Channel estimation method, device, terminal and network side equipment

Abstract

The present application discloses a channel estimation method, device, terminal and network side equipment, belonging to the field of communication technology. The channel estimation method includes: the terminal receives a pilot signal sent by a network side device, and the resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different; according to the pilot signal on the first time domain transmission unit and the second time domain transmission unit, the channel estimation is performed on the third time domain transmission unit. The technical solution of the embodiment of the present application can reduce the pilot overhead, improve the accuracy of channel estimation and prediction, and improve the performance of the communication system.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a channel estimation method, apparatus, terminal and network-side equipment. Background Art

[0002] The massive antenna array formed by massive MIMO technology can support more users to send and receive signals at the same time, thereby increasing the channel capacity and data traffic of the mobile network by dozens of times or more, while drastically reducing interference between multiple users.

[0003] However, in massive MIMO systems, as the antenna size increases dramatically, the pilot overhead and the complexity of channel estimation increase by orders of magnitude, which has become one of the key bottlenecks restricting the large-scale commercial use of massive MIMO.

[0004] In existing communication systems, pilot channel state information-reference signal (CSI-RS) transmission is based on resource block radio bearer (RB) as the basic unit. Each resource block RB contains 12 subcarriers in the frequency domain and 6-7 OFDM symbols in the time domain. CSI-RS is sent on each RB, which will lead to large pilot overhead and affect the performance of the communication system. Summary of the invention

[0005] The embodiments of the present application provide a channel estimation method, apparatus, terminal, and network-side equipment, which can reduce pilot overhead and improve the performance of a communication system.

[0006] In a first aspect, an embodiment of the present application provides a channel estimation method, including:

[0007] The terminal receives a pilot signal sent by a network-side device, and resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different;

[0008] Channel estimation is performed on the third time domain transmission unit according to the pilot signals on the first time domain transmission unit and the second time domain transmission unit.

[0009] In a second aspect, an embodiment of the present application provides a channel estimation method, including:

[0010] The network side device sends a pilot signal, and resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different.

[0011] In a third aspect, an embodiment of the present application provides a channel estimation device, applied to a terminal, the device comprising:

[0012] A receiving module, configured to receive a pilot signal sent by a network-side device, wherein resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different;

[0013] The channel estimation module is used to perform channel estimation on the third time domain transmission unit according to the pilot signals on the first time domain transmission unit and the second time domain transmission unit.

[0014] In a fourth aspect, an embodiment of the present application provides a channel estimation device, which is applied to a network side device, and the device includes:

[0015] The sending module is used to send a pilot signal, and the resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different.

[0016] In a fifth aspect, a terminal is provided, comprising a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the method described in the first aspect.

[0017] In the sixth aspect, a terminal is provided, comprising a processor and a communication interface, wherein the communication interface is used to receive a pilot signal sent by a network side device, and the resource blocks RB occupied by the pilot signals in the first time domain transmission unit and the second time domain transmission unit are at least partially different; the processor is used to perform channel estimation on the third time domain transmission unit based on the pilot signals on the first time domain transmission unit and the second time domain transmission unit.

[0018] In the seventh aspect, a network side device is provided, which includes a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the method described in the second aspect.

[0019] In an eighth aspect, a network side device is provided, comprising a processor and a communication interface, wherein the communication interface is used to send a pilot signal, and resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different.

[0020] In a ninth aspect, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented, or the steps of the method described in the second aspect are implemented.

[0021] In the tenth aspect, a chip is provided, comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run a program or instruction to implement the method described in the first aspect, or to implement the method described in the second aspect.

[0022] In the eleventh aspect, a computer program / program product is provided, wherein the computer program / program product is stored in a non-volatile storage medium, and the program / program product is executed by at least one processor to implement the steps of the method described in the first aspect or the second aspect.

[0023] In an embodiment of the present application, a network side device sends a pilot signal to a terminal, and resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different. In this way, the pilot signal is sent only on part of the RBs, which can reduce the pilot overhead and improve the performance of the communication system. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic diagram showing a wireless communication system;

[0025] Figure 2 A schematic diagram showing a flow chart of a channel estimation method performed by a terminal according to an embodiment of the present application;

[0026] Figure 3 A schematic diagram showing a flow chart of a channel estimation method performed by a network side device according to an embodiment of the present application;

[0027] Figure 4 and Figure 5 A schematic diagram showing the sending of CSI-RS in an embodiment of the present application;

[0028] Figure 6 A schematic diagram showing the structure of a channel estimation device applied to a terminal according to an embodiment of the present application;

[0029] Figure 7 A schematic diagram showing the structure of a channel estimation device applied to a network side device according to an embodiment of the present application;

[0030] Figure 8 A schematic diagram showing the composition of a communication device according to an embodiment of the present application;

[0031] Fig. 9 A schematic diagram showing the composition of a terminal according to an embodiment of the present application;

[0032] Fig.10 A schematic diagram showing the composition of a network side device according to an embodiment of the present application;

[0033] Fig.11 Represents the informer network structure diagram;

[0034] Fig.12 Schematic diagram showing the multi-head attention mechanism;

[0035] Fig.13 Schematic diagram showing the masked multi-head attention mechanism. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of this application.

[0037] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way are interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first" and "second" are generally of the same type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally represents that the objects associated with each other are in an "or" relationship.

[0038] It is worth noting that the technology described in the embodiments of the present application is not limited to the Long Term Evolution (LTE) / LTE-Advanced (LTE-A) system, but can also be used in other wireless communication systems, such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single-carrier Frequency Division Multiple Access (SC-FDMA) and other systems. The terms "system" and "network" in the embodiments of the present application are often used interchangeably, and the described technology can be used for the above-mentioned systems and radio technologies as well as for other systems and radio technologies. The following description describes a new radio (NR) system for example purposes, and NR terms are used in most of the following descriptions, but these technologies can also be applied to applications other than NR system applications, such as the 6th Generation (6G) communication system.

[0039] Figure 1A block diagram of a wireless communication system applicable to an embodiment of the present application is shown. The wireless communication system includes a terminal 11 and a network side device 12. Among them, the terminal 11 can also be called a terminal device or a user terminal (User Equipment, UE), and the terminal 11 can be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer) or a notebook computer, a personal digital assistant (Personal Digital Assistant, PDA), a handheld computer, a netbook, an ultra-mobile personal computer (ultra-mobile personal computer, UMPC), a mobile Internet device (Mobile Internet Device, MID), a wearable device (Wearable Device) or a vehicle-mounted device (VUE), a pedestrian terminal (PUE) and other terminal side devices, and the wearable device includes: a smart watch, a bracelet, a headset, glasses, etc. It should be noted that the specific type of the terminal 11 is not limited in the embodiment of the present application. The network side device 12 can be a base station or a core network, wherein the base station can be called a node B, an evolved node B, an access point, a base transceiver station (Base Transceiver Station, BTS), a radio base station, a radio transceiver, a basic service set (Basic Service Set, BSS), an extended service set (Extended Service Set, ESS), a B node, an evolved B node (eNB), a home B node, a home evolved B node, a WLAN access point, a WiFi node, a transmitting and receiving point (Transmitting Receiving Point, TRP) or some other suitable term in the field. As long as the same technical effect is achieved, the base station is not limited to a specific technical vocabulary. It should be noted that in the embodiment of the present application, only the base station in the NR system is taken as an example, but the specific type of the base station is not limited. The core network device can be a location management device, for example, a location management function (LMF, E-SLMC), etc.

