A high dynamic channel communication and sensing multi-performance index dynamic compromise method

Abstract

The application discloses a kind of high dynamic channel under communication and sensing multi-performance index dynamic compromise method, it is related to wireless communication technical field.The method is based on target movement, geometric distance, size scale fading and Doppler frequency shift constructs high dynamic channel model;Spectrum efficiency, energy efficiency and speed sensing accuracy index are calculated and normalized;Channel aging factor is mapped as the discount factor of Rubinstein bargaining model, the weight of three kinds of indexes is determined by alternating offer, and a dynamic compromise weight distribution algorithm is constructed.Different from the existing quasi-static channel or fixed weight sensing optimization method, the application makes the weight of spectrum efficiency, energy efficiency and speed sensing accuracy index adaptively change with the degree of channel aging, realizes the dynamic coordination of communication rate, energy consumption and speed estimation error boundary, and provides basis for resource allocation in vehicle networking, low-altitude unmanned aerial vehicle and other scenarios.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a dynamic trade-off method for multiple performance indicators of communication and sensing under high dynamic channels. Background Technology

[0002] With the development of sixth-generation (6G) mobile communication systems, future wireless networks will break through traditional communication paradigms and evolve into ubiquitous intelligent information networks that deeply integrate communication, sensing, computing, and intelligence. Integrated sensing and communication (ISAC), by sharing spectrum resources and hardware platforms, can simultaneously achieve data transmission and proactive environmental perception, and is considered one of the key enabling technologies for 6G networks. Application scenarios such as connected vehicles, industrial IoT, drone networks, and low-altitude urban services place dual demands on efficient communication and accurate sensing. However, in these scenarios, targets such as ground vehicles and aerial drones typically exhibit high mobility, leading to significant time-varying characteristics in wireless channels. This causes Doppler shift, rapid channel fluctuations, and channel state information aging, which severely impacts communication links, parameter estimation, and resource allocation. Therefore, it is urgent to conduct performance index modeling and resource optimization research for 6G integrated sensing and communication networks under high-dynamic channel conditions to achieve synergistic improvements in spectrum efficiency, energy efficiency, and speed sensing accuracy.

[0003] Existing research on integrated sensing and communication networks has made some progress in dynamic channel modeling, channel aging suppression, waveform design, and resource allocation. For example, existing studies have used methods such as time-varying geometric parameters, cluster generation and extinction processes, target scattering models, and three-dimensional ray tracing to characterize the non-stationary characteristics of highly dynamic ISAC channels. Meanwhile, other studies have utilized methods such as channel prediction, reconfigurable smart surface-assisted transmission, and resource optimization to mitigate the impact of channel aging and balance communication and sensing performance. However, most existing studies only analyze a single indicator and lack a unified framework that can simultaneously describe the coupling relationship between highly dynamic channels, communication performance, and sensing performance. Furthermore, traditional resource allocation methods are usually based on quasi-static channel assumptions or fixed-weight strategies, making it difficult to adjust communication and sensing resources in real time according to Doppler shift and channel aging, thus hindering their adaptability to rapidly changing network environments.

[0004] In integrated sensing and communication networks (ISACs), communication and sensing functions share limited resources, leading to inherent competition between them in terms of performance objectives. Communication performance typically focuses on spectral efficiency and energy efficiency, while sensing performance relies more on metrics such as the Cramer-Rao bound. These metrics have different dimensions and optimization directions, and they dynamically change with the target's motion state and channel aging under highly dynamic channels, making direct joint optimization highly complex. Existing research lacks a multi-performance evaluation and dynamic trade-off mechanism for highly dynamic ISACs, resulting in a lack of comprehensive consideration of channel time-varying characteristics, communication-sensing coupling relationships, and the priority changes of heterogeneous performance metrics in system resource optimization. Therefore, it is necessary to construct a unified high-dynamic ISAC performance metric trade-off framework to achieve adaptive trade-offs and balances among spectral efficiency, energy efficiency, and velocity sensing accuracy metrics, while considering target motion and channel aging. Summary of the Invention

[0005] The technical problem this invention aims to solve is to overcome the shortcomings of existing technologies and provide a dynamic trade-off method for multiple performance indicators of communication and sensing under highly dynamic channels. This method addresses the Doppler frequency shift and channel state information aging problems caused by target motion in highly dynamic scenarios by establishing a dynamic trade-off relationship among three performance indicators: spectral efficiency, energy efficiency, and speed sensing accuracy. Furthermore, it maps the channel aging factor to a discount factor in the Rubinstein bargaining model, and adaptively determines the weight allocation results of spectral efficiency, energy efficiency, and speed sensing accuracy indicators based on the time-varying characteristics of different communication and sensing links. This method overcomes the problem that existing fixed-weight strategies are difficult to adapt to highly dynamic channel changes and can provide a theoretical basis for performance evaluation and resource allocation of 6G integrated communication and sensing networks under different mobility conditions.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0007] A dynamic trade-off method for multiple performance indicators of communication and sensing under high dynamic channel conditions, proposed according to the present invention, includes:

[0008] Step S1: Establish an integrated communication and sensing network model, determine the base station, communication target and sensing target in the integrated communication and sensing network model, and construct a high dynamic channel model that considers target motion and Doppler frequency shift based on the integrated communication and sensing network model;

[0009] Step S2: Based on the high dynamic channel model, calculate the spectral efficiency, energy efficiency, and speed sensing accuracy indicators respectively;

[0010] Step S3: Introduce the channel aging factor into the index trade-off mechanism to construct a dynamic weight allocation method for multiple performance index trade-offs of spectral efficiency, energy efficiency and speed sensing accuracy.

[0011] As a further optimization scheme for the dynamic trade-off method of multiple performance indicators of communication and sensing under high dynamic channel described in this invention, in step S1, the integrated communication and sensing network model includes base stations, UAVs, ground vehicles and quasi-static users; wherein, ground vehicles and quasi-static users are communication targets, and UAVs are sensing targets.

[0012] The origin of the coordinate system is set to the center of the road, and the ground is defined as... The plane is used to establish a three-dimensional Cartesian coordinate system; the base station location coordinates are... , , , These are the coordinates of the base station on the X, Y, and Z axes, respectively; the drone's position coordinates are... , and The coordinates of the drone on the X and Y axes, The altitude of the drone is expressed as [missing information], and the speed of the drone is expressed as [missing information]. , , The velocity components of the UAV in the X and Y axes; the position coordinates of the ground vehicle are... , and Let X be the coordinates of the UAV on the X and Y axes, and let Y be the velocity of the ground vehicle. , , The velocity components of the ground vehicle in the X and Y axes; the quasi-static user position coordinates are... , and The coordinates of the quasi-static user on the X and Y axes are obtained; the relative distance between the communication target and the base station coordinates is calculated. ;in, It is the relative distance between the communication target and the base station coordinates. These are the coordinates of the communication target. These are the coordinates of the base station.

[0013] As a further optimization scheme for the dynamic trade-off method of multiple performance indicators of communication and sensing under high dynamic channels described in this invention, the high dynamic channel model in step S1 includes:

[0014] Step S101: Based on the location coordinates of the base station and the communication target in the integrated communication and sensing network, calculate the relative distance between the communication target and the base station, and obtain the large-scale mobile fading based on the relative distance, as follows:

[0015] Based on geometric distance, large-scale moving fading is obtained. for

[0016] ;

[0017] in, For path loss, For shadow fading, the large-scale path loss PL is calculated using the following formula;

[0018] ;

[0019] in, For communication carrier frequency, For three-dimensional geometric distance, This represents large-scale path loss.

