Reference signal sequence generation method, apparatus, device, and medium

Abstract

The application discloses a reference signal sequence generation method, device, equipment and medium, and belongs to the technical field of communication. The reference signal sequence generation method of the application embodiment comprises the following steps: a first device acquires a target neural network; and the first device acquires a target reference signal sequence based on the target neural network. The target neural network comprises N target neuron groups, each target neuron group corresponds to a target weight parameter, and the target reference signal sequence comprises N target elements corresponding to the N target weight parameters one by one, wherein N is a positive integer.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of communication, and particularly relates to a reference signal sequence generation method and device, equipment and medium. BACKGROUND

[0002] At present, in new radio (NR), a device 1 (for example, a transmitting end) can generate a reference signal through a pseudo noise (PN) sequence (or a Zadoff-Chu sequence), and send the reference signal to a device 2 (for example, a receiving end) through a channel, so that the receiving end can perform channel estimation or channel sounding according to the reference signal, and perform wireless communication with the transmitting end according to the result of channel estimation or channel sounding.

[0003] However, since the reference signal may be affected by noise and interference of the channel when passing through the channel, the result of channel estimation or channel sounding of the receiving end may be inaccurate, which may cause the transmitting end to be unable to perform wireless communication with the receiving end, and thus the reliability of device communication is poor. SUMMARY

[0004] The embodiments of the present application provide a reference signal sequence generation method, device, equipment and medium, which can solve the problem of poor reliability of device communication.

[0005] In a first aspect, a reference signal sequence generation method is provided, which includes: a first device obtaining a target neural network; and the first device obtaining a target reference signal sequence based on the target neural network; wherein the target neural network includes N target neuron groups, each target neuron group corresponding to a target weight parameter; and the target reference signal sequence includes N target elements corresponding one by one to the N target weight parameters, N being a positive integer.

[0006] In a second aspect, a reference signal sequence generation device is provided, which is a first reference signal sequence generation device, and includes an obtaining module. The obtaining module is configured to obtain a target neural network, and obtain a target reference signal sequence based on the target neural network; wherein the target neural network includes N target neuron groups, each target neuron group corresponding to a target weight parameter; and the target reference signal sequence includes N target elements corresponding one by one to the N target weight parameters, N being a positive integer.

[0007] In a third aspect, a reference signal sequence generation method is provided. The method comprises: receiving, by a second device, at least one third reference signal sequence from at least one third device, the at least one third reference signal sequence comprising: a first reference signal sequence transmitted by a first device; performing, by the second device, a third operation based on each third reference signal sequence to obtain a corresponding fourth reference signal sequence; transmitting, by the second device, a second channel estimation value to the first device; wherein the second channel estimation value is a channel estimation value obtained by the second device based on the fourth reference signal sequence corresponding to the first device; the second channel estimation value is used for training a target neural network by the first device; and the target neural network is used for generating a target reference signal sequence.

[0008] In a fourth aspect, a reference signal sequence generation apparatus is provided. The reference signal sequence generation apparatus is a second reference signal sequence generation apparatus. The reference signal sequence generation apparatus comprises: a receiving module, an executing module, and a transmitting module. The receiving module is configured to receive at least one third reference signal sequence from at least one third reference signal sequence generation apparatus, the at least one third reference signal sequence comprising: a first reference signal sequence transmitted by a first reference signal sequence generation apparatus in the at least one third reference signal sequence generation apparatus. The executing module is configured to perform a third operation based on each third reference signal sequence received by the receiving module to obtain a corresponding fourth reference signal sequence. The transmitting module is configured to transmit a second channel estimation value to the first reference signal sequence generation apparatus; wherein the second channel estimation value is a channel estimation value obtained by the second reference signal sequence generation apparatus based on the fourth reference signal sequence corresponding to the first reference signal sequence generation apparatus; the second channel estimation value is used for training a target neural network by the first reference signal sequence generation apparatus; and the target neural network is used for generating a target reference signal sequence.

[0009] In a fifth aspect, a terminal is provided. The terminal comprises a processor, a memory, and a program or instructions stored in the memory and executable on the processor. When the program or instructions are executed by the processor, the steps of the method according to the first aspect or the steps of the method according to the third aspect are implemented.

[0010] In a sixth aspect, a network side device is provided. The network side device comprises a processor, a memory, and a program or instructions stored in the memory and executable on the processor. When the program or instructions are executed by the processor, the steps of the method according to the first aspect or the steps of the method according to the third aspect are implemented.

[0011] In a seventh aspect, a terminal is provided, which includes a processor and a communication interface, wherein the processor is configured to obtain a target neural network, and obtain a target reference signal sequence based on the target neural network; wherein the target neural network includes N target neuron groups, each target neuron group corresponding to a target weight parameter; and the target reference signal sequence includes N target elements corresponding to the N target weight parameters, N being a positive integer.

[0012] In an eighth aspect, a terminal is provided, which includes a processor and a communication interface, wherein the communication interface is configured to receive at least one third reference signal sequence from at least one third device, the at least one third reference signal sequence including a first reference signal sequence transmitted by a first device; the processor is configured to perform a third operation based on each third reference signal sequence to obtain a corresponding fourth reference signal sequence; and the communication interface is further configured to transmit a second channel estimation value to the first device, wherein the second channel estimation value is a channel estimation value obtained by the terminal based on the fourth reference signal sequence corresponding to the first device, and the second channel estimation value is used by the first device to train a target neural network, and the target neural network is used to generate a target reference signal sequence.

[0013] In a ninth aspect, a network-side device is provided, which includes a processor and a communication interface, wherein the processor is configured to obtain a target neural network, and obtain a target reference signal sequence based on the target neural network; wherein the target neural network includes N target neuron groups, each target neuron group corresponding to a target weight parameter; and the target reference signal sequence includes N target elements corresponding to the N target weight parameters, N being a positive integer.

[0014] In a tenth aspect, a network-side device is provided, which includes a processor and a communication interface, wherein the communication interface is configured to receive at least one third reference signal sequence from at least one third device, the at least one third reference signal sequence including a first reference signal sequence transmitted by a first device; the processor is configured to perform a third operation based on each third reference signal sequence to obtain a corresponding fourth reference signal sequence; and the communication interface is further configured to transmit a second channel estimation value to the first device, wherein the second channel estimation value is a channel estimation value obtained by the network-side device based on the fourth reference signal sequence corresponding to the first device, and the second channel estimation value is used by the first device to train a target neural network, and the target neural network is used to generate a target reference signal sequence.

[0015] In a eleventh aspect, a readable storage medium is provided, and the readable storage medium stores a program or instructions, and the program or instructions are executed by a processor to implement the steps of the method in the first aspect or implement the steps of the method in the third aspect.

[0016] In a twelfth aspect, a chip is provided, and the chip includes a processor and a communication interface, the communication interface is coupled with the processor, and the processor is configured to run a program or instructions to implement the method in the first aspect or implement the method in the third aspect.

[0017] In a thirteenth aspect, a computer program / program product is provided, and 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 in the first aspect or implement the steps of the method in the third aspect.

[0018] In the embodiments of the present application, the first device can obtain a target neural network, and obtain a target reference signal sequence based on the target neural network, the target neural network includes N target neuron groups, and N target weight parameters of the N target neuron groups correspond to N target elements of the target reference signal sequence one by one. Since the first device obtains the target reference signal sequence according to the target neural network, the N target weight parameters of the target neural network correspond to the N target elements one by one, and the reference signal sequence is not generated by a PN sequence (or a Zadoff-Chu sequence), the anti-noise capability and the anti-interference capability of the target reference signal sequence are stronger than the anti-noise capability and the anti-interference capability of the reference signal sequence, so that the influence of the target reference signal sequence when passing through a channel can be reduced, and thus the reliability of device communication can be improved. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is a block diagram of a wireless communication system provided by the embodiments of the present application;

[0020] Figure 2 is one of the schematic diagrams of the reference signal sequence generation method provided by the embodiments of the present application;

[0021] Figure 3 is the second schematic diagram of the reference signal sequence generation method provided by the embodiments of the present application;

[0022] Figure 4 is the third schematic diagram of the reference signal sequence generation method provided by the embodiments of the present application;

[0023] Figure 5 is a schematic diagram of the real part parameter and the imaginary part parameter of the element of the first reference sequence signal calculated by the embodiments of the present application;

[0024] Figure 6 FIG. 4 is a schematic diagram of a fourth reference signal sequence generation method according to an embodiment of the present application;

[0025] Figure 7 FIG. 5 is a schematic diagram of a fifth reference signal sequence generation method according to an embodiment of the present application;

[0026] Figure 8 FIG. 6 is a structural schematic diagram of a first reference signal sequence generation apparatus according to an embodiment of the present application;

[0027] Figure 9 FIG. 7 is a structural schematic diagram of a second reference signal sequence generation apparatus according to an embodiment of the present application;

[0028] Figure 10 FIG. 8 is a structural schematic diagram of a communication device according to an embodiment of the present application;

[0029] Figure 11 FIG. 9 is a hardware structural schematic diagram of a terminal according to an embodiment of the present application;

[0030] Figure 12 FIG. 10 is a hardware structural schematic diagram of a network side device according to an embodiment of the present application. DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the present application will be clearly described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art belong to the scope of protection of the present application.

[0032] The terms related to the embodiments of the present application will be described below.

[0033] 1. Neural network

[0034] The neural network is composed of a plurality of neurons, wherein each neuron in the plurality of neurons includes a weight parameter.

[0035] In the training process of the neural network, each neuron can receive input information and output processed information after being processed by an activation function respectively, so as to determine a loss function according to the deviation degree between the processed information and the real information, and then correct the weight parameter of each neuron through an optimization algorithm, and repeat the above process to obtain a trained neural network.

[0036] The optimization algorithm is an algorithm for minimizing or maximizing a loss function. The optimization algorithm can include any of the following: a gradient descent algorithm, a stochastic gradient descent algorithm, a mini-batch gradient descent algorithm, a momentum method, a stochastic gradient descent algorithm with momentum, an adaptive gradient descent algorithm, a root mean square error descent algorithm, an adaptive momentum estimation algorithm, and the like.

[0037] 2. Reference signal

[0038] A reference signal (RS), i.e., a reference signal sequence, is an important component of system design. The reference signal includes a downlink reference signal and an uplink reference signal.

[0039] The downlink reference signal is used for channel state information measurement, data demodulation, beam training, time-frequency parameter tracking, and the like.

[0040] The uplink reference signal is used for uplink and downlink channel measurement, data demodulation, and the like.

[0041] 3. Other terms

[0042] The terms "first", "second", and the like in the specification and claims of the present application are used to distinguish similar objects, and are not intended to describe a particular order or sequence. It should be understood that the terms used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second" are generally of a kind and are not limited in number, for example, the first object can be one or more. In addition, "and / or" in the specification and claims indicates at least one of the connected objects, and the character " / " generally indicates that the front and rear associated objects are in an "or" relationship.