[0040] The massive antenna array formed by massive multiple-input multiple-output (MIMO) technology can support more users to send and receive signals at the same time, thereby increasing the channel capacity and data traffic of the mobile network by dozens of times or more, and at the same time can achieve a sharp reduction in interference between multiple users. Therefore, it has been receiving continuous and high attention from researchers since it was proposed. In order to support broadband wireless communications, orthogonal frequency division multiplexing (OFDM) has become the underlying technology of mobile communications since 4G. This technology can effectively combat multipath interference and divide the frequency-domain frequency selective channel into multiple flat-fading sub-channels to support wireless transmission. OFDM combined with massive MIMO is already the basic framework for current and future wireless communications.

[0041] However, in massive MIMO systems, as the antenna size increases dramatically, the pilot overhead and the complexity of channel estimation increase by orders of magnitude, which has become one of the key bottlenecks restricting the large-scale commercial use of massive MIMO.

[0042] In existing communication systems, the pilot channel state information-reference signal (CSI-RS) is sent in resource block radio bearer (RB) as the basic unit. Each resource block RB contains 12 subcarriers in the frequency domain and 6-7 OFDM symbols in the time domain. CSI-RS is sent on each RB. In order to reduce pilot overhead, the pilot signal CSI-RS is not sent in each transmission time interval (TTI), but once every few TTIs. Therefore, in the TTI where CSI-RS is not sent, the real large-scale MIMO channel cannot be known. For this reason, the relevant technology generally uses the channel estimated by the TTI in which the CSI-RS was sent most recently as the channel of the current TTI, thereby finding the optimal precoding from the codebook. Since the channel is changing, there is a deviation between the channel estimated by the TTI in which the CSI-RS was sent most recently and the channel of the current TTI, and the faster the movement speed, the greater the deviation, which leads to deviations in the channel feedback and the precoding operation at the transmitting end, causing a decrease in system performance.

[0043] The present application embodiment provides a channel estimation method, which is executed by a terminal, such as Figure 2 As shown, the method includes:

[0044] Step 101: The terminal receives a pilot signal sent by a network-side device, and resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different;

[0045] Step 102: Perform channel estimation on the third time domain transmission unit according to the pilot signals on the first time domain transmission unit and the second time domain transmission unit.

[0046] Among them, the third time domain transmission unit can be the same as or different from the first time domain transmission unit; the third time domain transmission unit can be the same as or different from the second time domain transmission unit; the first time domain transmission unit and the second time domain transmission unit are different time domain transmission units.

[0047] In an embodiment of the present application, the terminal receives a pilot signal sent by a network side device, and the resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different, that is, the network side device only sends the pilot signal on part of the RBs, which can reduce the pilot overhead; in addition, if the number of RBs for sending pilot signals remains unchanged, since the pilot signal is only sent on part of the RBs in the time domain transmission unit, the number of time domain transmission units for sending pilot signals can be increased, and the interval between the time domain transmission units for sending pilot signals is reduced, which can improve the accuracy of channel estimation and improve the performance of the communication system.

[0048] In some embodiments, performing channel estimation on the third time domain transmission unit according to the pilot signals on the first time domain transmission unit and the second time domain transmission unit includes:

[0049] Channel estimation is performed based on the pilot signal on the first time domain transmission unit, the pilot signal on the second time domain transmission unit and a pre-trained neural network model to obtain channels on all RBs in the third time domain transmission unit.

[0050] In this embodiment, the pilot signal on the first time domain transmission unit and the pilot signal on the second time domain transmission unit can be used to obtain the channel on the RB that does not send the pilot signal. Since the RB that sends the pilot signal and the RB that does not send the pilot signal can be located in the same time domain transmission unit and have a strong correlation, the accuracy of channel estimation can be improved and the performance of the communication system can be improved.

[0051] In some embodiments, performing channel estimation on the third time domain transmission unit according to the pilot signals on the first time domain transmission unit and the second time domain transmission unit includes:

[0052] Channel estimation is performed based on the pilot signal on the first time domain transmission unit, the pilot signal on the second time domain transmission unit and a pre-trained neural network model to obtain channels on all RBs in L time domain transmission units after the current time domain transmission unit, where L is a positive integer.

[0053] This embodiment can also use the pilot signal on the first time domain transmission unit and the pilot signal on the second time domain transmission unit to predict future channels, and can predict future channels without pilot conditions, thereby greatly reducing large pilot overhead and improving system performance.

[0054] In some embodiments, all the RBs are RBs determined according to a predefined rule or a preconfigured manner.

[0055] In some embodiments, the RBs determined according to a predefined rule or a preconfigured manner include at least RBs occupied by the pilot signal in the first time slot transmission unit and RBs occupied by the pilot signal in the second time slot transmission unit.

[0056] In order to ensure the accuracy of channel estimation and channel prediction, the pilot signals on the nearest K time domain transmission units that send pilot signals are used for channel estimation, where K is a positive integer. That is, the channel estimation is performed using the current time domain transmission unit and the K-1 time domain transmission units that sent pilot signals before the current time domain transmission unit.

[0057] In some embodiments, the pilot signal is sent on half of the RB, and the terminal receives the pilot signal sent by the network side device, including:

[0058] The terminal receives a pilot signal sent by a network side device on a first RB of a first time domain transmission unit;

[0059] The terminal receives a pilot signal sent by a network side device on a second RB of a second time domain transmission unit;

[0060] The first RB is different from the second RB, the second time domain transmission unit is the time domain transmission unit that sends the pilot signal most recently after the first time domain transmission unit, and the second time domain transmission unit is spaced from the first time domain transmission unit by at least one time domain transmission unit.

[0061] In this embodiment, the pilot signal is not sent on all time domain transmission units, but is sent at intervals of one or more time domain transmission units. If the number of RBs for sending the pilot signal on each time domain transmission unit remains unchanged, the pilot overhead can be reduced.

[0062] In some embodiments, the sequence number of the first RB is an even number, and the sequence number of the second RB is an odd number; or

[0063] The sequence number of the first RB is an odd number, and the sequence number of the second RB is an even number.

[0064] For example, in a certain pilot signal sending time domain transmission unit, if you choose to receive the pilot signal on the RB with an odd sequence number, then in the next pilot signal sending time domain transmission unit, you choose to receive the pilot signal on the RB with an even sequence number; or, in a certain pilot signal sending time domain transmission unit, if you choose to receive the pilot signal on the RB with an even sequence number, then in the next pilot signal sending time domain transmission unit, you choose to receive the pilot signal on the RB with an odd sequence number.

[0065] This embodiment also includes a step of training to obtain a neural network model, wherein the step of training to obtain the neural network model includes at least one of the following:

[0066] Establish steps to build the initial model of neural network;

[0067] A training step, using training data to train the neural network initial model to obtain a trained neural network initial model; wherein the training data includes input data of the neural network initial model and corresponding actual channel data, the input data is the channel estimation results of K consecutive time domain transmission units that send pilot signals, the length of the channel estimation result of each time domain transmission unit is M×N, and the output of the neural network initial model is the channel estimation results of N antennas on all M RBs in the Kth time domain transmission unit that sends the pilot signal and the subsequent L consecutive time domain transmission units, where M, N, K, and L are positive integers;

[0068] The judgment step is as follows: if the mean square error between the output of the trained neural network initial model and the corresponding actual channel data is greater than or equal to a preset threshold or does not reach a preset number of iterations, turn to the training step; if the mean square error between the output of the trained neural network initial model and the corresponding actual channel data is less than a preset threshold or reaches a preset number of iterations, use the trained neural network initial model as the neural network model.