[0020] Step S102: Analyze the corresponding small-scale fading based on the environmental parameters and line-of-sight and non-line-of-sight components in the integrated communication and sensing network, as follows:

[0021] exist Small-scale decay of time for

[0022] ;

[0023] in, for The Rice factor at time t is calculated using the following formula:

[0024] ;

[0025] Where e is the natural base, and For environmental factors, The elevation angle of the communication target relative to the base station. and These are the sight distance and non-sight distance components, respectively, which are calculated using the following formula;

[0026] ;

[0027] ;

[0028] Where j is the imaginary unit, For time delay variables, Let be the Dirac function of the line-of-sight delay component. The Dirac function is the non-line-of-sight time delay component. , For line-of-sight path in The delay of time, It is the speed of light; in the non-line-of-sight component, The total number of clusters, For uniformly distributed random phases, and These are the power and delay in the base station and the c-th communication target, respectively. The delay is calculated using the following formula:

[0029] ;

[0030] in, and These are from the base station to the... Cluster distance and from the first The distance from the cluster to the communication target, For the latency of the virtual link;

[0031] Consider omnidirectional antennas and vertical polarization in the non-line-of-sight component;

[0032] Step S103: Analyze the channel aging effect caused by mobility based on small-scale fading, as follows:

[0033] exist Channel state at time for

[0034] ;

[0035] in, These are the weighting coefficients for independent new components. , This is the initial channel state. To indicate in The independent new weight of time , The time correlation coefficient, It is a zero-order Bessel function of the first kind. The sampling period is The maximum Doppler frequency shift is calculated using the following formula:

[0036] ;

[0037] in, , The radial velocity between the base station and the communication target. It is the elevation angle from the communication target to the base station. Let be the speed of the c-th communication target relative to the base station at time t;

[0038] Therefore, in The effective channel at time t is represented as at The channel estimated at that time; Estimate the delay for the channel;

[0039] ;

[0040] in, It is the time correlation coefficient of the c-th communication target under a time delay τ. It is the small-scale channel state estimated at time t - τ. These are the independent new component weighting coefficients corresponding to the time correlation coefficient of the c-th communication target under a time delay τ. As an independent new component;

[0041] Therefore, the high dynamic channel model can be expressed as the product of small-scale fading and large-scale fading, taking into account channel aging effects:

[0042] .

[0043] As a further optimization scheme for the dynamic trade-off method of multiple performance indicators of communication and sensing under high dynamic channel described in this invention, step S2 includes:

[0044] Step S201: Based on large-scale mobile fading and small-scale fading, the spectral efficiency of the network is calculated as follows:

[0045] The base stations in the network employ a maximum ratio transmission method based on estimated channel state information; within each time slot, the base station transmits beamforming vectors. Signal Send to the One communication target, among which , The normalized mathematical expectation of the average power of the communication signal; therefore, the signal received by the ground vehicle for

[0046] ;

[0047] in, This represents the large-scale moving fading coefficient. Let be the independent new component introduced by channel aging at time t. For the beamforming vector oriented towards the c-th communication target, The communication signal sent by the base station, with the superscript H indicating conjugate transpose; The transmit power allocated to communication by the base station. It is additive white Gaussian noise with a mean of zero and a variance of . Therefore, the signal-to-interference-to-noise ratio at a certain moment under a dynamic link The following formula is used to calculate:

[0048] ;

[0049] in, Let be the transmit power allocated by the base station to the j-th communication target. The beamforming vector is directed toward the j-th communication target;

[0050] A channel estimation delay is obtained by integrating the signal-to-interference-to-noise ratio at a given moment over the time slot. Overall spectral efficiency of internal networks :

[0051] ;

[0052] in, This represents the percentage of time allocated to communication in each synesthetic task cycle. Let C be the c-th communication target, and C be the set of all communication targets. Let be the signal-to-interference-to-noise ratio of the c-th communication target at time ξ, and let represent any time interval from t – τ to t.

[0053] Step S202: Based on the spectral efficiency obtained in step S201, the energy efficiency of the network is calculated as follows:

[0054] Network energy efficiency is the ratio of total data transmission rate to total energy consumption, where the total data transmission rate is numerically equal to the spectral efficiency; therefore, network energy efficiency... The following formula is used to calculate:

[0055] ;

[0056] in, The total energy consumption of the network. Composed of circuit energy consumption and transmission power, it can be calculated using the following formula:

[0057] ;

[0058] in, This represents the transmit power corresponding to the l-th transmitting node. For circuit power, Total transmission power, This is a constant for the efficiency of the power amplifier; circuit power is typically modeled as...

[0059] ;

[0060] in, The circuit power for each antenna, To and Irrelevant static circuit power terms The total number of drones, This represents the total number of ground vehicles. The total number of quasi-static ground users, For circuit power, This is the linear proportionality coefficient between circuit power consumption and transmission power;

[0061] Step S203: Based on large-scale motion fading and small-scale fading, the network's speed sensing accuracy index is calculated, as follows:

[0062] The expression for calculating the unit direction vector between the base station and the sensing target is as follows:

[0063] ;

[0064] in, Let be the unit direction vector from the base station to the d-th sensing target. This represents the component of the unit direction vector along the x-axis. The unit direction vector is in the y-axis direction, and the superscript T indicates transpose;

[0065] Therefore, the radial velocity is calculated by the following formula:

[0066] ;

[0067] in, Let represent the radial velocity of the d-th sensing target relative to the base station. Let be the unit direction vector between the base station and the d-th sensing target. Let d be the velocity vector of the d-th perceived target;

[0068] Considering additive Gaussian measurement error, from the first The base station to the The radial velocity observation model for a single target is given by the following equation:

[0069] ;

[0070] in, It is the observed radial velocity. It is the actual radial velocity. , It is observation noise. To observe the variance of the noise, the radial velocity observation model is extended to all base stations within the network to obtain the actual velocity vector:

[0071] ;

[0072] in, Let be the radial velocity observation vector corresponding to the d-th sensing target. The observation matrix consists of multiple unit direction vectors. Let d be the velocity vector of the d-th perceived target. For the observed noise vector;

[0073] The covariance matrix of the noise is the same; the relationship between the Doppler frequency shift and the radial velocity is expressed as follows:

[0074] ;

[0075] in, The Doppler frequency shift generated when the s-th base station observes the d-th sensing target. Let d be the radial velocity of the d-th sensing target relative to the s-th base station, and let the sensing carrier frequency be... The unit is Therefore, the relationship between the Doppler measurement noise variance and the radial velocity measurement noise variance is expressed as follows:

[0076] ;

[0077] in, The noise variance of the radial velocity measurement of the s-th base station to the d-th sensed target is given. The noise variance for Doppler measurements is calculated using the following formula:

[0078] ;

[0079] in, Represents system configuration constants. This represents the variance of the additive white Gaussian noise at the receiver. For wavelength, Radar cross section, Transmission power for target sensing, , for Time of the first The first goal and the first The distance between base stations Let be the location vector of the s-th base station. Let d be the position vector of the d-th sensing target at time t; considering the Doppler frequency shift interference caused by channel aging, the noise variance of the radial velocity measurement is further expressed as:

[0080] ;

[0081] in, The noise variance of the radial velocity measurement of the s-th base station to the d-th sensed target is given. Let be the transmit power used for the d-th sensing target. To detect the radar cross section of the target, , It is the time correlation coefficient; the expression for the radial velocity observation Fisher information matrix of the d-th sensing target is:

[0082] ;

[0083] in, Estimate the Fisher information matrix for the velocity of the d-th perceived target. Let be the radial velocity vector of the d-th sensing target relative to the base station. Let be the radial velocity observation vector corresponding to the d-th sensing target. and These represent the i-th and j-th components in the velocity vector, respectively.

