A deep learning-based full-duplex radar communication system joint transmit beamforming method

Abstract

The application discloses a full-duplex radar communication system joint transmitting beam forming method based on deep learning, first, the performance of the sending and receiving signals of the FD-ISAC system, radar sensing and communication rate is mathematically modeled. In the constructed model, the base station receives the signal and obtains the estimated user angle through the traditional channel estimation method. Secondly, based on the angle, the uplink historical channel state information and the downlink historical channel state information are calculated. Then, the cross attention-gated recurrent unit network (CAttn-GRU Net) is designed, which takes the uplink historical channel state information and the downlink historical channel state information as input, and then predicts the radar signal covariance matrix and the communication signal beam forming matrix, so as to be used for assisting the construction of the sensing signal and the beam forming design at the current time. In addition, in the training process, a multi-objective optimization loss function is adopted.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless transmission, and specifically relates to a joint transmit beamforming method for a full-duplex radar communication system based on deep learning. Background Art

[0002] Effective transmit beamforming design is key to unlocking the potential of multiple-input, multiple-output (MIMO) communication systems and MIMO radar systems. Many existing works have studied transmit design in multi-antenna ISAC systems by focusing on joint beamforming optimization. Specifically, beamforming is optimized by minimizing the beam pattern matching error while taking into account the individual signal-to-interference-plus-noise ratio (SINR) requirements of the communicating users. However, these works only design transmit beamforming without considering the reception of radar echoes.

[0003] The primary function of a radar system is to estimate target channel parameters, such as delay and Doppler frequency, from the received radar echo signal. Considering radar echo reception in an ISAC system, two scenarios are considered. The first corresponds to downlink ISAC, where radar sensing reuses the resources of downlink transmission, and the base station acts as both a radar transceiver and a communication transmitter. The transmitted downlink ISAC signal is known to the base station and can be used for receive processing for sensing. By applying a linear receive beamformer to the echo signal, the radar signal-to-interference-and-noise ratio (SINR) for target detection is explicitly obtained. The second scenario considers integrating sensing with uplink communication, where the base station can be considered as both a radar transceiver and a communication receiver. The radar receiver operates simultaneously with transmission, i.e., in full-duplex (FD) mode. Self-interference (SI) is a key issue in FD operation. In FD ISAC systems, SIC should be performed only for direct signal coupling between transceiver antennas, while retaining target reflections. However, for FD radars, integrated communication functions occur only in the downlink or uplink, operating in half-duplex (HD) mode. To achieve higher spectral efficiency, FD communication capabilities have also been considered, allowing the base station to function as both a radar transceiver and a communication transceiver. In this setup, not only does interference exist between sensing and communication functions, but there is also coupling between uplink and downlink transmissions, which greatly complicates ISAC design. Previous algorithms have not considered the impact of uplink communication, sending only a pure downlink sensing signal with a fixed uplink transmission power. Downlink communication is not considered, and no uplink transmission is designed.

[0004] The information disclosed in this background section is only intended to enhance understanding of the overall background of the invention and should not be considered as an admission or any form of suggestion that the information constitutes the prior art already known to a person of ordinary skill in the art. Summary of the Invention

[0005] This paper proposes a deep learning-based joint transmit beamforming method for full-duplex radar communication systems. Through innovative algorithm design, neural network architecture optimization, and effective integration with traditional methods, this method significantly reduces signal detection complexity and improves the system's signal perception accuracy and communication rate in complex environments.

[0006] A deep learning-based joint transmit beamforming method for a full-duplex radar communication system of the present invention comprises the following steps:

[0007] Step 1: The full-duplex base station receives the echo signal and estimates the angle of the user in each time slot through traditional channel estimation methods such as Multiple Signal Classification (MUSIC).

[0008] Step 2: construct a transmission and reception signal model of the FD-ISAC system, and calculate the uplink historical channel state information and the downlink historical channel state information according to the angle of the user in each time slot;

[0009] Step 3: Construct the CAttn-GRU neural network, including the cross-attention module, the gated recurrent unit module, and the fully connected network;

[0010] The estimated uplink and downlink historical channel state information are input into the cross-attention module in the CAttn-GRU neural network to obtain the attention-modeled channel state information. The attention-modeled channel state information is input into the gated recurrent unit module to obtain the global historical channel state information. The global historical channel state information is sent to the fully connected network to predict the radar signal covariance matrix and the communication signal beamforming matrix.

