Resource allocation and service value optimization method based on matching theory algorithm

Abstract

The invention provides a resource allocation and service value optimization method based on a matching theory algorithm. The method comprises the steps of S1, defining a matching matrix and an allocation matrix; s2, calculating a communication service value for each vehicle user; s3, calculating a perception service value for each vehicle user; s4, performing weighted summation on the communication service value and the sensing service value to obtain a total service value, constructing an original problem of a target function for maximizing the evaluation based on the total service value, and configuring a limiting condition for the original problem of the target function; and S5, decoupling the original problem of the target function into a vehicle user-base station association problem and a sub-channel allocation problem according to the variables of the matching matrix and the allocation matrix, and solving the vehicle user-base station association problem and the sub-channel allocation problem based on each limiting condition so as to optimize the total service value to be maximized. The method has the beneficial effects that sensing and communication performances are balanced at the same time, and the overall performance of the system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle networking communication and perception, and in particular to a resource allocation and service value optimization method based on a matching theory algorithm. Background Art

[0002] With the popularization of 5G mobile communication systems and the rapid development of Internet of Things (IoT) and Artificial Intelligence (AI) technologies, intelligent transportation systems (ITS) are rapidly emerging and gradually changing the way people travel. As an important application scenario of 5G communication, the emerging application scenarios of intelligent transportation systems require high-precision perception capabilities and high-speed data communication services under complex weather conditions.

[0003] Therefore, the base station needs to balance the perception and communication performance of vehicle users under limited wireless resources. To achieve this goal, researchers have proposed an Integrated Sensing and Communication (ISAC) system that allows simultaneous communication and perception on a unified wireless platform. The ISAC system is promising in the next generation of wireless communication networks. Radar and communication functions can be performed simultaneously on a single platform in a public frequency band. In recent years, it has inspired extensive research in industry and academia, such as the development of electronic warfare and intelligent transportation systems.

[0004] In the ISAC system, waveform design is crucial to improving radar and communication performance. So far, there are two main categories of ISAC waveform design, namely, multiplexed waveforms and homogeneous waveforms. For multiplexed waveforms, although radar and communication waveforms are multiplexed, the resource utilization efficiency is low. In contrast, for homogeneous waveforms, radar and communication functions share wireless resources, and the resource utilization efficiency is high. Therefore, joint allocation of resources for the dual functions can be achieved.

[0005] The hardware platform and radio resources need to be shared during the communication and perception processes. The perception process and data transmission are carried out simultaneously on the RF front end, which means that perception and communication are competing for resources to improve their own performance. Without a unified design goal, two independent resource allocation goals will lead to a one-sided emphasis on perception accuracy or communication data rate, thereby reducing the performance of the entire system. In addition, once the needs of vehicle users are met, allocating additional resources to achieve higher data transmission rates or perception accuracy will greatly reduce system performance. Therefore, there is an urgent need for a comprehensive evaluation indicator to balance the performance of perception and communication according to the needs of different users. Finally, designing a resource allocation algorithm based on performance indicators that evaluate dual functions is also challenging. Summary of the invention

[0006] The technical problem to be solved by the present invention is to simultaneously balance the performance of perception and communication and improve the overall performance of the system. In order to overcome the defects of the above-mentioned prior art (or related technology), the present invention provides a resource allocation and service value optimization method based on a matching theory algorithm.

[0007] The present invention provides a resource allocation and service value optimization method based on a matching theory algorithm, which pre-configures a group of base stations for a two-lane one-way highway scenario. All base stations m share the same orthogonal subchannel The vehicle users in the one-way highway scenario are The resource allocation and service value optimization method comprises the following steps:

[0008] Step S1, defining a matching matrix for characterizing the association between a vehicle user i and a base station m and an allocation matrix for characterizing the allocation of a subchannel v;

[0009] Step S2, for each of the vehicle users i, when the vehicle user i sends a task requirement to be associated with one of the base stations and the orthogonal subchannel is allocated to the vehicle user i, the communication service value of the vehicle user i is obtained according to the matching matrix, the allocation matrix and the interaction data of the process of associating the vehicle user i with the base station;

[0010] Step S3, for each of the vehicle users i, obtaining an OFDM integrated waveform generated when the vehicle user i performs radar detection on the vehicle user ahead and an integrated symbol of the OFDM integrated waveform on the orthogonal subchannel, obtaining an impulse response of the vehicle user i on the orthogonal subchannel and a received reflected signal, and obtaining the perceived service value of the vehicle user i according to the integrated symbol, the impulse response and the reflected signal;

[0011] Step S4, performing weighted summation of the communication service value and the perception service value to obtain a total service value, constructing an original objective function problem for maximizing the evaluation based on the total service value, and configuring a plurality of limiting conditions for the original objective function problem;

[0012] Step S5, decoupling the original objective function problem into a vehicle user-base station association problem and a sub-channel allocation problem according to the variables of the matching matrix and the variables of the allocation matrix, and solving the vehicle user-base station association problem and the sub-channel allocation problem respectively based on the respective limiting conditions to optimize the total service value to maximum.

[0013] Compared with the prior art, the resource allocation and service value optimization method based on matching theory algorithm in this application has the following advantages:

[0014] In the present application, the matching matrix and the allocation matrix are defined through step S1, the communication service value is calculated through step S2, the perception service value is calculated through step S3, the total service value is calculated and the original objective function problem is constructed through step S4, and the original objective function problem is decoupled and solved through step S5. The service value is used as a performance indicator to guide the matching algorithm to allocate resources for communication and perception. The matching algorithm includes two parts: the vehicle user-base station association problem and the sub-channel allocation problem, which effectively reduces the complexity of solving the original objective function problem, and associates different base stations and different sub-channels with different vehicle users. This differentiated resource allocation strategy helps to improve the overall performance of the system.