[0084] The relationship between the Fisher information matrix of the radial velocity observation and the Fisher information matrix of the actual velocity is expressed as follows:

[0085] ;

[0086] in, Estimate the Fisher information matrix for the velocity of the d-th perceived target. It is the perceived actual speed of the target. It is a Jacobian matrix. It is a zero-order Bessel function of the first kind. Let d be the radial velocity vector of the d-th sensing target relative to the base station;

[0087] Based on the relationship between the radial velocity and the actual velocity matrix, the Jacobian matrix is ​​calculated to be equal to... Therefore, the Fisher information matrix used to sense the actual velocity of the target is calculated using the following formula:

[0088] ;

[0089] Where D is the set of perceived targets. Let d be the velocity vector of the d-th perceived target. and These are the x- and y-axis components of the unit direction vector from the s-th base station to the d-th sensing target. Let V be the measurement noise variance when the s-th base station performs radial velocity observation on the d-th sensing target;

[0090] To simplify the calculation process, the items in the Fisher information matrix were simplified.

[0091] ;

[0092] in, This represents the Fisher information content of the d-th sensing target on the x-axis velocity component. This represents the amount of coupling information between the velocity components of the x-axis and y-axis. This represents the Fisher information content of the d-th sensing target on the y-axis velocity component;

[0093] Therefore, by inverting the total Fisher information matrix, the Cramer-Rao bound for velocity estimation is calculated using the following formula.

[0094] ;

[0095] Let Cramer-Rao lower bound matrix be the two-dimensional velocity estimation error of the d-th sensing target;

[0096] Since the Cramérob boundary is a vector, to unify it with spectral efficiency and energy efficiency in terms of dimensions; and The sum of the directional velocities and the Cramé-Rao bounds serves as an indicator of velocity perception accuracy; its expression is:

[0097] ;

[0098] Furthermore, to maintain consistency with communication metrics units, the expression for the average velocity sensing accuracy metric based on the Cramer-Rao boundary is as follows:

[0099] ;

[0100] in, This is a speed-sensing accuracy indicator for the network. Let τ be the percentage of time allocated to sensing in each sensing task cycle, D be the set of sensing targets, and τ be the channel estimation delay. Let be the variance of the equivalent velocity observation noise for the d-th sensing target at time ξ. Let be the lower bound scalar of the velocity estimation error of the d-th sensing target at time ξ.

[0101] As a further optimization scheme for the dynamic trade-off method of multiple performance indicators of communication and sensing under high dynamic channel described in this invention, step S3 includes:

[0102] Step S301: Based on the spectral efficiency, energy efficiency, and velocity sensing accuracy indices obtained in step S2, normalization processing is performed, as follows:

[0103] ;

[0104] in, This represents the normalized result of the Type I performance index. This represents the type I performance index function. This represents the minimum value of the Type I performance index. This represents the maximum value of the Class I performance index, where SE, EE, and Sen represent the spectral efficiency, energy efficiency, and speed sensing accuracy index, respectively.

[0105] Step S302: Based on the normalized spectral efficiency, energy efficiency, and velocity sensing accuracy indices obtained in step S301, a joint function for compromising multiple performance indices is constructed, as follows:

[0106] Given the weights, the optimization objective is written as

[0107] ;

[0108] in, Let Ω represent the resource variable to be optimized, and let Ω represent the set of feasible constraints for the resource variable. This represents the weighting coefficient corresponding to the performance index of type I. This represents the normalized result of the Type I performance index;

[0109] Step S303: Based on the normalized spectral efficiency, energy efficiency, and velocity sensing accuracy indices obtained in step S301, and the joint function for compromising multiple performance indices obtained in step S302, a dynamic weight allocation method for compromising multiple performance indices is constructed based on the Rubinstein bargaining model, as detailed below:

[0110] Based on the time correlation coefficients of the communication link and the sensing link, dynamic discount factors corresponding to the communication performance index and the sensing performance index are constructed respectively:

[0111] ;

[0112] in, This refers to the dynamic discount factor corresponding to the communication performance index. This refers to the dynamic discount factor corresponding to the perceived performance index. Let c be the maximum Doppler frequency shift of the communication target relative to the base station. The Doppler frequency shift when the s-th base station senses the d-th target. It is a zero-order Bessel function of the first kind; The channel sampling period;

[0113] During the Yth round of alternating bids, indicator I proposes a weighting allocation scheme:

[0114] ;

[0115] in, The weight allocation scheme proposed in round Y, , and Let be the weights corresponding to the spectral efficiency, energy efficiency, and speed sensing accuracy indicators in the Yth round, respectively, and satisfy the following:

[0116] ;

[0117] in, This represents the weighting share of the I-th performance indicator in the Y-th round of alternating bids;

[0118] Substituting the weight allocation scheme into the joint function in step S302, we obtain the compromise benefit corresponding to the scheme in round Y:

[0119] ;

[0120] in, The compromise return for the I-th performance metric in round Y. This is the normalized result of the I-th performance metric;

[0121] In the current round, for performance metrics that have not submitted bids, determine whether they accept the current weighting allocation scheme; if the following conditions are met:

[0122] ;

[0123] in, This represents the dynamic discount factor for the I-th performance indicator in the (Y+1)th round of alternating bids. This represents the expected weight share of the I-th performance indicator in the (Y+1)th round of alternating bids;

[0124] The current weighting scheme is accepted when all unbid performance metrics meet the above acceptance criteria; the current weighting scheme is rejected when at least one unbid performance metric does not meet the above acceptance criteria, and the process proceeds to the next round of alternating bids.

[0125] When the preset termination condition is met, a dynamic trade-off weight for multiple performance indicators is determined based on the current bidding priority; if spectral efficiency is the priority in bidding, the balance weight is expressed as:

[0126] ;

[0127] in, , and These represent the balanced weights of spectral efficiency, energy efficiency, and speed sensing accuracy when prioritizing spectral efficiency in the bidding process. This represents the dynamic discount factor corresponding to the perceived performance metric. This represents the dynamic discount factor corresponding to the communication performance index.

[0128] The closed-loop equilibrium weighting expression for prioritizing bids based on speed perception accuracy is as follows:

[0129] ;

[0130] in, , and These represent the balanced weights of spectral efficiency, energy efficiency, and speed perception accuracy when prioritizing bids based on speed perception accuracy.

[0131] After reaching a consensus during the alternating bidding process, the determined equilibrium weights are input into the multi-performance index joint function constructed in step S302, thus completing the dynamic weight allocation of the multi-performance indexes.

