Resource allocation method based on OTFS-NOMA system, storage medium and program product

Abstract

The invention discloses a resource allocation method based on an OTFS-NOMA system, a storage medium and a program product, which maximizes the total rate of the system by performing cross-domain resource reuse in a time delay-Doppler domain and a time-frequency domain and jointly optimizing user matching and power allocation. The method comprises the following steps: firstly, carrying out initialization modeling on users and resource blocks, and carrying out non-orthogonal multiplexing on high-speed users and low-speed user signals of a time-frequency domain after pairwise superposition of the high-speed users in a time delay-Doppler domain; decomposing a complex joint optimization problem into a user matching sub-problem and a power distribution sub-problem, and solving through alternate iteration: the user matching sub-problem adopts a KM algorithm to realize optimal matching of a high-speed user pair and a low-speed user pair; and the power distribution sub-problem is solved by using convex difference programming in combination with a logarithmic barrier function method. The spectrum efficiency and the system throughput are effectively improved, and the method is particularly suitable for high-mobility scenes such as Internet of Vehicles and unmanned aerial vehicle communication.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of resource allocation of NOMA systems, and in particular to a resource allocation method based on an OTFS-NOMA system, a computer readable storage medium and a program product. BACKGROUND

[0002] With the research and development of wireless communication technology, the number of user terminals is growing exponentially. In order to meet the growing demand for communication, researchers have achieved the standardization and commercialization of 5G mobile communication worldwide after years of research, and have shifted their focus to 6G, which will provide services for scenarios such as vehicle networking, drones, and low-orbit satellites. To achieve this goal, one of the key challenges is how to provide reliable communication in high mobility environments. For this purpose, OTFS (Orthogonal Time Frequency Space) has received widespread attention as a modulation method that can provide efficient and reliable communication in high mobility scenarios.

[0003] However, there are usually multiple types of users in real scenarios, such as high-speed users (HSU) and low-speed users (LSU). If OTFS is simply used, LSU will have to wait for HSU communication before communicating. Through the Non-Orthogonal Multiple-Access (NOMA) technology, different types of users can be placed in the Delay Doppler (DD) domain and the Time Frequency (TF) domain to improve the frequency utilization of the system and improve system performance. Although existing methods consider scenarios with multiple user types under the combination of OTFS and NOMA, they ignore matching different types of users.

[0004] Therefore, the present application proposes the matching of HSU and LSU and the power allocation problem for the resource allocation problem of the OTFS-NOMA system to improve system performance. SUMMARY

[0005] In view of the problems in the prior art, the present application aims to provide a resource allocation method based on an OTFS-NOMA system, a computer readable storage medium and a program product, which jointly matches users and allocates power to maximize the total communication rate of all users.

[0006] To achieve the above-mentioned purpose, the technical solution adopted by the present application is as follows: A resource allocation method based on an OTFS-ISAC system, the method comprising the following steps: Step 1, in the OTFS-NOMA system, a set of HSU and a set of LSU are initialized; The HSU set and the LSU set are respectively denoted by and , the index symbol of HSU is denoted by , the index symbol of LSU is denoted by , and ; Step 2, the DD domain resource block and the TF domain resource block of the OTFS-NOMA system are defined; The frame length of OTFS is set to , the subcarrier is set to , there are frames and subcarriers in the system, the resource block is divided into groups, and the DD domain resource block used by the th group is defined as: , wherein, ; denotes the Doppler domain index, ; denotes the time delay domain index, ; denote the parameters for dividing and , respectively, and ; The TF domain resource block is defined as: , wherein, , , , ; Step 3, signal superposition; The HSU signals are superposed in pairs in the DD domain, and mapped into the HSU pair signal in the DD domain resource block, and then the HSU pair signal is superposed with the LSU signal in the TF domain, and mapped into the TF domain resource block; Specifically, the resource multiplexing relationship between HSU and LSU is defined by , if , it indicates that the th HSU pair and the th LSU match multiplexing the same resource, the HSU pair signal is converted into TF by inverse symplectic Fourier transform and superposed with the LSU signal to obtain the high-low speed user superposed TF domain signal; if , it indicates that the th HSU pair and the th LSU do not match. Step 4: The high- and low-speed user superimposed TF domain signals obtained after matching are converted into time domain signals by Heisenberg transformation and then transmitted. Step 5: Calculate the sum of communication rates for all users in the OTFS-NOMA system, list the constraints the system must satisfy, and describe the optimization model as follows to maximize the total communication rate for all users: ; (1) No. In HSUs, those with higher allocated power are called strong users. Conversely, it is called a weak user. ;No. The signal-to-interference-plus-noise ratio of each HSU is: ,in They respectively represent the assignments to the first HSU, the first The HSU and the first The power of each LSU Indicates the first The noise term of the HSU; The SINR of each HSU is: ,in Indicates the first The noise term of the HSU; The SINR of each LSU is: ,in ; (2) Constraints: All HSUs and LSUs must meet the corresponding signal-to-interference-plus-noise ratio thresholds. and The sum of the transmit power of the HSU and LSU shall not exceed the maximum transmit power. At the same time, it needs to meet the following requirements. HSUs must be paired and stacked; an LSU can only be matched with a pair of HSUs. (3) The system performance of the OTFS-NOMA system is measured by the sum of the communication rates of all users. The optimization model is expressed as follows: :