[0069] In some embodiments, the step of training to obtain the neural network model includes at least one of the following:

[0070] A training step, using the training data to train the neural network model obtained in the last training to obtain a trained neural network model; wherein the training data includes the input data of the neural network model obtained in the last training and the corresponding actual channel data, the input data is the channel estimation results of K consecutive time domain transmission units that send pilot signals, the length of the channel estimation result of each time domain transmission unit is M×N, and the output of the neural network model obtained in the last training is the channel estimation results of N antennas on all M RBs in the Kth time domain transmission unit that sends the pilot signal and the subsequent L consecutive time domain transmission units, and M, N, K, and L are positive integers;

[0071] The judgment step is as follows: if the mean square error between the output of the trained neural network model and the corresponding actual channel data is greater than or equal to a preset threshold or has not reached a preset number of iterations, turn to the training step; if the mean square error between the output of the trained neural network model and the corresponding actual channel data is less than a preset threshold or has reached a preset number of iterations, use the trained neural network model as the neural network model.

[0072] In which, the neural network model includes a first neural network model and a second neural network model. In the input data for training the first neural network model, the pilot signal in the first time domain transmission unit among the K consecutive time domain transmission units for sending pilot signals is sent on the RB with an even number; in the input data for training the second neural network model, the pilot signal in the first time domain transmission unit among the K consecutive time domain transmission units for sending pilot signals is sent on the first RB with an odd number.

[0073] Since the pilot signal in the first time domain transmission unit can be sent on RBs with even numbers or on RBs with odd numbers, in order to improve the accuracy of signal estimation and channel prediction, it is necessary to train neural network models separately for these two cases. After training the neural network models, the two neural network models can be used alternately for channel estimation and prediction.

[0074] In some embodiments, performing channel estimation according to the neural network model includes:

[0075] If the pilot signal in the first received time domain transmission unit is sent on an RB with an even number, the first neural network model is used to perform channel estimation; if the pilot signal in the first received time domain transmission unit is sent on an RB with an odd number, the second neural network model is used to perform channel estimation.

[0076] In this embodiment, the pilot signal includes at least one of the following:

[0077] Channel State Information Reference Signal CSI-RS, Sounding Reference Signal SRS, Demodulation Reference Signal DMRS. The time domain transmission unit includes any one of the following: Transmission Time Interval TTI, subframe, millisecond, time slot and symbol.

[0078] The present application also provides a channel estimation method, which is performed by a network side device, such as Figure 3 As shown, the method includes:

[0079] Step 201: A network-side device sends a pilot signal, and resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different.

[0080] In the embodiment of the present application, in order to save pilot overhead, the network side device does not send a pilot signal on the RB of each time domain transmission unit, but sends a pilot signal on part of the RB. In each time domain transmission unit, the transmission of the pilot signal is based on the resource block RB as the basic unit. In addition, if the number of RBs for sending pilot signals remains unchanged, since the pilot signal is only sent on part of the RBs in the time domain transmission unit, the time domain transmission units for sending pilot signals can be increased, and the interval between the time domain transmission units for sending pilot signals is reduced, which can improve the accuracy of channel estimation and improve the performance of the communication system.

[0081] In addition, in order to further save pilot overhead, the network side device may also send a pilot signal once every several time domain transmission units. The network side device sending the pilot signal includes:

[0082] Sending a pilot signal on a first RB of a first time domain transmission unit;

[0083] Sending a pilot signal on a second RB of a second time domain transmission unit;

[0084] The first RB is different from the second RB, the second time domain transmission unit is the time domain transmission unit that sends the pilot signal most recently after the first time domain transmission unit, and the second time domain transmission unit is spaced from the first time domain transmission unit by at least one time domain transmission unit.

[0085] In some embodiments, the sequence number of the first RB is an even number, and the sequence number of the second RB is an odd number; or

[0086] The sequence number of the first RB is an odd number, and the sequence number of the second RB is an even number.

[0087] That is, if in a certain pilot signal transmission time domain transmission unit, the pilot signal is selected to be sent on an odd-numbered RB, then in the next pilot signal transmission time domain transmission unit, the pilot signal is selected to be sent on an even-numbered RB. And so on, in each different pilot signal transmission time domain transmission unit, the network side device alternately selects to send the pilot signal on an odd-numbered RB and an even-numbered RB.

[0088] In this embodiment, the pilot signal includes at least one of the following:

[0089] Channel State Information Reference Signal CSI-RS, Sounding Reference Signal SRS, Demodulation Reference Signal DMRS. The time domain transmission unit includes any one of the following: Transmission Time Interval TTI, subframe, millisecond, time slot and symbol.

[0090] Taking the pilot signal as CSI-RS and the time domain transmission unit as TTI as an example, this embodiment is based on the OFDM-based large-scale MIMO system, and there are N antennas at the transmitting end. In order to enable the terminal user to obtain the channel information of the downlink large-scale MIMO, the network side device sends the pilot CSI-RS of N ports. In order to reduce the pilot overhead, the pilot signal CSI-RS is not sent in every TTI, but once every S TTI. In each TTI, the pilot CSI-RS is sent based on RB as the basic unit. Each RB contains 12 subcarriers in the frequency domain and multiple OFDM symbols in the time domain (equal to the OFDM symbols contained in each TTI). Assuming that each TTI has M RBs and each RB has a sequence number, the pilot design and resource allocation are performed based on RB as the basic unit. In this embodiment, Figure 4 As shown in the figure, the pilot CSI-RS is not sent on every RB, but only on half of the RBs. If in a certain CSI-RS transmission TTI, the CSI-RS is sent on an odd-numbered RB, then in the next CSI-RS transmission TTI, the CSI-RS is sent on an even-numbered RB. And so on, in each different CSI-RS transmission TTI, the network side device alternately selects to send CSI-RS on an odd-numbered RB and an even-numbered RB.

[0091] After the receiving end (i.e., the terminal) receives the signal on the TTI where the network side device sends CSI-RS, it can use the CSI-RS therein to estimate the channels of the N antennas on the corresponding RB, but it cannot accurately know the channels on the RBs where CSI-RS is not sent on the TTI. In addition, the channels of the N antennas on all RBs in the S TTIs between the TTI and the next TTI where CSI-RS is sent cannot be accurately known. This embodiment uses the channels on each RB estimated from the previous K TTIs where CSI-RS was sent. The neural network model based on the attention mechanism can not only estimate the channels of the N antennas on the RBs where CSI-RS is not sent on the current TTI (represented by the Kth TTI), but can also predict the channels on all RBs in the next L TTIs where CSI-RS is not sent, such as Figure 5 shown.

[0092] In this embodiment, the neural network model adopts an informer network structure based on the attention mechanism. The neural network model adopts an encoder-decoder architecture, such as Fig.11 shown.

[0093] Among them, the encoder consists of multiple layers of multi-head ProbSparse Self-attention and Self-attention Distilling.

[0094] For multi-head ProbSparse Self-attention, such as Fig.12 As shown, its input is three vectors Q, V and K, and its calculation process is as follows:

[0095] 1. Q, V, K are transformed n times linearly to obtain n groups of Q, K, V, where n corresponds to n-head.