[0011] The radar signal covariance matrix and the communication signal beamforming matrix are used to assist in the current moment perception signal and beamforming design.

[0012] Furthermore, in step 2, the receiving signal model of the FD-ISAC system is constructed, and the calculation formula of the base station receiving signal is expressed as follows:

[0013]

[0014] in, represents the antenna gain, γ k,n represents the uplink communication channel coefficient, γ l,n represents the downlink communication channel coefficient, γ 0,n represents the downlink communication channel coefficient of the sensing target, τ 0,n and ν 0,n They represent the time delay and Doppler shift of the perceived target, τ l,n and ν l,n represent the time delay and Doppler shift of downlink communication users respectively; represents the beamforming vector associated with downlink communication user l;

[0015] l∈{1,…,L}, Represents the unit power data symbol of downlink communication user l, that is,

[0016] Represents a covariance matrix dedicated radar signal; assuming the unit power signal of downlink communication user l and dedicated radar signals Independent of each other, Indicates the unit power data symbol of uplink communication user k, a t (θ l,n ) represents the steering vector of the transmitting array in the direction of downlink communication user l, a r (θ l,n ) represents the steering vector of the receiving array in the direction of downlink communication user l, a r (θ 0,n ) represents the steering vector of the receiving array to the sensing target, z represents the undesired interference, and noise represents the noise;

[0017] In the above formula, downlink beamforming is designed by and F0; once F0 is determined, a dedicated radar signal is generated where a t (θ k,n ) represents the steering vector of the transmitting array in the θ direction, which can be written as:

[0018]

[0019] a r (θ k,n ) represents the steering vector of the receiving array in the θ direction, which can be written as:

[0020]

[0021] Among them, N t Indicates the number of base station transmitting antennas, N r Indicates the number of base station receiving antennas;

[0022] Assume that there are I signal-uncorrelated interference sources, located at angles and θ i,n ≠θ 0,n , these interferences will also reflect the sensing signal to the BS, generating undesired interference z, the calculation formula of z is as follows:

[0023]

[0024] In addition, considering the total transmit power constraint Among them, P max represents the maximum available power budget of the BS;

[0025] definition represents the uplink channel state information between uplink communication user k and BS;

[0026] Applying the arrival angle AOA estimation technology at the base station, we can get the angle of arrival AOA. k,n The estimated results Substitute it into the uplink channel state information expression to obtain the uplink channel estimation result Thus, the uplink historical channel state information described in step 2 is obtained.

[0027] Furthermore, in step 2, the signal transmission model of the FD-ISAC system is constructed, and the downlink signal received by the user is expressed as:

[0028]

[0029] in, represents the downlink channel state information between downlink communication user l and BS, where represents the transmitting antenna gain, γ l,n represents the downlink communication channel coefficient, represents other downlink communication user signals other than l, f l′ represents the downlink beamforming vector related to other communication users other than l;

[0030] Apply the arrival angle AOA estimation technology at the base station to obtain the angle of arrival l,n The estimated results Substitute it into the downlink channel state information expression to obtain the downlink channel estimation result Thus, the downlink historical channel state information described in step 2 is obtained.

[0031] Furthermore, the non-perceiving radar signal echo is filtered to estimate the angle of the perceived target. Obtain the echo of the sensing radar signal in the nth time slot from the received signal Expressed as

[0032]

[0033] in represents the receive beamforming vector for spatial filtering, The existing angle of arrival (AoA) estimation technology is used to estimate θ 0,n Estimate of; Since the receiving beamforming vectors of spatial filtering are orthogonal, means the mean is zero and the variance is Then we use the classic matched filtering method to obtain the estimated time delay of the perceived target and Doppler shift

[0034]

[0035] Where ΔT d is the length of time the signal is received; according to the estimated value and Assume that the remaining interference is ideally eliminated; thus derive the perceived radar signal echo for

[0036]

[0037] Among them G m is the matched filter gain, Express obedience The noise of the distribution has a variance of f 0,n represents the beamforming vector associated with the sensing target;

[0038] definition and Respectively represent τ 0,n and ν 0,n These errors are related to the signal-to-noise ratio (SNR), specifically:

[0039]

[0040] where ρ τ and ρ υ is a constant determined by a specific system, represents the variance of the noise in the perceived radar signal echo, μ 0,n Indicates the Doppler parameter in the perceived radar signal echo.