[0043] It is worth noting that the technology described in the embodiments of the present application is not limited to Long Term Evolution (LTE) / LTE-Advanced (LTE-A) systems, 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 in the above-mentioned systems and radio technologies, as well as in other systems and radio technologies. The following description describes a New Radio (NR) system for example purposes, and NR terminology is used in most of the following description, but these technologies can also be applied outside the NR system application, such as in a 6th Generation (6G) communication system. th

[0044] Figure 1 ​A block diagram of a wireless communication system to which embodiments of the present application can be applied is shown. The wireless communication system includes a terminal 11 and a network side device 12. The terminal 11 can be a terminal side device such as a mobile phone, a tablet personal computer, a laptop computer, a personal digital assistant (PDA), a palm computer, a netbook, an ultra-mobile personal computer (UMPC), a mobile Internet device (MID), an augmented reality (AR) / virtual reality (VR) device, a robot, a wearable device, a vehicle-mounted device (VUE), a pedestrian terminal (PUE), a smart home (a home device with a wireless communication function such as a refrigerator, a television, a washing machine, or furniture), and the like. The wearable device includes a smart watch, a smart bracelet, a smart earphone, smart glasses, smart jewelry (a smart bracelet, a smart necklace, a smart ring, a smart necklace, a smart anklet, a smart necklace, and the like), a smart wristband, smart clothing, a game console, and the like. It should be noted that the specific type of the terminal 11 is not limited in the embodiments of the present application. The network side device 12 can be a base station or a core network. The base station can be referred to as a node B, an evolved node B, an access point, a base transceiver station (BTS), a radio base station, a radio transceiver, a basic service set (BSS), an extended service set (ESS), a node B, an evolved node B (eNB), a home node B, a home evolved node B, a WLAN access point, a WiFi node, a transmitting receiving point (TRP), or some other appropriate terminology in the art, as long as the same technical effects are achieved. The base station is not limited to a specific technical term, and it should be noted that only a base station in an NR system is taken as an example in the embodiments of the present application, but the specific type of the base station is not limited.

[0045] The reference signal sequence generation method provided by the embodiments of the present application will be described in detail below in combination with the accompanying drawings and some embodiments and application scenarios.

[0046] Figure 2 A flowchart of a reference sequence generation method provided by the embodiments of the present application is shown. As shown in FIG. 1, the reference sequence generation method includes the following steps. Figure 2As shown, the reference sequence generation method provided by the embodiments of the present application can include the following steps 101 and 102.

[0047] Step 101, the first device acquires a target neural network.

[0048] Optionally, in the embodiments of the present application, the first device can be a user equipment (User Equipment, UE) or a network side device.

[0049] Optionally, in the embodiments of the present application, the target neural network is obtained based on channel information of the first device.

[0050] Optionally, in the embodiments of the present application, the target neural network can be any one of the following: a fully connected neural network, a convolutional neural network, a recurrent neural network, a residual neural network, a generative adversarial neural network, etc. It should be noted that the embodiments of the present application take the fully connected neural network as an example for illustration, and of course the target neural network can also be other neural networks, which can be selected according to the needs of those skilled in the art, and the embodiments of the present application do not limit this.

[0051] Optionally, in the embodiments of the present application, the target neural network can be a pre-stored neural network in the first device. Specifically, the pre-stored neural network can be a neural network trained by the first device.

[0052] Further optionally, in the embodiments of the present application, the target neural network can be a neural network determined by the first device from the pre-stored trained neural network in the first device. It can be understood that at least one trained neural network can be pre-stored in the first device, and a neural network in the at least one trained neural network can be selected according to the needs.

[0053] The following will illustrate how the first device determines the target neural network from the pre-stored neural network.

[0054] Optionally, in the embodiments of the present application, the step 101 can be implemented by the following step 101a.

[0055] Step 101a, the first device determines the target neural network according to a second physical parameter.

[0056] In the embodiments of the present application, the second physical parameter includes at least one of the following:

[0057] Bundling size;

[0058] Resource block RB;

[0059] Physical resource block PRB;

[0060] A number of users, MUs;

[0061] A density of a time-frequency domain pattern of the reference signal sequence in the time-frequency domain.

[0062] Further optionally, in embodiments of the present application, the first device can determine, according to X first corresponding relationships, a target neural network corresponding to the second physical parameter from at least one third neural network. Each first corresponding relationship is a corresponding relationship between one physical parameter and one third neural network, and X is a positive integer. The at least one third neural network is a pre-stored trained neural network in the first device.

[0063] Specifically, the first device can determine one physical parameter matching the second physical parameter from the X physical parameters, and determine a third neural network corresponding to the one physical parameter as the target neural network.

[0064] In embodiments of the present application, when the first device is in an inference stage (i.e., needs to use a reference signal sequence for wireless communication), the parameters (such as length, etc.) of the reference signal sequence that the first device needs to use can be different, so the first device can determine the target neural network that needs to be used according to the demand from the pre-stored at least one third neural network.

[0065] As can be seen, since the target neural network that needs to be used can be determined according to the second physical parameter, and is not based on a fixed neural network to obtain a reference signal sequence, the accuracy of obtaining the reference signal sequence required by the first device can be improved.

[0066] In embodiments of the present application, the target neural network includes N target neuron groups, each target neuron group in the N target neuron groups corresponds to a target weight parameter, and N is a positive integer.

[0067] Optionally, in embodiments of the present application, for each target weight parameter in the N target weight parameters, the target weight parameter can be a real part parameter (i.e., a real number) or an unquantized parameter.

[0068] Further optionally, in embodiments of the present application, each target weight parameter in the N target weight parameters can be a complex number.

[0069] Optionally, in embodiments of the present application, each target neuron group in the N target neuron groups includes at least one neuron. Each neuron includes a sub-weight parameter, and the target weight parameter corresponding to each target neuron group is determined according to the sub-weight parameters of the neurons included in the each target neuron group.

[0070] Exemplarily, assuming that the N target neuron groups include a target neuron group 1 and a target neuron group 2, the target neuron group 1 includes neurons 1, 2 and 3, and the target neuron group 2 includes neurons 4, 5 and 6, the target weight parameter corresponding to the target neuron group 1 is determined according to the sub weight parameter of the neuron 1, the sub weight parameter of the neuron 2 and the sub weight parameter of the neuron 3, and the target weight parameter corresponding to the target neuron group 2 is determined according to the sub weight parameter of the neuron 4, the sub weight parameter of the neuron 5 and the sub weight parameter of the neuron 6.

[0071] Further optionally, in the embodiment of the present application, each of the N target neuron groups includes a part of neurons whose sub weight parameters are real part parameters and another part of neurons whose sub weight parameters are imaginary part parameters, so that the real part parameter of the target weight parameter corresponding to each of the target neuron groups can be determined according to the sub weight parameters of the part of neurons, and the imaginary part parameter of the target weight parameter corresponding to each of the target neuron groups can be determined according to the sub weight parameters of the other part of neurons.

[0072] Hereinafter, a part of neurons whose sub weight parameters are real part parameters are taken as L first neurons, and another part of neurons whose sub weight parameters are imaginary part parameters are taken as M second neurons as an example for illustration.

[0073] Optionally, in the embodiment of the present application, each of the N target neuron groups includes L first neurons and M second neurons, each of the L first neurons corresponds to a first real part parameter, each of the M second neurons corresponds to a first imaginary part parameter, and L and M are positive integers.

[0074] It can be understood that for each of the L first neurons, the sub weight parameter of a first neuron can be specifically the corresponding first real part parameter, and for each of the M second neurons, the sub weight parameter of a second neuron can be specifically the corresponding second real part parameter.

[0075] In the embodiment of the present application, the target weight parameter corresponding to each of the target neuron groups is determined according to the L first real part parameters and the M first imaginary part parameters.

[0076] Further optionally, in the embodiment of the present application, L=M.

[0077] Further optionally, in embodiments of the present application, the first device can determine, according to the L first real part parameters, the real part parameters of the target weight parameter corresponding to each target neuron group by using a first algorithm, and determine, according to the M second real part parameters, the imaginary part parameters of the target weight parameter corresponding to each target neuron group by using a second algorithm, thereby obtaining the target weight parameter corresponding to each target neuron group.

[0078] Specifically, the first algorithm can be at least one of addition, subtraction, multiplication, division, average calculation, etc., and the second algorithm can be at least one of addition, subtraction, multiplication, division, average calculation, etc. The second algorithm can be the same as or different from the first algorithm.

[0079] In embodiments of the present application, the channel information includes any of the following: a channel vector, a channel matrix.

[0080] Optionally, in a possible implementation manner of embodiments of the present application, the channel information is determined by the first device according to the data sequence received by the first device.

[0081] Optionally, in another possible implementation manner of embodiments of the present application, the channel information is received by the first device from the second device.

[0082] Optionally, in embodiments of the present application, the channel information can include at least one element, and each element in the at least one element can be any of the following: a real number, an imaginary number, a complex number.

[0083] In step 102, the first device obtains a target reference signal sequence based on a target neural network.

[0084] In embodiments of the present application, the target reference signal sequence includes N target elements corresponding to the N target weight parameters.

[0085] Optionally, in embodiments of the present application, the first device can determine the N target weight parameters of the N target neuron groups as the corresponding target elements respectively, to generate the target reference signal sequence.

[0086] Further optionally, in embodiments of the present application, the first device can determine the target weight parameter of the first target neuron group as a first target element, determine the target weight parameter of the second target neuron group as a second target element, determine the target weight parameter of the third target neuron group as a third target element, and so on, until the target weight parameter of the last target neuron group is determined as a last target element, to obtain the target reference signal sequence.

[0087] It can be understood that the number of the N target neuron groups of the target neural network is the same as the length of the target reference signal sequence; and the number of neurons of the target neural network is at least one time of the length of the target reference signal sequence.

[0088] Exemplarily, the number of neurons of the target neural network is two times of the length of the target reference signal sequence.

[0089] Optionally, in the embodiment of the present application, after the first device acquires the target reference signal sequence, the first device can further execute the above-mentioned steps 101 and 102 again according to the requirement, that is, acquire other neural networks and acquire other reference signal sequences based on the other neural networks.

[0090] In the embodiment of the present application, the target neural network is pre-stored in the first device, and the target neural network is trained based on the channel information of the first device, so that the first device can acquire the target neural network, and determine the target weight parameters corresponding to the target neuron groups of the target neural network as the target elements of the reference signal sequence to generate the target reference signal sequence.