[0132] Compared with the prior art, the present invention, employing the above technical solution, has the following technical effects:

[0133] This invention considers the Doppler frequency shift and channel state information aging problems caused by target motion in highly dynamic scenarios, establishes a dynamic trade-off relationship among three performance indicators: spectral efficiency, energy efficiency, and speed sensing accuracy, and adaptively determines the weight allocation results of spectral efficiency, energy efficiency, and speed sensing accuracy based on the Rubinstein bargaining model. This invention can provide guidance for the performance evaluation and resource allocation of 6G integrated sensing networks under different mobility conditions. Attached Figure Description

[0134] Figure 1 This is a flowchart of the method of the present invention;

[0135] Figure 2 It represents the trade-off between spectral efficiency and sensing rate under different total transmit powers;

[0136] Figure 3 It represents a trade-off between energy efficiency and sensing rate under different total transmit powers;

[0137] Figure 4 It represents the spectral efficiency at different ground vehicle speeds;

[0138] Figure 5 It refers to the sensing rate at different drone speeds;

[0139] Figure 6 This refers to the energy efficiency under different numbers of base stations. Detailed Implementation

[0140] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0141] like Figure 1 As shown, a dynamic trade-off method for multiple performance indicators of communication and sensing under high dynamic channels includes:

[0142] Step S1: Establish an integrated communication and sensing network model, determine the base station, communication target and sensing target in the integrated communication and sensing network model, and construct a high dynamic channel model that considers target motion and Doppler frequency shift based on the integrated communication and sensing network model;

[0143] Step S2: Based on the high dynamic channel model, calculate the spectral efficiency, energy efficiency, and speed sensing accuracy indicators respectively;

[0144] Step S3: Introduce the channel aging factor into the index trade-off mechanism to construct a dynamic weight allocation method for multiple performance index trade-offs of spectral efficiency, energy efficiency and speed sensing accuracy.

[0145] The integrated communication and sensing network model includes base stations, drones, ground vehicles, and quasi-static users; among them, ground vehicles and quasi-static users are communication targets, and drones are sensing targets.

[0146] The origin of the coordinate system is set to the center of the road, and the ground is defined as... The plane is used to establish a three-dimensional Cartesian coordinate system; the base station location coordinates are... , , , These are the coordinates of the base station on the X, Y, and Z axes, respectively; the drone's position coordinates are... , and Let X and Y be the coordinates of the drone. The altitude of the drone is expressed as [missing information], and the speed of the drone is expressed as [missing information]. , , The velocity components of the UAV in the X and Y axes; the position coordinates of the ground vehicle are... , and Let X be the coordinates of the UAV on the X and Y axes, and let Y be the velocity of the ground vehicle. , , The velocity components of the ground vehicle in the X and Y axes; the quasi-static user position coordinates are... , and The coordinates of the quasi-static user on the X and Y axes are obtained; the relative distance between the communication target and the base station coordinates is calculated. ;in, It is the relative distance between the communication target and the base station coordinates. These are the coordinates of the communication target. These are the coordinates of the base station.

[0147] Step S1 includes:

[0148] Step S101: Based on the location coordinates of the base station and the communication target in the integrated communication and sensing network, calculate the relative distance between the communication target and the base station, and obtain the large-scale mobile fading based on the relative distance, as follows:

[0149] Based on geometric distance, large-scale moving fading is obtained. for

[0150] ;

[0151] in, For path loss, For shadow fading, the large-scale path loss PL is calculated using the following formula;

[0152] ;

[0153] in, For communication carrier frequency, For three-dimensional geometric distance, This represents large-scale path loss.

[0154] Step S102: Analyze the corresponding small-scale fading based on the environmental parameters and line-of-sight and non-line-of-sight components in the integrated communication and sensing network, as follows:

[0155] exist Small-scale decay of time for

[0156] ;

[0157] in, for The Rice factor at time t is calculated using the following formula:

[0158] ;

[0159] Where e is the natural base, and For environmental factors, The elevation angle of the communication target relative to the base station. and These are the sight distance and non-sight distance components, respectively, which are calculated using the following formula;

[0160] ;

[0161] ;

[0162] Where j is the imaginary unit, For time delay variables, Let be the Dirac function of the line-of-sight delay component. The Dirac function is the non-line-of-sight time delay component. , For line-of-sight path in The delay of time, It is the speed of light; in the non-line-of-sight component, The total number of clusters, For uniformly distributed random phases, and These are the power and delay of the communication between the base station and the c-th communication target, respectively. The delay is calculated using the following formula;

[0163] ;

[0164] in, and These are from the base station to the... Cluster distance and from the first The distance from the cluster to the communication target. For the latency of the virtual link;

[0165] Consider omnidirectional antennas and vertical polarization in the non-line-of-sight component;

[0166] Step S103: Analyze the channel aging effect caused by mobility based on small-scale fading, as follows:

[0167] exist Channel state at time for

[0168] ;

[0169] in, The weighting coefficients for independent new components. , This is the initial channel state. To indicate in The independent new weight of time , The time correlation coefficient, It is a zero-order Bessel function of the first kind. The sampling period is The maximum Doppler frequency shift is calculated using the following formula:

[0170] ;

[0171] in, , The radial velocity between the base station and the communication target. It is the elevation angle from the communication target to the base station. Let be the speed of the c-th communication target relative to the base station at time t;

[0172] Therefore, in The effective channel at time t is represented as at The channel estimated at that time; Estimate the delay for the channel;

[0173] ;

[0174] in, It is the time correlation coefficient of the c-th communication target under a time delay τ. It is the small-scale channel state estimated at time t - τ. These are the independent new component weight coefficients corresponding to the time correlation coefficient of the c-th communication target under a time delay τ. It is a new independent component.

[0175] Therefore, the high dynamic channel model can be expressed as the product of small-scale fading and large-scale fading, taking into account channel aging effects:

[0176] ;

[0177] Step S2 includes:

[0178] Step S201: Based on large-scale mobile fading and small-scale fading, the spectral efficiency of the network is calculated as follows:

[0179] The base stations in the network employ a maximum ratio transmission method based on estimated channel state information; within each time slot, the base station transmits beamforming vectors. Signal Send to the One communication target, among which , The normalized mathematical expectation of the average power of the communication signal; therefore, the signal received by the ground vehicle for

[0180] ;

[0181] in, This represents the large-scale moving fading coefficient. Let be the independent new component introduced by channel aging at time t. For the beamforming vector directed towards the c-th communication target, The superscript H indicates the conjugate transpose of the communication signal sent by the base station; The transmit power allocated to communication by the base station. It is additive white Gaussian noise with a mean of zero and a variance of . Therefore, the signal-to-interference-to-noise ratio at a certain moment under a dynamic link The following formula is used to calculate:

[0182] ;

[0183] in, Let be the transmit power allocated by the base station to the j-th communication target. Let be the beamforming vector for the j-th communication target.

[0184] A channel estimation delay is obtained by integrating the signal-to-interference-to-noise ratio at a given moment over the time slot. Overall spectral efficiency of internal networks :

[0185] ;

[0186] in, This represents the percentage of time allocated to communication in each synesthetic task cycle. Let C be the c-th communication target, and C be the set of all communication targets. Let be the signal-to-interference-to-noise ratio of the c-th communication target at time ξ, which is used to represent any time within the time interval from t – τ to t.