[0096] 2. For each set of Q, K, V, the corresponding output Head is obtained through Attition (scaled Dot-product Attention). The calculation method is as follows:

[0097]

[0098] 3. Concatenate all heads and linearly map them to the final output.

[0099] For Self-attention Distilling, self-attention concentration is to find that as the number of Encoder layers increases, the length of the input sequence can be shortened because the output of each position in the sequence already contains the information of other elements in the sequence (the responsibility of self-attention). Therefore, the Encoder has a pyramid-like structure.

[0100] As a natural result of ProbSparse Self-attention, the feature map of the encoder will bring redundant combinations, and distilling is used to privilege the dominant features with dominant features and generate focus self-attention feature maps in the next layer. Inspired by the dilated convolution, the "distilling" process here is advanced from the jth layer to j+1 as follows:

[0101]

[0102] Where [.]AB contains the basic operations in the multi-head ProbSparse self-attention and attention blocks. Conv1d(·) uses the ELU(·) activation function to perform a one-dimensional convolution filter (kernel width = 3) in the time dimension. A maximum pooling layer with stride 2 and down-sample is added here.

[0103] The decoder consists of a single layer of multi-head ProbSparse Self-attention and Masked multi-headattention.

[0104] like Fig.12 As shown, the input of multi-head attention is three vectors Q, V and K, and the calculation process is as follows:

[0105] 1. Q, V, K are transformed n times linearly to obtain n groups of Q, K, V, where n corresponds to n-head.

[0106] 2. For each set of Q, K, V, the corresponding output Head is obtained through Attition (scaled Dot-product Attention). The calculation method is as follows

[0107]

[0108] 3. Concatenate all heads and linearly map them to the final output.

[0109] Masked multi-head attention is applied to ProbSparse self-attention calculation. It prevents each position from paying attention to future positions, thus avoiding autoregression. Fig.13 As shown, the result of the dot product is set to negative infinity.

[0110] The input of the neural network model in this embodiment is K sequences X t , each sequence length is M×N, the element value is the channel of N antennas on M RBs estimated by CSI-RS, and the channel value on RBs where CSI-RS is not sent is replaced by 0. Therefore, X t is a MN×K matrix.

[0111] If, among the K TTIs used for prediction, the CSI-RS in the first TTI is sent on an odd-numbered RB, X t It can be expressed as:

[0112]

[0113] in Indicates the channel of the jth antenna on the i-th RB in the k-th TTI. t i For X t The i-th column of , that is, the channel estimation result on the i-th TTI.

[0114] If, among the K TTIs used for prediction, the CSI-RS in the first TTI is sent on an even-numbered RB, X t It can be expressed as:

[0115]

[0116] The output of the neural network model is L+1 sequences Y t , are the estimated and predicted values ​​of the channels of N antennas on all M RBs in the Kth TTI and the subsequent L consecutive TTIs.

[0117]

[0118] in It represents the predicted value of the channel of the jth antenna on the i-th RB on the k-th TTI.

[0119] In this embodiment, when the neural network model is used for channel estimation and prediction, first, the channel vector sequence X estimated by the CSI-RS on K TTIs is t The encoder is input into the encoder, which includes a multi-layer attention mechanism composite network, each layer of which is composed of multi-header ProbSparse self-attention and Self-attention Distilling. After passing through the multi-layer attention mechanism composite network, the encoder output B is generated t Then B t Input to the decoder. The decoder has only one layer, which consists of Masked ProbSparse multi-head attention and multi-head attention. The input of Masked ProbSparse multi-head attention is a vector sequence X of length K / 2+L+1. t , its first K / 2 sequences are X t In the second half of , the last L+1 vector sequences are all 0 vectors. The output of Masked ProbSparse multi-head attention is passed to the multi-head attention together with the output of the encoder to generate the output of multi-head attention. The output of multi-head attention is then passed through a fully connected network to output the prediction result Y t .

[0120] In order to use the informer network structure based on the attention mechanism for channel estimation and prediction, the network parameters of the informer network need to be trained. During training, the X t The CSI-RS channel estimation from each TTI is matched to the actual channel H from K to K+L TTI. t The informer network is trained using multiple sets of training data. Each set of training data includes the input data of the informer network (CSI-RS channel estimation results at each TTI) and the corresponding actual channel data H t , using the channel estimation results of K consecutive TTIs for training, the training optimization target is the output Y of the neural network model t and the actual channel H t The mean square error between is the smallest.

[0121]

[0122] Alternatively, the goal of training optimization may be to achieve a preset number of iterations, such as 100.

[0123]

[0124] Since the pilot signal in the first TTI can be sent on RBs with even numbers or on RBs with odd numbers, in order to improve the accuracy of signal estimation and channel prediction, it is necessary to train neural network models separately for these two cases. After training the neural network models, the two neural network models can be used alternately for channel estimation and prediction.

[0125] This embodiment can be applied to application scenarios where LTE, GSM and CDMA technologies adopt large-scale MIMO. By using the channel estimation and prediction method based on the attention mechanism of this embodiment and the known channel information, it is possible to predict future channels without pilot signals, thereby greatly reducing the pilot overhead and improving the performance of the system.

[0126] It should be noted that the channel estimation method provided in the embodiment of the present application can be executed by a channel estimation device or a module in the channel estimation device for executing the loaded channel estimation method. In the embodiment of the present application, the channel estimation method provided in the embodiment of the present application is described by taking the channel estimation device executing the loaded channel estimation method as an example.

[0127] The embodiment of the present application provides a channel estimation device, which is applied to a terminal 300. Figure 6 As shown, the device comprises:

[0128] The receiving module 310 is configured to receive a pilot signal sent by a network-side device, wherein resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different;

[0129] The channel estimation module 320 is used to perform channel estimation on the third time domain transmission unit according to the pilot signals on the first time domain transmission unit and the second time domain transmission unit.

[0130] In an embodiment of the present application, the terminal receives a pilot signal sent by a network side device, and the resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different, that is, the network side device only sends the pilot signal on part of the RB, which can reduce the pilot overhead and improve the performance of the communication system.

[0131] In some embodiments, the channel estimation module is specifically used to perform channel estimation based on the pilot signal on the first time domain transmission unit, the pilot signal on the second time domain transmission unit and a pre-trained neural network model to obtain the channel on all RBs in the third time domain transmission unit.

[0132] In some embodiments, the channel estimation module is specifically used to perform channel estimation based on the pilot signal on the first time domain transmission unit, the pilot signal on the second time domain transmission unit and a pre-trained neural network model to obtain the channels on all RBs in L time domain transmission units after the current time domain transmission unit, where L is a positive integer.

[0133] In some embodiments, all the RBs are RBs determined according to a predefined rule or a preconfigured manner.

[0134] In some embodiments, the RBs determined according to a predefined rule or a preconfigured manner include at least RBs occupied by the pilot signal in the first time slot transmission unit and RBs occupied by the pilot signal in the second time slot transmission unit.

[0135] In some embodiments, the channel estimation module is specifically configured to perform channel estimation using pilot signals on the most recent K time domain transmission units that send pilot signals, where K is a positive integer.

[0136] In some embodiments, the receiving module is specifically used to receive a pilot signal sent by a network side device on a first RB of a first time domain transmission unit; receive a pilot signal sent by a network side device on a second RB of a second time domain transmission unit;

[0137] The first RB is different from the second RB, the second time domain transmission unit is the time domain transmission unit that sends the pilot signal most recently after the first time domain transmission unit, and the second time domain transmission unit is spaced from the first time domain transmission unit by at least one time domain transmission unit.

