.

[0122] Figure 8An example comparison of throughput performance between the present solution and the existing solution by CDF is shown in FIG800. The inventors analyze the throughput performance as a function of the number of users using different CSI acquisition schemes, assuming an SNR of 15 dB and considering 2 users. Figure 8 As shown in , 810 represents the mode 1 case with a compression value of 8 in the proposed scheme, 820 represents the OMP case with a compression value of 9 in the existing scheme, 830 represents the mode 0 case with a compression value of 8 in the proposed scheme, 840 represents the mode 1 case with a compression value of 16 in the proposed scheme, 850 represents the OMP case with a compression value of 17 in the existing scheme, 860 represents the mode 0 case with a compression value of 16 in the proposed scheme, and 870 represents the perfect case. It can be observed that the proposed scheme in mode 0 outperforms the schemes in mode 1 and the OMP scheme with similar compression values.

[0123] Figure 9 The figure shows an example comparison of the throughput performance and the number of terminal devices between the present solution and the existing solution through CDF (assuming the SNR is 15dB). Figure 9 As shown in , 910 represents the mode 0 case with a compression value of 8 in this scheme, 920 represents the mode 1 case with a compression value of 16 in this scheme, 930 represents the mode 1 case with a compression value of 8 in this scheme, 940 represents the beamformed CSI-RS case, 950 represents the OMP case with a compression value of 17 in the existing scheme, 960 represents the mode 0 case with a compression value of 16 in this scheme, 970 represents the perfect case, and 980 represents the OMP case with a compression value of 9 in the existing scheme.

[0124] Similar results are shown when considering different numbers of users. The beamformed CSI-RS case is also considered, where a DFT-based codebook is used for the beamforming process. The performance loss is due to the separate "two-stage" precoding, where the "first stage" is beamforming for the CSI-RS and the "second stage" precoding is based on the beamformed CSI design.

[0125] In addition, the inventors performed an overhead analysis. The main overhead in the system is determined by CSI-RS and CSI feedback, as shown in Table 3. The overhead of transmitting CSI-RS depends on the number of CSI-RS transmitted, and therefore on the number of transmit antenna ports. For the proposed scheme, due to the presence of transmit compression, the CSI-RS overhead is O(N T ) and in our example N T =8. For the OMP method, no transmission compression is performed, so the overhead increases with M T And the expansion and in our example M T =32. As for the overhead of CSI feedback, since the quantization bit is represented by Q, the number of CSI feedback bits used for the proposed scheme can be expressed by Calculate, where is the compression value. In this example, it is 2Q·8 or 2Q·16. The OMP-based scheme in the simulation requires 2Q·9 or 2Q·17, where L=9 or 17. In addition, in the proposed method for Mode 1 and the OMP-based scheme, the index of the compression matrix needs to be fed back. The compression matrix is ​​usually selected from a codebook, and its size is determined by the number of antennas. The proposed Mode 1 requires bits to feed back the index of the codebook, which is 1 or 2 bits in our example. For the OMP algorithm, it requires bits are used to feed back the index, which requires 1·log2(32·4)=7 bits.

[0126] Table 2: Cost Analysis

[0127]

[0128] Based on the calculation, the overhead analysis of the schemes considered in the case is shown in Table 4. It can be observed that the proposed scheme significantly reduces the overhead of both CSI-RS and codebook indication.

[0129] Table 3: Cost analysis and comparison of different solutions

[0130]

[0131]

[0132] In summary, based on the performance in terms of estimation error, throughput, and overhead, it can be concluded that the proposed ML-based CSI recovery method from limited CSI feedback shows good performance and reduces the overhead of both CSI-RS and CSI feedback.

[0133] In some embodiments, an apparatus capable of executing method 300 (e.g., network device 110) may include components for executing corresponding steps of method 300. The components may be implemented in any suitable form. For example, the components may be implemented in a circuit system or a software module.

