Roadside unit sensing and communication integrated system and hybrid beamforming method

Abstract

The application belongs to the technical field of wireless communication, and discloses a roadside unit integrated sensing system and a hybrid beamforming method. In the method, the roadside unit integrated sensing system (RSU-ISAC) is used as a hardware device for sensing fusion, a plurality of vehicle targets on the road surface are sensed, the road traffic condition is analyzed, and a millimeter wave communication link is established to provide real-time communication services for users. In order to further reduce the multi-layer interference between communication users and between communication users and radar sensing, based on the Cramer-Rao bound of the arrival angle of the multi-vehicle target sensing and the overall power constraint condition of the system, the weighted sum rate of the communication users is maximized, and the block coordinate descent algorithm is used to design the digital beamforming matrix at the baseband and the analog beamforming matrix in the radio frequency domain. The method can achieve a good performance compromise between the communication rate and the radar sensing accuracy.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of wireless communication, and particularly relates to a millimeter wave integrated sensing and communication system for vehicle networking and a hybrid beamforming method. BACKGROUND

[0002] Vehicle networking is an important part of future intelligent transportation systems, such as cellular vehicle-to-everything (C-V2X) technology. C-V2X technology solves the problems of high reliability, low latency communication and high-precision sensing between vehicles and roads by building vehicle-road cooperative perception, communication and information interaction. Vehicles are equipped with communication transceivers and various sensors to extract environmental information and exchange information with road side units, other vehicles and even pedestrians. Integrated sensing and communication (ISAC) assisted V2X network can integrate communication and sensing functions into the same hardware platform under the signaling framework, thereby improving the spectrum and hardware efficiency of the system, obtaining integration gain and helping the development of C-V2X technology.

[0003] Millimeter wave (mmWave) has a large bandwidth of 30-300GHz, which can provide Gbit / s high-capacity communication and centimeter-level high-precision sensing services for the system. In order to further improve the performance of the ISAC assisted vehicle networking system, a scheme is disclosed in the literature

P. Kumari, J. Choi, N. González-Prelcic and R. W. Heath, "IEEE 802.11ad-Based Radar: An Approach to Joint Vehicular Communication-Radar System," in IEEE Transactions on Vehicular Technology, vol. 67, no. 4, pp. 3012-3027, April 2018

[0004] Considering that there is serious path loss and rain attenuation in millimeter wave, a large-scale multiple-input multiple-output (mMIMO) technology is needed to provide high-gain directional beams to make up for the deficiency of millimeter wave communication. If a full-digital beamformer with a separate radio frequency chain for each antenna is used, it will result in high hardware cost and power consumption. In order to balance system performance and cost, a hybrid beamforming technology has been proposed in millimeter wave MIMO communication. For example, a scheme for a millimeter wave sensing and communication integrated system based on a hybrid beamforming structure is proposed in the literature

X. Wang, Z. Fei, J. A. Zhang and J. Xu, "Partially-Connected Hybrid Beamforming Design for Integrated Sensing and Communication Systems," in IEEE Transactions on Communications, vol. 70, no. 10, pp. 6648-6660, Oct. 2022

[0005] The purpose of the present application is to propose a sensing and communication integrated system and a hybrid beamforming method for solving the sensing and communication fusion problem of high-rate communication and high-precision sensing and positioning in the vehicle-to-road (V2I) scenario of vehicle networking.

[0006] To achieve the above problems, the present application adopts the following scheme:

[0007] A road side unit sensing and communication integrated system (RSU-ISAC) for vehicle networking, comprising a digital beamformer, a radio frequency chain (RF Chain), an analog beamformer, a transmitting antenna array, a receiving antenna array, a matched filter and a sensing signal processor, the radio frequency chain is electrically connected to the digital beamformer and the analog beamformer, the analog beamformer is electrically connected to the transmitting antenna array, the transmitting antenna array is used for transmitting signals, the receiving antenna array is used for receiving echo signals in response to the transmitting signals, the matched filter receives and transmits the filtered echo signals to the sensing signal processor, and the sensing signal processor obtains vehicle information based on the received filtered signals, the vehicle information including at least one of vehicle angle, time delay parameter or distance parameter.