[0007] No. Communication rate per HSU , No. Communication rate per HSU , No. Communication rate per LSU , Indicates bandwidth. Indicates the first The HSU and the first The HSUs are divided into a pair and are related to the first LSU matching; Step 6, HSU pairing: the first HSU and the first user are matched as a pair, thus obtaining Q pairs of HSU, i.e. ; when the pairing result of HSU is given, can be converted into ; Step 7, dividing into user matching and power allocation two sub-models for solving, initializing the transmission power of two HSU and LSU, setting the iteration precision of alternating optimization , and the number of alternating iterations ; , ; Step 8, solving the user resource multiplexing sub-model in the first iteration , according to the transmission power of two HSU and LSU obtained in the first iteration , the optimal user matching result in the first iteration is obtained through the KM algorithm; Step 9, solving the power allocation sub-model in the first iteration , according to the optimal matching result obtained in step 8, adjusting the transmission power of the system to obtain the optimal power solution in the first iteration ; Step 10, according to the result of the first iteration , calculating the sum of the communication rates of all users, if , the iteration is ended, and the optimal matching result and the optimal power solution are output, otherwise, returning to step 8 for iteration.

[0008] The step 8 is specifically as follows: 8-1, first, two HSU are matched as a pair, a bipartite graph model is constructed, the vertex set of HSU pair and the vertex set of LSU are established, and the edge weight between vertices is calculated , which represents the sum of the communication rates of the first pair of HSU and the first LSU: , wherein represents the communication rate of the first pair of HSU, specifically as follows:

[0009] Set A virtual LSU vertex is set to make the number of HSU pairs consistent with the number of LSU vertices, and its virtual edge weight is set to 0; 8-2, use KM algorithm to solve, and obtain the optimal matching result.

[0010] 3. The resource allocation method based on the OTFS-ISAC system according to claim 1, wherein the step 9 is specifically as follows: Step 9-1, set the maximum number of iterations allowed in the surrogate function approximation stage and the logarithmic barrier function stage And the maximum convergence error ; Step 9-2, surrogate function approximation stage; Initialize the HSU and LSU transmit power vectors of this stage ; Then express the objective function as the difference form of the biconvex function:

[0011] Wherein, P represents the power allocation vector obtained at the current iteration of the surrogate function approximation stage, n represents the iteration number, P represents the power allocation vector at the th iteration of the surrogate function approximation stage, represents the gradient of ; Step 9-3, logarithmic barrier function stage: initialize the HSU and LSU transmit power vectors of the logarithmic barrier function stage , set the initial precision of the logarithmic barrier function and the adjustment parameter ; Step 9-4, construct the logarithmic barrier function to convert the constrained convex optimization problem into an unconstrained optimization model :

[0012] Wherein, n represents the iteration number of the logarithmic barrier function stage, P represents the power vector at the th iteration of the logarithmic barrier function stage, and the logarithmic barrier function is defined as: , represents the constraint: , , , , , , ; Based on the optimal matching result obtained in step 8, the golden section algorithm is used to optimize the unconstrained model under constraints. The optimal power allocation between the HSU and LSU is obtained by solving the problem. Step 9-5, when Or the number of iterations of the logarithmic barrier function Then stop the iteration and let the iteration in step 9-2 stop. Equal to the current power vector Otherwise, adjust the precision of the logarithmic barrier function: Continue with step 9-4; when Alternatively, the number of iterations in the approximation stage can be replaced. If the result is positive, stop the iteration and output the current power vector, i.e., the optimal power allocation; otherwise, return to step 9-2 and let the result in step 9-2 be positive. Equal to the current power vector .