[0091] Optionally, in the embodiment of the present application, the first device can acquire the target neural network based on the channel information of the first device. Figure 2 As shown in the above-mentioned step 102, the reference signal sequence generation method provided by the embodiment of the present application can further include the following step 103. Figure 3

[0092] Step 103, the first device sends first signaling to the second device.

[0093] Optionally, in the embodiment of the present application, the above-mentioned first signaling includes at least one of the following:

[0094] Radio Resource Control (RRC) signaling;

[0095] Layer 1 signaling of a Physical Downlink Control Channel (PDCCH);

[0096] Information of a Physical Downlink Shared Channel (PDSCH);

[0097] Signaling of a Medium Access Control (MAC) control element;

[0098] System Information Block (SIB);

[0099] Layer 1 signaling of a Physical Uplink Control Channel (PUCCH);

[0100] Target message information of a Physical Random Access Channel (PRACH);

[0101] Information of a Physical Uplink Shared Channel (PUSCH);

[0102] XN interface signaling; ​

[0103] PC5 interface signaling;

[0104] Sidelink interface instructions;

[0105] In the embodiments of the present application, the target message includes at least one of the following: MSG 1 information, MSG 2 information, MSG 3 information, MSG 4 information, MSG A information, and MSG B information.

[0106] In the embodiments of the present application, the first signaling carries: a target reference signal sequence, or an identifier of the target reference signal sequence, or information obtained by compressing or processing the target reference signal sequence.

[0107] In the embodiments of the present application, the second device can be specifically: a UE or a network-side device.

[0108] Further optionally, in the embodiments of the present application, when the first device is a UE, the second device is a network-side device; and when the first device is a network-side device, the second device is a UE.

[0109] It can be understood that the first device can send the target reference signal sequence (or the identifier of the target reference signal sequence) to the second device, so that the second device can perform channel estimation or signal detection related operations according to the target reference signal sequence.

[0110] Further optionally, in the embodiments of the present application, after the first device generates the other reference signal sequence, the first device can perform the step 103 again. It can be understood that when the reference signal sequence generated by the first device changes, the first device can send the changed reference signal sequence to the second device through the first signaling again.

[0111] As can be seen, since the first device can send the first signaling carrying the target reference signal sequence (or the identifier of the target reference signal sequence) to the second device, so that the second device can perform channel estimation according to the target reference signal sequence, the reliability of communication of the first device can be improved.

[0112] The reference signal sequence generation method provided in this application embodiment allows a first device to acquire a target neural network and, based on the target neural network, acquire a target reference signal sequence. The target neural network includes N target neuron groups, and the N target weight parameters of these N target neuron groups correspond one-to-one with the N target elements of the target reference signal sequence. Since the first device acquires the target reference signal sequence based on the target neural network, where the N target weight parameters correspond one-to-one with the N target elements, and not through a PN sequence (or Zadoff-Chu sequence), the noise immunity and interference immunity of this target reference signal sequence are stronger than those of a standard reference signal sequence. This reduces the impact on the target reference signal sequence when passing through a channel, thereby improving the reliability of the first device's communication.

[0113] The following example illustrates how the first device obtains the target neural network.

[0114] Optionally, in the embodiments of this application, combined with Figure 2 ,like Figure 4 As shown, prior to step 101 above, the reference signal sequence generation method provided in this application embodiment may further include step 201 below.

[0115] Step 201: The first device uses channel information to train the first neural network to obtain the target neural network.

[0116] Further optionally, in this embodiment of the application, the first neural network can be: an untrained neural network pre-stored in the first device.

[0117] Further optionally, in the embodiments of this application, the first neural network may include: one neural network or multiple neural networks.

[0118] It should be noted that the above statement "the first neural network includes multiple neural networks" can be understood as: the multiple neural networks included in the first neural network can output processed information based on the input information, that is, the multiple neural networks are set up relatively independently.

[0119] The following example illustrates how the first device determines the first neural network.

[0120] Optionally, in the embodiments of this application, before step 201 above, the reference signal sequence generation method provided in the embodiments of this application may further include the following step 301.

[0121] Step 301: The first device determines the first neural network based on the first physical parameters.

[0122] In the embodiments of the present application, the first physical parameter comprises at least one of the following:

[0123] a bundling size;

[0124] an RB;

[0125] a PRB;

[0126] a number of MUs;

[0127] a density of a time-frequency domain pattern of a reference signal sequence in a time-frequency domain.

[0128] Further optionally, in the embodiments of the present application, the first device can determine the first neural network corresponding to the first physical parameter from the at least one fourth neural network according to Y second corresponding relationships. Each second corresponding relationship is a corresponding relationship between one physical parameter and one fourth neural network, and Y is a positive integer.

[0129] Specifically, the first device can determine one physical parameter matching the first physical parameter from the Y physical parameters, and determine the fourth neural network corresponding to the one physical parameter as the first neural network.

[0130] Therefore, different neural networks can be obtained, so that the first device can obtain different reference signal sequences based on the different neural networks, and thus the accuracy of generating the reference signal sequence required by the first device can be improved.

[0131] Optionally, in the embodiments of the present application, the channel information comprises Q first elements, and Q is a positive integer. Specifically, the above step 201 can be implemented through the following steps 201a to 201c.

[0132] Step 201a, the first device inputs the Q first elements into the first neural network to generate the first reference signal sequence.

[0133] Optionally, in the embodiments of the present application, the first neural network comprises N first neuron groups, each of the N first neuron groups corresponds to a first weight parameter; and the first reference signal sequence is calculated by multiplying the Q first elements and the N first weight parameters.

[0134] It should be noted that the multiplication operation described above is not limited to multiplication of real number parameters, but can also be multiplication of complex number parameters, and is not limited to multiplication of elements, but can be multiplication of any generalized range, such as inner product, outer product, multiplication of matrices in a matrix set, Hadamard product of matrices, Kronecker product of matrices, outer product of tensors, tensor product of tensors, and point-by-point product.

[0135] For example, taking Q first elements as a first element (for example, h) and N first weight parameters as a first weight parameter (for example, p) as an example for description. As shown in Figure 5 h includes a real part parameter (for example, Re{h}) and an imaginary part parameter (for example, Im{h}), and p includes a real part parameter (for example, Re{p}) and an imaginary part parameter (for example, Im{p}), so that the first device can calculate the real part parameter and the imaginary part parameter of the element of the first reference sequence signal according to p and h by using a multiplication operation to obtain the first reference signal sequence. Wherein, Re{} represents taking the real part of the element in the parentheses, Im{} represents taking the imaginary part of the element in the parentheses, "x" represents point multiplication operation, "+" represents addition operation, and "-" represents subtraction operation.

[0136] In the case that the first neural network includes different numbers of neural networks, the process of generating the first reference signal sequence by the first device can be different, and the following will be described by taking the case that the first neural network includes one neural network and the case that the first neural network includes multiple neural networks as examples.

[0137] For the case that the first neural network includes one neural network:

[0138] Optionally, in the embodiments of the present application, each of the Q first elements corresponds to a second real part parameter and a second imaginary part parameter; and the first reference signal sequence includes N second elements. Specifically, the step 201a can be implemented by the following steps 201a1 and 201a2.

[0139] In step 201a1, the first device inputs the Q second real part parameters and the Q second imaginary part parameters into the first neural network respectively to obtain N third real part parameters and N third imaginary part parameters output by the first neural network.

[0140] Further optionally, in the embodiments of the present application, after the first device inputs the Q second real part parameters and the Q second imaginary part parameters into the first neural network, the first neural network can calculate the N third real part parameters and the N third imaginary part parameters by using a multiplication operation based on the Q second real part parameters and the Q second imaginary part parameters.

[0141] It can be understood that the N third real part parameters are real part parameters of the elements of the first reference sequence signal, and the N third imaginary part parameters are imaginary part parameters of the elements of the first reference sequence signal.

[0142] In step 201a2, the first device determines N second elements according to the N third real part parameters and the N third imaginary part parameters, respectively.

[0143] Further optionally, in the embodiment of the present application, the first device can determine one second element according to one third real part parameter and one third imaginary part parameter, and determine another second element according to another third real part parameter and another third imaginary part parameter, and so on, until the last second element is determined according to the last third real part parameter and the last third imaginary part parameter, to determine the N second elements, to obtain the first reference signal sequence.

[0144] As can be seen, since the first device can input the Q second real part parameters and the Q second imaginary part parameters into the first neural network to obtain the real part parameters and the imaginary part parameters of the elements of the first reference signal sequence, i.e., the real part parameters and the imaginary part parameters of the elements of the reference signal sequence after passing through the channel, to obtain the first reference signal sequence after passing through the channel, the first device can train the first neural network based on the first reference signal sequence to obtain the target neural network that can generate strong anti-noise capability and anti-interference capability, so that the reliability of the communication of the first device can be improved.

[0145] For the case that the first neural network includes multiple neural networks:

[0146] Optionally, in the embodiment of the present application, the first neural network includes R second neural networks, and R is a positive integer greater than 1. Specifically, the above step 201a can be implemented through the following steps 201a3 to 201a5.

[0147] In step 201a3, the first device determines R element groups according to the Q first elements.

[0148] In the embodiment of the present application, different element groups include different first elements.

[0149] Further optionally, in the embodiment of the present application, the first device can determine a first value (for example, k) according to the number (i.e., Q) of the Q first elements and the number (i.e., R) of the R second neural networks, and then determine each k first elements as an element group to obtain the R element groups; wherein the first value is an integer.

[0150] It can be understood that each element group in the R element groups includes at least one first element, and the number of the first elements included in each element group can be the same or different.

[0151] Specifically, the first device can determine the first value according to a ratio of Q and R.

[0152] For example, assuming that the Q first elements include element 1, element 2, element 3, element 4, element 5 and element 6, and the R second neural networks include neural network 1 and neural network 2, the first device can determine the first value according to a ratio of the number of the Q first elements (i.e. 6) and the number of the R second neural networks (i.e. 2), i.e. 6 / 2=3, and then determine each 3 first elements as an element group, i.e. determine element 1, element 2 and element 3 as an element group, and determine element 4, element 5 and element 6 as another element group, to obtain 2 element groups.

[0153] For another example, assuming that the Q first elements include element 1, element 2, element 3, element 4 and element 5, and the R second neural networks include neural network 1 and neural network 2, the first device can determine the first value according to a ratio of the number of the Q first elements (i.e. 5) and the number of the R second neural networks (i.e. 2), i.e. 5 / 2=2.5, and then determine each 3 first elements as an element group, i.e. determine element 1, element 2 and element 3 as an element group, and determine element 4 and element 5 as another element group, to obtain 2 element groups.

[0154] Step 201a4, the first device inputs each element group into each second neural network respectively, generates a second reference signal sequence corresponding to each second neural network, and obtains R second reference signal sequences.