[0041] Furthermore, the performance of the CAttn-GRU neural network is evaluated using the uplink communication user rate, downlink communication user rate and Cramer-Rao lower bound (CRB). The signal to interference plus noise ratio (SINR) corresponding to the uplink communication user k is expressed as:

[0042]

[0043] Q represents the covariance matrix of the downlink ISAC signal, CQC H I represents the interference caused to uplink communication user k due to the presence of downlink signal; Nr Indicates size N r ×N r The identity matrix, Indicates the variance of the noise in the received uplink signal;

[0044] At this time, the uplink communication user rate is

[0045] The signal-to-interference-and-noise ratio (SINR) of downlink communication user l is expressed as:

[0046]

[0047] F0 represents the radar signal covariance matrix, Indicates the variance of the noise in the received downlink signal;

[0048] At this time, the downlink communication user rate is

[0049] Calculate the distance d to the perceived target based on the Fish information matrix 0,n and angle θ 0,n The Cramer-Rao lower bound of is as follows:

[0050]

[0051] Where c represents the speed of light, Represents the noise variance contained in the perceived radar echo signal.

[0052] Furthermore, the CAttn-GRU neural network is trained using a multi-objective loss function:

[0053]

[0054] Among them, λ1, λ2 and λ3 represent the weights of the uplink communication user rate constraint item, the downlink communication user rate constraint item and the power constraint item respectively, R UL Indicates the target value of the uplink communication user rate (artificially set constant), R DL Indicates the target value of the downlink communication user rate (artificially set constant).

[0055] Beneficial effects: To achieve higher spectrum efficiency, the base station acts as both a radar transceiver and a communication transceiver. The BS receives and transmits signals from multiple uplink and downlink communication users, reusing the same time and frequency resources. The downlink transmission signal is an ISAC signal that is used to transmit information to downlink communication users and perform sensing tasks for point target detection. The BS also simultaneously receives uplink communication signals and processes radar echo signals. However, in this setting, not only is there interference between the sensing and communication functions, but there is also coupling between uplink and downlink transmissions, which greatly increases the complexity of the communication and perception ISAC design. By adopting SIC technology for the ISAC system, it is assumed that the SI at the BS, or more precisely, the direct signal coupling link between transceivers, is suppressed to an acceptable level.

[0056] Furthermore, in traditional approaches, effective predictive beamforming schemes require high computational complexity due to the need to continuously track the motion parameters of dynamic targets. To address this challenge, a deep learning (DL)-based framework has been introduced. Full-duplex radar communication systems based on DL have shown great potential in signal processing and beamforming. Deep learning algorithms can automatically extract deep signal features, improving the accuracy of target recognition and classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a flow chart of the method of the present invention;

[0058] Figure 2 It is a schematic diagram of uplink and downlink signals;

[0059] Figure 3 This is the CAttn-GRU neural network structure diagram. DETAILED DESCRIPTION

[0060] A deep learning-based joint transmit beamforming method for a full-duplex radar communication system includes the following steps:

[0061] Step 1: Build an ISAC system

[0062] Consider an ISAC system, which is equipped with two dual-function full-duplex base stations FD BS with two uniform linear arrays (ULAs) to receive the communication signals from K single-antenna uplink communication users and send downlink ISAC signals through the same time-frequency resources. t The downlink ISAC signal transmitted by the uniform linear array ULA of the unit communicates with L single antenna downlink communication users at the same time to sense and detect dynamic targets. The radar echo signal and the uplink communication signal are transmitted through the N r The receiving uniform linear array (ULA) of each unit is received at the BS end.

