RSMA-OTFS-based multi-user-multiple-input single-output (MU-MISO) car networking downlink power allocation and beam forming joint optimization method

Abstract

The invention relates to an MU-MISO Internet of Vehicles downlink power allocation and beam forming joint optimization method based on RSMA-OTFS. The method comprises the following steps: S1, splitting user information to be transmitted into public sub-information and private sub-information by a base station; s2, the demodulation processes of all users are the same and do not interfere with each other, signals received by the users are recovered to a DD domain through OTFS demodulation, private data corresponding to the users are recovered by adopting a staged recovery strategy, and then recovery of corresponding original data is completed in combination with recovered public data; and S3, by calculating signal to interference plus noise ratios (SINR) of different users in each SIC stage, constructing an optimization problem with maximization of a system sum rate as a target, and calculating to obtain an optimal power allocation and beam forming vector. According to the method, the RSMA-OTFS system power allocation and beam forming strategies are jointly optimized, higher user access fairness and resource utilization rate are achieved, the system, rate and communication reliability can be remarkably improved, and the method is particularly suitable for heterogeneous user coexistence scenes.

Description

Technical Field

[0001] This invention relates to a joint optimization method for downlink power allocation and beamforming in MU-MISO vehicular networks based on RSMA-OTFS, belonging to the field of wireless communication technology. Background Technology

[0002] The continuous development of vehicle-to-everything (V2X) technology has brought serious impacts due to its high mobility and high density environment. Building efficient and reliable vehicle communication links has become a core requirement for ensuring efficient transportation and vehicle safety. The integration of Rate Division Multiple Access (RSMA) and Orthogonal Time-Frequency Space (OTFS) modulation technology provides a feasible path for improving V2X performance in high-mobility scenarios. Meanwhile, Multi-User Multiple-Input Single-Output (MU-MISO) technology is widely used in V2X because it can serve multiple users in parallel. RSMA, based on a "rate division + power domain superposition" mechanism, can divide data into public shared data and dedicated private data, used for low-rate high-reliability services and high-rate personalized services respectively. The receiver decodes the public and private data sequentially using Serial Interference Cancellation (SIC) technology, improving system spectral efficiency and enhancing user access and anti-interference capabilities while balancing flexibility and fairness. OTFS technology utilizes delay-Doppler (DD) domain characteristics to process signals, effectively combating problems such as rapid channel time-varying and Doppler frequency shift under high-speed movement, perfectly adapting to highly dynamic mobile environments. For the downlink of vehicular networks based on RSMA-OTFS, convex optimization algorithms can be used to further optimize the design around the key performance of the system, and obtain a joint optimization scheme of power allocation and beamforming strategy, which is crucial for comprehensively improving the key performance of the system.

[0003] While typical non-orthogonal multiple access (NOMA-OTFS) schemes can improve user fairness and minimum rate performance to some extent, NOMA power domain multiplexing also has its own limitations: First, in high-speed scenarios, the rapid time-varying channel makes it difficult to accurately obtain channel state information (CSI), and residual interference is difficult to eliminate, leading to performance degradation for low-power users; second, channel state fluctuations cause frequent resource scheduling updates, increasing overhead; and third, the lack of multi-domain resource coordination makes spectrum efficiency improvement a bottleneck and unable to meet high bandwidth requirements.

[0004] With the rapid evolution of vehicle-to-everything (V2X) technology and the continuously rising demand for big data services in dense user scenarios, effectively improving the service efficiency and transmission quality of V2X communication systems has become a critical issue that urgently needs to be addressed. Against this backdrop, researching and designing a joint optimization scheme for power allocation and beamforming based on RSMA-OTFS for V2X downlink is of great significance. Summary of the Invention

[0005] To address the shortcomings of existing technologies and aim at improving the downlink transmission efficiency and reliability of vehicular networks, this invention proposes a joint optimization method for MU-MISO vehicular network downlink power allocation and beamforming based on RSMA-OTFS. At a transmit signal-to-noise ratio of 25 dB, this method can improve the downlink system speed and efficiency of vehicular networks by approximately 11.2% compared to a typical NOMA-OTFS scheme.