[0013] A computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the resource allocation method based on the OTFS-ISAC system as described above. A computer program product includes a computer program / instructions that, when executed by a processor, implement the steps of the resource allocation method based on the OTFS-ISAC system as described above. By adopting the above scheme, this invention enables the BS to provide services to both types of users simultaneously in the presence of both HSU and LSU, using NOMA (Normally Oscillating Analysis). It also considers the matching problem between HSU pairs and LSUs, establishing a model to maximize the communication rate for all users, involving user resource reuse and transmit power allocation. Specifically, this invention divides this into two sub-problems: user resource reuse and power allocation. Combining the idea of ​​alternating optimization, it first obtains the optimal matching result between HSU pairs and LSUs, and then uses convexity programming to solve for the optimal transmit power for each user. Compared with previous methods, the method proposed in this invention considers the matching problem between HSU and LSU, achieving joint optimization of user resource reuse and power allocation, significantly improving the communication rate for all users. Furthermore, this invention enables HSU and LSU to communicate simultaneously, improving spectrum utilization. Attached Figure Description

[0014] Figure 1 A model diagram of the OTFS-NOMA system; Figure 2 This is a schematic diagram of the user access process described in this invention; Figure 3The power allocation algorithm flowchart of the present application; Figure 4 The algorithm flowchart of the present application. DETAILED DESCRIPTION

[0015] The present application relates to a resource allocation method based on an OTFS-ISAC system, in the case of high-speed users (HSU) and low-speed users (LSU), both of which are placed in two different domains for communication through NOMA, improving the system spectrum utilization. Under the premise of ensuring the communication quality of HSU and LSU, the sum of the communication rates of all users is maximized through the joint optimization of user grouping matching and transmit power allocation.

[0016] As Figure 1 shown, the present application considers an OTFS downlink communication scenario, in which there is a BS (base station) with a coverage of , which provides services for HSU and LSU. The set of HSU is defined as , and the set of LSU is defined as . Since HSU has a serious Doppler shift problem, the signal is placed in the DD domain (Delay-Doppler Domain), and LSU has a slow moving speed, so the Doppler effect is not obvious, and the LSU signal is placed in the TF domain (Time Frequency Domain). Since the corresponding TF domain resource is occupied when the HSU uses the DD domain resource for communication, the LSU cannot communicate, so NOMA technology is used to provide services for both, and the user resource reuse and power allocation problems are considered.

[0017] The resource method of the present application specifically includes the following steps: Step 1, in the OTFS-NOMA system, a set of HSU set and LSU set is initialized and generated.

[0018] The HSU set and the LSU set are represented by and respectively, the index symbol of HSU is represented by , the index symbol of LSU is represented by , and .

[0019] Step 2, define the DD domain resource block and the TF domain resource block of the OTFS-NOMA system.

[0020] Set the frame length of OTFS as , the subcarrier as , and the system exists frames and subcarriers, the resource blocks are divided into groups, the group uses resource blocks in the delay-Doppler (DD) domain defined as: , where denotes the Doppler domain index, denotes the delay domain index, denote the parameters that partition and respectively, and .

[0021] resource blocks in the time-frequency (TF) domain are defined as: , where, , , , .

[0022] Step 3, signal superposition.

[0023] As shown in Figure 2 , the HSU signals are superimposed in pairs in the DD domain, and mapped into the DD domain resource blocks to obtain the HSU pair signals, and then the HSU pair signals are converted into TF domain signals by inverse symplectic Fourier transform (ISFFT) and superimposed with the LSU signals in the TF domain, and mapped into the TF domain resource blocks.

[0024] To maximize the system performance, define denotes the resource multiplexing relationship of the HSU and the LSU, if , it means that the pair of HSU and the LSU are matched to multiplex the same resource, the DD domain signal of the HSU pair is converted into the TF domain signal by inverse symplectic Fourier transform (ISFFT) and superimposed with the LSU signal to obtain the superimposed TF domain signal of high and low speed users; if , it means that the pair of HSU and the LSU are not matched.