[0008] A transmit signal model of a communication and perception integrated signal is constructed and transmitted to a downlink multi-user and multi-vehicle target through the roadside unit communication and perception integrated system. The echo signal returned by the multi-vehicle target is received by a receiving antenna array and transmitted back to a matched filter. The matched filter transmits the matched filtered signal to a perception signal processor. The perception signal processor outputs vehicle angle, time delay parameter and distance parameter. A capacity expression of a communication user is derived according to a receiving signal model of each communication user in the multi-user. The roadside unit communication and perception integrated system allocates a weighting factor according to the strength of the communication user channel and takes the weighted sum rate of all communication users as a communication performance indicator. The Fisher information of the vehicle angle and time delay parameter is derived through the echo signal to obtain the CRB of the vehicle angle and time delay parameter estimation. The angle estimation CRB of the multi-vehicle is taken as the perception performance indicator of the roadside unit communication and perception integrated system. The digital beamformer and the analog beamformer are designed under the constraint of meeting the perception performance indicator. The optimization problem is set as: optimizing the digital beamformer and the analog beamformer to maximize the multi-user weighted sum rate under the constraint of the CRB of the multi-vehicle angle estimation, the total power budget of the system and the single mode of the analog beamformer. The BCD algorithm is used to solve the optimization problem: first, the problem of maximizing the multi-user weighted sum rate is equivalent to a minimization problem by using the weighted least mean square error algorithm. The Riemann conjugate gradient algorithm under the exact penalty function is used to optimize and solve the analog beamforming matrix. The successive convex approximation algorithm is used to optimize the digital beamforming matrix. The analog beamforming matrix and the digital beamforming matrix are alternately optimized until the BCD algorithm converges.

[0009] Preferably, the digital beamformer is located at the baseband.

[0010] Preferably, the roadside unit communication and perception integrated system forms a transmitted communication and perception integrated signal at the transmitting antenna array. The communication and perception integrated signal is transmitted to the downlink multi-user and multi-vehicle target through the transmitting antenna array.

[0011] The embodiment of the present application proposes a hybrid beamforming method using the above roadside unit communication and perception integrated system. The method comprises the following steps:

[0012] Step S1: constructing a transmit signal model and transmitting to a downlink multi-user and multi-vehicle target;

[0013] Step S2: constructing a millimeter wave communication channel model and deriving a capacity expression of a communication user according to a receiving signal model of each communication user in the multi-user. The roadside unit communication and perception integrated system allocates a weighting factor according to the strength of the communication user channel and takes the weighted sum rate of all communication users as a communication performance indicator.

[0014] Step S3: The perception signal processor obtains the CRB of the vehicle angle and time delay parameter estimation based on the received echo signal of the vehicle target response and using a perception target echo model, takes the CRB of the multi-vehicle angle estimation as the perception performance index of the road side unit integrated sensing and communication system, and designs the digital beamformer and the analog beamformer under the constraint of the index; Step S4: based on the road side unit integrated sensing and communication system, the communication performance index and the perception performance index, an optimization problem is modeled; the optimization problem is to optimize the digital beamformer and the analog beamformer to maximize the multi-user weighted sum rate under the constraints of the CRB of the multi-vehicle angle estimation, the total power budget of the system and the single mode of the analog beamformer;

[0015] Step S5: the BCD algorithm is used to solve the optimization problem.

[0016] In the method, a hybrid beamforming structure composed of the digital beamformer and the analog beamformer is designed, so that the communication rate and the perception accuracy can be maximized under the conditions of reducing the cost and the power consumption; and the model and the technical support are provided for the road side unit actually deployed to simultaneously complete the downlink communication and the vehicle target perception detection.

[0017] Preferably, the step S1 comprises:

[0018] The communication symbol is pre-encoded by the digital beamformer, then is subjected to the up-conversion processing of the radio frequency link and the phase modulation processing of the analog beamformer, and finally forms the transmitted communication and sensing integrated signal at the transmitting antenna array, and the communication and sensing integrated signal is transmitted to the downlink multi-user and the multi-vehicle target through the transmitting antenna array.

[0019] Preferably, the communication symbol is subjected to the amplitude and phase pre-encoding processing in the digital beamformer, then is subjected to the up-conversion to 60GHz through the radio frequency link, is subjected to the single mode phase modulation by the analog beamformer, and finally forms the communication and sensing integrated transmitting signal at the transmitting antenna array, and the transmitting signal adopts the IEEE 802.11ad waveform structure. Compared with the existing communication and radar sensing spectrum coexistence scheme, the communication and sensing of the application can be realized through the IEEE 802.11ad waveform of the shared millimeter wave frequency band, so that the two are integrated into the same hardware platform, share the spectrum resource, and obtain the integration gain; compared with the existing hybrid beam scheme, the application adopts the full-connection hybrid beamforming structure, so that the system can have greater performance gain compared with the partial-connection structure without increasing the power consumption.

[0020] Preferably, in the hybrid beamforming method, the capacity expression of the communication user is derived according to the received signal model of each communication user in the multi-user, and the CRB of the sensing angle is derived using the perception target echo model.