[0134] In some embodiments, the apparatus may include: a component for sending, at a network device, a channel state information reference signal associated with a first number of ports at the network device to a terminal device; a component for receiving, from the terminal device, compressed channel state information generated based on the capabilities of the terminal device; and a component for recovering, from the compressed channel state information, channel state information associated with a second number of ports at the network device, the second number of ports being not less than the first number of ports.

[0135] In some embodiments, the component for sending may include: a component for compressing the original channel state information reference signal associated with the second number of ports into a channel state information reference signal associated with the first number of ports at the network device; and a component for sending the compressed channel state information reference signal to the terminal device.

[0136] In some embodiments, the component for recovery includes: a component for determining whether the compression mode of the channel state information is a first compression mode that performs compression based on antenna characteristics at a terminal device; and a component for recovering the channel state information based on mapping the compressed channel state information to a first predetermined model of the channel state information according to determining that the compression mode of the channel state information is the first compression mode.

[0137] In some embodiments, the apparatus further comprises: means for receiving an indication of a codebook for compressing the channel state information from the terminal device. In some embodiments, the means for recovering comprises: means for determining whether a compression mode of the channel state information is a second compression mode in which compression is performed based on channel propagation characteristics at the terminal device; and means for generating processed channel state information based on the codebook and recovering the channel state information based on mapping the processed channel state information to a second predetermined model of the channel state information in accordance with determining that the compression mode of the channel state information is the second compression mode.

[0138] In some embodiments, the apparatus further comprises: means for transmitting an indicator indicating a compression mode for compressing the channel state information to the terminal device based on a compression mode determined by the network device; and means for receiving an indicator indicating a compression mode for compressing the channel state information from the terminal device based on a compression mode determined by the terminal device. In some further embodiments, the means for recovering comprises means for recovering the channel state information based on the compression mode.

[0139] In some embodiments, an apparatus capable of executing method 500 (e.g., terminal device 120) may include components for executing corresponding steps of method 500. The components may be implemented in any suitable form. For example, the components may be implemented in a circuit system or a software module.

[0140] In some embodiments, the apparatus may include: a component for receiving, at a terminal device and from a network device, a channel state information reference signal associated with a first number of ports at the network device; a component for determining channel state information based on the channel state information reference signal; a component for compressing the channel state information based on a capability of the terminal device; and a component for sending compressed channel state information to the network device for recovering channel state information associated with a second number of ports at the network device, the second number of ports being not less than the first number of ports.

[0141] In some embodiments, the means for receiving includes means for receiving, from a network device, compressed channel state information reference signals associated with a first number of ports, the compressed channel state information reference signals associated with the first number of ports being generated from original channel state information reference signals associated with a second number of ports.

[0142] In some embodiments, the means for compressing includes: a means for determining whether the terminal device enables a first compression mode; and a means for compressing channel state information based on antenna characteristics at the terminal device in response to determining that the terminal device enables the first compression mode.

[0143] In some embodiments, the means for compressing includes: means for determining whether the terminal device has enabled a second compression mode; and means for compressing the channel state information based on channel propagation characteristics at the terminal device in accordance with the determination that the terminal device has enabled the second compression mode. In some further embodiments, the apparatus further includes: means for sending an indication of a codebook used to compress the channel state information to a network device.

[0144] In some embodiments, the apparatus further comprises: means for transmitting an indicator indicating a compression mode for compressing the channel state information to a network device based on a compression mode determined by the terminal device; and means for receiving an indicator indicating a compression mode from the network device based on a compression mode for compressing the channel state information determined by the network device. In some further embodiments, the means for compressing comprises means for compressing the channel state information based on the mode.

[0145] Figure 10 is a simplified block diagram of a device 1000 suitable for implementing embodiments of the present disclosure. The device 1000 may be provided to implement a network device or a terminal device, such as Figure 1 As shown, the device 1000 includes one or more processors 1010, one or more memories 1020 coupled to the processors 1010, and one or more communication modules 1040 (such as transmitters and / or receivers) coupled to the processors 1010.