[0187] Step S202: Based on the spectral efficiency obtained in step S201, the energy efficiency of the network is calculated as follows:

[0188] Network energy efficiency is the ratio of total data transmission rate to total energy consumption, where the total data transmission rate is numerically equal to the spectral efficiency; therefore, network energy efficiency... The following formula is used to calculate:

[0189] ;

[0190] in, The total energy consumption of the network. Composed of circuit energy consumption and transmission power, it can be calculated using the following formula:

[0191] ;

[0192] in, This represents the transmit power corresponding to the l-th transmitting node. For circuit power, Total transmission power, This is a constant for the efficiency of the power amplifier; circuit power is typically modeled as...

[0193] ;

[0194] in, The circuit power for each antenna, To and Irrelevant static circuit power terms The total number of drones, This represents the total number of ground vehicles. The total number of quasi-static ground users, For circuit power, This is the linear proportionality coefficient between circuit power consumption and transmission power;

[0195] Step S203: Based on large-scale motion fading and small-scale fading, the network's speed sensing accuracy index is calculated, as follows:

[0196] The expression for calculating the unit direction vector between the base station and the sensing target is as follows:

[0197] ;

[0198] in, Let be the unit direction vector from the base station to the d-th sensing target. Let x be the component of the unit direction vector along the x-axis. The unit direction vector is in the y-axis direction, and the superscript T indicates transpose;

[0199] Therefore, the radial velocity is calculated by the following formula:

[0200] ;

[0201] in, Let represent the radial velocity of the d-th sensing target relative to the base station. Let be the unit direction vector between the base station and the d-th sensing target. Let d be the velocity vector of the d-th perceived target.

[0202] Considering additive Gaussian measurement error, from the first The base station to the The radial velocity observation model for a single target is given by the following equation:

[0203] ;

[0204] in, It is the observed radial velocity. This is the actual radial velocity. , It is observation noise. To observe the variance of the noise, the radial velocity observation model is extended to all base stations within the network to obtain the actual velocity vector:

[0205] ;

[0206] in, Let be the radial velocity observation vector corresponding to the d-th sensing target. The observation matrix consists of multiple unit direction vectors. Let d be the velocity vector of the d-th perceived target. This is the observed noise vector.

[0207] The covariance matrix of the noise is the same; the relationship between the Doppler frequency shift and the radial velocity is expressed as follows:

[0208] ;

[0209] in, The Doppler frequency shift generated when the s-th base station observes the d-th sensing target. Let d be the radial velocity of the d-th sensing target relative to the s-th base station, and let the sensing carrier frequency be... The unit is Therefore, the relationship between the Doppler measurement noise variance and the radial velocity measurement noise variance is expressed as follows:

[0210] ;

[0211] in, The noise variance of the radial velocity measurement of the s-th base station to the d-th sensed target is given. The noise variance for Doppler measurements is calculated using the following formula:

[0212] ;

[0213] in, Represents system configuration constants. This represents the variance of the additive white Gaussian noise at the receiver. For wavelength, Radar cross section, Transmission power for target sensing, , for Time of the first The first goal and the first The distance between base stations Let be the location vector of the s-th base station. Let d be the position vector of the d-th sensing target at time t; considering the Doppler frequency shift interference caused by channel aging, the noise variance of the radial velocity measurement is further expressed as:

[0214] ;

[0215] in, The noise variance of the radial velocity measurement of the s-th base station to the d-th sensed target is given. Let be the transmit power used for the d-th sensing target. To detect the radar cross section of the target, , It is the time correlation coefficient; the expression for the radial velocity observation Fisher information matrix of the d-th sensing target is:

[0216] ;

[0217] in, Estimate the Fisher information matrix for the velocity of the d-th perceived target. Let be the radial velocity vector of the d-th sensing target relative to the base station. Let be the radial velocity observation vector corresponding to the d-th sensing target. and These represent the i-th and j-th components in the velocity vector, respectively.

[0218] The relationship between the Fisher information matrix of the radial velocity observation and the Fisher information matrix of the actual velocity is expressed as follows:

[0219] ;

[0220] in, Estimate the Fisher information matrix for the velocity of the d-th perceived target. It is the perceived actual speed of the target. It is a Jacobian matrix. It is a zero-order Bessel function of the first kind. Let d be the radial velocity vector of the d-th sensing target relative to the base station;

[0221] Based on the relationship between the radial velocity and the actual velocity matrix, the Jacobian matrix is ​​calculated to be equal to... Therefore, the Fisher information matrix used to sense the actual velocity of the target is calculated using the following formula:

[0222] ;

[0223] Where D is the set of perceived targets. Let d be the velocity vector of the d-th perceived target. and These are the x- and y-axis components of the unit direction vector from the s-th base station to the d-th sensing target. Let be the measurement noise variance when the s-th base station performs radial velocity observation on the d-th sensing target.

[0224] To simplify the calculation process, the items in the Fisher information matrix were simplified.

[0225] ;

[0226] in, This represents the Fisher information content of the d-th sensing target on the x-axis velocity component. This represents the amount of coupling information between the velocity components of the x-axis and y-axis. This represents the Fisher information content of the d-th sensing target on the y-axis velocity component.

[0227] Therefore, by inverting the total Fisher information matrix, the Cramer-Rao bound for velocity estimation is calculated using the following formula.

[0228] ;

[0229] Let Cramer-Rao lower bound matrix be the two-dimensional velocity estimation error of the d-th sensing target;

[0230] Since the Cramérob boundary is a vector, to unify it with spectral efficiency and energy efficiency in terms of dimensions; and The sum of the directional velocities and the Cramé-Rao bounds serves as an indicator of velocity perception accuracy; its expression is:

[0231] ;

[0232] Furthermore, to maintain consistency with communication metrics units, the expression for the average velocity sensing accuracy metric based on the Cramer-Rao boundary is as follows:

[0233] ;

[0234] in, This is a speed-sensing accuracy indicator for the network. Let τ be the percentage of time allocated to sensing in each sensing task cycle, D be the set of sensing targets, and τ be the channel estimation delay. Let be the variance of the equivalent velocity observation noise for the d-th sensing target at time ξ. Let be the lower bound scalar of the velocity estimation error of the d-th sensing target at time ξ.

[0235] Step S3 includes:

[0236] Step S301: Based on the spectral efficiency, energy efficiency, and velocity sensing accuracy indices obtained in step S2, normalization processing is performed, as follows:

[0237] ;

[0238] in, This represents the normalized result of the Type I performance index. This represents the type I performance index function. This represents the minimum value of the Type I performance index. represents the maximum value of the Class I performance index, and SE, EE, and Sen represent the spectral efficiency, energy efficiency, and speed sensing accuracy index, respectively.

[0239] Step S302: Based on the normalized spectral efficiency, energy efficiency, and velocity sensing accuracy indices obtained in step S301, a joint function for compromising multiple performance indices is constructed, as follows:

[0240] Given the weights, the optimization objective is written as

[0241] ;

[0242] in, Let Ω represent the resource variable to be optimized, and let Ω represent the set of feasible constraints for the resource variable. This represents the weighting coefficient corresponding to the performance index of type I. This represents the normalized result of the Type I performance index.

[0243] Step S303: Based on the normalized spectral efficiency, energy efficiency, and velocity sensing accuracy indices obtained in step S301, and the joint function for compromising multiple performance indices obtained in step S302, a dynamic weight allocation method for compromising multiple performance indices is constructed based on the Rubinstein bargaining model, as detailed below:

[0244] Based on the time correlation coefficients of the communication link and the sensing link, dynamic discount factors corresponding to the communication performance index and the sensing performance index are constructed respectively:

[0245] ;

[0246] in, This refers to the dynamic discount factor corresponding to the communication performance index. This refers to the dynamic discount factor corresponding to the perceived performance index. Let c be the maximum Doppler frequency shift of the communication target relative to the base station. The Doppler frequency shift when the s-th base station senses the d-th target. It is a zero-order Bessel function of the first kind. The channel sampling period.

