CN-121487010-B - Dynamic channel awareness-based general sense resource scheduling method, network equipment and system

CN121487010BCN 121487010 BCN121487010 BCN 121487010BCN-121487010-B

Abstract

The invention relates to the technical field of new generation wireless communication, in particular to a general sense resource scheduling method, network equipment and a system based on dynamic channel sensing. In order to solve the problems of hysteresis and resource mismatch in the dynamic scene of the existing reactive scheduling, the invention adopts a prospective scheduling method. According to the method, a general digital twin model is constructed and updated by continuously collecting historical channels and position data, and an intelligent prediction system is utilized to deduce the digital twin model so as to predict channel quality and target state at future time. And the scheduler performs cross-period universal resource pre-allocation and optimization decision-making according to the prediction result. The scheduling mechanism is improved from instant decision based on the current state to prospective planning based on future state prediction, so that timeliness and accuracy of resource scheduling are improved, communication reliability is ensured, and meanwhile, perception performance of a moving target is enhanced.

Inventors

LAI SIJIA
HE JIAN
LI XUEFENG
ZHANG PEICHANG
HUANG LEI
LIN HUIPENG
YE JUNJIE

Assignees

深圳大学

Dates

Publication Date: 20260508
Application Date: 20260107

Claims (10)

1. A method for scheduling communication resources based on dynamic channel perception is characterized by comprising the following steps: s1, constructing a probabilistic digital twin model, namely constructing and maintaining the probabilistic digital twin model in a probability distribution form in real time, wherein the probabilistic digital twin model comprises uncertainty of future channel state evolution and multiple possibilities of a dynamic target motion trail; S2, generating candidate scheduling schemes, namely generating candidate scheduling schemes with different design ideas according to service requirements of a current network and surrounding mutually conflicting optimization targets, wherein the range of the candidate scheduling schemes covers different strategy trends from pursuing the highest performance to pursuing the lowest energy consumption, and each scheme is a general sense resource allocation scheme formulated for a future scheduling period; s3, simulation deduction and performance evaluation, namely placing the candidate scheduling schemes into a virtual future constructed by the probabilistic digital twin model, checking performance of the candidate scheduling schemes under various possibilities by executing the simulation deduction, and generating a detailed performance analysis report comprising expected benefits and potential risks for each scheme; And S4, final decision of risk perception, namely comprehensively evaluating and sequencing the candidate scheduling schemes by adopting a risk perception utility function according to the performance analysis report, and selecting a scheme with optimal utility as a final decision to issue and execute.
2. The method for scheduling the universal resources based on the dynamic channel awareness according to claim 1, wherein the probabilistic digital twin model is constructed and maintained in real time, and the implementation comprises the following two parts: The method comprises the steps of firstly tracking and predicting a dynamic target, tracking the target through a prediction-correction iteration loop by adopting a kinematic modeling framework, firstly carrying out preliminary prediction based on the position of the last moment in each loop by the kinematic modeling framework, then correcting the preliminary prediction by utilizing a newly received perceived echo, continuously updating the state probability distribution of the target, and simultaneously outputting a covariance matrix for measuring the reliability of the preliminary prediction, and secondly, modeling and predicting a wireless channel, fitting the complex time-varying characteristic of the channel by adopting a multi-mode statistical method, continuously adjusting the parameters and weights of internal probability components through an iterative optimization algorithm according to the actually measured channel state information to form a comprehensive probability prediction model for deducing the overall evolution trend of the future channel state.
3. The method for scheduling generic resources based on dynamic channel awareness according to claim 1, wherein generating the candidate scheduling scheme comprises: The method comprises the steps of decomposing comprehensive service demands of a network into clear, independent and quantifiable optimization targets, wherein the optimization targets comprise maximization of communication throughput, minimization of perceived positioning errors, minimization of transmission delay and minimization of system energy consumption, adopting an intelligent body with efficient scheduling capability, and learning how to make efficient general-sense scheduling schemes by repeatedly testing and evaluating different resource scheduling decisions in an offline simulation environment, and delivering different strategy instructions to the intelligent body to guide optimization focuses to be respectively aligned to core targets in decision making, so that candidate scheduling schemes embodying different communication priorities, perceived priorities, delay priorities and energy-saving priorities are generated under the same network condition.
4. The method for scheduling universal resources based on dynamic channel awareness according to claim 1, wherein the specific implementation of parallel execution of multiple simulation deductions comprises: Starting parallel simulation threads aiming at candidate scheduling schemes to be evaluated, randomly sampling from probability distribution defined by the probabilistic digital twin model at the initial stage of each simulation thread, generating future event tracks comprising specific channel evolution and target motion details for the simulation threads, and simulating and executing a complete general sense task process in the future event tracks exclusive to the simulation threads according to resource allocation regulations of the candidate scheduling schemes and the corresponding communication and perception performance indexes under the future event tracks in each simulation thread.