[0015] In a possible implementation manner, in step S1, Y is defined as [y i,m ] I×M As the matching matrix, define X = [x v,i ] V×I is the allocation matrix. When the vehicle user i is associated with the base station m, then y i,m =1, and zero if not associated; when the orthogonal subchannel v is allocated to the vehicle user i, then x v,i =1 if not assigned and zero if not assigned.

[0016] In a possible implementation, step S2 includes:

[0017] Step S21, for each of the vehicle users i, when the vehicle user i sends a task requirement to be associated with one of the base stations m and the orthogonal subchannel v is allocated to the vehicle user i, based on the Shannon theorem, the transmission rate generated by the vehicle user i is obtained according to the matching matrix, the allocation matrix and the interaction data:

[0018]

[0019] in, represents the bandwidth allocated by base station m to vehicle user i on subchannel v; i，m represents the matching matrix of the vehicle user i; x v，i represents the allocation matrix of the vehicle user i; i represents the power when the vehicle user i sends the task requirement; represents the small-scale Rayleigh fading coefficient between the base station m and the vehicle user i when sending the task requirement, which obeys the complex Gaussian distribution CN(0,1); d i,m represents the distance between the vehicle user i and the base station m; α represents the large-scale fading factor; represents the power of additive Gaussian white noise; i′ represents the interfering vehicle; x v，i′ represents the allocation matrix of the interfering vehicle i′; Represents the gain of interference signal transmission; P i′ Indicates the transmission power of the interference echo;

[0020] Step S22, calculating the transmission rate that the vehicle user i can achieve:

[0021]

[0022] Step S23, calculating the communication service value of the vehicle user i:

[0023]

[0024] Among them, VC i Indicates the communication service value; Indicates the preset threshold value of communication service value.

[0025] In a possible implementation, step S3 includes:

[0026] Step S31, for each vehicle user i, obtaining an OFDM integrated waveform generated when the vehicle user i performs radar detection on the vehicle user in front and K consecutive integrated symbols of the OFDM integrated waveform on the orthogonal subchannel v:

[0027]

[0028] in, Represents the integrated symbol; f c represents the center frequency of the orthogonal subchannel v; T0 represents the duration of a single integrated symbol with a periodic prefix, T0 = T + T g , T = 1 / Δf represents the duration of the basic code element of the integrated symbol, Tg represents a cyclic prefix; represents the amplitude of the waveform of the vehicle user i on the orthogonal sub-channel v; represents the phase encoding of the modulation symbol τ of the vehicle user i on the orthogonal subchannel v; Δf represents the subcarrier spacing of the duration of the OFDM symbol T; rect[x] represents a rectangular function, and its value is 1 when 0≤x≤1, otherwise it is 0; t represents the current time;

[0029] Step S32, setting the impulse response of the vehicle user i on the orthogonal subchannel v is a Gaussian random process, and the reflected signal received by the vehicle user i on the orthogonal sub-channel v is expressed as:

[0030]

[0031] Among them, y i (t) represents the reflected signal; ψ is the intermediate variable of time t; d is the differential symbol; z i (t) represents additive white Gaussian noise;

[0032] Step S33, for the detection and evaluation of radar targets, conditional MI is used to evaluate the estimation accuracy of the target impulse response. When the orthogonal subchannel v of the base station m is allocated to the vehicle user i, the conditional mutual information of the vehicle user i on the orthogonal subchannel v is:

[0033]

[0034] Among them, T p =KT0 represents the total duration of the OFDM integrated waveform; K represents the number of OFDM symbols; express Fourier transform of P i represents the transmission power of the vehicle user i; i′ represents the interfering vehicle; P i′ Indicates the transmission power of the interference echo; Indicates the gain of the interference echo; represents the power of additive white Gaussian noise;

[0035] Step S34, within the total duration of the OFDM integrated waveform, the average mutual information that the vehicle user i can achieve is:

[0036]

[0037] in, represents the average mutual information;

[0038] Step S34, calculating the perceived service value of the vehicle user i:

[0039]

[0040] Among them, VS i represents said perceived service value; Indicates the preset threshold of perceived service value.

[0041] In a possible implementation manner, in step S4, the total service value is obtained by the following calculation formula:

[0042]

[0043] Wherein, VoS(Y, X) represents the total service value; α i The weight representing the value of communication services; VC i Indicates the value of the communication service; β i Represents the perceived service value weight; VS i represents the perceived service value; α i +β i =1.

[0044] In a possible implementation, in step S4, the objective function original problem is represented by the following expression:

[0045]

[0046] In a possible implementation, in step S4, each of the limiting conditions is represented by the following expression:

[0047]

[0048]

[0049]

[0050]

[0051]

[0052]

[0053]

[0054] Among them, C1 and C2 are used to limit the minimum perception performance and communication performance of each vehicle; C3 and C4 are used to limit that each vehicle user i can only match one orthogonal subchannel v; C5 and C6 are used to limit that each vehicle user i can only match one base station m; C7 is used to limit that a maximum of I can be associated within the communication range of the base station m. m Car.