[0247] During the Yth round of alternating bids, indicator I proposes a weighting allocation scheme:

[0248] ;

[0249] in, The weight allocation scheme proposed in round Y, , and Let be the weights corresponding to the spectral efficiency, energy efficiency, and speed sensing accuracy indicators in the Yth round, respectively, and satisfy the following:

[0250] ;

[0251] in, This represents the weighting percentage of the I-th performance metric in the Y-th round of alternating bids.

[0252] Substituting the weight allocation scheme into the joint function in step S302, we obtain the compromise benefit corresponding to the scheme in round Y:

[0253] ;

[0254] in, The compromise return for the I-th performance metric in round Y. This is the normalized result for the I-th performance metric. These are the resource variables to be optimized.

[0255] In the current round, for performance metrics that have not submitted bids, determine whether they accept the current weighting allocation scheme. If the following conditions are met:

[0256] ;

[0257] in, This represents the dynamic discount factor for the I-th performance indicator in the (Y+1)th round of alternating bids. This represents the expected weight share of the I-th performance indicator in the Y+1 round of alternating bids.

[0258] The current weighting scheme is accepted when all unbid performance metrics meet the above acceptance criteria; the current weighting scheme is rejected when at least one unbid performance metric does not meet the above acceptance criteria, and the process proceeds to the next round of alternating bids.

[0259] Upon reaching the preset termination condition, a dynamic trade-off weight for multiple performance indicators is determined based on the current bidding priority. If spectral efficiency is prioritized in the bidding, the balance weight is expressed as follows:

[0260] ;

[0261] in, , and These represent the balanced weights of spectral efficiency, energy efficiency, and speed sensing accuracy when prioritizing spectral efficiency in the bidding process. This represents the dynamic discount factor corresponding to the perceived performance metric. This represents the dynamic discount factor corresponding to the communication performance index.

[0262] The closed-loop equilibrium weighting expression for prioritizing bids based on speed perception accuracy is as follows:

[0263] ;

[0264] in, , and These represent the balanced weights of spectral efficiency, energy efficiency, and speed perception accuracy when prioritizing bids based on speed perception accuracy.

[0265] After reaching a consensus during the alternating bidding process, the determined equilibrium weights are input into the multi-performance index joint function constructed in step S302, thus completing the dynamic weight allocation of the multi-performance indexes.

[0266] Example 1:

[0267] S1. Consider a city scenario with a service radius of 220 m, a communication frequency of 3.5 GHz, and a sensing frequency of 4.0 GHz. There are 4 base stations, with the center of the service area as the origin. The coordinates of the 4 base stations are (150, -150), (150, 150), (-150, 150), and (-150, -150), respectively, and the base station height is 60 m. There are 8 ground vehicles with a speed of 20 m / s and random initial positions and orientations. There are 8 quasi-static users with random positions. There are 8 drones with a speed of 10 m / s and random initial positions and orientations, and the drone height is 15 m. The total transmit power on the base station side is 10W, of which the communication power accounts for 0.8. When performing communication and sensing tasks, the time allocated to communication in one task cycle is 0.4%, and the time allocated to sensing is 0.6%.

[0268] A high dynamic channel model is constructed using the parameters described above.

[0269] S2. The corresponding spectral efficiency, energy efficiency, and speed sensing accuracy are calculated based on the high dynamic channel model.

[0270] S3. Based on the calculated spectral efficiency, energy efficiency, and speed sensing accuracy, a dynamic trade-off model with multiple performance indicators is constructed using the Rubinstein bargaining model.

[0271] The trade-off between spectral efficiency and sensing rate under different total transmit powers is analyzed and plotted. Specific results are shown below. Figure 2 As shown. By Figure 2It can be seen that as the sensing rate increases, the spectral efficiency generally decreases, indicating a resource competition relationship between communication performance and sensing performance. Meanwhile, the higher the transmit power, the more the curve shifts to the upper right. For example, with a transmit power of 20 W, the system can achieve higher spectral efficiency and sensing rate, indicating that increasing the transmit power is beneficial for improving the trade-off between communication and sensing performance.

[0272] The trade-off between energy efficiency and sensing rate under different total transmit powers is analyzed and plotted. Specific results are shown below. Figure 3 As shown. By Figure 3 It can be seen that as the sensing rate increases, the overall energy efficiency decreases, indicating that the system needs to consume more resources to improve sensing performance, thus resulting in energy efficiency loss. A comparison of different transmit power curves shows that higher transmit power can improve the overall energy efficiency of the system and expand the achievable sensing rate range, but there is still a clear trade-off between sensing and energy consumption.

[0273] The impact of ground vehicle speed on spectral efficiency was analyzed and plotted. Specific results are shown below. Figure 4 As shown. By Figure 4 It can be seen that as the speed of ground vehicles increases from 0 m / s to 30 m / s, the spectral efficiency gradually decreases, with a more significant decrease occurring at low to medium speeds. This is because increased vehicle speed enhances Doppler shift and channel aging effects, leading to a decline in communication link quality. Meanwhile, the spectral efficiency remains consistently high for higher total transmit power, indicating that increasing transmit power can mitigate the communication performance loss caused by high-speed movement to some extent.

[0274] The impact of UAV speed on sensing rate was analyzed and plotted, with specific results as follows: Figure 5 As shown. By Figure 5 It can be seen that as the drone's speed increases from 0 m / s to 24 m / s, the overall sensing rate decreases, indicating that the high-speed movement of the drone exacerbates Doppler frequency shift and velocity measurement errors, thereby reducing the accuracy of velocity sensing. A comparison of the transmission power curves shows that the sensing rate under the 20 W condition is consistently higher than that under the 12 W and 4 W conditions, indicating that increasing the transmission power can improve the quality of the sensing echo, thus improving sensing performance in high-dynamic scenarios.

[0275] The impact of the number of base stations on energy efficiency was analyzed and plotted. Specific results are shown below. Figure 6 As shown. By Figure 6It can be seen that as the number of base stations increases, energy efficiency does not show a monotonically increasing trend, but rather exhibits an overall decrease followed by local rebounds. This indicates that while increasing the number of base stations can improve link coverage and spatial diversity conditions, it also increases circuit power consumption and total system energy consumption. Therefore, in highly dynamic sensing integrated networks, it is necessary to comprehensively consider communication performance, sensing performance, and energy consumption to rationally configure the number of base stations.

[0276] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A dynamic trade-off method for multiple performance indicators of communication and sensing under high dynamic channels, characterized in that, include: Step S1: Establish an integrated communication and sensing network model, determine the base station, communication target and sensing target in the integrated communication and sensing network model, and construct a high dynamic channel model that considers target motion and Doppler frequency shift based on the integrated communication and sensing network model; Step S2: Based on the high dynamic channel model, calculate the spectral efficiency, energy efficiency, and speed sensing accuracy indicators respectively; Step S3: Introduce the channel aging factor into the index trade-off mechanism to construct a dynamic weight allocation method for multiple performance index trade-offs of spectral efficiency, energy efficiency and speed sensing accuracy.