5. The method for scheduling universal resources based on dynamic channel awareness according to claim 1, wherein the specific implementation of comprehensive evaluation and ranking based on the utility function of risk awareness comprises: The method comprises the steps of obtaining performance results of each candidate scheduling scheme in multiple simulation deduction, carrying out statistics processing to obtain performance distribution results capable of reflecting expected values and distribution forms of key performance indexes, carrying out quantitative evaluation on the performance distribution results of each candidate scheme by adopting a risk-aware utility function, designing comprehensive evaluation items reflecting average performance levels of the utility function, combining risk items for evaluating performance under a few worst scenes, sequencing all candidate scheduling schemes according to calculation scores of the utility function, and selecting an optimal scheme to be determined as a final scheduling decision.
6. The method for dynamic channel aware-based sensory resource scheduling of claim 4, wherein evaluating perceived performance comprises: in independent simulation deduction, virtual echo signals are simulated and generated according to resource configuration set by candidate scheduling schemes to be evaluated and combining with future event tracks, key information matrixes are calculated based on the virtual echo signals and known emission signal waveforms and used for measuring the information quantity about a target state in the echo signals, and inverse matrixes of the key information matrixes, namely theoretical minimum error values which can be achieved, are used as final indexes for judging the perception performance of the candidate scheduling schemes under deduction.
7. The dynamic channel aware-based sensory resource scheduling method of claim 6, wherein evaluating the performance of the communication comprises: In the independent simulation deduction, according to the resource configuration set by the candidate scheduling scheme to be evaluated, combining with the future channel evolution track, continuously calculating the definition index of the communication signal in the standard time segment, simulating the complete transmission process of the data packet based on the dynamic change of the definition index, wherein the process comprises the strategy of adjusting the data packaging mode in real time according to the signal definition and an intelligent retransmission mechanism for ensuring the accurate data transmission, and counting the simulation transmission results of all the data packets in the whole scheduling period to obtain the communication performance index capable of reflecting the service quality.
8. A sensory resource scheduling network device based on dynamic channel awareness, comprising a processor, and a memory having a computer program stored therein, wherein the computer program is configured to implement the method of any one of claims 1 to 7 when executed by the processor.
9. A system for scheduling communication resources based on dynamic channel awareness, based on the communication resource scheduling method based on dynamic channel awareness according to any one of claims 1 to 7, characterized in that the system comprises: An environment probe beam generating module configured to generate and control an active probe beam with a specific space and frequency structure, which emits low power, in a gap where a conventional sense task is performed, the active probe beam being used to excite a specific multipath scattering response in a wireless environment; The differential response analysis module is configured to accurately synchronize and capture differential echo signals excited by the active detection beam and containing the information of small environmental changes, and extract key environmental features about the structural changes of non-line-of-sight paths and the existence of dynamic scatterers in potential perception dead zones from the extremely weak differential signals through high-sensitivity correlation analysis; The characteristic-driven scheduling decision module is configured to take the environmental characteristics extracted by the differential response analysis module as key context information indicating inductive resource scheduling, and directly drive a scheduling strategy to carry out real-time self-adaptive adjustment towards the direction capable of actively utilizing the newly discovered environmental characteristics, so that the overall adaptability and performance of the system to complex dynamic environments are improved on the premise of not increasing conventional perception expenditure.
10. The dynamic channel aware-based sensory resource scheduling system of claim 9, wherein the implementation of the environmental probe beam generation module and the differential response analysis module comprises: Generating a pair of baseband signals to which mutually orthogonal code sequences are applied in phase, wherein one signal serves as a reference signal and the other signal serves as a disturbance signal; the baseband signal is processed by adopting a beam forming weight to form two detection beams which are completely overlapped in space for transmission, the detection beams are controlled to be transmitted at a power level lower than background noise, and the scheduling gap of a communication task is utilized for transmission, so that the low-interference active detection of the environment is realized; The method comprises the steps of receiving mixed echo signals, carrying out correlation operation on the mixed echo signals and local orthogonal coding sequences for generating reference beams and disturbance beams respectively to separate two paths of independent echo signals, carrying out coherent integration processing on the two paths of separated echo signals, wherein symmetrical echo components generated by static environment reflection are restrained, and asymmetrical echo components caused by dynamic environment change form detectable energy leakage, and extracting key environment characteristic information about weak dynamic events in the environment by detecting and analyzing the characteristics of the energy leakage.