[0055] In a possible implementation, step S5 includes:

[0056] Step S51, decoupling the original objective function problem into the vehicle user-base station association problem and the sub-channel allocation problem according to the variables of the matching matrix and the variables of the allocation matrix;

[0057] Step S52, solving the vehicle user-base station association problem based on the limiting conditions C2, C5, C6 and C7 to optimize Y;

[0058] Step S53, solving the subchannel allocation problem based on the limiting conditions C1, C2, C3 and C4 to optimize X;

[0059] Step S54, setting a maximum number of iterations, repeatedly executing the steps S52 and S53 to optimize X and Y, and stopping the optimization after reaching the maximum number of iterations.

[0060] In a possible implementation, step S52 includes:

[0061] Step S521, solving the vehicle user-base station association problem, the solution formula is expressed as:

[0062]

[0063] st:C2,C5,C6,C7

[0064] From the perspective of the base station, define E m is the set of users accessing the mth base station, then P1.1 can be equivalently rewritten as:

[0065]

[0066]

[0067]

[0068] C7-1:|E m |≤I m ;

[0069] in, Transform the optimization of Y into optimization of Em Optimization;

[0070] Step S522, define a base station set of M base stations as B m represents the mth base station; similarly, the vehicle user set of I vehicle users is defined as V i represents the i-th vehicle user, and matches the vehicle user set and the base station set through a many-to-one matching function model, and the expression of the many-to-one matching function model is:

[0071] M(x){V1,V2,...V i ...,V I}∪{B1,B2,...B m ...,B M},x{V1,V2,...V i ...,V I}∪{B1,B2,...B m ...,B M}

[0072] in,

[0073] And |M(V i )|=1;

[0074] and|M(B m )|≤I m ;

[0075] M(V i )=B m , if and only if V i ∈M(B m );

[0076] Step S523: for the vehicle user V i , which is about any of the base stations B m The preference degree is defined as when the vehicle user V i Access the base station B m The base station B m The performance gain is expressed as:

[0077] G i,m =O m (E′ m ∪i)-O m (E′ m )

[0078] Among them, E′ m In addition to V iOther access base stations B m The set of vehicle users;

[0079] Step S524: each of the base stations B m Initialize the set of vehicle users that access itself: And each vehicle user V i Calculate its own m The preference of the available base station set can be initialized as and not connected to base station B m The vehicle user set is

[0080] Step S525: each non-connected base station B m Vehicle users V i Calculate the value of itself and the base station B m The preference between them is the highest and the available base station B with the highest preference is m Make an access request;

[0081] Step S526: Based on the received access request, each base station B m Accept the vehicle user V with the highest preference i And add it to the review set and reject other vehicle users V i At the same time, the base station B m Accessed vehicle users V i The number increases by 1. If a base station B m Access vehicle user V i If the number reaches the upper limit, the base station B is added to the available base station set. m Remove;

[0082] Step S527, each rejected vehicle user V i The preference is recalculated and updated, and the process returns to step S525. When the remaining vehicle user set becomes an empty set, the iteration is terminated to obtain the optimal allocation.

[0083] In a possible implementation, step S53 includes:

[0084] Step S531, solving the sub-channel allocation problem, the solution formula is expressed as:

[0085]

[0086] stC 1~C4

[0087] From the perspective of orthogonal subchannels, define A v is the set of users served by the vth orthogonal subchannel, then the problem Can be rewritten equivalently as:

[0088]

[0089] stC1-C2

[0090]

[0091] in, Transform the optimization of X into A v Optimization;

[0092] Step S532, define a vehicle user set of I vehicle users as u i = {u1, u2, ..., u I},u i represents the i-th vehicle user; similarly, the orthogonal subchannel set of V orthogonal subchannels is defined as c v ={c1, c2, ..., c V},c v represents the Vth orthogonal subchannel, and the vehicle user set and the orthogonal subchannel set are matched by a many-to-one matching function model, and the expression of the many-to-one matching function model is:

[0093]

[0094] in,

[0095] M(u i )∈{c1,c2,...,c V}, and |M(u i )|=1;

[0096] M(c v )∈{u1,u2,...,u I};

[0097] M(u i )=c v If and only if u i ∈M(c v );

[0098] Step S533: for the vehicle user u i , which is related to any of the orthogonal subchannels c v The preference degree of the vehicle user u is defined as i Access the orthogonal subchannel c v When the orthogonal subchannel c v The performance gain is expressed as:

[0099] G i,v =O v (Av ′∪u i )-O ν (A′ ν )

[0100] Among them, A v ′ means that except for vehicle user u i Other access orthogonal subchannels c v The set of vehicle users;

[0101] Step S534: each of the orthogonal sub-channels c v Initialize the vehicle user u that accesses itself i , each of the vehicle users u i Based on the performance and constraints, it calculates the c for all the orthogonal sub-channels v The preference of the orthogonal subchannel set is The set of vehicle users that do not access the orthogonal subchannel is initialized as

[0102] Step S535: each vehicle user u that is not connected to the orthogonal sub-channel i Towards the orthogonal subchannel c with the largest preference v Make an allocation request;

[0103] Step S536: each orthogonal subchannel c v Upon receiving the vehicle user u i The access request allows the vehicle user u with a high preference degree i Access and reject other vehicle users u i After each successful access, the vehicle user set that has not accessed will assign the successfully allocated vehicle user u i Remove, if the vehicle user set is an empty set, the loop ends;

[0104] Step S537: Based on the latest access and pairing results, each non-accessed vehicle user recalculates the orthogonal sub-channel c v preference, and returns to step S535. BRIEF DESCRIPTION OF THE DRAWINGS

[0105] Figure 1 A schematic diagram of a one-way highway scenario of the present invention;

[0106] Figure 2 is a flow chart of the steps of the present invention;

[0107] Figure 3 This is a schematic diagram of the total service value under different numbers of vehicle users in Example 1 of the present invention. DETAILED DESCRIPTION

[0108] First, those skilled in the art should understand that these implementations are only used to explain the technical principles of the embodiments of the present application, and are not intended to limit the protection scope of the embodiments of the present application. Those skilled in the art can make adjustments to them as needed to adapt to specific application scenarios.