[0155] Further optionally, in the embodiment of the application, the first device can input the first element group into the first second neural network to generate a first second reference signal sequence, the elements of the first second reference signal sequence being output by the first second neural network, and input the second element group into the second second neural network to generate a second second reference signal sequence, the elements of the second second reference signal sequence being output by the second second neural network, and so on, until inputting the last element group into the last second neural network to generate a last second reference signal sequence, the elements of the last second reference signal sequence being output by the last second neural network, to obtain R second reference signal sequences.

[0156] Step 201a5, the first device performs a second operation based on the R second reference signal sequences, and generates a first reference signal sequence.

[0157] In the embodiment of the application, the second operation includes at least one of the following:

[0158] adding noise, superimposing the R second reference signal sequences, and splicing the R second reference signal sequences.

[0159] Thus, it can be known that, since the first device can divide the Q first elements into R element groups, input the R element groups into the R second neural networks, generate the R second reference signal sequences, and perform the adding noise (and / or superimposing the R second reference signal sequences, and / or splicing the R second reference signal sequences) operation based on the R second reference signal sequences to obtain the first reference signal sequence after the different channels, therefore, the first device can train the first neural network based on the first reference signal sequence to obtain the target neural network which can generate strong anti-noise capability and strong anti-interference capability, and thus the reliability of the communication of the first device can be improved.

[0160] In step 201b, the first device determines a target loss function based on a target channel estimation value.

[0161] In the embodiments of the present application, the target channel estimation value is obtained by channel estimation based on the first reference signal sequence.

[0162] Further optionally, in the embodiments of the present application, the target channel estimation value can be obtained by channel estimation based on the first reference signal sequence by the first device (or other devices (such as the second device in the following embodiments)).

[0163] Optionally, in the embodiments of the present application, the target channel estimation value includes any of the following:

[0164] a first channel estimation value obtained by the first device based on the first reference signal sequence;

[0165] a second channel estimation value received by the first device from the second device.

[0166] In the embodiments of the present application, the second channel estimation value is obtained by channel estimation based on the first reference signal sequence by the second device in the case that the second device receives the first reference signal sequence from the first device.

[0167] Further optionally, in the embodiments of the present application, the first device can obtain the target channel estimation value based on the first channel estimation value obtained by channel estimation based on the first reference signal sequence by the fifth neural network.

[0168] It can be understood that the output of the first neural network is connected with the input of the fifth neural network.

[0169] Specifically, the fifth neural network and the first neural network can be regarded as one large untrained neural network, and the one large untrained neural network is all deployed in the first device.

[0170] Further optionally, in embodiments of the present application, the second channel estimation value is: in a case that the second device receives the first reference signal sequence from the first device, the second device adopts the sixth neural network to perform channel estimation according to the first reference signal sequence to obtain the second channel estimation value.

[0171] It can be understood that the output of the first neural network is connected with the input of the sixth neural network.

[0172] Specifically, the sixth neural network and the first neural network can be regarded as one large untrained neural network, a part of which is deployed in the first device and another part of which is deployed in the second device.

[0173] Specifically, the first device can send the first reference signal sequence to the second device, so that the second device performs channel estimation according to the first reference signal sequence in a case that the first reference signal sequence is received, and sends channel feedback information to the first device, the channel feedback information including the second channel estimation value, so that the first device can recover the second channel estimation value according to the channel feedback information to obtain the target channel estimation value.

[0174] In embodiments of the present application, the above target loss function is used to represent the degree of deviation between the target channel estimation value and the channel true value, and the channel true value is determined according to the channel information.

[0175] Optionally, in embodiments of the present application, the above target loss function includes a first parameter between the target channel estimation value and the channel true value.

[0176] In embodiments of the present application, the first parameter includes at least one of the following: mean square error, normalized mean square error, norm, correlation coefficient, cosine similarity.

[0177] It can be understood that the first device can construct an optimization target (i.e., the target loss function) during supervised learning according to the target channel estimation value, and the optimization target can be one or more of the following:

[0178] The mean square error between the target channel estimation value and the channel true value;

[0179] The normalized mean square error between the target channel estimation value and the channel true value;

[0180] The norm between the target channel estimation value and the channel true value;

[0181] The correlation index between the target channel estimation value and the channel true value;

[0182] The correlation index includes at least one of the following: correlation coefficient, cosine similarity, etc.

[0183] Optionally, in embodiments of the present application, the target loss function comprises a second parameter obtained after performing a first operation using the target channel estimation value.

[0184] In embodiments of the present application, the first operation is a subsequent operation of channel estimation in the wireless communication system.

[0185] Optionally, in embodiments of the present application, the first operation comprises any one of the following: signal detection, precoding, and beamforming.

[0186] The second parameter is used to represent at least one of the following: wireless communication transmission correctness rate and wireless communication transmission efficiency.

[0187] Optionally, in embodiments of the present application, the second parameter comprises at least one of the following: wireless communication transmission error bit rate, wireless communication transmission error block probability, wireless communication transmission signal-to-noise ratio, wireless communication transmission signal-to-interference-and-noise ratio, wireless communication transmission spectral efficiency, and wireless communication transmission throughput.

[0188] The first operation comprises but is not limited to the following: signal detection, precoding, and beamforming.

[0189] It can be understood that the first device (or other device (such as the second device in the following embodiments)) can construct an optimization target (i.e., a target loss function) in unsupervised learning according to the target channel estimation value, and the optimization target can be one or more of the following:

[0190] Error bit probability after performing a signal detection related operation (such as signal detection, precoding, and beamforming) using the target channel estimation value;

[0191] Error block probability after performing a signal detection related operation (such as signal detection, precoding, and beamforming) using the target channel estimation value;

[0192] Signal-to-noise ratio after performing a signal detection related operation (such as signal detection, precoding, and beamforming) using the target channel estimation value;

[0193] Signal-to-interference-and-noise ratio after performing a signal detection related operation (such as signal detection, precoding, and beamforming) using the target channel estimation value;

[0194] Spectral efficiency after performing a signal detection related operation (such as signal detection, precoding, and beamforming) using the target channel estimation value;

[0195] Throughput after performing a signal detection related operation (such as signal detection, precoding, and beamforming) using the target channel estimation value.

[0196] Further optionally, in embodiments of the present application, the first device (or other device) can obtain the second parameter after performing the first operation according to the target channel estimation value.

[0197] Specifically, in a possible implementation, the first device can send the first reference signal sequence to the second device, so that the second device performs channel estimation according to the first reference signal sequence upon receiving the first reference signal sequence, and sends channel feedback information to the first device, the channel feedback information including the second channel estimation value, so that the first device can recover the second channel estimation value according to the channel feedback information, to obtain the target channel estimation value, and perform the first operation using the target channel estimation value, to obtain the second parameter, so that the first device can obtain the target loss function.

[0198] Specifically, in another possible implementation, the first device can send the first reference signal sequence to the second device, so that the second device performs channel estimation according to the first reference signal sequence upon receiving the first reference signal sequence, to obtain the second channel estimation value, to obtain the target channel estimation value, and perform the first operation using the target channel estimation value, to obtain the second parameter, to obtain the target loss function, so that the second device can send the target loss function to the first device, so that the first device can obtain the target loss function.

[0199] Step 201c, the first device trains the first neural network according to the target loss function, to obtain the target neural network.

[0200] Further optionally, in embodiments of the present application, the first device can train the first neural network according to the target loss function using an optimization algorithm.

[0201] As can be seen, since the first device can input the Q first elements into the first neural network to generate the first reference signal sequence after the channel, and determine the target loss function based on the target channel estimation value obtained by the first device (or other device) according to the channel estimation of the first reference signal sequence, and train the first neural network according to the target loss function, the target neural network that can generate strong anti-noise capability and anti-interference capability can be obtained, and thus the reliability of communication of the first device can be improved.

[0202] Optionally, in embodiments of the present application, in combination with Figure 4 As shown in Figure 6 The above step 201 can be implemented by the following step 201d.

[0203] Step 201d, the first device trains the first neural network according to the target constraint condition using the channel information, to obtain the target neural network.

[0204] In the embodiments of the present application, the target constraint condition includes any one of the following:

[0205] The power value corresponding to each target weight parameter in the N target weight parameters is a first preset value;

[0206] The total power value corresponding to the N target weight parameters is less than or equal to a second preset value.

[0207] In the actual communication system, the value of the element of the reference signal sequence is generally limited, and not any value can be taken. The specific limitations include, for example, a constant modulus limitation (for example, the power of each element of the reference signal sequence is 1), a total power limitation (for example, the total power of the reference signal sequence is less than or equal to a specific value), and the like.

[0208] When the constant modulus limitation is adopted, the first device can calculate the power of the target element corresponding to each target weight parameter according to the real part parameter and the imaginary part parameter of the target weight parameter, and then normalize the real part parameter and the imaginary part parameter.

[0209] When the total power limitation is adopted, the first device can calculate the total power of the entire target reference signal sequence (i.e., the N target weight parameters), and adjust the real part parameter and the imaginary part parameter of each target weight parameter in combination with the total power limitation.

[0210] It should be noted that the description of training the first neural network by the first device using the channel information can refer to the specific description in the above embodiments, and the present application embodiments will not be repeated here.

[0211] As can be seen, since the first device can train the first neural network based on the target constraint condition, the situation that the value of the element of the target reference signal sequence exceeds the limitation can be avoided, and thus the reliability of the communication of the first device can be improved.

[0212] It should be noted that the reference signal sequence generation method provided by the present application can be executed by the first device, or a control module in the first device for executing the reference signal sequence generation method. In the embodiments of the present application, the first device executes the reference signal sequence generation method as an example to illustrate the first device provided by the present application.

[0213] Figure 7 A flowchart of a reference sequence generation method provided by an embodiment of the present application is shown. As shown in Figure 7 The reference sequence generation method provided by the present application can include the following steps 401 to 403.

[0214] Step 401, the second device receives at least one third reference signal sequence from at least one third device.

[0215] In the embodiment of the application, the at least one third reference signal sequence comprises a first reference signal sequence sent by a first device in the at least one third device.

[0216] Optionally, in the embodiment of the application, each third device in the at least one third device can be a UE or a network side device.

[0217] Optionally, in the embodiment of the application, each third device in the at least one third device corresponds to at least one third reference signal sequence respectively.

[0218] It can be understood that the number of third devices is less than or equal to the number of third reference signal sequences.

[0219] For each third device in the at least one third device, the second device can receive one third reference signal sequence or multiple third reference signal sequences from one third device.

[0220] Step 402, the second device respectively performs a third operation based on each third reference signal sequence to obtain a corresponding fourth reference signal sequence.