[0021] Preferably, the maximization of the multi-user weighted sum rate and solving by using a preset algorithm until the algorithm converges comprises:

[0022] optimizing the digital beamformer and the analog beamformer under the constraints of the CRB of the multi-vehicle angle estimation, the total power budget of the system, and the analog beamformer to maximize the multi-user weighted sum rate,

[0023] The BCD algorithm is used for solving. The BCD algorithm uses a perception target echo model to derive the CRB of the multi-target angle and time delay parameter estimation, and maximizes the weighted sum rate of multi-user communication under the constraint of the CRB of the multi-vehicle target angle parameter. The model considered is more in line with the actual scenario of multi-communication users and multi-vehicle perception targets in vehicle networking. The BCD algorithm is used to optimize the digital beamforming matrix and the analog beamforming matrix, and the designed algorithm has low complexity and can quickly converge. The designed algorithm can achieve a good performance compromise between communication rate and radar perception accuracy

[0024] Preferably, the solving process in step S5 comprises:

[0025] The weighted least mean square error algorithm is used to equivalently convert the maximization of the multi-user weighted sum rate into a minimization problem,

[0026] The Riemann conjugate gradient algorithm under the design of the exact penalty function is used to optimize and solve the analog beamforming matrix,

[0027] The successive convex approximation algorithm is used to optimize the digital beamforming matrix, and the analog beamforming matrix and the digital beamforming matrix are alternately optimized until the BCD algorithm converges.

[0028] Preferably, in the hybrid beamforming method,

[0029] The weighted least mean square error algorithm is used to equivalently convert the maximization of the multi-user weighted sum rate into a minimization problem,

[0030] The converted minimization problem is decomposed into a first sub-problem and a second sub-problem,

[0031] The BCD algorithm framework is used to solve the first sub-problem and the second sub-problem;

[0032] The first sub-problem is to solve the analog beamforming matrix, and the second sub-problem is to solve the digital beamforming matrix;

[0033] To the first sub-problem, due to the existence of single-mode constraint, the analog beamforming matrix is vectorized and defined in the Riemannian manifold space, and the CRB of the target angle and the system power constraint are taken as the penalty function of the objective function of the first sub-problem, so that the first sub-problem is converted into an unconstrained optimization problem, and the Riemannian conjugate gradient algorithm is used to optimize and solve the analog beamforming matrix;

[0034] To the second sub-problem, due to the existence of non-convex constraint of the CRB of the target angle estimation, the first-order Taylor expansion approximation of the constraint is converted into a convex constraint, and then the SCA algorithm is used to iteratively optimize the digital beamforming matrix. By alternately optimizing the above-mentioned analog beamforming matrix and digital beamforming matrix, the BCD algorithm converges. In the method, the initial values of the analog beamforming matrix and the digital beamforming matrix are set, and the first sub-problem and the second sub-problem are alternately iteratively solved until the following termination condition is met: the number of iterations reaches the maximum set number, or the difference between the objective functions calculated in the adjacent two iterations is less than the set threshold error.

[0035] Advantages

[0036] In the mixed beamforming method of the present application, the CRB of the multi-target angle and time delay parameter estimation is derived by utilizing the perception target echo model, and the weighted sum rate of multi-user communication is maximized under the constraint condition of the multi-vehicle target angle parameter CRB. The model considered is more in line with the actual scene of multi-communication user and multi-vehicle perception target in vehicle networking. The BCD algorithm is used to optimize the digital forming beam matrix and the analog forming beam matrix, the designed algorithm has low complexity and can quickly converge. The designed algorithm can achieve a good performance compromise between communication rate and radar perception accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 It is a roadside unit sensing and communication integrated system result schematic diagram of the embodiment of the present application;

[0038] Figure 2 It is a scheme flow chart of the embodiment of the present application;

[0039] Figure 3 It is a performance comparison chart of the algorithm of the embodiment of the present application and the existing algorithm;

[0040] Figure 4 It is a communication performance and sensing performance compromise simulation chart of the embodiment of the present application;

[0041] Figure 5 It is a sensing angle beam chart of the embodiment of the present application. DETAILED DESCRIPTION

[0042] In order to make the purposes, technical solutions and advantages of the present application clearer, further detailed description of the present application will be given below with reference to the drawings. Hereinafter, the terms "first", "second", etc. are only used for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second", etc. can explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise specified, the meaning of "several" is two or more.

[0043] The present application discloses a vehicle networking millimeter wave sensing and communication integrated hybrid beamforming method and device, which takes a roadside unit sensing and communication integrated system as a sensing and communication integrated hardware device, senses multiple vehicle targets on the road surface, analyzes the road traffic condition, and establishes a millimeter wave communication link to provide real-time communication services for users. In order to further reduce the multi-layer interference between communication users and between communication users and radar sensing, the weighted sum rate of communication users is maximized based on the Cramer-Rao bound of the angle of arrival of the multiple vehicle target sensing and the overall power constraint condition of the system, and a block coordinate descent algorithm is used to design the digital beamforming matrix at the baseband and the analog beamforming matrix in the radio frequency domain. A good performance compromise between communication rate and radar sensing accuracy can be achieved.