[0063] Step 2, full-duplex signal model, such as Figure 2 As shown, including uplink communication and downlink communication

[0064] Consider the uplink signal, such as Figure 1 As shown, the base station received signal includes the communication user uplink communication signal passing through the channel, the base station downlink communication signal echo, the base station perception signal echo, undesired interference and noise. The calculation formula of the base station received signal is:

[0065]

[0066] in, represents the antenna gain, γ k,n represents the uplink communication channel coefficient, γ l,n represents the downlink communication channel coefficient, γ 0,n represents the downlink communication channel coefficient of the sensing target, τ 0,n and ν 0,n They represent the time delay and Doppler shift of the perceived target, τ l,n and ν l,n They represent the time delay and Doppler shift of downlink communication users respectively. represents the beamforming vector associated with downlink communication user l, l∈{1,…,L}, Represents the unit power data symbol of downlink communication user l, that is, here, Represents a covariance matrix The dedicated radar signal is used to expand the DoF of the transmitted signal x to achieve enhanced perception performance. Assume that the unit power signal of the downlink communication user l is and dedicated radar signals Independent of each other, Indicates the unit power data symbol of uplink communication user k, a t (θ l,n ) represents the steering vector of the transmitting array in the direction of downlink communication user l, a r (θ l,n ) represents the steering vector of the receiving array in the direction of downlink communication user l, a r (θ 0,n ) represents the steering vector of the receiving array to the perceived target, z represents the unwanted interference, and noise represents the noise.

[0067] In the above formula, downlink beamforming is designed by Once F0 is determined, a dedicated radar signal can be generated. where a t (θ k,n ) represents the steering vector of the transmitting array in the θ direction, which can be written as:

[0068]

[0069] a r (θ k,n ) represents the steering vector of the receiving array in the θ direction, which can be written as:

[0070]

[0071] Among them, N t Indicates the number of base station transmitting antennas, N r Indicates the number of base station receiving antennas.

[0072] Assume that there are I signal-correlated uncorrelated interference sources, located at angles and θ i,n ≠θ 0,n , these interferences will also reflect the sensing signal to the BS, generating undesired interference z, the calculation formula of z is as follows:

[0073]

[0074] In addition, considering the total transmit power constraint Among them, P max represents the maximum available power budget of the BS. When the FD BS sends x, it receives the uplink communication signal and the target reflection at the same time.

[0075] Considering the downlink signal of the full-duplex radar, it consists of four parts: the downlink communication signal of the base station, the signal interference between communication users, the perception signal of the dedicated radar received by the communication user, and noise. The downlink signal received by the communication user is

[0076]

[0077] in, represents the downlink channel between downlink communication user l and BS, where represents the transmitting antenna gain, γ l,n Represents the downlink communication channel coefficient.

[0078] Step 3: Parameter estimation using traditional channel estimation methods:

[0079] Using spatial filtering technology, the echo of the point target radar in the nth time slot can be obtained from the reflected signal echo, which can be expressed as

[0080]

[0081] in represents the receive beamforming vector for spatial filtering, The existing angle of arrival (AoA) estimation technology is used to estimate θ 0,nTo simplify the formula, we assume that item It means the mean is zero and the variance is σ 2 Then we use the classic matched filtering method to obtain the estimated time delay of the perceived target. and Doppler shift

[0082]

[0083] Where, ΔT d is the length of time it takes to receive the signal. and Assuming that the rest of the interference can be eliminated ideally, the angle θ of the perceived target can be deduced. 0,n The measurement model is

[0084]

[0085] G m is the matched filter gain, Express obedience The noise of the distribution has a variance of

[0086] and Respectively represent τ 0,n and ν 0,n These errors are related to the signal-to-noise ratio (SNR), specifically:

[0087]

[0088] where ρ τ and ρ υ is a constant determined by the specific system, the beamforming vector f 0,n Noise variance and Therefore, we use a deep learning-based method to predict state evolution and optimize the beamforming matrix f 0,n , thereby improving the estimation accuracy.

[0089] Step 4: System performance evaluation

[0090] The system performance is evaluated using the uplink communication user rate, downlink communication user rate and Cramér-Rao lower bound (CRB). The signal to interference plus noise ratio (SINR) corresponding to uplink communication user k is expressed as:

[0091]

[0092] in, represents the uplink channel between uplink communication user k and the BS.