[0006] Terminology Explanation: 0. MU-MISO: Multi-User Multiple-Input Single-Output (MU-MISO) systems are a typical multi-antenna configuration in the field of wireless communication. Its architectural features include: a base station equipped with multiple transmit antennas, while multiple vehicle user terminals are each equipped with a single receive antenna. Through base station beamforming design, information is transmitted in parallel to different user terminals, thereby significantly improving system spectrum utilization and overall capacity.

[0007] 1. RSMA: Rate Division Multiple Access (RSMA) is a multiple access technology that supports parallel communication by multiple users on the same time and frequency resources. It divides user data into public data (for all users) and private data (for individual corresponding users) in the rate domain, and distinguishes users by power allocation for different user data, thus achieving efficient sharing of the same time and frequency resources by multiple users. RSMA, combined with SiC, beamforming, and power allocation technologies, can significantly improve system spectral efficiency and flexibly adapt to different channel conditions in heterogeneous networks, meeting the different Quality of Service (QoS) requirements of users and providing more balanced communication guarantees and service capabilities for all users.

[0008] 2. OTFS: Orthogonal Time-Frequency Space (OTFS) is a high-efficiency modulation technique. Its core idea is to fully exploit the orthogonal characteristics of the signal in the time and frequency domains to achieve efficient signal transmission. This technique is particularly suitable for high-speed mobile scenarios, effectively mitigating two main negative impacts of mobile channels: Doppler shift and rapid time-varying fading. Unlike the processing paradigm of traditional modulation techniques, OTFS maps data symbols to the DD domain and time-frequency domain for transmission. This unique multi-domain mapping mechanism gives it excellent resistance to channel fading, ultimately ensuring the stability and reliability of communication transmission.

[0009] 3. Beamforming: This is a spatial signal processing technique based on array antennas. Its core principle is to apply specific amplitude weighting and phase compensation to the transmitted and received signals of each element in the antenna array, thereby achieving coherent superposition of electromagnetic wave energy in the target direction and suppressing signals in the interference direction. This technique can effectively improve the received signal-to-noise ratio of the target link, reduce multipath and co-channel interference, and thus improve communication reliability by enhancing the directionality of signal transmission.

[0010] The technical solution of this invention is as follows: The joint optimization method of downlink power allocation and beamforming for MU-MISO vehicle networking based on RSMA-OTFS is implemented through the downlink of MU-MISO vehicle networking based on RSMA-OTFS. The MU-MISO vehicular network downlink based on RSMA-OTFS includes a base station with multiple transmit antennas, multiple distributed vehicle units with single antennas, and corresponding signal processing modules. Each distributed vehicle unit corresponds to multiple high-speed vehicle users, and signal modulation is performed in the DD domain. The signal processing module integrates SiC and frequency-domain linear equalizer FD-LE. Simultaneously, the noise at the vehicle user receiver is additive white Gaussian noise. This includes: Step S1: The base station splits the user information to be transmitted into public sub-information and private sub-information. All user public sub-information is aggregated into a public data stream and transmitted uniformly after being modulated by OTFS. Each user's private sub-information is sent as an independent private data stream after being modulated by OTFS. Step S2: All users undergo the same demodulation process without interference. i The received signal is demodulated by OTFS and recovered to the DD domain. A phased recovery strategy is adopted: first, the common data stream is restored through FD-LE; then, SIC is used to eliminate the interference of the common data stream on the current user's private data stream, thereby accurately restoring the user's private data stream. i The corresponding private data is then combined with the recovered public data to restore the original data. Step S3: By calculating the signal-to-interference-plus-noise ratio (SINR) of different users in each SIC stage, an optimization problem is constructed with the goal of maximizing the system and rate. The power allocation and beamforming strategies are jointly optimized. Then, the original optimization problem is decomposed into two sub-optimization problems, power allocation and beamforming, by using an alternating optimization algorithm. Finally, the optimal power allocation and beamforming vectors are obtained by fusing continuous convex approximation and semidefinite programming algorithms.