[0044] Embodiment one

[0045] A vehicle networking millimeter wave sensing and communication integrated hybrid beamforming method, comprising the following steps:

[0046] Step S1: constructing a transmission signal model based on a roadside unit sensing and communication integrated system (ISAC) to transmit a signal; in the present embodiment, the frequency band of the transmission signal adopts a 60GHz IEEE 802.11ad waveform structure, and when the signal is transmitted, the communication symbol is pre-encoded in amplitude and phase in the digital beamformer, then the signal is up-converted to 60GHz through the radio frequency link, and finally the signal is single-mode phase modulated by the analog beamformer to form a sensing and communication integrated transmission signal at the transmission antenna array.

[0047] Step S2: constructing a millimeter wave communication channel model, which includes deriving a capacity expression of each communication user according to the received signal model of each communication user in the multi-user; the roadside unit sensing and communication integrated system allocates a weighting factor according to the strength of the communication user channel, and takes the weighted sum rate of all communication users as a communication performance indicator.

[0048] Step S3: data processing based on the received echo signal, which is returned to the integrated sensing and communication system through the channel model, the sensing signal processor derives the Fisher information of the vehicle angle and the time delay parameter based on the received echo signal and obtains the angle estimation CRB (Cramér-Rao bound) information of the vehicle angle and time delay parameter estimation; and the angle estimation CRB of the multi-vehicle is taken as the sensing performance index of the integrated sensing and communication system of the road side unit; under the constraints, the digital beamformer and the analog beamformer are designed;

[0049] Step S4: based on the integrated sensing and communication system of the road side unit, the communication performance index and the sensing performance index, an optimization problem is modeled; the optimization problem is to optimize the digital beamformer and the analog beamformer to maximize the multi-user weighted sum rate under the constraints of the CRB of the multi-vehicle angle estimation, the total power budget of the system and the single mode of the analog beamformer;

[0050] Step S5: the BCD algorithm is used to solve the optimization problem, and the solving process includes:

[0051] The WMMSE (Weighted Minimum Mean Square Error) algorithm is used to convert the maximum multi-user weighted sum rate problem into a minimum problem,

[0052] The Riemann conjugate gradient algorithm under the design of the exact penalty function is used to optimize and solve the analog beamforming matrix,

[0053] Then, the SCA (Successive Convex Approximation) algorithm is used to optimize the digital beamforming matrix, and the analog beamforming matrix and the digital beamforming matrix are alternately optimized until the BCD algorithm converges.

[0054] In an embodiment, in step S1, the transmission signal model can be expressed as the communication data of the kth user at the lth moment is expressed as a k (l), using the IEEE 802.11ad waveform structure, the baseband signal in one pulse period is expressed as: Wherein the element s k (t) is expressed as:

[0055]

[0056] The transmission signal model of the integrated sensing and communication signal is built, and the signal received by the kth communication user is expressed as:

[0057]

[0058] nk (t) is Gaussian white noise with variance

[0059] According to the above signal model received by the communication user, the capacity of the kth user is R k = log2(1+γ k ), where γ k is the signal-to-noise ratio; the weighted sum rate of all users is α k is the weighting factor; the weighted sum rate is designed as the communication performance index to design the hybrid beam matrix.

[0060] g(t) is a unit energy pulse with power spectral density |G(f)| 2 s N s is the number of communication symbols in a period, T s is the symbol duration, and the total time of a period is T0≈N s T RF F D s(t).

[0061] In an embodiment, in S2, the channel between the transmit antenna array of the roadside unit sensing integrated system and each communication user is modeled as a Saleh-Valenzuela millimeter wave channel, and the channel of the kth communication user is denoted as h k .

[0062] In an embodiment, in S3, N vehicle targets are set at different angles θ n , and the echo signal of the target is modeled as where ξ n includes the path loss and target cross section of the nth vehicle target, τ n is the delay parameter of the nth vehicle target, and z(t) is Gaussian white noise, which is subject to a circularly symmetric complex Gaussian distribution are the transmit and receive steering vectors, respectively; due to the increasing orthogonality of the steering vectors in different directions with the increasing number of transmit antennas, the Fisher information matrix of all target angles and delay parameters is approximately equivalent to the Fisher matrix of each target; set η n = [θ n , τ n ] T , then each element in the Fisher matrix is expressed as:

[0063]

[0064] where, ​​denote the expectation and the real part, respectively, The CRB of the target angle and the time delay parameter estimation can be obtained by inverting the Fisher matrix. In this embodiment, the CRB of the angle parameter estimation is selected as the performance parameter of the target perception. Preferably, in order to achieve high-precision perception of the vehicle target, the CRB needs to be lower than a given threshold κ n .