[0146] The communication module 1040 is used for two-way communication. The communication module 1040 has at least one antenna to facilitate communication. The communication interface can represent any interface required to communicate with other network elements.

[0147] Processor 1010 may be of any type suitable for the local technology network and may include one or more of the following: as non-limiting examples, a general purpose computer, a special purpose computer, a microprocessor, a digital signal processor (DSP), and a processor based on a multi-core processor architecture. Device 1000 may have multiple processors, such as application-specific integrated circuit chips, that are time-controlled by a clock synchronized with a main processor.

[0148] The memory 1020 may include one or more non-volatile memories and one or more volatile memories. Examples of non-volatile memories include, but are not limited to, read-only memory (ROM) 1024, electrically programmable read-only memory (EPROM), flash memory, hard disks, compact disks (CDs), digital video disks (DVDs), and other magnetic and / or optical storage devices. Examples of volatile memories include, but are not limited to, random access memory (RAM) 1022 and other volatile memories that do not persist during a power outage.

[0149] Computer program 1030 includes computer-executable instructions executed by associated processor 1010. Program 1030 may be stored in ROM 1024. Processor 1010 may perform any suitable actions and processes by loading program 1030 into RAM 1022.

[0150] The embodiment of the present disclosure can be implemented with the aid of the program 1030 so that the device 1000 can execute the following steps: Figures 3 to 6 Any process of the present disclosure discussed. The embodiments of the present disclosure may also be implemented by hardware or by a combination of software and hardware.

[0151] In some embodiments, program 1030 may be tangibly embodied in a computer-readable medium that may be included in device 1000 (such as in memory 1020) or other storage device accessible to device 1000. Device 1000 may load program 1030 from the computer-readable medium into RAM 1022 for execution. Computer-readable media may include any type of tangible, non-volatile storage device, such as ROM, EPROM, flash memory, hard disk, CD, DVD, and the like. Figure 11 An example of a computer readable medium 1100 is shown in the form of a CD or DVD. The computer readable medium has a program 1030 stored thereon.

[0152] In general, various embodiments of the present disclosure may be implemented in hardware or dedicated circuits, software, logic, or any combination thereof. Some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software, which may be executed by a controller, microprocessor, or other computing device. Although various aspects of the embodiments of the present disclosure are illustrated and described as block diagrams, flow charts, or using some other graphical representations, it should be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuits or logic, general-purpose hardware or a controller or other computing device, or some combination thereof, as non-limiting examples.

[0153] The present disclosure also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions (such as computer-executable instructions included in program modules) that are executed in a device on a target real or virtual processor to perform the operations described above with reference to Figures 3 to 6 Methods 300 to 600 are described. Generally speaking, program modules include routines, programs, libraries, objects, classes, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of the program modules can be combined or split between program modules as desired. Machine-executable instructions for program modules can be executed on a local device or distributed devices. In distributed devices, program modules can be located in both local media and remote storage media.

[0154] The program code for executing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partially on the machine as a standalone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0155] In the context of the present disclosure, computer program codes or related data may be carried by any suitable carrier to enable a device, apparatus or processor to perform various processes and operations as described above. Examples of carriers include signals, computer-readable media, etc.

[0156] The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of computer readable storage media would include an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0157] In addition, when operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in a continuous order, or that all illustrated operations be performed or that desired results be achieved. In some cases, multitasking and parallel processing can be advantageous. Similarly, although some specific implementation details are contained in the above discussion, these should not be interpreted as limiting the scope of the present disclosure, but rather as describing features that are specific to a particular embodiment. Certain features described in the context of a separate embodiment can also be implemented in a combination of a single embodiment. On the contrary, the various features described in the context of a single embodiment can also be implemented individually in multiple embodiments or in any suitable subcombination.

[0158] Although the disclosure has been described in language specific to structural features and / or methodological acts, it should be understood that the disclosure defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.