[0011] According to a preferred embodiment of the present invention, the specific implementation process of step S1 includes: In the MU-MISO vehicle-to-everything (V2X) downlink based on RSMA-OTFS, a base station with multiple transmit antennas simultaneously serves multiple randomly distributed high-speed vehicle users with single antennas. All users are served in the DD domain, and their signals are OTFS modulated; assuming there exists K There are users, and there are K≤M ,in M For the number of OTFS subcarriers, the number of users to be transmitted i Information m i Split into common sub-information m c,iand private sub-information m p,i ,in i= 1, 2, … , K The base station aggregates all user public sub-information into public information. m c and public information m c Encoded as a public data stream x c Each user's private sub-information is then encoded into a private data stream. x 1, x 2, ⋯, x K}; The base station transmits public and private data streams to users after power allocation and beamforming preprocessing. Base station number v The time-frequency domain signal transmitted by the transmitting antenna is: ,in, p c and p i They are public data streams and users, respectively. i Private data stream power allocation coefficient, and These represent the beamforming vectors of the common data stream. and users i Private data stream beamforming vector At the base station v The weight at the root transmitting antenna, and Representing public data streams and users respectively. i The time-frequency domain signal of the private data stream.

[0012] According to a preferred embodiment of the present invention, the specific implementation process of step S2 includes: The user receiver operates in the DD domain, first demodulating the public data stream and treating the user's private data stream as interference; for the user... i The receive vector is: ,in, V This refers to the number of base station antennas. For base station number v root antenna to user i The DD domain block cyclic channel matrix has a dimension of NM MN , and Representing the common data signal vector and the user respectively j Private data signal vector,j= 1, 2, … , K , For users i The additive white Gaussian noise vector at the location; user i Further employing FD-LE to reduce inter-symbol interference and restore common data signals; setting the equalization matrix as follows. ,in and They are respectively N Dot and M Point Fourier transform matrix, It is a diagonal matrix and the (th)th line on its diagonal is kM + l +1) elements are ,in Indicates the first v The first transmitting antenna m The subcarrier n Time-domain channel impulse response amplitude of one OTFS symbol k and l These represent the Doppler index and the time delay index in the DD domain two-dimensional grid, respectively. right After equalization, the equalized signal is as follows: ,in, The first term on the right side of the equation is the common data signal term, and the second term is the user term. i The third term in the private data signal interference term is additive white Gaussian noise. Assuming all user signals have the same amplitude and all noise power is normalized; in this case, the user i The transmitted signal-to-noise ratio is: Thus, for users i Received public data, SINR is ,in, N For the number of OTFS symbols, For multi-antenna base stations to users i The channel impulse response matrix.

[0013] According to a preferred embodiment of the present invention, the specific implementation process of step S3 includes: Assuming the common data stream signal is ideally demodulated and eliminated, then the user's... i The private data signal reception SINR is: The first term in the denominator is the value divided by the user. iThe interference term caused by other user private data signals, and the second term is Gaussian white noise; under the conditions of meeting user QoS requirements, perfectly executing SIC and total power constraints, the optimization problem established to maximize the downlink system and rate of RSMA-OTFS MU-MISO vehicular network is as follows: ; Among them, constraint (a) is the perfect SIC constraint, which means that each user at the receiving end can correctly decode the public information; constraint (b) is the user QoS constraint, which means that the total information rate of all users is not lower than a preset value. R 0, R 0 represents the minimum information rate required to meet user QoS; constraint (c) represents the power constraint for beamforming of public information and each user's private information; constraint (d) represents the system transmit power constraint. P This represents the total transmit power of the base station. R ci For users i The rate of public information.