[0025] Step 4, the superimposed TF domain signal of high and low speed users obtained after matching is converted into a time domain signal by Heisenberg transform for transmission.

[0026] Step 5, the sum of the communication rates of all users in the OTFS-NOMA system is calculated, and the constraint conditions that the system should satisfy are listed, and the optimization model is described as .

[0027] (1) the In HSUs, those with higher allocated power are called strong users. Conversely, it is called a weak user. , No. The signal-to-interference-plus-noise ratio (SINR) of each HSU is: ,in They respectively represent the assignments to the first The HSU, the first The HSU and the first The power of each LSU Indicates the first The noise term of each HSU is specifically expressed as follows: ,in diagonal matrix The One element, the specific expression is: , It is the first Channel matrix of HSU Located in the first column, number The elements of the row. The SINR of each HSU is: ,in Indicates the first The noise term of each HSU, ,in , It is the first Channel matrix of HSU Located in the first column, number The elements of the row. The SINR of each LSU is: ,in .

[0028] (2) Constraints: All HSUs and LSUs must meet the corresponding signal-to-interference-plus-noise ratio thresholds. and The sum of the transmit power of the HSU and LSU shall not exceed the maximum transmit power. At the same time, it needs to meet the following requirements. HSUs must be paired and stacked; an LSU can only be matched with a pair of HSUs.

[0029] (3) The system performance of the OTFS-NOMA system is measured by the sum of the communication rates of all users. The optimization model is expressed as follows: :

[0030] Among them, the Communication rate per HSU , No. Communication rate per HSU , No. Communication rate per LSU , Indicates bandwidth. Indicates the first The HSU and the first The HSUs are divided into a pair and are related to the first One LSU match.

[0031] Step 6, HSU pair: Based on an important conclusion in NOMA, the greater the difference in channel conditions between two users, the higher the performance gain. The specific method is defined as follows:

[0032] The upcoming The HSU and the first Each user is divided into pairs, thus obtaining Q pairs of HSUs. Given the pairwise results of HSUs... It can be converted into .

[0033] Step 7, as follows Figure 3 and Figure 4 As shown, combining the idea of ​​alternating optimization, Divided into user matching and power distribution Solve the two sub-models, initialize the transmit power of the two HSUs and LSUs, and set the iterative accuracy for alternating optimization. Number of alternating iterations ; , ; Step 8, Solve the first... User resource reuse sub-model in the next iteration According to the The transmit powers of the two HSUs and LSUs obtained in the second iteration are used to obtain the third iteration using the KM algorithm. The optimal user matching result of the next iteration The details are as follows: 8-1. First, divide the two HSUs into a pair, simplifying the problem to a one-to-one matching problem between a pair of HSUs and an LSU. Construct a bipartite graph model, establish the vertex sets of the HSU pair and the LSU vertex sets, and calculate the edge weights between the vertices. It indicates the first For HSU and the The sum of the communication rates of each LSU: ,in Indicates the first Communication rate to HSU Specifically as follows:

[0034] set up A set of virtual LSU vertices is created to make the number of HSU pairs and LSU vertex sets the same, and the virtual edge weights are set to 0.

[0035] 8-2. Solve using the KM algorithm to obtain the optimal matching result.

[0036] Step 9, Solve the first... Power allocation submodel in the next iteration Based on the optimal matching result obtained in step 8 The system's transmission power is adjusted to obtain the first... The optimal power solution for the next iteration is as follows: 9-1. Set the maximum number of iterations allowed in the substitution function approximation stage and the logarithmic barrier function stage. and maximum convergence error .

[0037] 9-2, Approximation stage of substitution function; Initialize the HSU and LSU transmit power vectors during this phase. The objective function is expressed in the form of the difference (DC) of biconvex functions, i.e. Based on the concept of DC programming, a first-order Taylor expansion is used to construct... substitution function Minimize convex problems It iteratively approximates the non-convex objective function. :

[0038] in, This represents the power allocation vector obtained in the current iteration of the substitution function approximation phase. Indicates the number of iterations. The substitution function approximation stage is represented by the first... The power allocation vector at the next iteration express The gradient.

[0039] 9-3. Logarithmic Barrier Function Phase: Initializing the transmit power vectors of HSU and LSU during the logarithmic barrier function phase. Set the initial precision of the logarithmic barrier function. and adjusting parameters .