[0093] At this time, the uplink communication user rate is

[0094] The signal-to-interference-and-noise ratio (SINR) of downlink communication user l is expressed as:

[0095]

[0096] At this time, the downlink communication user rate is

[0097] The Cramer-Rao lower bound (CRB) is widely used to characterize the estimation error. The smaller the value, the higher the estimation accuracy. The distance d to the perceived target is calculated based on the Fish information matrix. 0,n and angle θ 0,n The Cramer-Rao lower bound of is as follows:

[0098]

[0099] Where c represents the speed of light.

[0100] The uplink communication user rate is

[0101] The downlink communication user rate is

[0102] Step 5: Objective function construction and optimization problem solving

[0103] The loss function is a penalty method that transforms the constrained optimization problem into an equivalent unconstrained problem, so that we can use DL technology to solve the resulting unconstrained problem, which is expressed as

[0104]

[0105] Among them, CRB(d 0,n ,f l,n ) and CRB(θ 0,n ,f l,n ) is responsible for ensuring that the output of the neural network can provide perception performance with a sufficiently low error lower bound, while providing uplink and downlink communication rates that exceed the threshold.

[0106] The constructed CAttn-GRU neural network includes a cross-attention module, a gated recurrent unit module and a fully connected network; Figure 3As shown, each channel historical state information dataset is input into the cross-attention module in the designed network (CAttn-GRU net) to obtain the attention modeling channel state information; each attention modeling channel state information is input into the gated recurrent unit module in the designed network to obtain the global historical channel state information; the global historical channel state information is sent to the fully connected network to predict the radar signal covariance matrix and the communication signal beamforming matrix; the radar signal covariance matrix and the communication signal beamforming matrix will be used to assist in the current moment perception signal and beamforming design;

[0107] During the training of the CAttn-GRU neural network, a multi-objective optimization loss function is used to ensure that both uplink and downlink rates and transmit power meet requirements while maintaining perception accuracy. A deep learning-based approach is used to optimize the beamforming matrix and improve system estimation accuracy.

[0108] Step 6: Channel state estimation and signal processing

[0109] The neural network is as follows: Based on the full-duplex signal model, the historical information of the uplink and downlink channel states of each communication user and the sensing target in each time slot is estimated to obtain the uplink channel information estimation results of each time slot And the downlink channel information estimation results of each time slot Where λ is the length of historical information, which determines how much historical channel estimation information the designed neural network needs to obtain.

[0110] Taking the estimated uplink and downlink channel state history information as input, the designed neural network attention module A i Get attention modeling channel state information H i , where i∈[1,λ]. The formula is as follows:

[0111]

[0112] Among them, d n Represents a vector The size of , softmax(·) represents the flexible maximum transfer function.

[0113] After the previous step, the attention modeling channel state information Η in each historical time slot can be obtained i , which will be fed into a network consisting of λ gated recurrent units, and then the global historical channel state information C is obtained. For the tth gated recurrent unit S t , whose input signal, input hidden state, reset gate and output signal are h t-1 r t z t h t : The formula is as follows:

[0114]

[0115] Among them, W z W r W h Η is the model parameter, [·,·] represents the matrix operation, σ(·) represents the S-type growth function, and tanh(·) is the hyperbolic tangent function. i After passing through the gated cyclic unit group, the global historical channel state information C is obtained, and then the required radar signal covariance matrix F0 and communication signal beamforming matrix f are output through a layer of fully connected network.

[0116] This paper designs a Cross-Attention-Gated Recurrent Unit (CAttn-GRU) network. This network takes a dataset of historical channel state information as input and feeds the global historical channel state information into a fully connected network to predict the radar signal covariance matrix and the communication signal beamforming matrix, respectively, to assist in current-time signal perception and beamforming design. During training, a multi-objective optimization loss function is employed to ensure that both uplink and downlink rates and transmit power meet requirements while maintaining perception accuracy. In terms of detection performance, higher transmit power improves the echo reception signal-to-noise ratio (SNR), and the CRB value decreases with increasing antenna number, achieving excellent perception performance.