[0065] In an embodiment, in the step S4, according to the established roadside unit integrated sensing and communication system, the communication and perception performance indicators, the CRB threshold of the multi-vehicle angle estimation, the system total power and the single-mode constraint of the analog beamforming matrix, the weighted sum rate of the communication users is maximized, and the optimization problem P1 is as follows:

[0066]

[0067] The optimization problem P1 is equivalent to a minimization problem by the WMMSE algorithm, and then the converted minimization problem is decomposed into a first sub-problem and a second sub-problem, and then the first sub-problem and the second sub-problem are solved by using the BCD (Block Coordinate Descent) algorithm framework. Specifically, for the first sub-problem, since there is a single-mode constraint, the analog beamforming matrix is vectorized and defined in the Riemannian manifold space, and the CRB of the target angle and the system power constraint are used as the penalty function of the objective function of the first sub-problem, and the first sub-problem is converted into an unconstrained optimization problem, and the Riemannian conjugate gradient algorithm is used to optimize and solve the analog beamforming matrix. For the second sub-problem, since the CRB of the target angle estimation is a non-convex constraint, the first-order Taylor expansion approximation is used to convert the constraint into a convex constraint, and then the SCA algorithm is used to iteratively optimize the digital beamforming matrix. The analog beamforming matrix and the digital beamforming matrix are alternately optimized until the BCD algorithm converges.

[0068] In the optimization problem P1, the initial values of the analog beamforming matrix and the digital beamforming matrix are set, and the first sub-problem and the second sub-problem are alternately iteratively solved until the following termination condition is met: the number of iterations reaches a maximum set number of times, or the difference between the objective functions calculated in the adjacent two iterations is less than a set threshold error; after executing all the above steps, the optimal digital beam matrix and the analog beam matrix

[0069] In an embodiment, the step S1 further includes configuring a roadside unit integrated sensing and communication system (RSU-ISAC, hereinafter referred to as the system). The system is as shown in Figure 1As shown, it comprises: a digital beamformer, a radio frequency chain (RF Chain), an analog beamformer composed of full-connection phase shifters, a transmitting antenna array, a receiving antenna array, a matched filter, and a sensing signal processor.

[0070] In the system, a communication symbol is pre-encoded by the digital beamformer, and then sequentially subjected to up-conversion processing of the radio frequency chain and phase modulation processing of the analog beamformer, to form a transmitted sensing-integrated signal at the transmitting antenna array. The sensing-integrated signal is transmitted by the transmitting antenna array to a downlink multi-user and multi-vehicle target. The echo signal returned by the multi-vehicle target is received by the receiving antenna array and transmitted back to the matched filter. The matched filter transmits the matched filtered signal to the sensing signal processor, which outputs the vehicle angle, time delay parameter, and distance parameter.

[0071] In an embodiment, the road side unit sensing-integrated system is configured with N RF radio frequency chains, N t transmitting antennas in the transmitting antenna array, and the transmitting antenna array adopts a uniform linear array (ULA). K communication users are set, and the digital beamforming matrix at the digital beamformer is represented as The analog beamforming matrix at the analog beamformer is represented as Each element in the analog beam matrix has a single mode constraint.

[0072] Embodiment Two

[0073] A vehicle networking millimeter wave sensing-integrated hybrid beamforming method, comprising the following steps:

[0074] Sa1: configuring a road side unit sensing-integrated system (RSU-ISAC), which comprises: a digital beamformer at a baseband, a radio frequency chain (RF Chain), an analog beamformer composed of full-connection phase shifters, a transmitting antenna array, a receiving antenna array, a matched filter, and a sensing signal processor. A communication symbol is pre-encoded by the digital beamformer, and then sequentially subjected to up-conversion processing of the radio frequency chain and phase modulation processing of the analog beamformer, to form a transmitted sensing-integrated signal at the transmitting antenna array. The sensing-integrated signal is transmitted by the transmitting antenna array to a downlink multi-user and multi-vehicle target. The echo signal returned by the multi-vehicle target is received by the receiving antenna array and transmitted back to the matched filter. The matched filter transmits the matched filtered signal to the sensing signal processor, which outputs the vehicle angle, time delay parameter, and distance parameter. In this embodiment, the number of communication users K is not less than 4, the radio frequency chain is connected to the transmitting antenna array in a full-connection manner, the number of radio frequency chains N RFThe range of K is K ≤ N RF The range of N is N ≥ 1 t The transmitting antenna and the receiving antenna adopt a uniform linear array structure and have the same number, i.e., N t The range of N is N ≥ 1 r The range of N is N ≥ 1 The digital beam matrix to be optimized at the digital beamformer is represented as The analog beam matrix of the analog beamformer is represented as