[0221] Optionally, in the embodiment of the application, the second device can perform the third operation based on a first third reference signal sequence to obtain a fourth reference signal sequence corresponding to the first third reference signal sequence, perform the third operation based on a second third reference signal sequence to obtain a fourth reference signal sequence corresponding to the second third reference signal sequence, and so on, until performing the third operation based on a last third reference signal sequence to obtain a fourth reference signal sequence corresponding to the last third reference signal sequence, to obtain at least one fourth reference signal sequence.

[0222] Optionally, in the embodiment of the application, the third operation comprises at least one of the following:

[0223] adding noise, superimposing the at least one third reference signal sequence, and splicing the at least one third reference signal sequence.

[0224] Step 403, the second device sends a second channel estimation value to the first device.

[0225] In the embodiment of the application, the second channel estimation value is a channel estimation value obtained by the second device performing channel estimation according to the fourth reference signal sequence corresponding to the first device; the second channel estimation value is used for the first device to train a target neural network; and the target neural network is used to generate a target reference signal sequence.

[0226] The reference signal sequence generation method provided in this application embodiment allows a second device to receive at least one third reference signal sequence from at least one third device, and perform a third operation based on each third reference signal sequence to obtain a corresponding fourth reference signal sequence. This allows the second device to send a second channel estimate to a first device among the at least one third device. The second channel estimate is obtained by the second device performing channel estimation based on the fourth reference signal sequence corresponding to the first device. This second channel estimate is used by the first device to train a target neural network, which is then used to generate a target reference signal sequence. Since the second device can receive at least one third reference signal sequence and obtain a corresponding fourth reference signal sequence based on each third reference signal sequence, the second device can send a second channel estimate to the first device for training a target neural network (which generates the target reference signal sequence). This allows the first device to train the target neural network and obtain the target reference signal sequence based on it, instead of generating the reference signal sequence through a PN sequence (or Zadoff-Chu sequence). Therefore, the noise immunity and interference immunity of the target reference signal sequence are stronger than those of the reference signal sequence, thereby reducing the impact on the target reference signal sequence when passing through the channel. This improves the reliability of the second device's communication.

[0227] It should be noted that the reference signal sequence generation method provided in this application embodiment can be executed by a second device, or by a control module within the second device for executing the reference signal sequence generation method. This application embodiment uses the execution of the reference signal sequence generation method by a second device as an example to illustrate the second device provided in this application embodiment.

[0228] Figure 8 This illustration shows a possible structural diagram of a reference signal sequence generation device involved in an embodiment of this application, wherein the reference signal sequence generation device is a first reference signal sequence generation device. Figure 8 As shown, the first reference signal sequence generation device 50 may include: an acquisition module 51.

[0229] The acquisition module 51 is used to acquire the target neural network and, based on the target neural network, acquire the target reference signal sequence. The target neural network comprises N target neuron groups, each corresponding to a target weight parameter; the target reference signal sequence comprises N target elements, each corresponding one-to-one with one of the N target weight parameters, where N is a positive integer.

[0230] In a possible implementation, each of the N target neuron groups comprises L first neurons and M second neurons, each first neuron corresponding to a first real part parameter, each second neuron corresponding to a first imaginary part parameter, L and M being positive integers. The target weight parameter corresponding to each of the target neuron groups is determined according to the L first real part parameters and the M first imaginary part parameters.

[0231] In a possible implementation, the target neural network is trained based on channel information of the first reference signal sequence generation apparatus 50. The channel information includes any of the following: a channel vector and a channel matrix.

[0232] In a possible implementation, the first reference signal sequence generation apparatus 50 further comprises a training module. The training module is configured to train the first neural network based on the channel information to obtain the target neural network.

[0233] In a possible implementation, the channel information includes Q first elements, Q being a positive integer. The first reference signal sequence generation apparatus 50 provided in the embodiments of the present application further comprises a generation module and a determination module. The generation module is configured to input the Q first elements into the first neural network to generate the first reference signal sequence. The determination module is configured to determine a target loss function based on a target channel estimation value, the target channel estimation value being obtained by performing channel estimation on the first reference signal sequence. The training module is specifically configured to train the first neural network based on the target loss function determined by the determination module. The target loss function is used to represent a degree of deviation between the target channel estimation value and a channel true value, and the channel true value is determined based on the channel information.

[0234] In a possible implementation, the first neural network comprises N first neuron groups, each first neuron group corresponding to a first weight parameter. The first reference signal sequence is obtained by performing multiplication operation on the Q first elements and the N first weight parameters.

[0235] In a possible implementation, each first element corresponds to a second real part parameter and a second imaginary part parameter. The first reference signal sequence includes N second elements. The generation module is specifically configured to input the Q second real part parameters and the Q second imaginary part parameters into the first neural network to obtain N third real part parameters and N third imaginary part parameters, and determine the N second elements based on the N third real part parameters and the N third imaginary part parameters, respectively.

[0236] In a possible implementation, the target loss function comprises a first parameter between the target channel estimation value and the channel true value. The first parameter comprises at least one of a mean square error, a normalized mean square error, a norm, a correlation coefficient, and a cosine similarity.

[0237] In a possible implementation, the target loss function comprises a second parameter obtained after a first operation is performed on the target channel estimation value. The first operation is a subsequent operation of channel estimation in the wireless communication system, and the second parameter is used to represent at least one of a wireless communication transmission accuracy and a wireless communication transmission efficiency.

[0238] In a possible implementation, the first neural network comprises R second neural networks, where R is a positive integer greater than 1. The generation module is specifically configured to determine R element groups including different first elements according to the Q first elements, input each element group into each second neural network respectively, generate a second reference signal sequence corresponding to each second neural network, and obtain R second reference signal sequences; and perform a second operation based on the R second reference signal sequences to generate the first reference signal sequence. The second operation comprises at least one of adding noise, superimposing the R second reference signal sequences, and splicing the R second reference signal sequences.

[0239] In a possible implementation, the target channel estimation value comprises any one of a first channel estimation value obtained by the first reference signal sequence generation apparatus 50 performing channel estimation on the first reference signal sequence, and a second channel estimation value received by the first reference signal sequence generation apparatus 50 from the second reference signal sequence generation apparatus. The second channel estimation value is a channel estimation value obtained by the second reference signal sequence generation apparatus performing channel estimation on the first reference signal sequence in a case where the second reference signal sequence generation apparatus receives the first reference signal sequence from the first reference signal sequence generation apparatus 50.

[0240] In a possible implementation, the training module is specifically configured to train the first neural network based on a target constraint condition and the channel information. The target constraint condition comprises at least one of a first preset value of a power value corresponding to each target weight parameter, and a second preset value of a total power value corresponding to the N target weight parameters.

[0241] In a possible implementation, the first reference signal sequence generation apparatus 50 provided by the embodiment of the present application further includes a determination module. The determination module is configured to determine the first neural network according to a first physical parameter. The first physical parameter includes at least one of the following: a bundling size, a RB, a PRB, a number of MUs, and a density of a time-frequency domain pattern of the reference signal sequence in a time-frequency domain.

[0242] In a possible implementation, the channel information is received by the first reference signal sequence generation apparatus 50 from a second reference signal sequence generation apparatus.

[0243] In a possible implementation, the acquisition module 51 is specifically configured to determine the target neural network according to a second physical parameter. The second physical parameter includes at least one of the following: a bundling size, a RB, a PRB, a number of MUs, and a density of a time-frequency domain pattern of the reference signal sequence in a time-frequency domain.

[0244] In a possible implementation, the first reference signal sequence generation apparatus 50 provided by the embodiment of the present application further includes a sending module. The sending module is configured to send, to a second reference signal sequence generation apparatus, first signaling. The first signaling carries: a target reference signal sequence, an identifier of the target reference signal sequence, or information obtained by compressing the target reference signal sequence.

[0245] In a possible implementation, the first signaling includes at least one of the following: RRC signaling, layer 1 signaling of a PDCCH, information of a PDSCH, signaling of a MAC, SIB, layer 1 signaling of a PUCCH, target message information of a PRACH, information of a PUSCH, XN interface signaling, PC5 interface signaling, and Sidelink interface instruction. The target message includes at least one of the following: MSG 1 information, MSG 2 information, MSG 3 information, MSG 4 information, MSG A information, and MSG B information.

[0246] The reference signal sequence generation apparatus provided by the embodiment of the present application has the following advantages. The first reference signal sequence generation apparatus acquires the target reference signal sequence according to the target neural network. The N target weight parameters of the target neural network are in one-to-one correspondence with the N target elements, rather than generating the reference signal sequence by using a PN sequence (or a Zadoff-Chu sequence). Therefore, the anti-noise capability and the anti-interference capability of the target reference signal sequence are stronger than those of the reference signal sequence. As a result, the influence on the target reference signal sequence when passing through a channel can be reduced, and thus the reliability of communication of the first reference signal sequence generation apparatus can be improved.

[0247] The first reference signal sequence generation apparatus in the embodiments of the present applicationapplicationbe a device, a device with an operating system, or an electronic device, andapplicationalso be a component in a terminal, an integrated circuit, or a chip. The device or the electronic deviceapplicationbe a mobile terminal or a non-mobile terminal. Exemplarily, the mobile terminalapplicationinclude, but is not limited to, the types of the terminal 11 listed above, and the non-mobile terminalapplicationbe a server, a network attached storage (NAS), a personal computer (PC), a television (TV), a teller machine, or a self-service machine, and the like, without specific limitation in the embodiments of the present application.

[0248] The first reference signal sequence generation apparatus provided in the embodiments of the present applicationapplicationimplement the method embodiments of the present applicationapplicationimplement the various processes, and achieve the same technical effects. To avoid repetition, details are not described herein. Figures 1 to 6

[0249] Figure 9 A possible structural schematic diagram of a reference signal sequence generation apparatus involved in the embodiments of the present application is shown, which is a second reference signal sequence generation apparatus. As shown in Figure 9 The second reference signal sequence generation apparatus 60applicationinclude a receiving module 61, an executing module 62, and a sending module 63. The receiving module 61applicationbe configured to receive at least one third reference signal sequence from at least one third reference signal sequence generation apparatus, the at least one third reference signal sequence including a first reference signal sequence sent by a first reference signal sequence generation apparatus in the at least one third reference signal sequence generation apparatus. The executing module 62applicationbe configured to perform a third operation on each third reference signal sequence received by the receiving module 61, respectively, to obtain a corresponding fourth reference signal sequence. The sending module 63applicationbe configured to send a second channel estimation value to the first reference signal sequence generation apparatus. The second channel estimation valueapplicationbe a channel estimation value obtained by the second reference signal sequence generation apparatus 60 according to the fourth reference signal sequence corresponding to the first reference signal sequence generation apparatus, and the second channel estimation valueapplicationbe used for training a target neural network by the first reference signal sequence generation apparatus, and the target neural networkapplicationbe used for generating a target reference signal sequence.