[0014] A further preferred approach is to decompose the optimization problem into two sub-optimization problems: power allocation and beamforming, and solve them separately; including: For users i First, fix the beamforming vector. and Solve for the power distribution vector p The details are as follows: First, the power allocation optimization subproblem is obtained as follows: ; The power allocation optimization subproblem is further transformed into a concavity function: ; A continuous convex approximation algorithm is used for further transformation, to... and Transform it into a convex function to ensure the concavity of the objective function; At this point, since constraints (a) and (b) are still non-convex, they are transformed into: and ; At this point, the power allocation optimization subproblem has become a standard convex problem, which can be solved using CVX; Secondly, fixed base station power allocation vector p Solve for beamforming vectors and Then the beamforming optimization subproblem becomes: ; Since the objective function is non-convex, we can introduce auxiliary variables... and The objective function is transformed into: ; Further by fixing variables and The objective function is transformed into Since the objective function is a non-convex fraction, it cannot be solved using traditional convex optimization algorithms. Therefore, auxiliary variables are introduced. and The original objective function is transformed into: The above objective function is used in optimizing beamforming vectors; therefore, fixing p and private information beamforming vector Optimize public information beamforming vector The subproblems are: ; The objective function of this problem is non-concave, and both constraints are non-convex. Therefore, we transform this problem into a convex problem using the following transformation: let... ,matrix It can be decomposed into the outer product of a column vector and a row vector, hence the matrix Since the rank is 1 and the constraint is non-convex, the constraint can be relaxed, thus transforming the original optimization problem into: ; in For trace function, For matrix The main diagonal elements; fixed p , and Optimized The subproblems are as follows: ; Similarly, let After relaxing the constraints on the rank-1 matrix, we have: ; in For matrix The main diagonal element.

[0015] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the above-described RSMA-OTFS-based MU-MISO vehicle networking downlink power allocation and beamforming joint optimization method.

[0016] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described RSMA-OTFS-based MU-MISO vehicle networking downlink power allocation and beamforming joint optimization method.

[0017] The beneficial effects of this invention are as follows: 4. Compared with typical beamforming design schemes based on NOMA-OTFS systems, this invention jointly optimizes the power allocation and beamforming strategies of RSMA-OTFS systems, resulting in higher user access fairness and resource utilization. It can significantly improve system speed and communication reliability, and is especially suitable for heterogeneous user coexistence scenarios.

[0018] The power allocation and beamforming joint optimization method based on convex optimization proposed in this invention can effectively ensure that the system and rate converge to a predetermined range within 15 iterations, and can better meet the system's performance requirements for the sum and rate. Attached Figure Description

[0019] Figure 1 This is a schematic block diagram of the downlink of the MU-MISO vehicle-to-everything (V2X) network based on RSMA-OTFS of the present invention; Figure 2 In the number of carriers M = 4, the number of subcarriers N = 4, number of base station antennas V Simulation comparison of the sum and speed performance of the method of the present invention and the NOMA-OTFS system under the configuration of =4; Figure 3 This is a convergence graph of the optimization algorithm in this invention; Figure 4 This is a schematic diagram of a joint optimization method for power allocation and beamforming. Detailed Implementation

[0020] The present invention will be further defined below with reference to the accompanying drawings and embodiments, but is not limited thereto.

[0021] Example 1 A joint optimization method for downlink power allocation and beamforming in MU-MISO vehicular networks based on RSMA-OTFS, such as Figure 4 As shown, the MU-MISO vehicular network downlink based on RSMA-OTFS is implemented. The RSMA-OTFS-based MU-MISO vehicular network downlink includes a base station with multiple transmit antennas, multiple distributed vehicle units with single antennas, and corresponding signal processing modules. Each distributed vehicle unit corresponds to multiple high-speed vehicle users, and the signal modulation process is completed in the DD domain. The signal processing module integrates SiC and frequency domain linear equalizer FD-LE. Meanwhile, the noise at the vehicle user receiver is additive white Gaussian noise. Figure 1 This is a schematic block diagram of the downlink of the MU-MISO vehicle-to-everything (V2X) network based on RSMA-OTFS of the present invention; wherein, Figure 1 In section (a), the process of base station processing transmitted signals and user vehicle receiving and recovering signals is described. Figure 1 (b) shows the detailed flowchart of the rate segmentation operation; including: Step S1: The base station splits the user information to be transmitted into public sub-information and private sub-information. All user public sub-information is aggregated into a public data stream and transmitted uniformly after being modulated by OTFS. Each user's private sub-information is sent as an independent private data stream after being modulated by OTFS. Step S2: All users undergo the same demodulation process without interference. i The received signal is demodulated by OTFS and recovered to the DD domain. A phased recovery strategy is adopted: first, the common data stream is restored through FD-LE; then, SIC is used to eliminate the interference of the common data stream on the current user's private data stream, thereby accurately restoring the user's private data stream. i The corresponding private data is then combined with the recovered public data to restore the original data. Step S3: By calculating the signal-to-interference-plus-noise ratio (SINR) of different users in each SIC stage, an optimization problem is constructed with the goal of maximizing the system and rate. The power allocation and beamforming strategies are jointly optimized. Then, the original optimization problem is decomposed into two sub-optimization problems, power allocation and beamforming, by using an alternating optimization algorithm. Finally, the optimal power allocation and beamforming vectors are obtained by fusing continuous convex approximation and semidefinite programming algorithms.