[0075] Sa2: Construct a transmitting signal model and determine the waveform structure of the transmitting signal; in this step, the IEEE 802.11ad waveform structure with a frequency band of 60 GHz is adopted, and the baseband signal in one pulse period can be represented as: wherein the element s k (t) is represented as:

[0076]

[0077] wherein a k (l) is the communication data of the kth user at the lth moment, g(t) is a unit energy pulse with a power spectral density of |G(f)| 2 N s is the number of communication symbols in one period, T s is the symbol duration, and the total time of one period is T0≈N s T s The number of communication symbols N s = 30 in one period, and the bandwidth W of the baseband signal is 125 MHz; the signal transmitted at the transmitting antenna array in the road unit integrated sensing system is represented as x(t) = F RF F D s(t);

[0078] Sa3: Construct a millimeter wave communication channel model; in this step, the Saleh-Valenzuela millimeter wave channel is adopted, and the channel of the kth user is represented as: L k denotes the number of channel paths, and v l,k denotes the channel attenuation coefficient of the lth path; wherein the value of L k = 5, χ = 61.4 + 20log 10 (d) + ξ, wherein d is the distance from the RSU to the communication user, a(θ) is the array steering vector, and the signal received by the kth communication user is:

[0079]

[0080] wherein the noise In this example The signal-to-interference-and-noise ratio of the received signal is:

[0081]

[0082] The weighted sum rate of all communication users is: α k is the weighting factor;

[0083] Sa4: Set the number of perceived target vehicles N to 3, assume that they are located at -40°, 0° and 40° respectively, and the echo signals of the targets are The path loss of the nth vehicle target is |ξ n | 2 = -90 dB, τ n is the time delay parameter of the nth vehicle target, and the noise The CRB of parameter estimation can be obtained by inverting the Fisher information matrix of the target angle and time delay parameter, which is:

[0084]

[0085]

[0086] The CRB of angle parameter estimation is selected as the performance parameter of the perceived target, and under the given CRB threshold κ n , the constraint is:

[0087] Sa5: According to the established roadside unit sensing and communication integrated system, communication and sensing performance indicators, under the constraints of the CRB of vehicle angle estimation, system power budget and single-mode constraint of simulated beamforming matrix, the problem of maximizing the weighted sum rate of communication users is optimized, and the optimization problem is represented as:

[0088]

[0089]

[0090]

[0091] where The weighting factor of user rate is set as α k = 1 / K, The total system power budget is set as P T = 30 dBm.

[0092] The BCD algorithm is used to solve the optimization problem P1 established in step Sa5. First, the maximum multi-user weighted sum rate problem is converted into a minimum problem by using the weighted minimum mean square error (WMMSE) algorithm:

[0093]

[0094]

[0095]

[0096] wherein, u k is the weight factor of the MSE of the kth user;

[0097] the fixed digital beam matrix F D and the analog beam matrix F RF , the update coefficient factor

[0098] The objective function of the minimum problem P2 is derived with respect to the variables , and the derivative is set to zero to obtain the update expression of the coefficient factor:

[0099]

[0100] the fixed digital beam matrix F D and the coefficient factor the optimal analog beam matrix F RF :

[0101] The optimal coefficient factor in P1 is substituted into the minimum problem P2, and the objective function and the constraint condition of the minimum problem P2 are transformed into a matrix to convert it into a first sub-problem:

[0102]

[0103]

[0104]

[0105] wherein, x=vec(F RF ),

[0106] Due to the existence of a single-mode non-convex constraint, the first sub-problem P3-1 is defined in the Riemannian manifold Then, by using the CRB constraint of the target angle and the power constraint of the system as the penalty function of the objective function of problem P3-1, the first subproblem P3-1 is transformed into an unconstrained optimization problem P3-2:

[0107]

[0108] in As the penalty factor, the Riemann conjugate gradient algorithm is used to solve P3-2;

[0109] Fixed analog beam matrix F RF Sum of coefficients and factors Optimize digital beam matrix F D :

[0110] Substituting the aforementioned optimal coefficient factors into the minimization problem P2, and performing matrix transformations on the objective function and constraints of minimization problem P2, we can transform it into the second subproblem P4:

[0111]

[0112]

[0113] in and To handle the non-convex CRB constraint for target angle estimation in P4, the SCA method is used to transform it into a convex constraint, and then the CVX solver is used to obtain the optimal solution iteratively.