[0109] The present application is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0110] See also Figure 2 The present application embodiment discloses a resource allocation and service value optimization method based on a matching theory algorithm, including:

[0111] Step S1, defining a matching matrix for characterizing the association between a vehicle user i and a base station m and an allocation matrix for characterizing the allocation of a subchannel v;

[0112] Step S2, for each vehicle user i, when the vehicle user i sends a task requirement to be associated with one of the base stations and the orthogonal subchannel is allocated to the vehicle user i, the communication service value of the vehicle user i is obtained according to the matching matrix, the allocation matrix and the interaction data of the process of associating the vehicle user i with the base station;

[0113] Step S3, for each vehicle user i, obtaining the OFDM integrated waveform generated when the vehicle user i performs radar detection on the vehicle user in front and the integrated symbol of the OFDM integrated waveform on the orthogonal subchannel, obtaining the impulse response of the vehicle user i on the orthogonal subchannel and the received reflected signal, and obtaining the perceived service value of the vehicle user i according to the integrated symbol, the impulse response and the reflected signal;

[0114] Step S4, performing weighted summation of the communication service value and the perception service value to obtain a total service value, constructing an original objective function problem for maximizing the evaluation based on the total service value, and configuring multiple limiting conditions for the original objective function problem;

[0115] Step S5, decouple the original objective function problem into a vehicle user-base station association problem and a sub-channel allocation problem according to the variables of the matching matrix and the variables of the allocation matrix, and solve the vehicle user-base station association problem and the sub-channel allocation problem respectively based on various limiting conditions to optimize the total service value to the maximum.

[0116] Continue to see Figure 2 In this application, the performance of perception and communication is weighed at the same time, and the overall performance of the system is improved. The scenarios considered are as follows Figure 1 The picture shows a two-lane one-way highway scene, where vehicles are driving from right to left. The scene contains a group of base stations. All base stations share the same orthogonal subchannels Vehicle user devices (VUEs) Assume that all VUEs have ISAC technology; VUEs can use ISAC signals to sense the vehicle in front, and can also communicate with the base station (vehicle-to-infrastructure, V2I). Considering that each VUE is associated with a computing task, VUEs need to use various sensors (including radar) to sense the surrounding vehicle conditions and environment, and upload the sensed data by wireless communication; in particular, this application also defines the process of vehicle user i to base station m as a service request, and establishes communication service value and perception service value based on communication and perception respectively, and finally integrates them into the total service value (VoS, value of service) of the system. The ultimate goal is to maximize the total service value of the system.

[0117] In the calculation process of communication service value, define Y = [y i,m ] I×M is the matching matrix. If vehicle user i is associated with base station m, then y i,m =1, otherwise zero; define X = [x v,i ] V×I is the allocation matrix. If the orthogonal subchannel v is allocated to vehicle user i, then x v,i =1; otherwise zero;

[0118] The system considered in this application is a wireless transmission system based on OFDM (Orthogonal Frequency Division Multiplexing). When vehicle user i is associated with base station m, the orthogonal subchannel v is allocated to vehicle user i. According to Shannon's theorem, the resulting transmission rate is:

[0119]

[0120] Among them, P i is the power when vehicle user i sends the task requirement; represents the small-scale Rayleigh fading coefficient between the base station m and the vehicle user i when sending the task demand, which obeys the complex Gaussian distribution CN(0,1); d i,m represents the distance between vehicle user i and base station m; α is the large-scale fading factor; is the power of additive white Gaussian noise; is the gain of interference signal transmission, where VUE i can achieve a transmission rate of:

[0121]

[0122] In order to describe the communication process of vehicle user i, the communication service value of vehicle user i is defined as:

[0123]

[0124] In the process of calculating the perceived service value, each vehicle user VUEi in the present application can radiate an OFDM integrated waveform through its dual-function transmitter. The integrated synaesthesia integrated waveform can perform radar detection on the vehicle user i in front. On the orthogonal subchannel v, the waveform used by VUEi has K consecutive integrated symbols, which can be described as:

[0125]

[0126] Where T = 1 / Δf is the duration of the basic code element of the OFDM symbol; f c is the center frequency of the orthogonal subchannel v; T0 is the duration of a single OFDM symbol with a cyclic prefix, that is, the time length of a complete OFDM symbol; specifically, T0 = T + T g , T g is a cyclic prefix (CP); represents the amplitude of the waveform of VUEi on the orthogonal subchannel v; represents the phase encoding of the modulation symbol τ of VUEi on the orthogonal subchannel v; rect[x] is a rectangular function, and its value is 1 when 0≤x≤1, otherwise it is 0;

[0127] Assume that the impulse response of VUE i on the orthogonal subchannel v is is a Gaussian random process, then In orthogonal subchannels The reflected signal received on can be expressed as:

[0128]

[0129] The received echo signal is subjected to additive white Gaussian noise z i (t) impact;

[0130] For the detection and evaluation of radar targets, conditional MI can be used to evaluate the estimation accuracy of the target impulse response. In the process of perception, the orthogonal subchannel is shared with the communication, so only the allocation process of the orthogonal subchannel needs to be considered. When the orthogonal subchannel of the base station is allocated to the vehicle user, the conditional mutual information of the vehicle user on the subchannel orthogonality is:

[0131]

[0132] Where T p =KT0 represents the total duration of the OFDM signal; yes Fourier transform of P i is the transmission power of vehicle user i; the interference source is the interference of radar signal, i′ represents the interfering vehicle; P i′ is the transmission power of the interference echo; is the gain of the interference echo; is the power of additive white Gaussian noise; the average mutual information that vehicle user i can achieve during the total duration of the OFDM signal is:

[0133]

[0134] Similar to the communication service value, the perceived service value can also be defined as:

[0135]

[0136] In the calculation process of the total service value, the total service value is obtained by weighted summation of the communication service value and the perception service value, so the total service value of the system is:

[0137]

[0138] The objective function of the original problem is:

[0139]

[0140] To ensure the minimum perception and communication performance of each vehicle:

[0141]

[0142]

[0143] Each vehicle user can only match one sub-channel:

[0144]

[0145]

[0146] Each vehicle user can only match one base station:

[0147]

[0148]

[0149] Within the communication range of base station m, at most I m Vehicles:

[0150]

[0151] In this application, the service value is used as a performance indicator to guide the matching algorithm to allocate resources for communication and perception. The matching algorithm includes two parts: vehicle user-base station association and subchannel allocation. It has the following advantages:

[0152] 1) In the stage of measuring the performance of communication and perception, this application introduces the Value of Service (VoS) as a criterion for resource allocation;

[0153] 2) In the resource allocation stage, different base stations and sub-channels are associated with different vehicle users according to their preferences after access. This differentiated resource allocation strategy helps improve the overall performance of the system.

[0154] 3) In addition, the present application decouples the original problem into two sub-problems based on two variables, which effectively reduces the complexity of solving the original problem.

[0155] The following is a detailed description of the process of solving the original problem:

[0156] 1. First, the original problem is decoupled into the vehicle user-base station association problem and the sub-channel allocation problem according to the matching matrix variables and the allocation matrix variables.

[0157] 2. Solve the first sub-problem "vehicle user-base station association problem", the formula is expressed as:

[0158]

[0159] st:C2,C5,C6,C7

[0160] From the perspective of the base station, define E m is the set of users accessing the mth base station, then Equivalently rewritten as:

[0161]

[0162]

[0163]

[0164] C7-1:|E m |≤I m ;

[0165] in Therefore, the original user matching problem is transformed from optimizing Y to optimizing E. m Optimization, in this application, the set of M base stations is defined as B m represents the mth base station; similarly, the set of I users is defined as Vi represents the i-th vehicle user;

[0166] Definition 1 (many-to-one matching model): Define V i is the i-th vehicle user, B m is the mth base station, then the matching between the vehicle user set and the base station set involves a many-to-one matching function model:

[0167] M(x){V1,V2,...V i ...,V I}∪{B1,B2,...B m ...,B M},x{V1,V2,...V i ...,V I}∪{B1,B2,...B m ...,B M}

[0168] The many-to-one matching function model is defined as follows:

[0169] And |M(V i )|=1;

[0170] and|M(B m )|≤I m ;

[0171] M(V i )=B m , if and only if V i ∈M(B m );

[0172] Definition 2 (Preference): For vehicle user V i , which is about any base station B m The preference of vehicle user V is defined as i Access base station B m The gain brought to the performance of the base station when

[0173] G i,m =O m (E′ m ∪i)-O m (E′ m )

[0174] Where E′ m Except for V i Other access base stations B m A collection of users;

[0175] Then proceed as follows:

[0176] The first step is to initialize. Each base station initializes the user set it accesses to And each vehicle user calculates its preference for the base station (based on performance and whether the constraints are met), and the set of available base stations is initialized as The set of vehicle users that are not connected to the base station is

[0177] In the second step, each vehicle user that has not accessed the base station calculates the preference between itself and the base station, and makes an access request to the available base station with the largest preference value;

[0178] In the third step, based on the requests received, each base station accepts the user with the highest preference for itself and adds it to the review set, and "rejects" the access applications of other users. At the same time, the number of vehicle users accessing the base station is increased by 1. If the number of users accessing a base station reaches the upper limit, the base station is removed from the set of available base stations.

[0179] In the fourth step, each "rejected" vehicle user recalculates and updates the preference based on the above formula and returns to the second step. When the remaining vehicle user set becomes an empty set, the iteration ends and the optimal allocation is obtained.