[0022] Example 2 The difference between the joint optimization method for downlink power allocation and beamforming in MU-MISO vehicular networking based on RSMA-OTFS described in Example 1 and the method described in Example 1 is as follows: The specific implementation process of step S1 includes: In the MU-MISO vehicle-to-everything (V2X) downlink based on RSMA-OTFS, a base station with multiple transmit antennas simultaneously serves multiple randomly distributed high-speed vehicle users (users) with single antennas. All users are served in the DD domain, and their signals are OTFS modulated; assuming there exists K There are users, and there are K≤M ,in M For the number of OTFS subcarriers, the number of users to be transmitted i Information m i Split into common sub-information m c,i and private sub-information m p,i ,in i= 1, 2,… , K The base station aggregates all user public sub-information into public information. m c and public information m c Encoded as a public data stream x c Each user's private sub-information is then encoded into a private data stream. x 1, x 2, ⋯, x K}; The base station transmits public and private data streams to users after power allocation and beamforming preprocessing. Base station number v The time-frequency domain signal transmitted by the transmitting antenna is: ,in, p c and p i They are public data streams and users, respectively. i Private data stream power allocation coefficient, and These represent the beamforming vectors of the common data stream. and users i Private data stream beamforming vector At the base station v The weight at the root transmitting antenna, and Representing public data streams and users respectively. i The time-frequency domain signal of the private data stream.

[0023] The specific implementation process of step S2 includes: The user receiver operates in the DD domain, first demodulating the public data stream and treating the user's private data stream as interference; for the user... i The receive vector is: ,in, V This refers to the number of base station antennas. For base station number v root antenna to user i The DD domain block cyclic channel matrix has a dimension of NM MN , and Representing the common data signal vector and the user respectively j Private data signal vector, j= 1, 2, … , K , For users iThe additive white Gaussian noise vector at the location; user i Further employing FD-LE to reduce inter-symbol interference and restore common data signals; setting the equalization matrix as follows. ,in and They are respectively N Dot and M Point Fourier transform matrix, It is a diagonal matrix and the (th)th line on its diagonal is kM + l +1) elements are ,in Indicates the first v The first transmitting antenna m The subcarrier n Time-domain channel impulse response amplitude of one OTFS symbol k and l These represent the Doppler index and the time delay index in the DD domain two-dimensional grid, respectively. right After equalization, the equalized signal is as follows: ,in, The first term on the right side of the equation is the common data signal term, and the second term is the user term. i The third term in the private data signal interference term is additive white Gaussian noise. Assuming all user signals have the same amplitude and all noise power is normalized; in this case, the user i The transmitted signal-to-noise ratio is: Thus, for users i Received public data, SINR is ,in, N For the number of OTFS symbols, For multi-antenna base stations to users i The channel impulse response matrix.

[0024] The specific implementation process of step S3 includes: Interference caused by public data signals is handled through the SIC strategy, accurately detecting and outputting user-private data signals. This leads to the construction of an optimization problem aimed at maximizing system performance and data rate, achieving joint optimization design of beamforming and power allocation. i After equalization, the common data stream signal in the DD domain is first detected and its corresponding common data signal is obtained. Then, interference in the common data stream signal is eliminated through SIC, and then the user... i The private data signal is recovered. Assuming the public data stream signal is ideally demodulated and eliminated, the user's private data signal is then obtained. i The private data signal reception SINR is: The first term in the denominator is the value divided by the user.i Interference from other users' private data signals is considered, with the second term being Gaussian white noise. Under the constraints of user QoS requirements, perfect SIC execution, and total power limitations, the optimization problem aims to maximize the downlink system and rate of the RSMA-OTFS MU-MISO vehicular network, thereby achieving efficient utilization of spectrum resources and optimizing user experience. The established optimization problem is as follows:

[0025] ; Among them, constraint (a) is the perfect SIC constraint, which means that each user at the receiving end can correctly decode the public information; constraint (b) is the user QoS constraint, which means that the total information rate of all users is not lower than a preset value. R 0, R 0 represents the minimum information rate required to meet user QoS; constraint (c) represents the power constraint for beamforming of public information and each user's private information; constraint (d) represents the system transmit power constraint. P This represents the total transmit power of the base station. R ci For users i The rate of public information.