Claims

1. A communication method, comprising: At a network device, sending, to a terminal device, channel state information reference signals associated with a first number of ports at the network device; receiving, from the terminal device, compressed channel state information generated based on capabilities of the terminal device, wherein a compression mode of the channel state information comprises one of the following: a first compression mode in which the compression is performed based on antenna characteristics at the terminal device, and a second compression mode in which the compression is performed based on channel propagation characteristics at the terminal device; as well as Channel state information associated with a second number of ports at the network device is recovered from the compressed channel state information, the second number of ports being no less than the first number of ports.

2. The method of claim 1 , wherein the sending comprises: compressing original channel state information reference signals associated with the second number of ports into the channel state information reference signals associated with the first number of ports at the network device; as well as Sending the compressed channel state information reference signal to the terminal device.

3. The method of claim 1 , wherein the restoring comprises: determining whether the compression mode of the channel state information is the first compression mode in which the compression is performed based on antenna characteristics at the terminal device; as well as According to determining that the compression mode of the channel state information is the first compression mode, the channel state information is restored based on mapping the compressed channel state information to a first predetermined model of the channel state information.

4. The method according to claim 1, further comprising: receiving from the terminal device an indication of a codebook for compressing the channel state information, and The recovery includes: determining whether the compression mode of the channel state information is the second compression mode in which the compression is performed based on channel propagation characteristics at the terminal device; and According to determining that the compression mode of the channel state information is the second compression mode, generating processed channel state information based on the codebook; and The channel state information is restored based on mapping the processed channel state information to a second predetermined model of the channel state information.

5. The method according to claim 1, further comprising: The network device determines a compression mode of the compressed channel state information, and sends an indicator indicating the mode to the terminal device; as well as The mode of compression of the compressed channel state information is determined by the terminal device, an indicator indicating the mode is received from the terminal device, and The recovery includes: The channel state information is recovered based on a predetermined model associated with the mode, the predetermined model mapping the compressed channel state information to the channel state information.

6. A communication method, comprising: receiving, at a terminal device, from a network device, channel state information reference signals associated with a first number of ports at the network device; determining channel state information based on the channel state information reference signal; compressing the channel state information based on a capability of the terminal device, wherein a compression mode of the channel state information comprises one of the following: a first compression mode in which the compression is performed based on antenna characteristics at the terminal device, and a second compression mode in which the compression is performed based on channel propagation characteristics at the terminal device; as well as The compressed channel state information is sent to the network device for use in recovering channel state information associated with a second number of ports at the network device, the second number of ports being not less than the first number of ports.

7. The method of claim 6, wherein the receiving comprises: Compressed channel state information reference signals associated with the first number of ports are received from the network device, the compressed channel state information reference signals associated with the first number of ports being generated from original channel state information reference signals associated with the second number of ports.

8. The method of claim 6, wherein the compressing comprises: Determining whether the terminal device enables the first compression mode; as well as According to determining that the terminal device enables the first compression mode, the channel state information is compressed based on antenna characteristics at the terminal device.

9. The method of claim 6, wherein the compressing comprises: Determining whether the terminal device enables the second compression mode; as well as compressing the channel state information based on channel propagation characteristics at the terminal device according to determining that the terminal device enables the second compression mode, and The method further comprises: An indication of a codebook for compressing the channel state information is sent to the network device.

10. The method according to claim 6, further comprising: According to the compression mode of the compressed channel state information being determined by the terminal device, an indicator indicating the mode is sent to the network device; as well as A compression mode of the compressed channel state information is determined by the network device, an indicator indicating the mode is received from the network device, and The compression includes: The channel state information is compressed based on the pattern.

11. A network device comprising: at least one processor; as well as at least one memory including computer program code; The at least one memory and the computer program code are configured to, with the at least one processor, cause the network device to: At the network device, sending, to a terminal device, channel state information reference signals associated with a first number of ports at the network device; receiving, from the terminal device, compressed channel state information generated based on capabilities of the terminal device, wherein a compression mode of the channel state information comprises one of the following: a first compression mode in which the compression is performed based on antenna characteristics at the terminal device, and a second compression mode in which the compression is performed based on channel propagation characteristics at the terminal device; as well as Channel state information associated with a second number of ports at the network device is recovered from the compressed channel state information, the second number of ports being no less than the first number of ports.