[0040] 9-4. Constructing a logarithmic barrier function transforms a constrained convex optimization problem into an unconstrained optimization model. :

[0041] in, This represents the number of iterations in the logarithmic barrier function phase. Represents the logarithmic barrier function stage. The power vector of the next iteration, the logarithmic barrier function is defined as: , Representing constraints: , , , , , , ; Based on the optimal matching result obtained in step 8, the golden section algorithm is used to optimize the unconstrained model under constraints. The optimal power allocation between the HSU and LSU is obtained by solving the problem.

[0042] 9-5, when Or the number of iterations of the logarithmic barrier function Then stop the iteration and let the iteration in step 9-2 stop. Equal to the current power vector Otherwise, adjust the precision of the logarithmic barrier function: Proceed to step 9-4.

[0043] when Alternatively, the number of iterations in the approximation stage can be replaced. If the result is positive, stop the iteration and output the current power vector, i.e., the optimal power allocation; otherwise, return to step 9-2 and let the result in step 9-2 be positive. Equal to the current power vector .

[0044] Step 10, according to the The result of this iteration is the sum of the communication rates for all users. ,like If the iteration ends, output the optimal matching result and the optimal power solution; otherwise, return to step 8 to continue the iteration.

[0045] This invention, based on the OTFS-NOMA system, considers the matching and power allocation of HSUs and LSUs. First, HSUs are paired, and the KM algorithm is used to obtain the optimal HSU pairing and LSU matching results. Then, the optimal transmit power for each user is solved using the concept of convexity programming, improving spectrum utilization while maximizing the sum of communication rates for all users. This invention effectively improves spectrum efficiency and system throughput, and is particularly suitable for high-mobility scenarios such as vehicle-to-everything (V2X) and drone communication.

[0046] The present invention also provides a computer-readable medium, which may be included in the electronic device described in the above embodiments; or it may exist alone and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs, which, when executed by an electronic device, cause the electronic device to perform the methods described in the above embodiments. It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units. From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the method according to the embodiments of this disclosure.

Claims

1. A resource allocation method based on the OTFS-ISAC system, characterized in that, The method includes the following steps: Step 1: In the OTFS-NOMA system, initialize and generate a set of HSU sets and LSU sets; The HSU set and LSU set are respectively used and Indicates, using Indicates the index symbol of HSU, The index symbol representing the LSU, and ; Step 2: Define the DD domain resource blocks and TF domain resource blocks of the OTFS-NOMA system; Set the OTFS frame length to Subcarrier is The system exists Frames and Subcarriers divide the resource block into Group, No. The DD domain resource block used by the group is defined as follows: ， in, ; Indicates a Doppler field index. ; Indicates the delay field index. ; Representing division and The parameters, and ; TF domain resource blocks are defined as follows: ， in, , , , ; Step 3: Signal superposition; HSU signals are superimposed in pairs in the DD domain and mapped to the DD domain resource block to obtain HSU pair signals. Then, the HSU pair signals and LSU signals are superimposed in the TF domain and mapped to the TF domain resource block. Specifically, define This indicates the resource reuse relationship between HSU and LSU. , indicating the first For HSU and the Each LSU (Local Subsystem Unit) multiplexes the same resource. The DD (Distributed Domain) signal of the HSU pair is converted into a TF (Transformed Domain) signal through inverse sine Fourier transform and then superimposed with the LSU signal to obtain the superimposed TF signal for high-speed and low-speed users. Then it means the first For HSU and the One LSU does not match; Step 4: The high- and low-speed user superimposed TF domain signals obtained after matching are converted into time domain signals by Heisenberg transformation and then transmitted. Step 5: Calculate the sum of communication rates for all users in the OTFS-NOMA system, list the constraints the system must satisfy, and describe the optimization model as follows to maximize the total communication rate for all users: ; (1) No. In HSUs, those with higher allocated power are called strong users. Conversely, it is called a weak user. ;No. The signal-to-interference-plus-noise ratio of each HSU is: ,in They respectively represent the assignments to the first HSU, the first The HSU and the first The power of each LSU Indicates the first The noise term of the HSU; The SINR of each HSU is: ,in Indicates the first The noise term of the HSU; The SINR of each LSU is: ,in ; (2) Constraints: All HSUs and LSUs must meet the corresponding signal-to-interference-plus-noise ratio thresholds. and The sum of the transmit power of the HSU and LSU shall not exceed the maximum transmit power. At the same time, it needs to meet the following requirements. HSUs must be paired and stacked; an LSU can only be matched with a pair of HSUs. (3) The system performance of the OTFS-NOMA system is measured by the sum of the communication rates of all users. The optimization model is expressed as follows: : No. Communication rate per HSU , No. Communication rate per HSU , No. Communication rate per LSU , Indicates bandwidth. Indicates the first The HSU and the first The HSUs are divided into a pair and are related to the first One LSU match; Step 6: Perform HSU pairing: (The text appears to be incomplete and contains several grammatical errors. A more The HSU and the first Each user is divided into a pair, thus obtaining Q pairs of HSUs, i.e. When the paired results of HSU are given, It can be converted into ; Step 7, Divided into user matching and power distribution Solve the two sub-models, initialize the transmit power of the two HSUs and LSUs, and set the iterative accuracy for alternating optimization. Number of alternating iterations ; ， ； Step 8, Solve the first... User resource reuse sub-model in the next iteration According to the The transmit powers of the two HSUs and LSUs obtained in the second iteration are used to obtain the third iteration using the KM algorithm. The optimal user matching result of the next iteration ; Step 9, Solve the first... Power allocation submodel in the next iteration Based on the optimal matching result obtained in step 8 The system's transmission power is adjusted to obtain the first... The optimal power solution for the next iteration; Step 10, according to the The result of this iteration is the sum of the communication rates for all users. ,like If the iteration ends, output the optimal matching result and the optimal power solution; otherwise, return to step 8 to continue the iteration.