[0026] The aforementioned joint optimization problem of beamforming and power allocation is non-convex, and , and p =[ p c , p 1, p 2, …, p K The two problems are mutually coupled, making direct solutions difficult. Therefore, the optimization problem is decomposed into two sub-optimization problems: power allocation and beamforming, which are solved separately. These include: The problem-solving process is identical and independent for all users. i First, fix the beamforming vector. and Solve for the power distribution vector p The details are as follows: First, the power allocation optimization subproblem is obtained as follows: ; The power allocation optimization subproblem is further transformed into a concavity function: ; To address the aforementioned issues, a continuous convex approximation algorithm is employed for further transformation, resulting in... and Transform it into a convex function to ensure the concavity of the objective function; At this point, since constraints (a) and (b) are still non-convex, they are transformed into: and ; At this point, the power allocation optimization subproblem has become a standard convex problem, which can be solved using CVX; Secondly, fixed base station power allocation vector p Solve for beamforming vectors and Then the beamforming optimization subproblem becomes: ; Since the objective function is non-convex, we can introduce auxiliary variables... and The objective function is transformed into: ; Further by fixing variables and The objective function is transformed into Since the objective function is a non-convex fraction, it cannot be solved using traditional convex optimization algorithms. Therefore, auxiliary variables are introduced. and The original objective function is transformed into: The above objective function is used in optimizing beamforming vectors; therefore, fixing p and private information beamforming vector Optimize public information beamforming vector The subproblems are: ; The objective function of this problem is non-concave, and both constraints are non-convex. Therefore, we transform this problem into a convex problem using the following transformation: let... ,matrix It can be decomposed into the outer product of a column vector and a row vector, hence the matrix Since the rank is 1 and the constraint is non-convex, the constraint can be relaxed, thus transforming the original optimization problem into: ; in For trace function, For matrix The main diagonal elements; since this problem is convex and its solution may not satisfy the rank-1 constraint, Gaussian randomization can be used to obtain its suboptimal solution.

[0027] fixed p , and Optimized The subproblems are as follows: ; Similarly, let After relaxing the constraints on the rank-1 matrix, we have: ; in For matrix The main diagonal elements. At this point, the transformed problem has become a convex problem, and its solution may not satisfy the rank-1 constraint. Therefore, Gaussian randomization is used to obtain a suboptimal solution. For Optimization steps and similar.

[0028] Set the maximum number of iterations, test the algorithm's convergence, obtain the optimal or suboptimal solutions for power allocation and beamforming for each user, and test the system and rate under different total transmit powers to evaluate the system's optimization performance.

[0029] After the system optimization algorithm converges, online deployment is implemented, with the system set to have 4 users, 4 base station antennas, 4 OTFS symbols, 4 subcarriers, and user QoS constraints. R 0 = 1 bps / Hz, modulation order is 4QAM, user moving speed is 120 km / h, maximum number of iterations is set to 15, and convergence error threshold is 0.001. Based on Shannon's limit theorem and optimized output... p , and The maximum sum rate of the system can be calculated, and the performance of the system can be further evaluated by referring to the NOMA-OTFS system.

[0030] Figure 2 In the number of carriers M = 4, the number of subcarriers N = 4, number of base station antennas V The simulation comparison of the sum-rate performance of the proposed method and the NOMA-OTFS system under a configuration of SNR=4 is shown in the figure. The horizontal axis represents the signal-to-noise ratio (SNR), and the vertical axis represents the system sum-rate. Compared to NOMA, RSMA effectively improves the system transmission rate by segmenting information. At SNR=25, the RSMA-OTFS scheme improves the system sum-rate by approximately 11.2% compared to the NOMA-OTFS scheme.