[0114] The above steps are iterated alternately until the difference between two adjacent iterations of the objective function in minimizing problem P2 is less than the set threshold error τ. Preferably, τ1 is taken as 1 × 10⁻¹⁰. -5 After performing all the above steps, the analog beam matrix of the roadside unit sensing integrated system of this application can be obtained. and digital beam matrix The specific execution steps of the above system modeling and iterative algorithm are summarized as follows: Figure 2 The implementation scheme flowchart.

[0115] Based on the parameters set in the example, the effectiveness of the proposed vehicle-to-everything (V2X) millimeter-wave integrated sensing hybrid beamforming method is further verified through simulation.

[0116] like Figure 3 The figure shows the situation when the number of communication users K=4 and the number of radio frequency links N. RF Simulation results obtained under settings of 4, 8, and 16 are shown, comparing the simulation results of the algorithm designed in this application with those of existing "two-stage" algorithms. The "two-stage" approach specifically refers to first optimizing the all-digital beamforming matrix, and then designing a hybrid beamforming matrix to approximate the optimal all-digital beamforming matrix. From... Figure 3It can be seen that the user weighted sum rate of the two schemes increases with the increase of the transmit power, but under the condition of three different values of the number of radio frequency links, the method designed in the application can obtain higher communication user weighted sum rate than the "two-stage" method, and the system has higher performance gain.

[0117] Figure 4 The simulation diagram obtained under the condition that the number of communication users is 4, 6 and 8 respectively, and the number of radio frequency links is 12 and 16 respectively, from which it can be seen that the user weighted sum rate of the two schemes increases with the increase of the transmit power, but under the condition of three different values of the number of radio frequency links, the method designed in the application can obtain higher communication user weighted sum rate than the "two-stage" method, and the system has higher performance gain. Figure 4 It can be seen that the weighted sum rate of the communication users increases with the increase of the CRB threshold of the vehicle target angle estimation, and then tends to be flat, reaching the upper bound of the communication performance. The smaller the CRB threshold, the higher the accuracy requirement for target perception, and the lower the weighted sum rate of the communication users. This phenomenon reveals the flow of the performance boundary of the integrated sensing and communication system from the optimal communication to the optimal perception. The algorithm designed in the application can achieve a good performance compromise between communication rate and radar perception accuracy.

[0118] Figure 5 The simulation diagram obtained under the condition that the number of communication users is 4 and the number of vehicle perception targets is 3, which shows the spatial beam pattern of the transmitted signal, from which it can be seen that the spatial beam pattern of the transmitted signal is a main lobe beam with a width of 80 degrees, which is consistent with the preset target direction, and the remaining side lobes have random fluctuation phenomenon, providing information transmission for the communication users. Figure 5 It can be seen that the integrated sensing and communication system of the roadside unit of the application can form the expected main lobe beam at angles -40°, 0° and 40°, which meets the preset target direction, and the remaining side lobes have random fluctuation phenomenon, providing information transmission for the communication users.

[0119] The application takes the integrated sensing and communication system of the roadside unit (RSU-ISAC) as the hardware device of the integrated sensing and communication, perceives the multiple vehicle targets on the road surface, analyzes the road traffic conditions, and establishes a millimeter wave communication link to provide real-time communication services for users (pedestrian or vehicle users). In order to further reduce the multi-layer interference between the communication users and between the communication users and the radar perception, the digital beamforming matrix at the baseband and the analog beamforming matrix in the radio frequency domain of the integrated sensing and communication system of the roadside unit are designed.

[0120] Based on the CRB of the arrival angle of the multiple vehicle targets and the overall power resource constraint condition of the system, the application maximizes the weighted sum rate of the communication users, and adopts the block coordinate descent algorithm (BCD) to optimize the design of the hybrid beamforming matrix. The vehicle networking millimeter wave integrated sensing and communication hybrid beamforming method of the application can be used in the vehicle networking millimeter wave integrated sensing and communication hybrid beamforming device.

[0121] The above is the preferred embodiment of the application. It should be noted that for those skilled in the art, without departing from the principles of the application, a number of improvements and refinements can be made, which should also be considered as the protection scope of the application.