Claims

1. A joint transmit beamforming method for a full-duplex radar communication system based on deep learning, characterized in that: The following steps are involved: Step 1: The full-duplex base station receives the echo signal and estimates the angle of the user in each time slot using the traditional channel estimation method; Step 2: construct a transmission and reception signal model of the FD-ISAC system, and calculate the uplink historical channel state information and the downlink historical channel state information according to the angle of the user in each time slot; Step 3: Construct the CAttn-GRU neural network, including the cross-attention module, the gated recurrent unit module, and the fully connected network; The estimated uplink and downlink historical channel state information are input into the cross-attention module in the CAttn-GRU neural network to obtain the attention-modeled channel state information. The attention-modeled channel state information is input into the gated recurrent unit module to obtain the global historical channel state information. The global historical channel state information is sent to the fully connected network to predict the radar signal covariance matrix and the communication signal beamforming matrix. The radar signal covariance matrix and the communication signal beamforming matrix are used to assist in the current moment perception signal and beamforming design.

2. The method for joint transmit beamforming of a full-duplex radar communication system based on deep learning according to claim 1, characterized in that: In step 2, the receiving signal model of the FD-ISAC system is constructed. The calculation formula of the base station receiving signal is expressed as follows: Where K represents the total number of uplink communication users, k∈{1,2,…,K} represents the kth uplink communication user; L represents the total number of downlink communication users, l∈{1,…,L} represents the lth downlink communication user, and n represents the current nth time slot. represents the antenna gain, γ k,n represents the uplink communication channel coefficient, γ l,n represents the downlink communication channel coefficient, γ 0,n represents the downlink communication channel coefficient of the sensing target, τ 0,n and ν 0,n They represent the time delay and Doppler shift of the perceived target, τ l,n and ν l,n They represent the time delay and Doppler shift of downlink communication users respectively; represents the beamforming vector associated with downlink communication user l; l∈{1,…,L}, Represents the unit power data symbol of downlink communication user l, that is, Represents a covariance matrix dedicated radar signal; assuming the unit power signal of downlink communication user l and dedicated radar signals Independent of each other, Indicates the unit power data symbol of uplink communication user k, a t (θ l,n ) represents the steering vector of the transmitting array in the direction of downlink communication user l, a r (θ l,n ) represents the steering vector of the receiving array in the direction of downlink communication user l, a r (θ 0,n ) represents the steering vector of the receiving array to the sensing target, z represents the undesired interference, and noise represents the noise; In the above formula, downlink beamforming is designed by and F0; once F0 is determined, a dedicated radar signal is generated where a t (θ k,n ) represents the steering vector of the transmitting array in the θ direction, which can be written as: a r (θ k,n ) represents the steering vector of the receiving array in the θ direction, which can be written as: Among them, N t Indicates the number of base station transmitting antennas, N r Indicates the number of base station receiving antennas; Assume that there are I signal-uncorrelated interference sources, located at angles and θ i,n ≠θ 0,n , these interferences will also reflect the sensing signal to the BS, generating undesired interference z, the calculation formula of z is as follows: In addition, considering the total transmit power constraint Among them, P max represents the maximum available power budget of the BS; definition represents the uplink channel state information between uplink communication user k and BS; Applying the arrival angle AOA estimation technology at the base station, we can get the angle of arrival AOA. k,n The estimated results Substitute it into the uplink channel state information expression to obtain the uplink channel estimation result Thus, the uplink historical channel state information described in step 2 is obtained.

3. The method for joint transmit beamforming of a full-duplex radar communication system based on deep learning according to claim 2, characterized in that: In step 2, the signal transmission model of the FD-ISAC system is constructed. The downlink signal received by the user is expressed as: in, represents the downlink channel state information between downlink communication user l and BS, where represents the transmitting antenna gain, γ l,n represents the downlink communication channel coefficient, represents other downlink communication user signals other than l, f l′ represents the downlink beamforming vector related to other users other than l; Apply the arrival angle AOA estimation technology at the base station to obtain the angle of arrival l,n The estimated results Substitute it into the downlink channel state information expression to obtain the downlink channel estimation result Thus, the downlink historical channel state information described in step 2 is obtained.