[0138] In some embodiments, the sequence number of the first RB is an even number, and the sequence number of the second RB is an odd number; or

[0139] The sequence number of the first RB is an odd number, and the sequence number of the second RB is an even number.

[0140] In some embodiments, a training module is further included, for training to obtain the neural network model;

[0141] The step of training to obtain the neural network model includes at least one of the following:

[0142] Establish steps to build the initial model of neural network;

[0143] A training step, using training data to train the neural network initial model to obtain a trained neural network initial model; wherein the training data includes input data of the neural network initial model and corresponding actual channel data, the input data is the channel estimation results of K consecutive time domain transmission units that send pilot signals, the length of the channel estimation result of each time domain transmission unit is M×N, and the output of the neural network initial model is the channel estimation results of N antennas on all M RBs in the Kth time domain transmission unit that sends the pilot signal and the subsequent L consecutive time domain transmission units, where M, N, K, and L are positive integers;

[0144] The judgment step is as follows: if the mean square error between the output of the trained neural network initial model and the corresponding actual channel data is greater than or equal to a preset threshold or does not reach a preset number of iterations, turn to the training step; if the mean square error between the output of the trained neural network initial model and the corresponding actual channel data is less than a preset threshold or reaches a preset number of iterations, use the trained neural network initial model as the neural network model.

[0145] In some embodiments, the training module is specifically configured to perform at least one of the following:

[0146] A training step, using the training data to train the neural network model obtained in the last training to obtain a trained neural network model; wherein the training data includes the input data of the neural network model obtained in the last training and the corresponding actual channel data, the input data is the channel estimation results of K consecutive time domain transmission units that send pilot signals, the length of the channel estimation result of each time domain transmission unit is M×N, and the output of the neural network model obtained in the last training is the channel estimation results of N antennas on all M RBs in the Kth time domain transmission unit that sends the pilot signal and the subsequent L consecutive time domain transmission units, and M, N, K, and L are positive integers;

[0147] The judgment step is as follows: if the mean square error between the output of the trained neural network model and the corresponding actual channel data is greater than or equal to a preset threshold or has not reached a preset number of iterations, turn to the training step; if the mean square error between the output of the trained neural network model and the corresponding actual channel data is less than a preset threshold or has reached a preset number of iterations, use the trained neural network model as the neural network model.

[0148] In some embodiments, the neural network model includes a first neural network model and a second neural network model, and in the input data for training the first neural network model, the pilot signal in the first time domain transmission unit among the K consecutive time domain transmission units that send pilot signals is sent on an RB with an even number; in the input data for training the second neural network model, the pilot signal in the first time domain transmission unit among the K consecutive time domain transmission units that send pilot signals is sent on the first RB with an odd number.

[0149] In some embodiments, the channel estimation module is specifically used to perform channel estimation using the first neural network model if the pilot signal in the first received time domain transmission unit is sent on an RB with an even number; and to perform channel estimation using the second neural network model if the pilot signal in the first received time domain transmission unit is sent on an RB with an odd number.

[0150] In some embodiments, the pilot signal includes at least one of the following:

[0151] Channel state information reference signal CSI-RS, sounding reference signal SRS, demodulation reference signal DMRS.

[0152] In some embodiments, the time domain transmission unit includes any one of the following: transmission time interval TTI, subframe, millisecond, time slot and symbol.

[0153] The channel estimation device in the embodiment of the present application can be a device, a device or an electronic device with an operating system, or a component, an integrated circuit, or a chip in a terminal. The device or electronic device can be a mobile terminal or a non-mobile terminal. Exemplarily, the mobile terminal can include but is not limited to the types of terminals 11 listed above, and the non-mobile terminal can be a server, a network attached storage (Network Attached Storage, NAS), a personal computer (personal computer, PC), a television (television, TV), a teller machine or a self-service machine, etc., which is not specifically limited in the embodiment of the present application.

[0154] The channel estimation device provided in the embodiment of the present application can achieve Figure 2The various processes implemented by the method embodiment and achieving the same technical effect are not described here to avoid repetition.

[0155] The embodiment of the present application provides a channel estimation device, which is applied to a network side device 400, such as Figure 7 As shown, the device comprises:

[0156] The sending module 410 is configured to send a pilot signal, wherein resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different.

[0157] In the embodiment of the present application, in order to save pilot overhead, the network side device does not send pilot signals on the RB of each time domain transmission unit, but sends pilot signals on part of the RB. In each time domain transmission unit, the pilot signal is sent based on the resource block RB as the basic unit.

[0158] In addition, in order to further save pilot overhead, the network side device may also send a pilot signal once every several time domain transmission units. In some embodiments, the sending module is specifically used to send a pilot signal on a first RB of a first time domain transmission unit; send a pilot signal on a second RB of a second time domain transmission unit;

[0159] The first RB is different from the second RB, the second time domain transmission unit is the time domain transmission unit that sends the pilot signal most recently after the first time domain transmission unit, and the second time domain transmission unit is spaced from the first time domain transmission unit by at least one time domain transmission unit.

[0160] In some embodiments, the sequence number of the first RB is an even number, and the sequence number of the second RB is an odd number; or

[0161] The sequence number of the first RB is an odd number, and the sequence number of the second RB is an even number.

[0162] In some embodiments, the pilot signal includes at least one of the following:

[0163] Channel State Information Reference Signal CSI-RS, Sounding Reference Signal SRS, Demodulation Reference Signal DMRS. The time domain transmission unit includes any one of the following: Transmission Time Interval TTI, subframe, millisecond, time slot and symbol.

[0164] The channel estimation device provided in the embodiment of the present application can achieve Figure 3 The various processes implemented by the method embodiment and achieving the same technical effect are not described here to avoid repetition.

[0165] Optional, such as Figure 8As shown, the embodiment of the present application also provides a communication device 500, including a processor 501, a memory 502, and a program or instruction stored in the memory 502 and executable on the processor 501. For example, when the communication device 500 is a terminal, the program or instruction is executed by the processor 501 to implement the various processes of the channel estimation method embodiment applied to the terminal, and the same technical effect can be achieved. When the communication device 500 is a network side device, the program or instruction is executed by the processor 501 to implement the various processes of the channel estimation method embodiment applied to the network side device, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0166] An embodiment of the present application also provides a terminal, including a processor and a communication interface, wherein the communication interface is used to receive a pilot signal sent by a network side device, and the resource blocks RB occupied by the pilot signals in the first time domain transmission unit and the second time domain transmission unit are at least partially different; the processor is used to perform channel estimation on the third time domain transmission unit based on the pilot signals on the first time domain transmission unit and the second time domain transmission unit.

[0167] Specifically, Fig. 9 A schematic diagram of the hardware structure of a terminal for implementing an embodiment of the present application.

[0168] The terminal 1000 includes but is not limited to: a radio frequency unit 1001, a network module 1002, an audio output unit 1003, an input unit 1004, a sensor 1005, a display unit 1006, a user input unit 1007, an interface unit 1008, a memory 1009, and at least some of the components of a processor 1010.

[0169] Those skilled in the art will appreciate that the terminal 1000 may also include a power source (such as a battery) for supplying power to various components, and the power source may be logically connected to the processor 1010 through a power management system, thereby implementing functions such as managing charging, discharging, and power consumption management through the power management system. Fig. 9 The terminal structure shown in the figure does not constitute a limitation on the terminal. The terminal may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently, which will not be described in detail here.