Claims

1. A roadside unit sensing integrated system for vehicle networking, characterized in that, include: Digital beamformers, RF links, analog beamformers, transmit antenna arrays, receive antenna arrays, matched filters, and sensing signal processors. The radio frequency link is electrically connected to the digital beamformer and the analog beamformer. The simulated beamformer is electrically connected to the transmitting antenna array. The transmitting antenna array is used to transmit signals. The receiving antenna array is used to receive the echo signal in response to the transmitted signal. The matched filter receives and responds to the echo signal, which is then filtered by the matched filter and transmitted to the sensing signal processor. The sensing signal processor calculates vehicle information based on the received filtered signal, and the vehicle information includes at least one of vehicle angle, time delay parameter, or distance parameter. The steps for achieving hybrid beamforming in the roadside unit integrated sensing system are as follows: Step S1: Construct a transmission signal model and transmit it to multiple users and multiple vehicles via the roadside unit sensing integrated system; Step S2: Construct a millimeter-wave communication channel model and derive the capacity expression of the communication user based on the received signal model of each communication user in the multi-user system; the roadside unit integrated sensing system allocates weighting factors according to the strength of the communication user channel and uses the weighted sum rate of all communication users as the communication performance index. Step S3: The sensing signal processor calculates the estimated CRB of vehicle angle and time delay parameters based on the received echo signal of the vehicle target response and using the sensing target echo model. The CRB of the multi-vehicle angle estimation is used as the sensing performance index of the roadside unit integrated sensing system. Under its constraints, a digital beamformer and an analog beamformer are designed. Step S4: Based on the roadside unit integrated sensing system, communication performance indicators, and sensing performance indicators, model the optimization problem; The optimization problem is to optimize the digital beamformer and the analog beamformer to maximize the multi-user weighted sum rate under the constraints of CRB for multi-vehicle angle estimation, total system power budget, and single-mode constraints of the analog beamformer. Step S5: Solve the optimization problem using the BCD algorithm.

2. The integrated roadside unit sensing system as described in claim 1, characterized in that, The digital beamformer is located at the baseband.

3. The integrated roadside unit sensing system as described in claim 1, characterized in that, A synthetic signal is generated at the transmitting antenna array and transmitted down to multiple user and multiple vehicle targets via the transmitting antenna array.

4. A hybrid beamforming method using the roadside unit inductive integrated system according to any one of claims 1-3.

5. The hybrid beamforming method as described in claim 4, characterized in that, Step S1 includes: the communication symbols are pre-encoded by a digital beamformer, and then sequentially up-converted by the radio frequency link and phase-modulated by the analog beamformer to form a transmitted integrated communication and sensing signal at the transmitting antenna array. The integrated communication and sensing signal is then transmitted downlink to multiple user and multiple vehicle targets via the transmitting antenna array.

6. The hybrid beamforming method as described in claim 5, characterized in that, The communication symbols are pre-coded for amplitude and phase in a digital beamformer, then up-converted to 60GHz via an RF link, and single-mode phase modulated by an analog beamformer to form a transceiver signal at the transmitting antenna array. This transceiver signal adopts the IEEE 802.11ad waveform structure.

7. The hybrid beamforming method as described in claim 4, characterized in that, Based on the received signal model of each communication user in a multi-user scenario, the capacity expression of the communication user is derived, and the CRB of the sensing angle is derived using the sensing target echo model.

8. The hybrid beamforming method as described in claim 4, characterized in that, The step of maximizing the multi-user weighted sum rate and solving it using a preset algorithm until the algorithm converges includes: Optimize digital and analog beamformers under multi-vehicle angle estimation CRB, system total power budget, and single-mode constraints of the analog beamformer to maximize multi-user weighted sum rate. The BCD algorithm is used to solve the problem.

9. The hybrid beamforming method as described in claim 8, characterized in that, In step S5, the solution process includes: The problem of maximizing the multi-user weighted sum rate is equivalently transformed into a minimization problem by using the weighted least mean square error algorithm. We design a Riemann conjugate gradient algorithm with an exact penalty function to optimize the solution of the simulated beamforming matrix. The digital beamforming matrix is ​​optimized using a successive convex approximation algorithm, and the analog beamforming matrix and the digital beamforming matrix are optimized alternately until the BCD algorithm converges.

10. The hybrid beamforming method as described in claim 8, characterized in that, The problem of maximizing the multi-user weighted sum rate is equivalently transformed into a minimization problem by using the weighted least mean square error algorithm. The transformed minimization problem is decomposed into a first subproblem and a second subproblem. The BCD algorithm framework is used to solve the first and second subproblems; The first subproblem is to solve for the analog beamforming matrix, and the second subproblem is to solve for the digital beamforming matrix. For the first subproblem, due to the existence of single-mode constraints, the simulated beamforming matrix is ​​vectorized and defined in the Riemannian manifold space. Then, the CRB of the target angle and the system power constraint are used as the penalty function of the objective function of the first subproblem, transforming the first subproblem into an unconstrained optimization problem. The Riemann conjugate gradient algorithm is used to optimize and solve the simulated beamforming matrix. For the second subproblem, since the CRB non-convex constraint exists for the target angle estimation, this constraint is approximately transformed into a convex constraint by a first-order Taylor expansion. Then, the SCA algorithm is used to iteratively optimize the digital beamforming matrix. By alternately optimizing the above-mentioned analog beamforming matrix and digital beamforming matrix, the BCD algorithm converges.

Images