Description

Dynamic channel awareness-based general sense resource scheduling method, network equipment and system Technical Field The invention relates to the technical field of new generation wireless communication, in particular to a general sense resource scheduling method, network equipment and a system based on dynamic channel sensing. Background The integrated sense of general is used as one of the key technologies of 6G, and when the same set of hardware equipment and wireless signals are utilized, two functions of communication and environment sensing (such as radar detection) are simultaneously realized, so that the frequency spectrum efficiency is greatly improved, and the hardware cost and the system energy consumption are reduced. In dynamic scenes, such as internet of vehicles or unmanned aerial vehicle cluster management, a core task is to efficiently schedule limited wireless resources (such as time, frequency, power and beam) so as to realize high-precision tracking of a high-speed moving target while guaranteeing high-reliability and low-delay communication. At present, most of mainstream resource scheduling methods adopt a reactive mechanism, namely, a decision of a next scheduling period is made in real time according to current channel state information and target state estimation. The mechanism can achieve a good effect in a quasi-static scene. Along with the increasing dynamic property of application scenes, in the scenes such as dense urban areas or highways, a large number of targets moving at high speed and having uncertain motion tracks exist, and the inherent defects of the traditional reactive scheduling mechanism are increasingly highlighted. Due to inherent processing delays throughout the link from sensing, reporting, decision making to execution, the actual state of the wireless channel and target may have changed significantly when the scheduling decision finally became effective. The "hysteresis" of such decisions results in "mismatch" between the resource allocation and the real demand of transient variability, especially in high-speed mobile scenarios, which may lead to interruption of the communication link due to burst occlusion or loss of perceived objectives due to failure to update beam pointing in time. Therefore, how to design a prospective resource scheduling mechanism capable of predicting and actively adapting to the future channel and target state changes so as to overcome the inherent delay and resource mismatch problems of reactive decision making, thereby improving the timeliness and the accuracy of scheduling is a key challenge facing the current field. Disclosure of Invention The present invention has been made in view of the above-described problems occurring in the prior art. Therefore, in order to solve the problems of hysteresis and resource mismatch existing in the existing reactive scheduling under the dynamic scene, the invention provides a general sense resource scheduling method, network equipment and a system based on dynamic channel perception. In order to solve the technical problems, the invention provides the following technical scheme: In a first aspect, the present invention provides a method for scheduling a generic resource based on dynamic channel awareness, which includes the following steps: S1, constructing a probabilistic digital twin model, namely constructing and maintaining the probabilistic digital twin model in a probability distribution form in real time, wherein the probabilistic digital twin model can simultaneously comprise uncertainty of future channel state evolution and multiple possibilities of a dynamic target motion trail; S2, generating candidate scheduling schemes, namely generating candidate scheduling schemes with different design ideas according to service requirements of a current network and surrounding mutually conflicting optimization targets, wherein the range of the candidate scheduling schemes covers different strategy trends from pursuing the highest performance to pursuing the lowest energy consumption, and each scheme is a general sense resource allocation scheme formulated for a future scheduling period; S3, simulation deduction and performance evaluation, namely placing each candidate scheduling scheme into a virtual future constructed by the probabilistic digital twin model, checking the performance of each candidate scheduling scheme under various possibilities by executing the simulation deduction, and generating an exhaustive performance analysis report comprising expected benefits and potential risks for each scheme; And S4, final decision of risk perception, namely comprehensively evaluating and sequencing the candidate scheduling schemes by adopting a risk perception utility function according to the performance analysis report, and selecting a scheme with optimal utility as a final decision to issue and execute. As a preferred scheme of the universal resource scheduling method based on dynamic channel awa