[0170] It should be understood that in the embodiment of the present application, the input unit 1004 may include a graphics processor (Graphics Processing Unit, GPU) 10041 and a microphone 10042, and the graphics processor 10041 processes the image data of the static picture or video obtained by the image capture device (such as a camera) in the video capture mode or the image capture mode. The display unit 1006 may include a display panel 10061, and the display panel 10061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 1007 includes a touch panel 10071 and other input devices 10072. The touch panel 10071 is also called a touch screen. The touch panel 10071 may include two parts: a touch detection device and a touch controller. Other input devices 10072 may include, but are not limited to, a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.

[0171] In the embodiment of the present application, the radio frequency unit 1001 receives downlink data from the network side device and sends it to the processor 1010 for processing; in addition, the uplink channel estimation is sent to the network side device. Generally, the radio frequency unit 1001 includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc.

[0172] The memory 1009 can be used to store software programs or instructions and various data. The memory 1009 can mainly include a program or instruction storage area and a data storage area, wherein the program or instruction storage area can store an operating system, an application program or instruction required for at least one function (such as a sound playback function, an image playback function, etc.), etc. In addition, the memory 1009 can include a high-speed random access memory, and can also include a non-volatile memory, wherein the non-volatile memory can be a read-only memory (Read-Only Memory, ROM), a programmable read-only memory (Programmable ROM, PROM), an erasable programmable read-only memory (Erasable PROM, EPROM), an electrically erasable programmable read-only memory (Electrically EPROM, EEPROM) or a flash memory. For example, at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices.

[0173] The processor 1010 may include one or more processing units; optionally, the processor 1010 may integrate an application processor and a modem processor, wherein the application processor mainly processes an operating system, a user interface, and application programs or instructions, etc., and the modem processor mainly processes wireless communications, such as a baseband processor. It is understandable that the modem processor may not be integrated into the processor 1010.

[0174] Among them, the processor 1010 is used to receive a pilot signal sent by a network side device, and the resource blocks RB occupied by the pilot signals in the first time domain transmission unit and the second time domain transmission unit are at least partially different; according to the pilot signals on the first time domain transmission unit and the second time domain transmission unit, channel estimation is performed on the third time domain transmission unit.

[0175] In some embodiments, the processor 1010 is used to perform channel estimation based on the pilot signal on the first time domain transmission unit, the pilot signal on the second time domain transmission unit and a pre-trained neural network model to obtain the channel on all RBs in the third time domain transmission unit.

[0176] In some embodiments, the processor 1010 is used to perform channel estimation based on the pilot signal on the first time domain transmission unit, the pilot signal on the second time domain transmission unit and a pre-trained neural network model to obtain channels on all RBs in L time domain transmission units after the current time domain transmission unit, where L is a positive integer.

[0177] In some embodiments, all the RBs are RBs determined according to a predefined rule or a preconfigured manner.

[0178] In some embodiments, the RBs determined according to a predefined rule or a preconfigured manner include at least RBs occupied by the pilot signal in the first time slot transmission unit and RBs occupied by the pilot signal in the second time slot transmission unit.

[0179] In some embodiments, the processor 1010 is specifically configured to perform channel estimation using pilot signals on the most recent K time domain transmission units that send pilot signals, where K is a positive integer.

[0180] In some embodiments, the processor 1010 is configured to receive a pilot signal sent by a network side device on a first RB of a first time domain transmission unit; receive a pilot signal sent by a network side device on a second RB of a second time domain transmission unit;

[0181] The first RB is different from the second RB, the second time domain transmission unit is the time domain transmission unit that sends the pilot signal most recently after the first time domain transmission unit, and the second time domain transmission unit is spaced from the first time domain transmission unit by at least one time domain transmission unit.

[0182] In some embodiments, the sequence number of the first RB is an even number, and the sequence number of the second RB is an odd number; or

[0183] The sequence number of the first RB is an odd number, and the sequence number of the second RB is an even number.

[0184] In some embodiments, the processor 1010 is used to train the neural network model, wherein the step of training the neural network model includes at least one of the following:

[0185] Establish steps to build the initial model of neural network;

[0186] A training step, using training data to train the neural network initial model to obtain a trained neural network initial model; wherein the training data includes input data of the neural network initial model and corresponding actual channel data, the input data is the channel estimation results of K consecutive time domain transmission units that send pilot signals, the length of the channel estimation result of each time domain transmission unit is M×N, and the output of the neural network initial model is the channel estimation results of N antennas on all M RBs in the Kth time domain transmission unit that sends the pilot signal and the subsequent L consecutive time domain transmission units, where M, N, K, and L are positive integers;

[0187] The judgment step is as follows: if the mean square error between the output of the trained neural network initial model and the corresponding actual channel data is greater than or equal to a preset threshold or does not reach a preset number of iterations, turn to the training step; if the mean square error between the output of the trained neural network initial model and the corresponding actual channel data is less than a preset threshold or reaches a preset number of iterations, use the trained neural network initial model as the neural network model.

[0188] In some embodiments, the step of training the neural network model includes one of the following:

[0189] A training step, using the training data to train the neural network model obtained in the last training to obtain a trained neural network model; wherein the training data includes the input data of the neural network model obtained in the last training and the corresponding actual channel data, the input data is the channel estimation results of K consecutive time domain transmission units that send pilot signals, the length of the channel estimation result of each time domain transmission unit is M×N, and the output of the neural network model obtained in the last training is the channel estimation results of N antennas on all M RBs in the Kth time domain transmission unit that sends the pilot signal and the subsequent L consecutive time domain transmission units, and M, N, K, and L are positive integers;

[0190] The judgment step is as follows: if the mean square error between the output of the trained neural network model and the corresponding actual channel data is greater than or equal to a preset threshold or has not reached a preset number of iterations, turn to the training step; if the mean square error between the output of the trained neural network model and the corresponding actual channel data is less than a preset threshold or has reached a preset number of iterations, use the trained neural network model as the neural network model.

[0191] In some embodiments, the neural network model includes a first neural network model and a second neural network model, and in the input data for training the first neural network model, the pilot signal in the first time domain transmission unit among the K consecutive time domain transmission units that send pilot signals is sent on an RB with an even number; in the input data for training the second neural network model, the pilot signal in the first time domain transmission unit among the K consecutive time domain transmission units that send pilot signals is sent on the first RB with an odd number.

[0192] In some embodiments, the processor 1010 is used to perform channel estimation using the first neural network model if the pilot signal in the first received time domain transmission unit is sent on an RB with an even sequence number; and to perform channel estimation using the second neural network model if the pilot signal in the first received time domain transmission unit is sent on an RB with an odd sequence number.

[0193] In some embodiments, the pilot signal includes at least one of the following:

[0194] Channel state information reference signal CSI-RS, sounding reference signal SRS, demodulation reference signal DMRS.

[0195] In some embodiments, the time domain transmission unit includes any one of the following: transmission time interval TTI, subframe, millisecond, time slot and symbol.

[0196] An embodiment of the present application also provides a network side device, including a processor and a communication interface, the communication interface is used to send a pilot signal, and the resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different.

[0197] This network side device embodiment corresponds to the above-mentioned network side device method embodiment. Each implementation process and implementation method of the above-mentioned method embodiment can be applied to this network side device embodiment and can achieve the same technical effect.