2. The method for dynamic trade-offs between multiple performance indicators of communication and sensing under high dynamic channels according to claim 1, characterized in that, In step S1, the integrated communication and sensing network model includes base stations, drones, ground vehicles, and quasi-static users; among them, ground vehicles and quasi-static users are communication targets, and drones are sensing targets. The origin of the coordinate system is set to the center of the road, and the ground is defined as... The plane is used to establish a three-dimensional Cartesian coordinate system; the base station location coordinates are... , , , These are the coordinates of the base station on the X, Y, and Z axes, respectively; the drone's position coordinates are... , and The coordinates of the drone on the X and Y axes, The altitude of the drone is expressed as [missing information], and the speed of the drone is expressed as [missing information]. , , The velocity components of the UAV in the X and Y axes; the position coordinates of the ground vehicle are... , and Let X be the coordinates of the UAV on the X and Y axes, and let Y be the velocity of the ground vehicle. , , The velocity components of the ground vehicle in the X and Y axes; the quasi-static user position coordinates are... , and The coordinates of the quasi-static user on the X and Y axes are obtained; the relative distance between the communication target and the base station coordinates is calculated. ;in, It is the relative distance between the communication target and the base station coordinates. These are the coordinates of the communication target. These are the coordinates of the base station.

3. The method for dynamically compromising multiple performance indicators of communication and sensing under high dynamic channels according to claim 2, characterized in that, The high dynamic channel model in step S1 includes: Step S101: Based on the location coordinates of the base station and the communication target in the integrated communication and sensing network, calculate the relative distance between the communication target and the base station, and obtain the large-scale mobile fading based on the relative distance, as follows: Based on geometric distance, large-scale moving fading is obtained. for ； in, For path loss, For shadow fading, the large-scale path loss PL is calculated using the following formula; ； in, For communication carrier frequency, For three-dimensional geometric distance, This represents large-scale path loss. Step S102: Analyze the corresponding small-scale fading based on the environmental parameters and line-of-sight and non-line-of-sight components in the integrated communication and sensing network, as follows: exist Small-scale decay of time for ； in, for The Rice factor at time t is calculated using the following formula: ； Where e is the natural base, and For environmental factors, The elevation angle of the communication target relative to the base station. and These are the sight distance and non-sight distance components, respectively, which are calculated using the following formula; ； ； Where j is the imaginary unit, For time delay variables, Let be the Dirac function of the line-of-sight delay component. The Dirac function is the non-line-of-sight time delay component. , For line-of-sight path in Delay of time, It is the speed of light; in the non-line-of-sight component, The total number of clusters, For uniformly distributed random phases, and These are the power and delay in the base station and the c-th communication target, respectively. The delay is calculated using the following formula: ； in, and These are from the base station to the... Cluster distance and from the first The distance from the cluster to the communication target. For the latency of the virtual link; Consider omnidirectional antennas and vertical polarization in the non-line-of-sight component; Step S103: Analyze the channel aging effect caused by mobility based on small-scale fading, as follows: exist Channel state at time for ； in, These are the weighting coefficients for independent new components. , This is the initial channel state. To indicate in The independent new weight of time , The time correlation coefficient, It is a zero-order Bessel function of the first kind. The sampling period is The maximum Doppler frequency shift is calculated using the following formula: ； in, , The radial velocity between the base station and the communication target. It is the elevation angle from the communication target to the base station. Let be the speed of the c-th communication target relative to the base station at time t; Therefore, in The effective channel at time t is represented as at The channel estimated at that time; Estimate the delay for the channel; ； in, It is the time correlation coefficient of the c-th communication target under a time delay τ. It is the small-scale channel state estimated at time t - τ. These are the independent new component weight coefficients corresponding to the time correlation coefficient of the c-th communication target under a time delay τ. As an independent new component; Therefore, the high dynamic channel model can be expressed as the product of small-scale fading and large-scale fading, taking into account channel aging effects: 。 4. The method for dynamic trade-offs between multiple performance indicators of communication and sensing under high dynamic channels according to claim 1, characterized in that, Step S2 includes: Step S201: Based on large-scale mobile fading and small-scale fading, the spectral efficiency of the network is calculated as follows: The base stations in the network employ a maximum ratio transmission method based on estimated channel state information; within each time slot, the base station transmits beamforming vectors. Signal Send to the One communication target, among which , The normalized mathematical expectation of the average power of the communication signal; therefore, the signal received by the ground vehicle for ； in, This represents the large-scale moving fading coefficient. Let be the independent new component introduced by channel aging at time t. For the beamforming vector oriented towards the c-th communication target, The communication signal sent by the base station, with the superscript H indicating conjugate transpose; The transmit power allocated to communication by the base station. It is additive white Gaussian noise with a mean of zero and a variance of . Therefore, the signal-to-interference-to-noise ratio at a certain moment under a dynamic link The following formula is used to calculate: ； in, Let be the transmit power allocated by the base station to the j-th communication target. The beamforming vector is directed toward the j-th communication target; A channel estimation delay is obtained by integrating the signal-to-interference-to-noise ratio at a given moment over the time slot. Overall spectral efficiency of internal networks : ； in, This represents the percentage of time allocated to communication in each synesthetic task cycle. Let C be the c-th communication target, and C be the set of all communication targets. Let be the signal-to-interference-to-noise ratio of the c-th communication target at time ξ, and let represent any time interval from t – τ to t. Step S202: Based on the spectral efficiency obtained in step S201, the energy efficiency of the network is calculated as follows: Network energy efficiency is the ratio of total data transmission rate to total energy consumption, where the total data transmission rate is numerically equal to the spectral efficiency; therefore, network energy efficiency... The following formula is used to calculate: ； in, The total energy consumption of the network. Composed of circuit energy consumption and transmission power, it can be calculated using the following formula: ； in, This represents the transmit power corresponding to the l-th transmitting node. For circuit power, Total transmission power, This is a constant for the efficiency of the power amplifier; circuit power is typically modeled as... ； in, The circuit power for each antenna, To and Irrelevant static circuit power terms The total number of drones, This represents the total number of ground vehicles. The total number of quasi-static ground users, For circuit power, This is the linear proportionality coefficient between circuit power consumption and transmission power; Step S203: Based on large-scale motion fading and small-scale fading, the network's speed sensing accuracy index is calculated, as follows: The expression for calculating the unit direction vector between the base station and the sensing target is as follows: ； in, Let be the unit direction vector from the base station to the d-th sensing target. Let x be the component of the unit direction vector along the x-axis. The unit direction vector is in the y-axis direction, and the superscript T indicates transpose; Therefore, the radial velocity is calculated by the following formula: ； in, Let represent the radial velocity of the d-th sensing target relative to the base station. Let be the unit direction vector between the base station and the d-th sensing target. Let d be the velocity vector of the d-th perceived target; Considering additive Gaussian measurement error, from the first The base station to the The radial velocity observation model for a single target is given by the following equation: ； in, It is the observed radial velocity. This is the actual radial velocity. , It is observation noise. To observe the variance of the noise, the radial velocity observation model is extended to all base stations within the network to obtain the actual velocity vector: ； in, Let be the radial velocity observation vector corresponding to the d-th sensing target. The observation matrix consists of multiple unit direction vectors. Let d be the velocity vector of the d-th perceived target. For the observed noise vector; The covariance matrix of the noise is the same; the relationship between the Doppler frequency shift and the radial velocity is expressed as follows: ； in, The Doppler frequency shift generated when the s-th base station observes the d-th sensing target. Let d be the radial velocity of the d-th sensing target relative to the s-th base station, and let the sensing carrier frequency be... The unit is Therefore, the relationship between the Doppler measurement noise variance and the radial velocity measurement noise variance is expressed as follows: ； in, The noise variance of the radial velocity measurement of the s-th base station to the d-th sensed target is given. The noise variance for Doppler measurements is calculated using the following formula: ； in, Represents system configuration constants. This represents the variance of the additive white Gaussian noise at the receiver. For wavelength, Radar cross section, Transmission power for target sensing, , for Time of the first The first goal and the first The distance between base stations Let be the location vector of the s-th base station. Let d be the position vector of the d-th sensing target at time t; considering the Doppler frequency shift interference caused by channel aging, the noise variance of the radial velocity measurement is further expressed as: ； in, The noise variance of the radial velocity measurement of the s-th base station to the d-th sensed target is given. Let be the transmit power used for the d-th sensing target. To detect the radar cross section of the target, , It is the time correlation coefficient; the expression for the radial velocity observation Fisher information matrix of the d-th sensing target is: ； in, Estimate the Fisher information matrix for the velocity of the d-th perceived target. Let be the radial velocity vector of the d-th sensing target relative to the base station. Let be the radial velocity observation vector corresponding to the d-th sensing target. and These represent the i-th and j-th components in the velocity vector, respectively. The relationship between the Fisher information matrix of the radial velocity observation and the Fisher information matrix of the actual velocity is expressed as follows: ； in, Estimate the Fisher information matrix for the velocity of the d-th perceived target. It is the perceived actual speed of the target. It is a Jacobian matrix. It is a zero-order Bessel function of the first kind. Let be the radial velocity vector of the d-th sensing target relative to the base station; Based on the relationship between the radial velocity and the actual velocity matrix, the Jacobian matrix is ​​calculated to be equal to... Therefore, the Fisher information matrix used to sense the actual velocity of the target is calculated using the following formula: ； Where D is the set of perceived targets. Let d be the velocity vector of the d-th perceived target. and These are the x- and y-components of the unit direction vector from the s-th base station to the d-th sensing target, respectively. Let V be the measurement noise variance when the s-th base station performs radial velocity observation on the d-th sensing target; To simplify the calculation process, the items in the Fisher information matrix were simplified. ； in, This represents the Fisher information content of the d-th sensing target on the x-axis velocity component. This represents the amount of coupling information between the velocity components of the x-axis and y-axis. This represents the Fisher information content of the d-th sensing target on the y-axis velocity component; Therefore, by inverting the total Fisher information matrix, the Cramer-Rao bound for velocity estimation is calculated using the following formula. ； Let Cramer-Rao lower bound matrix be the two-dimensional velocity estimation error of the d-th sensing target; Since the Cramérob boundary is a vector, to unify it with spectral efficiency and energy efficiency in terms of dimensions; and The sum of the directional velocities and the Cramé-Rao bounds serves as an indicator of velocity perception accuracy; its expression is: ； Furthermore, to maintain consistency with communication metrics units, the expression for the average velocity sensing accuracy metric based on the Cramer-Rao boundary is as follows: ； in, This is a speed-sensing accuracy indicator for the network. Let τ be the percentage of time allocated to sensing in each sensing task cycle, D be the set of sensing targets, and τ be the channel estimation delay. Let be the variance of the equivalent velocity observation noise for the d-th sensing target at time ξ. Let be the lower bound scalar of the velocity estimation error of the d-th sensing target at time ξ.