12. The network device according to claim 11, wherein the network device is caused to send the channel state information reference signal by: compressing original channel state information reference signals associated with the second number of ports into the channel state information reference signals associated with the first number of ports at the network device; and Sending the compressed channel state information reference signal to the terminal device.

13. The network device of claim 11 , wherein the network device is caused to recover the channel state information by: determining whether the compression mode of the channel state information is the first compression mode in which the compression is performed based on antenna characteristics at the terminal device; and According to determining that the compression mode of the channel state information is the first compression mode, the channel state information is restored based on mapping the compressed channel state information to a first predetermined model of the channel state information.

14. The network device of claim 11, wherein the network device is further configured to: receiving from the terminal device an indication of a codebook for compressing the channel state information, and The network device is configured to recover the channel state information by: determining whether the compression mode of the channel state information is the second compression mode in which the compression is performed based on channel propagation characteristics at the terminal device; as well as According to determining that the compression mode of the channel state information is the second compression mode, generating processed channel state information based on the codebook; as well as The channel state information is restored based on mapping the processed channel state information to a second predetermined model of the channel state information.

15. The network device of claim 11, wherein the network device is further configured to: According to the compression mode of the compressed channel state information being determined by the network device, an indicator indicating the mode is sent to the terminal device; and The mode of compression of the compressed channel state information is determined by the terminal device, an indicator indicating the mode is received from the terminal device, and The network device is configured to recover the channel state information by: The channel state information is recovered based on a predetermined model associated with the mode, the predetermined model mapping the compressed channel state information to the channel state information.

16. A terminal device comprising: at least one processor; as well as at least one memory including computer program code; The at least one memory and the computer program code are configured to, together with the at least one processor, cause the terminal device to: receiving, at the terminal device, from a network device, channel state information reference signals associated with a first number of ports at the network device; determining channel state information based on the channel state information reference signal; compressing the channel state information based on a capability of the terminal device, wherein a compression mode of the channel state information comprises one of the following: a first compression mode in which the compression is performed based on antenna characteristics at the terminal device, and a second compression mode in which the compression is performed based on channel propagation characteristics at the terminal device; as well as The compressed channel state information is sent to the network device for use in recovering channel state information associated with a second number of ports at the network device, the second number of ports being not less than the first number of ports.

17. The terminal device according to claim 16, wherein the terminal device is caused to receive the channel state information reference signal by: Compressed channel state information reference signals associated with the first number of ports are received from the network device, the compressed channel state information reference signals associated with the first number of ports being generated from original channel state information reference signals associated with the second number of ports.

18. The terminal device of claim 16, wherein the terminal device is caused to compress the channel state information by: determining whether the terminal device enables the first compression mode; and According to determining that the terminal device enables the first compression mode, the channel state information is compressed based on antenna characteristics at the terminal device.

19. The terminal device of claim 16, wherein the terminal device is caused to compress the channel state information by: determining whether the terminal device enables the second compression mode; and compressing the channel state information based on channel propagation characteristics at the terminal device according to determining that the terminal device enables the second compression mode, and The terminal device is further configured to: An indication of a codebook for compressing the channel state information is sent to the network device.

20. The terminal device according to claim 16, wherein the terminal device is further configured to: According to the compression mode of the compressed channel state information being determined by the terminal device, an indicator indicating the mode is sent to the network device; and A compression mode of the compressed channel state information is determined by the network device, an indicator indicating the mode is received from the network device, and The terminal device is configured to compress the channel state information by: The channel state information is compressed based on the pattern.

21. A non-transitory computer-readable medium comprising program instructions, wherein the program instructions are configured to cause an apparatus to execute the method according to any one of claims 1 to 5.

22. A non-transitory computer-readable medium comprising program instructions for causing an apparatus to execute the method according to any one of claims 6 to 10.

Images