4. The method for joint transmit beamforming of a full-duplex radar communication system based on deep learning according to claim 3, characterized in that: Filter non-perceiving radar signal echoes and estimate the angle of the perceived target Obtain the echo of the sensing radar signal in the nth time slot from the received signal Expressed as in represents the receive beamforming vector for spatial filtering, The existing angle of arrival (AoA) estimation technology is used to estimate θ 0,n Estimate of; Since the receiving beamforming vectors of spatial filtering are orthogonal, means the mean is zero and the variance is The complex Gaussian white noise is then used; then the estimated time delay of the perceived target is obtained through the classic matched filtering method. and Doppler shift Where, ΔT d is the length of time the signal is received; according to the estimated value and Assume that the remaining interference is ideally eliminated; thus derive the perceived radar signal echo for Among them G m is the matched filter gain, Express obedience The noise of the distribution has a variance of f 0,n represents the beamforming vector associated with the sensing target; definition and Respectively represent τ 0,n and ν 0,n These errors are related to the signal-to-noise ratio (SNR), specifically: where ρ τ and ρ υ is a constant determined by a specific system, represents the variance of the noise in the perceived radar signal echo, μ 0,n Indicates the Doppler parameter in the perceived radar signal echo.

5. The method for joint transmit beamforming of a full-duplex radar communication system based on deep learning according to claim 4, characterized in that: The performance of the CAttn-GRU neural network is evaluated using the uplink communication user rate, downlink communication user rate, and the Cramer-Rao lower bound (CRB). The signal to interference plus noise ratio (SINR) corresponding to the uplink user k is expressed as: Q represents the covariance matrix of the downlink ISAC signal, CQC H Indicates the interference caused to uplink communication due to the presence of downlink signals; Indicates size N r ×N r The identity matrix, Indicates the variance of the noise in the received uplink signal; At this time, the uplink communication user rate is The signal-to-interference-and-noise ratio (SINR) of downlink communication user 1 is expressed as: F0 represents the radar signal covariance matrix, Indicates the variance of the noise in the received downlink signal; At this time, the downlink communication user rate is Calculate the distance d to the perceived target based on the Fish information matrix 0,n and angle θ 0,n The Cramer-Rao lower bound of is as follows: Where c represents the speed of light, Represents the noise variance contained in the perceived target echo signal.

6. The method for joint transmit beamforming of a full-duplex radar communication system based on deep learning according to claim 5, characterized in that: The CAttn-GRU neural network is trained using a multi-objective loss function: Among them, λ1, λ2 and λ3 represent the weights of the uplink communication user rate constraint item, the downlink communication user rate constraint item and the power constraint item respectively, R UL Indicates the target value of the uplink communication user rate, R DL Indicates the target value of the downlink communication user rate.

7. The method for joint transmit beamforming of a full-duplex radar communication system based on deep learning according to claim 3, characterized in that: Based on the full-duplex signal model, the historical information of the uplink and downlink channel states of each communication user and the sensing target in each time slot is estimated to obtain the uplink channel information estimation results of each time slot And the downlink channel information estimation results of each time slot Where λ is the length of historical information, which determines how much historical channel estimation information the designed neural network needs to obtain; Taking the estimated uplink and downlink channel state history information as input, the CAttn-GRU neural network cross attention module A i Get attention modeling channel state information H i , where i∈[1,λ]; the formula is as follows: Among them, d n Represents a vector The size of , softmax(·) represents the flexible maximum transfer function; After the previous step, the attention modeling channel state information Η in each historical time slot is obtained i , which will be fed into a network consisting of λ gated recurrent units, and then the global historical channel state information C is obtained; for the tth gated recurrent unit S t , its input signal h t-1 , input hidden state r t , reset gate z t and the output signal h t They are as follows: Among them, W z W r W h are model parameters, [·,·] represents matrix operations, σ(·) represents the S-type growth function, and tanh(·) is the hyperbolic tangent function; Η i After passing through the gated cyclic unit group, the global historical channel state information C is obtained, and then the required radar signal covariance matrix F0 and communication signal beamforming matrix f are output through a layer of fully connected network.

Images