[0250] In a possible implementation, the third operationapplicationinclude at least one of the following: adding noise, superimposing at least one third reference signal sequence, and splicing at least one third reference signal sequence.

[0251] ​The reference signal sequence generation apparatus provided in the embodiments of the present application can receive at least one third reference signal sequence, and obtain a corresponding fourth reference signal sequence based on each third reference signal sequence, so that the second reference signal sequence generation apparatus can send the second channel estimation value for training a target neural network (the target neural network is used to generate a target reference signal sequence) to the first reference signal sequence generation apparatus, so that the first reference signal sequence generation apparatus can train the target neural network, and obtain the target reference signal sequence based on the target neural network, rather than generating the reference signal sequence through a PN sequence (or a Zadoff-Chu sequence), so that the anti-noise capability and the anti-interference capability of the target reference signal sequence are stronger than the anti-noise capability and the anti-interference capability of the reference signal sequence, thereby reducing the influence of the target reference signal sequence when passing through a channel, and thus the reliability of communication of the second reference signal sequence generation apparatus can be improved.

[0252] The second reference signal sequence generation apparatus in the embodiments of the present application can be a device, a device with an operating system, or an electronic device, and can also be a component in a terminal, an integrated circuit, or a chip. The device or the 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 the terminal 11 listed above, and the non-mobile terminal can be a server, a network attached storage (NAS), a personal computer (PC), a television (TV), a cashier machine, or a self-service machine, and the like, which are not limited in the embodiments of the present application.

[0253] The second reference signal sequence generation apparatus provided in the embodiments of the present application can implement the method embodiments Figure 7 The method embodiments implement various processes and achieve the same technical effects, and thus details are not repeated here.

[0254] Optionally, as shown in Figure 10 The embodiments of the present application also provide a communication device 70, which includes a processor 71, a memory 72, a program or instruction stored in the memory 72 and executable on the processor 71. For example, when the communication device 70 is a terminal, the program or instruction is executed by the processor 71 to implement various processes of the above-mentioned reference signal sequence generation method embodiments and achieve the same technical effects. When the communication device 70 is a network side device, the program or instruction is executed by the processor 71 to implement various processes of the above-mentioned reference signal sequence generation method embodiments and achieve the same technical effects. To avoid repetition, details are not repeated here.

[0255] The terminal provided in the embodiments of the present application also includes a processor and a communication interface. The processor is configured to acquire a target neural network, and acquire a target reference signal sequence based on the target neural network. The target neural network includes N target neuron groups, and each target neuron group corresponds to a target weight parameter. The target reference signal sequence includes N target elements corresponding to the N target weight parameters, and N is a positive integer.

[0256] The terminal embodiment corresponds to the terminal-side method embodiment described above. Each implementation process and implementation manner of the method embodiment can be applied to the terminal embodiment and achieve the same technical effects. Specifically, Figure 11 A hardware structure diagram of a terminal according to an embodiment of the present application is shown in FIG. 10.

[0257] The terminal 100 includes, but is not limited to, at least some of a radio frequency unit 101, a network module 102, an audio output unit 103, an input unit 104, a sensor 105, a display unit 106, a user input unit 107, an interface unit 108, a memory 109, and a processor 110.

[0258] Those skilled in the art can understand that the terminal 100 can also include a power supply (such as a battery) for powering each component. The power supply can be logically connected to the processor 110 through a power management system, so as to realize functions such as management of charging, discharging, and power consumption management through the power management system. Figure 11 The terminal structure shown in FIG. 10 does not constitute a limitation on the terminal. The terminal can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components, which will not be described here.

[0259] It should be understood that in the embodiments of the present application, the input unit 104 can include a graphics processing unit (GPU) 1041 and a microphone 1042. The graphics processing unit 1041 processes image data of a still picture or a video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The display unit 106 can include a display panel 1061, which can be configured in the form of a liquid crystal display, an organic light-emitting diode, etc. The user input unit 107 includes a touch panel 1071 and other input devices 1072. The touch panel 1071 is also called a touch screen. The touch panel 1071 can include a touch detection device and a touch controller. The other input devices 1072 can include, but are not limited to, a physical keyboard, function keys (such as volume control keys, on-off keys, etc.), a trackball, a mouse, a joystick, etc., which will not be described here.

[0260] In the embodiments of the present application, the radio frequency unit 101 receives the downlink data from the network side device, and then sends the data to the processor 110 for processing. In addition, the radio frequency unit 101 sends the uplink data to the network side device. Generally, the radio frequency unit 101 includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc.

[0261] The memory 109 can be used to store software programs or instructions and various data. The memory 109 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, application programs or instructions required by at least one function (such as a sound playing function, an image playing function, etc.), etc. In addition, the memory 109 can include a high-speed random access memory, and can also include a non-volatile memory, which can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. For example, at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state memory device.

[0262] The processor 110 can include one or more processing units; optionally, the processor 110 can 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 communication, such as a baseband processor. It can be understood that the above-mentioned modem processor can also not be integrated into the processor 110.

[0263] The processor 110 is configured to obtain a target neural network, and obtain a target reference signal sequence based on the target neural network.

[0264] The target neural network includes N target neuron groups, each target neuron group corresponding to a target weight parameter; the target reference signal sequence includes N target elements corresponding to the N target weight parameters, and N is a positive integer.

[0265] The terminal provided in the embodiments of the present application can reduce the influence of the target reference signal sequence when passing through a channel, thereby improving the reliability of terminal communication.

[0266] Optionally, in the embodiments of the present application, the processor 110 is further configured to train the first neural network by using the channel information to obtain the target neural network.

[0267] Optionally, in the embodiments of the present application, the channel information includes Q first elements, and Q is a positive integer.

[0268] The processor 110 is specifically configured to input the Q first elements into the first neural network to generate a first reference signal sequence, determine a target loss function based on a target channel estimation value, the target channel estimation value being obtained by performing channel estimation on the first reference signal sequence, and train the first neural network according to the target loss function.

[0269] The target loss function is used to represent the deviation degree between the target channel estimation value and a channel true value, and the channel true value is determined according to the channel information.

[0270] Optionally, in the embodiments of the present application, each first element corresponds to a second real part parameter and a second imaginary part parameter, and the first reference signal sequence includes N second elements.

[0271] The processor 110 is specifically configured to input the Q second real part parameters and the Q second imaginary part parameters into the first neural network respectively to obtain N third real part parameters and N third imaginary part parameters, and determine the N second elements according to the N third real part parameters and the N third imaginary part parameters respectively.

[0272] Optionally, in the embodiments of the present application, the first neural network includes R second neural networks, and R is a positive integer greater than 1.

[0273] The processor 110 is specifically configured to determine R element groups according to the Q first elements, different element groups including different first elements, input each element group into each second neural network respectively to generate a second reference signal sequence corresponding to each second neural network to obtain R second reference signal sequences, and perform a second operation based on the R second reference signal sequences to generate the first reference signal sequence.

[0274] The second operation includes at least one of adding noise, superimposing R second reference signal sequences, and splicing R second reference signal sequences.

[0275] Optionally, in embodiments of the present application, the processor 110 is specifically configured to train the first neural network based on a target constraint condition and channel information.

[0276] The target constraint condition includes any one of the following: a power value corresponding to each target weight parameter is a first preset value; and a total power value corresponding to N target weight parameters is less than or equal to a second preset value.

[0277] Optionally, in embodiments of the present application, the processor 110 is further configured to determine the first neural network according to a first physical parameter.

[0278] The first physical parameter includes at least one of the following: a bundling size, an RB, a PRB, a number of MUs, and a density of a time-frequency domain pattern of a reference signal sequence in a time-frequency domain.

[0279] Optionally, in embodiments of the present application, the processor 110 is specifically configured to determine the target neural network according to a second physical parameter.

[0280] The second physical parameter includes at least one of the following: a bundling size, an RB, a PRB, a number of MUs, and a density of a time-frequency domain pattern of a reference signal sequence in a time-frequency domain.

[0281] Optionally, in embodiments of the present application, the radio frequency unit 101 is configured to send first signaling to a second device, and the first signaling carries: a target reference signal sequence, an identifier of the target reference signal sequence, or information obtained by compressing the target reference signal sequence.

[0282] The embodiment of the present application further provides a terminal, comprising a processor and a communication interface, the communication interface is used for receiving at least one third reference signal sequence from at least one third device, the at least one third reference signal sequence comprises: a first reference signal sequence sent by a first device in the at least one third device; the processor is used for respectively performing a third operation based on each third reference signal sequence to obtain a corresponding fourth reference signal sequence; and the communication interface is further used for sending a second channel estimation value to the first device. Wherein, the second channel estimation value is: a channel estimation value obtained by the terminal according to the fourth reference signal sequence corresponding to the first device; the second channel estimation value is used for the first device to train a target neural network; and the target neural network is used for generating a target reference signal sequence. The terminal embodiment is corresponding to the terminal side method embodiment described above, each implementation process and implementation manner of the method embodiment can be applied to the terminal embodiment, and the same technical effects can be achieved. Specifically, Figure 11 A hardware structure diagram of a terminal for implementing the embodiment of the present application.

[0283] The terminal 100 comprises, but is not limited to, at least part of components such as a radio frequency unit 101, a network module 102, an audio output unit 103, an input unit 104, a sensor 105, a display unit 106, a user input unit 107, an interface unit 108, a memory 109, and a processor 110.

[0284] Those skilled in the art can understand that the terminal 100 can further comprise a power supply (such as a battery) for supplying power to each component, and the power supply can be logically connected to the processor 110 through a power management system, so as to realize functions such as management of charging, discharging, and power consumption management through the power management system. Figure 11 The terminal structure shown in the figure does not constitute a limitation on the terminal, and the terminal can comprise more or fewer components than those shown in the figure, or some components can be combined, or different components can be arranged, which will not be described here.

[0285] It should be understood that in the embodiments of the present application, the input unit 104 can include a graphics processing unit (GPU) 1041 and a microphone 1042. The graphics processing unit 1041 processes image data of a still picture or a video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The display unit 106 can include a display panel 1061, which can be configured in the form of a liquid crystal display, an organic light-emitting diode, etc. The user input unit 107 includes a touch panel 1071 and other input devices 1072. The touch panel 1071 is also called a touch screen. The touch panel 1071 can include two parts of a touch detection device and a touch controller. The other input devices 1072 can include, but are not limited to, a physical keyboard, function keys (such as volume control keys, on-off keys, etc.), a trackball, a mouse, a joystick, and the like, which will not be described here.

[0286] In the embodiments of the present application, the radio frequency unit 101 receives downlink data from a network side device and processes the data by the processor 110. In addition, the radio frequency unit 101 sends uplink data to the network side device. Generally, the radio frequency unit 101 includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc.