[0031] Figure 3 This is the convergence graph of the optimization algorithm in this embodiment; by Figure 3 As the number of iterations increases, the system and rate gradually increase, and convergence is achieved in about 15 iterations. Example 3 A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the RSMA-OTFS-based MU-MISO vehicle networking downlink power allocation and beamforming joint optimization method described in Embodiment 1 or 2.

[0032] Example 4 A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the RSMA-OTFS-based MU-MISO vehicle networking downlink power allocation and beamforming joint optimization method described in Embodiment 1 or 2.

Claims

1. A joint optimization method for downlink power allocation and beamforming in MU-MISO vehicular networks based on RSMA-OTFS, characterized in that, This is achieved through the MU-MISO vehicle-to-everything (V2X) downlink based on RSMA-OTFS. The MU-MISO vehicular network downlink based on RSMA-OTFS includes a base station with multiple transmit antennas, multiple distributed vehicle units with single antennas, and corresponding signal processing modules. Each distributed vehicle unit corresponds to multiple high-speed vehicle users, and signal modulation is performed in the DD domain. The signal processing module integrates SiC and frequency-domain linear equalizer FD-LE. Simultaneously, the noise at the vehicle user receiver is additive white Gaussian noise. This includes: Step S1: The base station splits the user information to be transmitted into public sub-information and private sub-information. All user public sub-information is aggregated into a public data stream and transmitted uniformly after being modulated by OTFS. Each user's private sub-information is sent as an independent private data stream after being modulated by OTFS. Step S2: All users undergo the same demodulation process without interference. i The received signal is demodulated by OTFS and recovered to the DD domain. A phased recovery strategy is adopted: first, the common data stream is restored through FD-LE; then, SIC is used to eliminate the interference of the common data stream on the current user's private data stream, thereby accurately restoring the user's private data stream. i The corresponding private data is then combined with the recovered public data to restore the original data. Step S3: By calculating the signal-to-interference-plus-noise ratio (SINR) of different users in each SIC stage, an optimization problem is constructed with the goal of maximizing the system and rate. The power allocation and beamforming strategies are jointly optimized. Then, the original optimization problem is decomposed into two sub-optimization problems, power allocation and beamforming, by using an alternating optimization algorithm. Finally, the optimal power allocation and beamforming vectors are obtained by fusing continuous convex approximation and semidefinite programming algorithms.

2. The joint optimization method for downlink power allocation and beamforming in MU-MISO vehicular networking based on RSMA-OTFS as described in claim 1, characterized in that, The specific implementation process of step S1 includes: In the MU-MISO vehicle-to-everything (V2X) downlink based on RSMA-OTFS, a base station with multiple transmit antennas simultaneously serves multiple randomly distributed high-speed vehicle users with single antennas. All users are served in the DD domain, and their signals are OTFS modulated; assuming there exists K There are users, and there are K≤M ,in M For the number of OTFS subcarriers, the number of users to be transmitted i Information m i Split into common sub-information m c,i and private sub-information m p,i ,in i= 1,2, … , K The base station aggregates all user public sub-information into public information. m c and public information m c Encoded as a public data stream x c Each user's private sub-information is then encoded into a private data stream. x 1, x 2, ⋯, x K }; The base station transmits public and private data streams to users after power allocation and beamforming preprocessing. Base station number v The time-frequency domain signal transmitted by the transmitting antenna is: ,in, p c and p i They are public data streams and users, respectively. i Private data stream power allocation coefficient, and These represent the beamforming vectors of the common data stream. and users i Private data stream beamforming vector At the base station v The weight at the root transmitting antenna, and Representing public data streams and users respectively. i The time-frequency domain signal of the private data stream.