2. The resource allocation method based on the OTFS-ISAC system according to claim 1, characterized in that, Step 8 is described in detail below: 8-1. First, divide the two HSUs into a pair, construct a bipartite graph model, establish the vertex set of the HSU pair and the vertex set of the LSU, and calculate the edge weights between the vertices. It indicates the first For HSU and the The sum of the communication rates of each LSU: ,in Indicates the first Communication rate to HSU Specifically as follows: set up A virtual LSU vertex is used to make the number of HSU pairs and LSU vertex sets the same, and its virtual edge weight is set to 0. 8-2. Solve using the KM algorithm to obtain the optimal matching result.

3. The resource allocation method based on the OTFS-ISAC system according to claim 1, characterized in that, Step 9 is described in detail below: Step 9-1: Set the maximum number of iterations allowed for the substitution function approximation stage and the logarithmic barrier function stage. and maximum convergence error ; Step 9-2: Substitution Function Approximation Stage; Initialize the HSU and LSU transmit power vectors during this phase. Then, the objective function is expressed as the difference form of a biconvex function: in, This represents the power allocation vector obtained in the current iteration of the substitution function approximation phase. Indicates the number of iterations. The substitution function approximation stage is represented by the first... The power allocation vector at the next iteration express The gradient; Step 9-3, Logarithmic Barrier Function Phase: Initialize the transmit power vectors of HSU and LSU during the logarithmic barrier function phase. Set the initial precision of the logarithmic barrier function. and adjusting parameters ; Step 9-4: Constructing a logarithmic barrier function transforms the constrained convex optimization problem into an unconstrained optimization model. : in, This represents the number of iterations in the logarithmic barrier function phase. Represents the logarithmic barrier function stage. The power vector of the next iteration, the logarithmic barrier function is defined as: , Representing constraints: ， ， ， ， ， ， ； Based on the optimal matching result obtained in step 8, the golden section algorithm is used to optimize the unconstrained model under constraints. The optimal power allocation between the HSU and LSU is obtained by solving the problem. Step 9-5, when Or the number of iterations of the logarithmic barrier function Then stop the iteration and let the iteration in step 9-2 stop. Equal to the current power vector Otherwise, adjust the precision of the logarithmic barrier function: Continue with step 9-4; when Alternatively, the number of iterations in the approximation stage can be replaced. If the result is positive, stop the iteration and output the current power vector, i.e., the optimal power allocation; otherwise, return to step 9-2 and let the result in step 9-2 be positive. Equal to the current power vector .

4. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the resource allocation method based on the OTFS-ISAC system as described in any one of claims 1-3.

5. A computer program product, comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the resource allocation method based on the OTFS-ISAC system as described in any one of claims 1-3.

Images