[0287] The memory 109 can be used to store software programs or instructions and various data. The memory 109 can mainly include a storage program or instruction area and a storage data area, wherein the storage program or instruction area can store an operating system, application programs or instructions required by at least one function (such as a sound playing function, an image playing function, etc.), etc. In addition, the memory 109 can include a high-speed random access memory, and can also include a non-volatile memory, which can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM) or a flash memory. For example, at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state memory device.

[0288] The processor 110 can include one or more processing units; optionally, the processor 110 can 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 communication, such as a baseband processor. It can be understood that the above-mentioned modem processor can also not be integrated into the processor 110.

[0289] The radio frequency unit 101 is configured to receive at least one third reference signal sequence from at least one third device, the at least one third reference signal sequence comprising a first reference signal sequence transmitted by a first device in the at least one third device.

[0290] The processor 110 is configured to perform a third operation on each third reference signal sequence to obtain a corresponding fourth reference signal sequence.

[0291] The radio frequency unit 101 is further configured to transmit a second channel estimation value to the first device.

[0292] The second channel estimation value is a channel estimation value obtained by performing channel estimation on the fourth reference signal sequence corresponding to the first device by the terminal; the second channel estimation value is used for training a target neural network by the first device; and the target neural network is used for generating a target reference signal sequence.

[0293] The terminal provided in the embodiments of the present application can receive at least one third reference signal sequence and obtain a corresponding fourth reference signal sequence based on each third reference signal sequence, so that the terminal can transmit a second channel estimation value for training a target neural network (the target neural network is used for generating a target reference signal sequence) to the first device, so that the first device can train the target neural network and obtain the target reference signal sequence based on the target neural network, rather than generating the reference signal sequence through a PN sequence (or a Zadoff-Chu sequence), so that the anti-noise capability and the anti-interference capability of the target reference signal sequence are stronger than those of the reference signal sequence, thereby reducing the influence of the target reference signal sequence when passing through a channel, so that the reliability of communication of the terminal can be improved.

[0294] The embodiment of the present application further provides a network side device, comprising a processor and a communication interface, the processor is configured to acquire a target neural network, and acquire a target reference signal sequence based on the target neural network; wherein the target neural network comprises N target neuron groups, each target neuron group corresponds to a target weight parameter; the target reference signal sequence comprises N target elements corresponding to the N target weight parameters, and N is a positive integer; or the communication interface is configured to receive at least one third reference signal sequence from at least one third device, the at least one third reference signal sequence comprises a first reference signal sequence sent by the first device. The processor is configured to perform a third operation based on each third reference signal sequence to obtain a corresponding fourth reference signal sequence. The communication interface is further configured to send a second channel estimation value to the first device. Wherein the second channel estimation value is a channel estimation value obtained by the network side device according to the fourth reference signal sequence corresponding to the first device; the second channel estimation value is used for the first device to train the target neural network; and the target neural network is used to generate the target reference signal sequence.

[0295] The network side device embodiment corresponds to the network side device method embodiment, and each implementation process and implementation manner of the method embodiment can be applied to the network side device embodiment and achieve the same technical effects.

[0296] Specifically, the embodiment of the present application further provides a network side device. As shown in the above Figure 12 The network side device 80 comprises an antenna 81, a radio frequency device 82 and a baseband device 83. The antenna 81 is connected with the radio frequency device 82. In the uplink direction, the radio frequency device 82 receives information through the antenna 81, and sends the received information to the baseband device 83 for processing. In the downlink direction, the baseband device 83 processes the information to be sent and sends it to the radio frequency device 82. The radio frequency device 82 processes the received information and sends it out through the antenna 81.

[0297] The above frequency band processing device can be located in the baseband device 83. The method executed by the network side device in the above embodiment can be implemented in the baseband device 83. The baseband device 83 comprises a processor 84 and a memory 85.

[0298] The baseband device 83 may, for example, comprise at least one baseband board, and a plurality of chips are arranged on the baseband board, as shown in the above Figure 12 One of the chips is, for example, the processor 84, which is connected with the memory 85 to call the program in the memory 85 and execute the operation of the network side device shown in the above method embodiment.

[0299] The baseband device 83 can also include a network interface 86 for interacting with the radio frequency device 82, which can be, for example, a common public radio interface (CPRI).

[0300] Specifically, the network side device of the embodiment of the present application further comprises instructions or programs stored on the memory 85 and executable on the processor 84, and the processor 84 invokes the instructions or programs in the memory 85 to execute the method performed by each module shown in the above embodiment and achieve the same technical effects. To avoid repetition, the details are not described here. Figure 12 The method performed by each module shown in the above embodiment and achieve the same technical effects. To avoid repetition, the details are not described here.

[0301] The embodiment of the present application also provides a readable storage medium, and the readable storage medium stores programs or instructions, which are executed by a processor to implement each process of the above-mentioned reference signal sequence generation method embodiment and achieve the same technical effects. To avoid repetition, the details are not described here.

[0302] The processor is the processor in the terminal in the above-mentioned embodiment. The readable storage medium includes a computer readable storage medium, such as a computer readable memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk, etc.

[0303] The embodiment of the present application further provides a chip, and the chip includes a processor and a communication interface. The communication interface is coupled with the processor, and the processor is used to run programs or instructions to implement each process of the above-mentioned reference signal sequence method embodiment and achieve the same technical effects. To avoid repetition, the details are not described here.

[0304] It should be understood that the chip mentioned in the embodiment of the present application can also be referred to as a system chip, a system chip, a chip system or a system on chip, etc.

[0305] It should be noted that, in the present document, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the element. Furthermore, it is to be understood that the method and apparatus of the present application can be carried out by more than one process, method, article, or apparatus either simultaneously, concurrently, or with intervening action that are carried out at the same time, in any order, or in an overlapping manner. For example, the described method can be performed in a different order or simultaneously, and the various steps can be combined or omitted, or additional steps can be added, without departing from the scope of the described method. Also, features described with respect to certain examples can be combined in other examples.

[0306] From the above description of the embodiments, it is apparent that the above-described method can be implemented by means of software and the requisite universal hardware platform, of course, but in many cases the former is the preferred implementation. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product stored in a storage medium (such as a ROM / RAM, a magnetic disk, an optical disk), including a number of instructions for causing a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network-side device, etc.) to execute the method described in the various embodiments of the present application.

[0307] 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-described specific embodiments, which are merely illustrative rather than restrictive, and a person of ordinary skill in the art can make many forms under the inspiration of the present application without departing from the scope of the present application and the protection scope of the claims.

Claims

1. A method for generating a reference signal sequence, characterized in that, The method includes: The first device acquires the target neural network; the target neural network is trained based on the channel information of the first device. The first device acquires a target reference signal sequence based on the target neural network; The target neural network includes N target neuron groups, and each target neuron group corresponds to a target weight parameter; The target reference signal sequence includes N target elements, each corresponding to one of the N target weight parameters, where N is a positive integer.

2. The method according to claim 1, characterized in that, Each of the N target neuron groups includes: L first neurons and M second neurons, each first neuron corresponding to a first real part parameter, and each second neuron corresponding to a first imaginary part parameter, where L and M are both positive integers; The target weight parameters corresponding to each target neuron group are determined based on L first real part parameters and M first imaginary part parameters.

3. The method according to claim 1, characterized in that, The channel information includes any one of the following: channel vector, channel matrix.

4. The method according to claim 3, characterized in that, Before the first device acquires the target neural network, the method further includes: The first device uses the channel information to train the first neural network to obtain the target neural network.

5. The method according to claim 4, characterized in that, The channel information includes: Q first elements, where Q is a positive integer; The first device uses the channel information to train the first neural network, including: The first device inputs Q first elements into the first neural network to generate a first reference signal sequence; The first device determines the target loss function based on the target channel estimate, which is obtained by channel estimation based on the first reference signal sequence. The first device trains the first neural network according to the target loss function; The target loss function is used to characterize the degree of deviation between the target channel estimate and the true channel value. The channel truth value is determined based on the channel information.

6. The method according to claim 5, characterized in that, The first neural network includes N first neuron groups, and each first neuron group corresponds to a first weight parameter; The first reference signal sequence is calculated by multiplying the Q first elements and N first weight parameters.

7. The method according to claim 5, characterized in that, Each first element corresponds to a second real part parameter and a second imaginary part parameter; The first reference signal sequence includes: N second elements; The first device inputs Q first elements into the first neural network to generate a first reference signal sequence, including: The first device inputs Q second real part parameters and Q second imaginary part parameters into the first neural network respectively to obtain N third real part parameters and N third imaginary part parameters output; The first device determines the N second elements based on the N third real part parameters and the N third imaginary part parameters, respectively.

8. The method according to claim 5, characterized in that, The target loss function includes: a first parameter between the target channel estimate and the true channel value; The first parameter includes at least one of the following: mean squared error, normalized mean squared error, norm, correlation coefficient, and cosine similarity.

9. The method according to claim 5, characterized in that, The target loss function includes: a second parameter obtained after performing the first operation using the target channel estimate; The first operation is: a subsequent operation of channel estimation in a wireless communication system; The second parameter is used to characterize at least one of the following: wireless communication transmission accuracy and wireless communication transmission efficiency.

10. The method according to claim 5, characterized in that, The first neural network includes: R second neural networks, where R is a positive integer greater than 1; The first device inputs Q first elements into the first neural network to generate a first reference signal sequence, including: The first device determines R element groups based on the Q first elements, and different element groups include different first elements; The first device inputs each element group into each second neural network to generate a second reference signal sequence corresponding to each second neural network, thereby obtaining R second reference signal sequences; The first device performs a second operation based on the R second reference signal sequences to generate the first reference signal sequence; The second operation includes at least one of the following: Add noise, superimpose the R second reference signal sequences, and splice the R second reference signal sequences.

11. The method according to claim 5, characterized in that, The target channel estimate includes any one of the following: The first channel estimate obtained by the first device through channel estimation based on the first reference signal sequence; The second channel estimate received by the first device from the second device; The second channel estimate is the channel estimate obtained by the second device performing channel estimation based on the first reference signal sequence when the second device receives the first reference signal sequence from the first device.

12. The method according to claim 4, characterized in that, The first device uses the channel information to train the first neural network, including: The first device trains the first neural network based on the target constraints and using the channel information; The target constraint includes any one of the following: The power value corresponding to each target weight parameter is the first preset value; The total power value corresponding to the N target weight parameters is less than or equal to the second preset value.

13. The method according to claim 4, characterized in that, Before the first device trains the first neural network using the channel information, the method further includes: The first device determines the first neural network based on the first physical parameters; The first physical parameter includes at least one of the following: Bundling size; Resource block (RB); Physical Resource Block (PRB); The number of MUs for multiple users; The density of the time-frequency domain pattern of the reference signal sequence in the time-frequency domain.