[0180] 3. Solve the second sub-problem "sub-channel allocation problem", the formula is expressed as:

[0181]

[0182] stC1~C4

[0183] From the perspective of the channel, define A v is the set of users served by the vth orthogonal subchannel, then the problem Can be rewritten equivalently as:

[0184]

[0185] stC1-C2

[0186]

[0187] in The overall system performance is calculated from the channel perspective, while It is calculated from the perspective of the vehicle user;

[0188] Definition 1 (many-to-one allocation model): The matching between the vehicle user set and the channel set involves a many-to-one matching function model:

[0189]

[0190] The many-to-one matching function model is defined as follows:

[0191] M(u i )∈{c1,c2,…,c V}, and |M(u i )|=1;

[0192] M(c v )∈{u1,u2,…,u I};

[0193] M(u i )=c v If and only if u i ∈M(c v );

[0194] Definition 2 (Preference): For vehicle user u i , its preference for any channel is defined as i Access c v The performance gain brought to the channel is:

[0195] G i,v =O v (A v ′∪u i )-O ν (A ν ′)

[0196] Among them A v ′ is except u i The set of users accessing orthogonal subchannel v other than ;

[0197] Then proceed as follows:

[0198] The first step is initialization: each channel initializes its own access user A v Each user calculates his preference for all channels based on performance and constraints. The set of channels is The set of users who have not accessed the channel is initialized to

[0199] In the second step, each user who has not accessed a channel makes an allocation request to the channel with the highest preference;

[0200] In the third step, each channel allows users with a high preference degree to access and rejects other users in the received user access request. Each time the access is successfully allowed, the user set that has not accessed removes the successfully allocated user. If the user set is an empty set, the loop ends.

[0201] Step 4: Loop: Based on the latest access and pairing results, each non-access user recalculates the preference for the channel, returns to step 2 and continues.

[0202] 4. Set a maximum number of iterations, optimize X and Y alternately, and calculate the optimal value of the corresponding optimization variable in each iteration. After reaching the maximum number of iterations, the optimization stops.

[0203] Example 1

[0204] For the communication model, it is assumed that the small-scale fading channel gain for communication follows unit-mean Rayleigh fading, and the uplink path loss is determined by the large-scale loss factor α = 2 and the distance d between the vehicle user and its associated BS. i,m Calculation,For the sensing model, it is assumed that the target frequency response follows a standard normal distribution, and other parameter settings are shown in Table 1 below:

[0205] Table 1 Simulation parameters

[0206]

[0207] This application compares the proposed (Matching algorithm of joint base station-UserAssociation and subchannel allocation, MUA) algorithm with the following three algorithms:

[0208] 1) MUA-FP (Matching algorithm ofjoint base station and UserAssociation-Fractional Programming): This method still uses the matching theory method to optimize the first subproblem, i.e., the base station-vehicle user pairing problem, while the fractional programming algorithm is used for the second subproblem, i.e., the subchannel allocation problem;

[0209] 2) IMUA-FP (Improved Matching algorithm of joint base station and UserAssociation-Fractional Programming): Compared with MUA-FP, this method introduces the Hungarian algorithm to directly solve the first sub-problem based on the original matching theory. This can reduce the number of iterations of the simulation process, but at the cost of poor performance. The second sub-problem still uses the fractional programming algorithm;

[0210] 3) GA (Genetic Algorithm): First, by encoding the user matching variable Y and the channel allocation variable X into the chromosome and defining the corresponding population and fitness value, natural selection, mating, mutation and exploration can be used to intelligently search for solutions. In this embodiment, the population size is defined as 100, the maximum number of evolutions is 300, the crossover probability and mutation probability are 0.0001-0.1 respectively, and the genetic algorithm toolbox of MATLAB is used for solving;

[0211] The total service value of the proposed solution in this application under different numbers of vehicle users is evaluated in the ISAC system, such as Figure 3 As shown in the figure, when the number of VUEs is less than 7, the four schemes all show a high VoS. When the number of VUEs is greater than 7, all schemes except MUA show a significant decline in VoS. When the number of VUEs exceeds 24, the VoS of MUA-FP, IMUA-FP and GA all drop below 0.7, while the service value of MUA remains above 0.85.

[0212] Because MUA can well consider the competition relationship between user devices related to the same resource (base station / channel) in both channel allocation and user-base station matching, while the other three methods cannot fully consider access relevance, the MUA mechanism can achieve the highest performance; GA can obtain higher service value because it can obtain results similar to brute force search through iterative evolution, continuous preservation of better variables and abandonment of overlapping variable selections; and the performance of the MUA-FP mechanism is slightly higher than that of the IMUA-FP mechanism. This is because for IMUA, in order to ensure that each new user can be associated with a base station in each iteration, some higher user-base station pairs can be prioritized.

[0213] The above experimental results show that MUA can effectively improve network performance and optimize resource utilization when dealing with complex network conditions, thereby providing strong support for the realization of intelligent transportation systems.

[0214] In the description of the present application, the description with reference to the terms "one embodiment", "some embodiments", "in the present embodiment", "specific example", or "some examples" etc. means that the specific features, mechanisms, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, mechanisms, materials or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0215] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A resource allocation and service value optimization method based on matching theory algorithm, characterized in that: Pre-configure a set of base stations for two-lane one-way highway scenarios All base stations m share the same orthogonal subchannel The vehicle users in the one-way highway scenario are The resource allocation and service value optimization method comprises the following steps: Step S1, defining a matching matrix for characterizing the association between a vehicle user i and a base station m and an allocation matrix for characterizing the allocation of a subchannel v; Step S2, for each of the vehicle users i, when the vehicle user i sends a task requirement to be associated with one of the base stations and the orthogonal subchannel is allocated to the vehicle user i, the communication service value of the vehicle user i is obtained according to the matching matrix, the allocation matrix and the interaction data of the process of associating the vehicle user i with the base station; Step S3, for each of the vehicle users i, obtaining an OFDM integrated waveform generated when the vehicle user i performs radar detection on the vehicle user ahead and an integrated symbol of the OFDM integrated waveform on the orthogonal subchannel, obtaining an impulse response of the vehicle user i on the orthogonal subchannel and a received reflected signal, and obtaining the perceived service value of the vehicle user i according to the integrated symbol, the impulse response and the reflected signal; Step S4, performing weighted summation of the communication service value and the perception service value to obtain a total service value, constructing an original objective function problem for maximizing the evaluation based on the total service value, and configuring a plurality of limiting conditions for the original objective function problem; Step S5, decoupling the original objective function problem into a vehicle user-base station association problem and a sub-channel allocation problem according to the variables of the matching matrix and the variables of the allocation matrix, and solving the vehicle user-base station association problem and the sub-channel allocation problem respectively based on the respective limiting conditions to optimize the total service value to maximum.