[0198] Specifically, the embodiment of the present application also provides a network side device. Fig.10 As shown, the network device 700 includes: an antenna 71, a radio frequency device 72, and a baseband device 73. The antenna 71 is connected to the radio frequency device 72. In the uplink direction, the radio frequency device 72 receives information through the antenna 71 and sends the received information to the baseband device 73 for processing. In the downlink direction, the baseband device 73 processes the information to be sent and sends it to the radio frequency device 72. The radio frequency device 72 processes the received information and sends it out through the antenna 71.

[0199] The frequency band processing device may be located in a baseband device 73 . The method executed by the network-side device in the above embodiment may be implemented in the baseband device 73 . The baseband device 73 includes a processor 74 and a memory 75 .

[0200] The baseband device 73 may include, for example, at least one baseband board on which a plurality of chips are arranged. Fig.10 As shown, one of the chips is, for example, a processor 74, which is connected to a memory 75 to call a program in the memory 75 to execute the network device operations shown in the above method embodiment.

[0201] The baseband device 73 may further include a network interface 76 for exchanging information with the radio frequency device 72 , and the interface may be, for example, a common public radio interface (CPRI for short).

[0202] Specifically, the network side device of the embodiment of the present invention further includes: instructions or programs stored in the memory 75 and executable on the processor 74, and the processor 74 calls the instructions or programs in the memory 75 to execute Figure 7 The methods executed by the modules shown achieve the same technical effects, and therefore will not be described here in detail to avoid repetition.

[0203] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, each process of the above-mentioned channel estimation method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0204] The processor is the processor in the terminal described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0205] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned channel estimation method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0206] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0207] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0208] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0209] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A channel estimation method, characterized in that: include: The terminal receives a pilot signal sent by a network-side device, and resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different; Performing channel estimation on a third time domain transmission unit according to pilot signals on the first time domain transmission unit and the second time domain transmission unit; The performing channel estimation on the third time domain transmission unit according to the pilot signals on the first time domain transmission unit and the second time domain transmission unit comprises any one of the following: Perform channel estimation according to the pilot signal on the first time domain transmission unit, the pilot signal on the second time domain transmission unit, and a pre-trained neural network model to obtain channels on all RBs in the third time domain transmission unit; Channel estimation is performed based on the pilot signal on the first time domain transmission unit, the pilot signal on the second time domain transmission unit and a pre-trained neural network model to obtain channels on all RBs in L time domain transmission units after the current time domain transmission unit, where L is a positive integer.

2. The channel estimation method according to claim 1, characterized in that All the RBs are RBs determined according to a predefined rule or a preconfigured manner.

3. The channel estimation method according to claim 2, characterized in that: The RBs determined according to the predefined rule or preconfiguration method at least include the RBs occupied by the pilot signal in the first time slot transmission unit and the RBs occupied by the pilot signal in the second time slot transmission unit.

4. The channel estimation method according to claim 1, characterized in that: Channel estimation is performed using pilot signals on the most recent K time-domain transmission units that send pilot signals, where K is a positive integer.

5. The channel estimation method according to claim 1, characterized in that: The terminal receiving a pilot signal sent by a network side device includes: The terminal receives a pilot signal sent by a network side device on a first RB of a first time domain transmission unit; The terminal receives a pilot signal sent by a network side device on a second RB of a second time domain transmission unit; The first RB is different from the second RB, the second time domain transmission unit is the time domain transmission unit that sends the pilot signal most recently after the first time domain transmission unit, and the second time domain transmission unit is spaced from the first time domain transmission unit by at least one time domain transmission unit.

6. The channel estimation method according to claim 5, characterized in that: The sequence number of the first RB is an even number, and the sequence number of the second RB is an odd number; or The sequence number of the first RB is an odd number, and the sequence number of the second RB is an even number.

7. The channel estimation method according to claim 1, characterized in that: Also includes: Training to obtain the neural network model; The step of training to obtain the neural network model includes at least one of the following: Establish steps to build the initial model of neural network; A training step, using training data to train the neural network initial model to obtain a trained neural network initial model; wherein the training data includes input data of the neural network initial model and corresponding actual channel data, the input data is the channel estimation results of K consecutive time domain transmission units that send pilot signals, the length of the channel estimation result of each time domain transmission unit is M×N, and the output of the neural network initial model is the channel estimation results of N antennas on all M RBs in the Kth time domain transmission unit that sends the pilot signal and the subsequent L consecutive time domain transmission units, where M, N, K, and L are positive integers; The judgment step is as follows: if the mean square error between the output of the trained neural network initial model and the corresponding actual channel data is greater than or equal to a preset threshold or does not reach a preset number of iterations, turn to the training step; if the mean square error between the output of the trained neural network initial model and the corresponding actual channel data is less than a preset threshold or reaches a preset number of iterations, use the trained neural network initial model as the neural network model.

8. The channel estimation method according to claim 7, characterized in that: The step of training to obtain the neural network model includes at least one of the following: A training step, using the training data to train the neural network model obtained in the last training to obtain a trained neural network model; wherein the training data includes the input data of the neural network model obtained in the last training and the corresponding actual channel data, the input data is the channel estimation results of K consecutive time domain transmission units that send pilot signals, the length of the channel estimation result of each time domain transmission unit is M×N, and the output of the neural network model obtained in the last training is the channel estimation results of N antennas on all M RBs in the Kth time domain transmission unit that sends the pilot signal and the subsequent L consecutive time domain transmission units, and M, N, K, and L are positive integers; The judgment step is as follows: if the mean square error between the output of the trained neural network model and the corresponding actual channel data is greater than or equal to a preset threshold or has not reached a preset number of iterations, turn to the training step; if the mean square error between the output of the trained neural network model and the corresponding actual channel data is less than a preset threshold or has reached a preset number of iterations, use the trained neural network model as the neural network model.

9. The channel estimation method according to claim 7, characterized in that: The neural network model includes a first neural network model and a second neural network model. In the input data for training the first neural network model, the pilot signal in the first time domain transmission unit among the K consecutive time domain transmission units for sending pilot signals is sent on the RB with an even number; in the input data for training the second neural network model, the pilot signal in the first time domain transmission unit among the K consecutive time domain transmission units for sending pilot signals is sent on the first RB with an odd number.

10. The channel estimation method according to claim 9, characterized in that: Channel estimation according to the neural network model includes: If the pilot signal in the first received time domain transmission unit is sent on an RB with an even number, the first neural network model is used to perform channel estimation; if the pilot signal in the first received time domain transmission unit is sent on an RB with an odd number, the second neural network model is used to perform channel estimation.

11. The channel estimation method according to claim 1, characterized in that: The pilot signal includes at least one of the following: Channel state information reference signal CSI-RS, sounding reference signal SRS, demodulation reference signal DMRS.

12. The channel estimation method according to claim 1, characterized in that: The time domain transmission unit includes any one of the following: transmission time interval TTI, subframe, millisecond, time slot and symbol.

13. A channel estimation method, characterized in that: include: The network side device sends a pilot signal, and resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different.

14. The channel estimation method according to claim 13, characterized in that: The network side device sending a pilot signal includes: Sending a pilot signal on a first RB of the first time domain transmission unit; Sending a pilot signal on a second RB of the second time domain transmission unit; The first RB is different from the second RB, the second time domain transmission unit is the time domain transmission unit that sends the pilot signal most recently after the first time domain transmission unit, and the second time domain transmission unit is spaced from the first time domain transmission unit by at least one time domain transmission unit.