14. The method according to claim 3, characterized in that, The channel information is received by the first device from the second device.

15. The method according to claim 1, characterized in that, The first device acquires the target neural network, including: The first device determines the target neural network based on the second physical parameters; The second physical parameter includes at least one of the following: bundling size; RB; PRB; The number of MUs; The density of the time-frequency pattern of the reference signal sequence in the time-frequency domain.

16. The method according to claim 1, characterized in that, After acquiring the target reference signal sequence, the method further includes: The first device sends a first signaling message to the second device. The first signaling message carries: the target reference signal sequence, or the identifier of the target reference signal sequence, or information obtained by compressing the target reference signal sequence.

17. The method according to claim 16, characterized in that, The first signaling includes at least one of the following: Radio Resource Control (RRC) signaling; Layer 1 signaling of the Physical Downlink Control Channel (PDCCH); Information about the Physical Downlink Shared Channel (PDSCH); Signaling of the Media Access Control Layer (MAC) control unit; System Information Block (SIB); Layer 1 signaling of the Physical Uplink Control Channel (PUCCH); Target message information for the Physical Random Access Channel (PRACH); Information about the Physical Uplink Shared Channel (PUSCH); XN interface signaling; PC5 interface signaling; Sidelink interface commands; The target message includes at least one of the following: message MSG 1, message MSG 2, message MSG 3, message MSG 4, message MSG A, and message MSG B.

18. A method for generating a reference signal sequence, characterized in that, The method includes: The second device receives at least one third reference signal sequence from at least one third device, the at least one third reference signal sequence comprising: a first reference signal sequence transmitted by the first device among the at least one third device; The second device performs a third operation based on each third reference signal sequence to obtain the corresponding fourth reference signal sequence; The second device sends a second channel estimate to the first device; Wherein, the second channel estimate is: the channel estimate obtained by the second device performing channel estimation based on the fourth reference signal sequence corresponding to the first device; the second channel estimate is used by: the first device to train the target neural network; The target neural network is used to generate a target reference signal sequence; The target neural network includes N target neuron groups, and each target neuron group corresponds to a target weight parameter; The target reference signal sequence includes N target elements, each corresponding to one of the N target weight parameters, where N is a positive integer.

19. The method according to claim 18, characterized in that, The third operation includes at least one of the following: Add noise, superimpose the at least one third reference signal sequence, and splice the at least one third reference signal sequence.

20. A reference signal sequence generation apparatus, wherein the reference signal sequence generation apparatus is a first reference signal sequence generation apparatus, characterized in that, The first reference signal sequence generation device includes: an acquisition module; The acquisition module is used to acquire a target neural network; and based on the target neural network, acquire a target reference signal sequence; the target neural network is trained based on the channel information of the first reference signal sequence generation device; The target neural network includes N target neuron groups, and each target neuron group corresponds to a target weight parameter; The target reference signal sequence includes N target elements, each corresponding to one of the N target weight parameters, where N is a positive integer.

21. The reference signal sequence generation apparatus according to claim 20, characterized in that, Each of the N target neuron groups includes: L first neurons and M second neurons, each first neuron corresponding to a first real part parameter, and each second neuron corresponding to a first imaginary part parameter, where L and M are both positive integers; The target weight parameters corresponding to each target neuron group are determined based on L first real part parameters and M first imaginary part parameters.

22. The reference signal sequence generation apparatus according to claim 20, characterized in that, The channel information includes any one of the following: channel vector, channel matrix.

23. The reference signal sequence generation apparatus according to claim 22, characterized in that, The first reference signal sequence generation device further includes: a training module; The training module is used to train the first neural network using the channel information to obtain the target neural network.

24. The reference signal sequence generation apparatus according to claim 23, characterized in that, The channel information includes: Q first elements, where Q is a positive integer; The first reference signal sequence generation device further includes: a generation module and a determination module; The generation module is used to input Q first elements into the first neural network to generate a first reference signal sequence; The determining module is used to determine a target loss function based on the target channel estimate, wherein the target channel estimate is obtained by channel estimation based on the first reference signal sequence; The training module is specifically used to train the first neural network according to the target loss function determined by the determining module; The target loss function is used to characterize the degree of deviation between the target channel estimate and the true channel value. The channel truth value is determined based on the channel information.

25. The reference signal sequence generation apparatus according to claim 24, characterized in that, The first neural network includes N first neuron groups, and each first neuron group corresponds to a first weight parameter; The first reference signal sequence is calculated by multiplying the Q first elements and N first weight parameters.

26. The reference signal sequence generation apparatus according to claim 24, characterized in that, Each first element corresponds to a second real part parameter and a second imaginary part parameter; the first reference signal sequence includes: N second elements; The generation module is specifically used to input Q second real part parameters and Q second imaginary part parameters into the first neural network respectively to obtain N third real part parameters and N third imaginary part parameters; and to determine the N second elements according to the N third real part parameters and the N third imaginary part parameters respectively.

27. The reference signal sequence generation apparatus according to claim 24, characterized in that, The target loss function includes: a first parameter between the target channel estimate and the true channel value; The first parameter includes at least one of the following: mean squared error, normalized mean squared error, norm, correlation coefficient, and cosine similarity.

28. The reference signal sequence generation apparatus according to claim 24, characterized in that, The target loss function includes: a second parameter obtained after performing the first operation using the target channel estimate; The first operation is: a subsequent operation of channel estimation in a wireless communication system; The second parameter is used to characterize at least one of the following: wireless communication transmission accuracy and wireless communication transmission efficiency.

29. The reference signal sequence generation apparatus according to claim 24, characterized in that, The first neural network includes: R second neural networks, where R is a positive integer greater than 1; The generation module is specifically configured to determine R element groups based on the Q first elements, wherein different element groups include different first elements; and input each element group into each second neural network to generate a second reference signal sequence corresponding to each second neural network, thereby obtaining R second reference signal sequences; and, based on the R second reference signal sequences, perform a second operation to generate the first reference signal sequence. The second operation includes at least one of the following: Add noise, superimpose the R second reference signal sequences, and splice the R second reference signal sequences.

30. The reference signal sequence generation apparatus according to claim 24, characterized in that, The target channel estimate includes any one of the following: The first reference signal sequence generating device obtains a first channel estimate value by performing channel estimation based on the first reference signal sequence; The second channel estimate received by the first reference signal sequence generating device from the second reference signal sequence generating device; The second channel estimate is the channel estimate obtained by the second reference signal sequence generating device performing channel estimation based on the first reference signal sequence when the second reference signal sequence generating device receives the first reference signal sequence from the first reference signal sequence generating device.

31. The reference signal sequence generation apparatus according to claim 23, characterized in that, The training module is specifically used to train the first neural network based on the target constraints and using the channel information. The target constraint includes any one of the following: The power value corresponding to each target weight parameter is the first preset value; The total power value corresponding to the N target weight parameters is less than or equal to the second preset value.

32. The reference signal sequence generation apparatus according to claim 23, characterized in that, The first reference signal sequence generation device further includes: a determination module; The determining module is used to determine the first neural network based on the first physical parameters; The first physical parameter includes at least one of the following: bundling size; RB; PRB; The number of MUs; The density of the time-frequency pattern of the reference signal sequence in the time-frequency domain.

33. The reference signal sequence generation apparatus according to claim 22, characterized in that, The channel information is received by the first reference signal sequence generating device from the second reference signal sequence generating device.

34. The reference signal sequence generation apparatus according to claim 20, characterized in that, The acquisition module is specifically used to determine the target neural network based on the second physical parameter; The second physical parameter includes at least one of the following: bundling size; RB; PRB; The number of MUs; The density of the time-frequency pattern of the reference signal sequence in the time-frequency domain.

35. The reference signal sequence generation apparatus according to claim 20, characterized in that, The first reference signal sequence generation device further includes: a transmission module; The sending module is used to send a first signaling to the second reference signal sequence generating device. The first signaling carries: the target reference signal sequence, or the identifier of the target reference signal sequence, or information obtained by compressing the target reference signal sequence.

36. The reference signal sequence generation apparatus according to claim 35, characterized in that, The first signaling includes at least one of the following: Radio Resource Control (RRC) signaling; Layer 1 signaling of the Physical Downlink Control Channel (PDCCH); Information about the Physical Downlink Shared Channel (PDSCH); Signaling of the Media Access Control Layer (MAC) control unit; System Information Block (SIB); Layer 1 signaling of the Physical Uplink Control Channel (PUCCH); Target message information for the Physical Random Access Channel (PRACH); Information about the Physical Uplink Shared Channel (PUSCH); XN interface signaling; PC5 interface signaling; Sidelink interface commands; The target message includes at least one of the following: message MSG 1, message MSG 2, message MSG 3, message MSG 4, message MSG A, and message MSG B.

37. A reference signal sequence generation apparatus, wherein the reference signal sequence generation apparatus is a second reference signal sequence generation apparatus, characterized in that, The second reference signal sequence generation device includes: a receiving module, an execution module, and a transmitting module; The receiving module is configured to receive at least one third reference signal sequence from at least one third reference signal sequence generating device, the at least one third reference signal sequence including: a first reference signal sequence sent by a first reference signal sequence generating device in the at least one third reference signal sequence generating device; The execution module is used to perform a third operation based on each third reference signal sequence received by the receiving module to obtain a corresponding fourth reference signal sequence. The transmitting module is used to transmit a second channel estimate to the first reference signal sequence generating device; Wherein, the second channel estimate is: the channel estimate obtained by the second reference signal sequence generating device performing channel estimation based on the fourth reference signal sequence corresponding to the first reference signal sequence generating device; the second channel estimate is used by: the first reference signal sequence generating device to train the target neural network; The target neural network is used to generate a target reference signal sequence; The target neural network includes N target neuron groups, and each target neuron group corresponds to a target weight parameter; The target reference signal sequence includes N target elements, each corresponding to one of the N target weight parameters, where N is a positive integer.

38. The reference signal sequence generation apparatus according to claim 37, characterized in that, The third operation includes at least one of the following: Add noise, superimpose the at least one third reference signal sequence, and splice the at least one third reference signal sequence.

39. A terminal, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein when the program or instructions are executed by the processor, they implement the steps of the reference signal sequence generation method as claimed in any one of claims 1 to 17, or implement the steps of the reference signal sequence generation method as claimed in any one of claims 18 or 19.

40. A network-side device, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein when the program or instructions are executed by the processor, they implement the steps of the reference signal sequence generation method as claimed in any one of claims 1 to 17, or implement the steps of the reference signal sequence generation method as claimed in any one of claims 18 or 19.

41. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the reference signal sequence generation method as described in any one of claims 1 to 17, or implement the steps of the reference signal sequence generation method as described in any one of claims 18 or 19.

Images