2. The resource allocation and service value optimization method according to claim 1, characterized in that: In step S1, define Y=[y i,m ] I×M As the matching matrix, define X = [x v,i ] V×I is the allocation matrix. When the vehicle user i is associated with the base station m, then y i,m =1, and zero if not associated; when the orthogonal subchannel v is allocated to the vehicle user i, then x v,i =1 if not assigned and zero if not assigned.

3. The resource allocation and service value optimization method according to claim 1, characterized in that: The step S2 comprises: Step S21, for each of the vehicle users i, when the vehicle user i sends a task requirement to be associated with one of the base stations m and the orthogonal subchannel v is allocated to the vehicle user i, based on the Shannon theorem, the transmission rate generated by the vehicle user i is obtained according to the matching matrix, the allocation matrix and the interaction data: in, represents the bandwidth allocated by base station m to vehicle user i on orthogonal subchannel v; i，m represents the matching matrix of the vehicle user i; x v，i represents the allocation matrix of the vehicle user i; i represents the power when the vehicle user i sends the task requirement; represents the small-scale Rayleigh fading coefficient between the base station m and the vehicle user i when sending the task requirement, which obeys the complex Gaussian distribution CN(0,1); d i,m represents the distance between the vehicle user i and the base station m; α represents the large-scale fading factor; represents the power of additive Gaussian white noise; i′ represents the interfering vehicle; x v，i′ The allocation matrix representing the interfering vehicle i′; Represents the gain of interference signal transmission; P i′ Indicates the transmission power of the interference echo; Step S22, calculating the transmission rate that the vehicle user i can achieve: Step S23, calculating the communication service value of the vehicle user i: Among them, VC i Indicates the communication service value; Indicates the preset threshold value of communication service value.

4. The resource allocation and service value optimization method according to claim 1, characterized in that: The step S3 comprises: Step S31, for each vehicle user i, obtaining an OFDM integrated waveform generated when the vehicle user i performs radar detection on the vehicle user in front and K consecutive integrated symbols of the OFDM integrated waveform on the orthogonal subchannel v: in, Represents the integrated symbol; f c represents the center frequency of the orthogonal subchannel v; T0 represents the duration of a single integrated symbol with a periodic prefix, T0 = T + T g , T = 1 / Δf represents the duration of the basic code element of the integrated symbol, T g represents a cyclic prefix; represents the amplitude of the waveform of the vehicle user i on the orthogonal sub-channel v; represents the phase encoding of the modulation symbol τ of the vehicle user i on the orthogonal subchannel v; Δf represents the subcarrier spacing of the duration of the OFDM symbol T; rect[x] represents a rectangular function, and its value is 1 when 0≤x≤1, otherwise it is 0; t represents the current time; Step S32, setting the impulse response of the vehicle user i on the orthogonal subchannel v is a Gaussian random process, and the reflected signal received by the vehicle user i on the orthogonal sub-channel v is expressed as: Among them, y i (t) represents the reflected signal; ψ is the intermediate variable of time t; d is the differential symbol; z i (t) represents additive white Gaussian noise; Step S33, for the detection and evaluation of radar targets, conditional MI is used to evaluate the estimation accuracy of the target impulse response. When the orthogonal subchannel v of the base station m is allocated to the vehicle user i, the conditional mutual information of the vehicle user i on the orthogonal subchannel v is: Among them, T p =KT0 represents the total duration of the OFDM integrated waveform; K represents the number of OFDM symbols; express Fourier transform of P i represents the transmission power of the vehicle user i; i′ represents the interfering vehicle; P i′ Indicates the transmission power of the interference echo; Indicates the gain of the interference echo; represents the power of additive white Gaussian noise; Step S34, within the total duration of the OFDM integrated waveform, the average mutual information that the vehicle user i can achieve is: in, represents the average mutual information; Step S34, calculating the perceived service value of the vehicle user i: Among them, VS i represents said perceived service value; Indicates the preset threshold of perceived service value.

5. The resource allocation and service value optimization method according to claim 1, characterized in that: In step S4, the total service value is obtained by the calculation formula: Wherein, VoS(Y, X) represents the total service value; α i The weight representing the value of communication services; VC i Indicates the value of the communication service; β i Represents the weight of perceived service value; VS i represents the perceived service value; α i +β i =1.

6. The resource allocation and service value optimization method according to claim 5, characterized in that: In step S4, the objective function original problem is expressed by the following expression:

7. The method for optimizing resource allocation and service value according to claim 6, characterized in that: In step S4, each of the limiting conditions is represented by the following expressions: Among them, C1 and C2 are used to limit the minimum perception performance and communication performance of each vehicle; C3 and C4 are used to limit that each vehicle user i can only match one orthogonal subchannel v; C5 and C6 are used to limit that each vehicle user i can only match one base station m; C7 is used to limit that a maximum of I can be associated within the communication range of the base station m. m Car.