5. The method for dynamically compromising multiple performance indicators of communication and sensing under high dynamic channels according to claim 1, characterized in that, Step S3 includes: Step S301: Based on the spectral efficiency, energy efficiency, and velocity sensing accuracy indices obtained in step S2, normalization processing is performed, as follows: ； in, This represents the normalized result of the Type I performance index. This represents the type I performance index function. This represents the minimum value of the Type I performance index. This represents the maximum value of the Class I performance index, where SE, EE, and Sen represent the spectral efficiency, energy efficiency, and speed sensing accuracy index, respectively. Step S302: Based on the normalized spectral efficiency, energy efficiency, and velocity sensing accuracy indices obtained in step S301, a joint function for compromising multiple performance indices is constructed, as follows: Given the weights, the optimization objective is written as ； in, Let Ω represent the resource variable to be optimized, and let Ω represent the set of feasible constraints for the resource variable. This represents the weighting coefficient corresponding to the performance index of type I. This represents the normalized result of the Type I performance index; Step S303: Based on the normalized spectral efficiency, energy efficiency, and velocity sensing accuracy indices obtained in step S301, and the joint function for compromising multiple performance indices obtained in step S302, a dynamic weight allocation method for compromising multiple performance indices is constructed based on the Rubinstein bargaining model, as detailed below: Based on the time correlation coefficients of the communication link and the sensing link, dynamic discount factors corresponding to the communication performance index and the sensing performance index are constructed respectively: ； in, This refers to the dynamic discount factor corresponding to the communication performance index. This refers to the dynamic discount factor corresponding to the perceived performance index. Let c be the maximum Doppler frequency shift of the communication target relative to the base station. The Doppler frequency shift when the s-th base station senses the d-th target. It is a zero-order Bessel function of the first kind; The channel sampling period; During the Yth round of alternating bids, indicator I proposes a weighting allocation scheme: ； in, The weight allocation scheme proposed in round Y, , and Let be the weights corresponding to the spectral efficiency, energy efficiency, and speed sensing accuracy indicators in the Yth round, respectively, and satisfy the following: ； in, This represents the weighting share of the I-th performance indicator in the Y-th round of alternating bids; Substituting the weight allocation scheme into the joint function in step S302, we obtain the compromise benefit corresponding to the scheme in round Y: ； in, The compromise return for the I-th performance metric in round Y. This is the normalized result of the I-th performance metric; In the current round, for performance metrics that have not submitted bids, determine whether they accept the current weighting allocation scheme; if the following conditions are met: ； in, This represents the dynamic discount factor for the I-th performance indicator in the (Y+1)th round of alternating bids. This represents the expected weight share of the I-th performance indicator in the (Y+1)th round of alternating bids; The current weighting scheme is accepted when all unbid performance metrics meet the above acceptance criteria; the current weighting scheme is rejected when at least one unbid performance metric does not meet the above acceptance criteria, and the process proceeds to the next round of alternating bids. When the preset termination condition is met, a dynamic trade-off weight for multiple performance indicators is determined based on the current bidding priority; if spectral efficiency is the priority in bidding, the balance weight is expressed as: ； in, , and These represent the balanced weights of spectral efficiency, energy efficiency, and speed sensing accuracy when prioritizing spectral efficiency in the bidding process. This represents the dynamic discount factor corresponding to the perceived performance metric. This represents the dynamic discount factor corresponding to the communication performance index. The closed-loop equilibrium weighting expression for prioritizing bids based on speed perception accuracy is as follows: ； in, , and These represent the balanced weights of spectral efficiency, energy efficiency, and speed perception accuracy when prioritizing bids based on speed perception accuracy. After reaching a consensus during the alternating bidding process, the determined equilibrium weights are input into the multi-performance index joint function constructed in step S302, thus completing the dynamic weight allocation of the multi-performance indexes.

Images