3. The joint optimization method for downlink power allocation and beamforming in MU-MISO vehicular networking based on RSMA-OTFS as described in claim 1, characterized in that, The specific implementation process of step S2 includes: The user receiver operates in the DD domain, first demodulating the public data stream and treating the user's private data stream as interference; for the user... i The receive vector is: ,in, V This refers to the number of base station antennas. For base station number v root antenna to user i The DD domain block cyclic channel matrix has a dimension of NM MN , and Representing the common data signal vector and the user respectively j Private data signal vector, j= 1, 2, … , K , For users i The additive white Gaussian noise vector at the location; user i Further employing FD-LE to reduce inter-symbol interference and restore common data signals; setting the equalization matrix as follows. ,in and They are respectively N Dot and M Point Fourier transform matrix, It is a diagonal matrix and the (th)th line on its diagonal is kM + l +1) elements are ,in Indicates the first v The first transmitting antenna m The subcarrier n Time-domain channel impulse response amplitude of one OTFS symbol k and l These represent the Doppler index and the time delay index in the DD domain two-dimensional grid, respectively. right After equalization, the equalized signal is as follows: ,in, The first term on the right side of the equation is the common data signal term, and the second term is the user term. i The third term in the private data signal interference term is additive white Gaussian noise. Assuming all user signals have the same amplitude and all noise power is normalized; in this case, the user i The transmitted signal-to-noise ratio is: Thus, for users i Received public data, SINR is ,in, N For the number of OTFS symbols, For multi-antenna base stations to users i The channel impulse response matrix.

4. The joint optimization method for downlink power allocation and beamforming in MU-MISO vehicular networking based on RSMA-OTFS according to any one of claims 1-3, characterized in that, The specific implementation process of step S3 includes: Assuming the common data stream signal is ideally demodulated and eliminated, then the user's... i The private data signal reception SINR is: The first term in the denominator is the value divided by the user. i The interference term caused by other user private data signals, and the second term is Gaussian white noise; under the conditions of meeting user QoS requirements, perfectly executing SIC and total power constraints, the optimization problem established to maximize the downlink system and rate of RSMA-OTFS MU-MISO vehicular network is as follows: ； Among them, constraint (a) is the perfect SIC constraint, which means that each user can correctly decode the public information at the receiving end; constraint (b) is the user QoS constraint, which means that the total information rate of all users is not lower than a preset value. R 0, R 0 represents the minimum information rate required to meet user QoS; constraint (c) represents the power constraint for beamforming of public information and each user's private information; constraint (d) represents the system transmit power constraint. P This represents the total transmit power of the base station. R ci For users i The rate of public information.

5. The joint optimization method for downlink power allocation and beamforming in MU-MISO vehicular networking based on RSMA-OTFS according to claim 4, characterized in that, The optimization problem is decomposed into two sub-optimization problems: power allocation and beamforming, which are solved separately. This includes: For users i First, fix the beamforming vector. and Solve for the power distribution vector p The details are as follows: First, the power allocation optimization subproblem is obtained as follows: ； The power allocation optimization subproblem is further transformed into a concavity function: ； A continuous convex approximation algorithm is used for further transformation, to... and Transform it into a convex function to ensure the concavity of the objective function; At this point, since constraints (a) and (b) are still non-convex, they are transformed into: and ; At this point, the power allocation optimization subproblem has become a standard convex problem, which can be solved using CVX; Secondly, fixed base station power allocation vector p Solve for beamforming vectors and Then the beamforming optimization subproblem becomes: ； Since the objective function is non-convex, we can introduce auxiliary variables... and The objective function is transformed into: ; Further by fixing variables and The objective function is transformed into Since the objective function is a non-convex fraction, it cannot be solved using traditional convex optimization algorithms. Therefore, auxiliary variables are introduced. and The original objective function is transformed into: The above objective function is used in optimizing beamforming vectors; therefore, fixing p and private information beamforming vector Optimize public information beamforming vector The subproblems are: ； The objective function of this problem is non-concave, and both constraints are non-convex. Therefore, we transform this problem into a convex problem using the following transformation: let... ,matrix It can be decomposed into the outer product of a column vector and a row vector, hence the matrix Since the rank is 1 and the constraint is non-convex, the constraint can be relaxed, thus transforming the original optimization problem into: ； in For trace function, For matrix The main diagonal elements; fixed p , and Optimized The subproblems are as follows: ； Similarly, let After relaxing the constraints on the rank-1 matrix, we have: ； in For matrix The main diagonal element.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the MU-MISO vehicle networking downlink power allocation and beamforming joint optimization method based on RSMA-OTFS as described in any one of claims 1-5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the joint optimization method for downlink power allocation and beamforming of MU-MISO vehicle networking based on RSMA-OTFS as described in any one of claims 1-5.

Images