8. The method for optimizing resource allocation and service value according to claim 7, characterized in that: The step S5 comprises: Step S51, decoupling the original objective function problem into the vehicle user-base station association problem and the sub-channel allocation problem according to the variables of the matching matrix and the variables of the allocation matrix; Step S52, solving the vehicle user-base station association problem based on the limiting conditions C2, C5, C6 and C7 to optimize Y; Step S53, solving the subchannel allocation problem based on the limiting conditions C1, C2, C3 and C4 to optimize X; Step S54, setting a maximum number of iterations, repeatedly executing the steps S52 and S53 to optimize X and Y, and stopping the optimization after reaching the maximum number of iterations.

9. The method for optimizing resource allocation and service value according to claim 8, characterized in that: The step S52 comprises: Step S521, solving the vehicle user-base station association problem, the solution formula is expressed as: st:C2,C5,C6,C7 From the perspective of the base station, define E m is the set of users accessing the mth base station, then Equivalently rewritten as: C7-1:|E m |≤I m ; in, Transform the optimization of Y into optimization of E m Optimization; Step S522, define a base station set of M base stations as B m represents the mth base station; similarly, the vehicle user set of I vehicle users is defined as V i represents the i-th vehicle user, and matches the vehicle user set and the base station set through a many-to-one matching function model, and the expression of the many-to-one matching function model is: M(x){V1,V2,...V i ...,V I }∪{B1,B2,...B m ...,B M },x{V1,V2,...V i ...,V I }∪{B1,B2,...B m ...,B M } in, And |M(V i )|=1; and|M(B m )|≤I m ; M(V i )=B m , if and only if V i ∈M(B m ); Step S523: for the vehicle user V i , which is about any of the base stations B m The preference degree is defined as when the vehicle user V i Access the base station B m The base station B m The performance gain is expressed as: G i,m =The m (AND' m ∪i)-O m (AND' m ) Among them, E′ m In addition to V i Other access base stations B m The set of vehicle users; Step S524: each of the base stations B m Initialize the set of vehicle users that access itself: And each vehicle user V i Calculate its own m The preference of the available base station set can be initialized as and not connected to base station B m The vehicle user set is Step S525: each non-connected base station B m Vehicle users V i Calculate itself and the base station B m The preference between them is the highest and the available base station B with the highest preference is m Make an access request; Step S526: Based on the received access request, each base station B m Accept the vehicle user V with the highest preference i And add it to the review set and reject other vehicle users V i At the same time, the base station B m Accessed vehicle users V i The number increases by 1. If a base station B m Access vehicle user V i If the number reaches the upper limit, the base station B is added to the available base station set. m Remove; Step S527, each rejected vehicle user V i The preference is recalculated and updated, and the process returns to step S525. When the remaining vehicle user set becomes an empty set, the iteration is terminated to obtain the optimal allocation.

10. The method for optimizing resource allocation and service value according to claim 8, characterized in that: The step S53 comprises: Step S531, solving the sub-channel allocation problem, the solution formula is expressed as: stC1~C4 From the perspective of orthogonal subchannels, define A v is the set of users served by the vth orthogonal subchannel, then the problem Can be rewritten equivalently as: stC1-C2 in, Transform the optimization of X into A v Optimization; Step S532, define a vehicle user set of I vehicle users as u i = {u1,u2,...,u I },u i represents the i-th vehicle user; similarly, the orthogonal subchannel set of V orthogonal subchannels is defined as c v ={c1, c2, ..., c V }, c v represents the Vth orthogonal subchannel, and the vehicle user set and the orthogonal subchannel set are matched by a many-to-one matching function model, and the expression of the many-to-one matching function model is: in, M(u i ) ∈ {c1, c2,..., c V}, and |M(u i )| = 1; M(c v )∈{u1,u2,...,u I }; M(u i )=c v If and only if u i ∈M(c v ); Step S533: for the vehicle user u i , which is related to any of the orthogonal subchannels c v The preference degree of the vehicle user u is defined as i Access the orthogonal subchannel c v When the orthogonal subchannel c v The performance gain is expressed as: G i ,grandmother v (THE v ′ ∪u i )-THE ν (THE ′ ν ) Among them, A v ′ Indicates that except for vehicle user u i Other access orthogonal subchannels c v The set of vehicle users; Step S534: each of the orthogonal sub-channels c v Initialize the vehicle user u that accesses itself i , each of the vehicle users u i Based on the performance and constraints, it calculates the c for all the orthogonal sub-channels v The preference degree of the orthogonal subchannels is V, and the set of vehicle users that are not connected to the orthogonal subchannels is initialized to u; Step S535: each vehicle user u that is not connected to the orthogonal sub-channel i Towards the orthogonal subchannel c with the largest preference v Make an allocation request; Step S536: each orthogonal subchannel c v Upon receiving the vehicle user u i The access request allows the vehicle user u with a high preference degree i Access and reject other vehicle users u i After each successful access, the vehicle user set that has not accessed will assign the successfully allocated vehicle user u i Remove, if the vehicle user set is an empty set, the loop ends; Step S537: Based on the latest access and pairing results, each non-accessed vehicle user recalculates the orthogonal sub-channel c v preference, and returns to step S535.

Images