15. The channel estimation method according to claim 14, characterized in that: The sequence number of the first RB is an even number, and the sequence number of the second RB is an odd number; or The sequence number of the first RB is an odd number, and the sequence number of the second RB is an even number.

16. The channel estimation method according to claim 13, characterized in that: The pilot signal includes at least one of the following: Channel state information reference signal CSI-RS, sounding reference signal SRS, demodulation reference signal DMRS.

17. The channel estimation method according to claim 13, characterized in that: The time domain transmission unit includes any one of the following: transmission time interval TTI, subframe, millisecond, time slot and symbol.

18. A channel estimation device, characterized in that: include: A receiving module, configured to receive a pilot signal sent by a network-side device, wherein resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different; a channel estimation module, configured to perform channel estimation on a third time domain transmission unit according to pilot signals on the first time domain transmission unit and the second time domain transmission unit; The channel estimation module is specifically used to perform channel estimation according to the pilot signal on the first time domain transmission unit, the pilot signal on the second time domain transmission unit and a pre-trained neural network model to obtain channels on all RBs in the third time domain transmission unit; or The channel estimation module is specifically used to perform channel estimation based on the pilot signal on the first time domain transmission unit, the pilot signal on the second time domain transmission unit and a pre-trained neural network model to obtain the channels on all RBs in L time domain transmission units after the current time domain transmission unit, where L is a positive integer.

19. The channel estimation device according to claim 18, characterized in that: All the RBs are RBs determined according to a predefined rule or a preconfigured manner.

20. The channel estimation device according to claim 19, characterized in that: The RBs determined according to the predefined rule or preconfiguration method at least include the RBs occupied by the pilot signal in the first time slot transmission unit and the RBs occupied by the pilot signal in the second time slot transmission unit.

21. The channel estimation device according to claim 18, characterized in that: The channel estimation module is specifically used to perform channel estimation using pilot signals on the most recent K time domain transmission units that send pilot signals, where K is a positive integer.

22. The channel estimation device according to claim 18, characterized in that: The receiving module is specifically used to receive a pilot signal sent by a network side device on a first RB of a first time domain transmission unit; and receive a pilot signal sent by a network side device on a second RB of a second time domain transmission unit; The first RB is different from the second RB, the second time domain transmission unit is the time domain transmission unit that sends the pilot signal most recently after the first time domain transmission unit, and the second time domain transmission unit is spaced from the first time domain transmission unit by at least one time domain transmission unit.

23. The channel estimation device according to claim 22, characterized in that: The sequence number of the first RB is an even number, and the sequence number of the second RB is an odd number; or The sequence number of the first RB is an odd number, and the sequence number of the second RB is an even number.

24. The channel estimation device according to claim 18, characterized in that: It also includes a training module, which is used to train and obtain the neural network model; The step of training to obtain the neural network model includes at least one of the following: Establish steps to build the initial model of neural network; A training step, using training data to train the neural network initial model to obtain a trained neural network initial model; wherein the training data includes input data of the neural network initial model and corresponding actual channel data, the input data is the channel estimation results of K consecutive time domain transmission units that send pilot signals, the length of the channel estimation result of each time domain transmission unit is M×N, and the output of the neural network initial model is the channel estimation results of N antennas on all M RBs in the Kth time domain transmission unit that sends the pilot signal and the subsequent L consecutive time domain transmission units, where M, N, K, and L are positive integers; The judgment step is as follows: if the mean square error between the output of the trained neural network initial model and the corresponding actual channel data is greater than or equal to a preset threshold or does not reach a preset number of iterations, turn to the training step; if the mean square error between the output of the trained neural network initial model and the corresponding actual channel data is less than a preset threshold or reaches a preset number of iterations, use the trained neural network initial model as the neural network model.

25. The channel estimation device according to claim 24, characterized in that: The training module is specifically configured to perform at least one of the following: A training step, using the training data to train the neural network model obtained in the last training to obtain a trained neural network model; wherein the training data includes the input data of the neural network model obtained in the last training and the corresponding actual channel data, the input data is the channel estimation results of K consecutive time domain transmission units that send pilot signals, the length of the channel estimation result of each time domain transmission unit is M×N, and the output of the neural network model obtained in the last training is the channel estimation results of N antennas on all M RBs in the Kth time domain transmission unit that sends the pilot signal and the subsequent L consecutive time domain transmission units, and M, N, K, and L are positive integers; The judgment step is as follows: if the mean square error between the output of the trained neural network model and the corresponding actual channel data is greater than or equal to a preset threshold or has not reached a preset number of iterations, turn to the training step; if the mean square error between the output of the trained neural network model and the corresponding actual channel data is less than a preset threshold or has reached a preset number of iterations, use the trained neural network model as the neural network model.

26. The channel estimation device according to claim 24, characterized in that: The neural network model includes a first neural network model and a second neural network model. In the input data for training the first neural network model, the pilot signal in the first time domain transmission unit among the K consecutive time domain transmission units for sending pilot signals is sent on the RB with an even number; in the input data for training the second neural network model, the pilot signal in the first time domain transmission unit among the K consecutive time domain transmission units for sending pilot signals is sent on the first RB with an odd number.

27. The channel estimation device according to claim 26, characterized in that: The channel estimation module is specifically used to perform channel estimation using the first neural network model if the pilot signal in the first time domain transmission unit received is sent on an RB with an even number; and to perform channel estimation using the second neural network model if the pilot signal in the first time domain transmission unit received is sent on an RB with an odd number.

28. The channel estimation device according to claim 18, characterized in that: The pilot signal includes at least one of the following: Channel state information reference signal CSI-RS, sounding reference signal SRS, demodulation reference signal DMRS.

29. The channel estimation device according to claim 18, characterized in that: The time domain transmission unit includes any one of the following: transmission time interval TTI, subframe, millisecond, time slot and symbol.

30. A channel estimation device, characterized in that: include: The sending module is used to send a pilot signal, and the resource blocks RB occupied by the pilot signal in the first time domain transmission unit and the second time domain transmission unit are at least partially different.

31. The channel estimation device according to claim 30, characterized in that: The sending module is specifically configured to send a pilot signal on a first RB of the first time domain transmission unit; and send a pilot signal on a second RB of the second time domain transmission unit; The first RB is different from the second RB, the second time domain transmission unit is the time domain transmission unit that sends the pilot signal most recently after the first time domain transmission unit, and the second time domain transmission unit is spaced from the first time domain transmission unit by at least one time domain transmission unit.

32. The channel estimation device according to claim 31, characterized in that: The sequence number of the first RB is an even number, and the sequence number of the second RB is an odd number; or The sequence number of the first RB is an odd number, and the sequence number of the second RB is an even number.

33. The channel estimation device according to claim 30, characterized in that: The pilot signal includes at least one of the following: Channel state information reference signal CSI-RS, sounding reference signal SRS, demodulation reference signal DMRS.

34. The channel estimation device according to claim 30, characterized in that: The time domain transmission unit includes any one of the following: transmission time interval TTI, subframe, millisecond, time slot and symbol.

35. A terminal, characterized in that: The method comprises a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the channel estimation method according to any one of claims 1 to 12.

36. A network side device, characterized in that: The method comprises a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the channel estimation method according to any one of claims 13 to 17.

37. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of the channel estimation method according to any one of claims 1 to 17 are implemented.

Images