36 results about "Energy constrained" patented technology

The invention relates to a Markov jump network system control method under DoS attack based on a phototropic growth algorithm. The method comprises the following steps: establishing a networked system model with Markov jump parameters; an energy-limited DoS attack model is constructed; introducing a dynamic event triggering mechanism, and determining a triggering moment based on a measurement error, a weighting coefficient, a design parameter and an internal dynamic variable; based on a networked system model and an energy-limited DoS attack model, a dynamic event triggering mechanism is fused to construct a closed-loop system dynamic model; establishing a hybrid anti-interference performance index; performing global optimization on gain parameters of the closed-loop system dynamic model by adopting a phototropic growth algorithm to obtain an optimal controller gain; and constructing a controller under the energy-limited DoS attack, and realizing the control of the Markov jump network system under the DoS attack. The finally constructed controller can realize high robustness, low communication burden and excellent dynamic performance control of the Markov hopping network system in an energy limited DoS attack scene.

The invention relates to an energy buffer assisted near-field MISO-SWIPT system. The system comprises a base station and user equipment, wherein the base station and the user equipment are distributed in a multipath non-line-of-sight dominant scene; the base station is used for transmitting a radio frequency signal on a downlink; the user equipment is configured with a power divider, an information decoder, an energy collector and an energy buffer, and is used for sending a signal to the base station on an uplink by using energy obtained by conversion based on a radio frequency signal; wherein the power divider is used for receiving a radio frequency signal, dividing the received radio frequency signal into a first power flow and a second power flow, and converting the first power flow and the second power flow into first energy and second energy; the information decoder is used for receiving the first energy and performing information decoding; the energy collector is used for collecting second energy and transmitting the second energy to the energy buffer; compared with the prior art, the communication reliability and energy utilization efficiency problems of the energy limited equipment in the near-field multipath environment are solved.

The present application relates to a kind of spectrum and energy limited network connection cooperative driving communication system resource allocation method, belong to the field of wireless short distance communication of Internet of Vehicles.The present application is directed to the problem of communication and computing resource limited in network connection cooperative driving communication system and link interruption under complex communication scenario, considering the maximum transmit power of vehicle end, maximum bandwidth, vehicle-to-base station communication maximum energy consumption, vehicle-to-vehicle communication minimum rate and other constraints, construct the task computing total delay minimization resource allocation model based on partial task offloading, use variable relaxation, continuous convex approximation, alternate optimization and variable substitution and other methods to convert mixed integer nonlinear programming resource allocation model into convex optimization model to solve.The present application can effectively reduce the vehicle-to-base station communication delay, while meeting the user service quality demand of vehicle-to-vehicle communication.

There is provided techniques for gesture recognition in a gesture tracking application. A method is performed by a controller device. The method comprises receiving user input from a user of the gesture tracking application. The user input identifies a set of gestures for the gesture tracking application. The method comprises training a BSMGRS and a CAMGRS to independently of each other recognize the set of gestures for the user. The method comprises establishing a database of gestures for the user. The database of gestures comprises a set of approved gestures comprises those of the gestures in the set of gestures for which the training of both the BSMGRS and the CAMGRS was successful. The method comprises performing gesture recognition of the user for the gesture tracking application. The gesture recognition is performed using the BSMGRS and in accordance with the established database of gestures.

The invention discloses a multi-robot task balanced distribution method under energy constraint based on segmented anchoring, which comprises the following steps of: firstly, distinguishing robot separation by marking path information with 0, 1 and-1, returning when the load is full or the electric quantity is insufficient, and generating an initial solution by hybrid coding of task number and completion proportion recorded by split information, and performing initial repair; in iteration, conventional population evolution is combined with probability to screen non-dominated solutions, and a charging interruption path is optimized through a sequence anchoring and balancing mechanism of segmented optimization, iterative anchoring and residual workload balancing; and finally, finishing time is finely adjusted by a re-balancing mechanism based on proportional splitting, and a non-dominated solution set is output. According to the method, the opportunity of converting passive charging constraint into active optimization is achieved, and the complex problem of multi-robot task allocation under energy limitation is systematically solved.

The invention provides an information updating method and system for a sensitivity-oriented wireless energy supply communication network, and the method comprises the steps: periodically obtaining the sensing data of each terminal node according to the current sampling frequency, recognizing a sensitive node of which the timeliness exceeds a threshold value or has an abnormal feature according to the historical sensing data and prediction data, and carrying out the recognition of the sensitive node. Serving as a combined scheduling target object; a scheduling optimization model is constructed to minimize the overall average AoI of the system under the condition of energy limitation, and meanwhile, timely transmission of node information with sensitivity higher than a set value is preferentially guaranteed to solve the model; and based on a solving result of the scheduling optimization model, generating a dynamic self-adaptive joint scheduling strategy, and flexibly adjusting the sampling frequency and the updating opportunity of the sensing data according to the current state of each joint scheduling target object. According to the invention, the priority acquisition and transmission capability of the key data can be improved, and the wireless energy supply communication network is promoted to develop towards the direction of higher energy efficiency and stronger perceptual intelligence.

The invention relates to a relay selection method based on Q-learning in an energy limited network, and belongs to the technical field of communication. Firstly, an energy-limited network energy consumption model is constructed; secondly, deriving the outage probability of the energy-limited network; and finally, introducing Q-Learning into a relay selection stage, and proposing a Q-Learning-based relay selection strategy (QRSS) under the energy-limited network to select an optimal relay, and the strategy enables an intelligent agent to learn an optimal relay selection scheme through training. Simulation results show that the method is superior to a random relay selection strategy (RSS) in throughput, data transmission reliability, node residual energy and other performances.

The present disclosure provides methods and apparatuses for signal transmission. In the method, an energy transmitter (ET) sequentially transmits a training signal over frequency sub-bands with a same transmit power, and an energy receiver (ER) transmits first information to the ET based on a received signal strength indicator (RSSI) of the training signal over the frequency sub-bands, the first information being used for indicating a frequency sub-band with a highest RSSI among the frequency sub-bands, thereby reducing feedback overhead from the ER to the ET. The ET transmits a first multi-carrier power signal using a waveform parameter, which may improve the efficiency of wireless power transmission and enable the ER to accept the first multi-carrier power signal. Therefore, the ER does not need to estimate channel state information between the ET and the ER. Thus, it is applicable to the ER with energy restrictions and hardware limitations.

The invention discloses a wireless energy supply OAM communication system design and optimization method based on a time switching protocol, and relates to the technical field of communication, and the method comprises the steps: obtaining array configuration data of a base station and energy limited equipment, calculating channel parameters, and determining power distribution; generating a modulation matrix according to the OAM mode order, and dividing a time period into an energy transmission time period and an information transmission time period; energy acquisition data and link capacity are calculated, an optimal time switching factor is determined by adopting a convex optimization method by taking maximization of uplink total capacity as a target under the condition that the constraint of a service quality threshold value is met, and finally, a base station is controlled to complete cooperative transmission of energy and information; the method has the advantages that through deep fusion of OAM mode multiplexing and TS protocols, dynamic optimal balance of energy transmission and information transmission is achieved on the premise that the service quality is guaranteed, and the spectrum efficiency and the energy efficiency of a communication system are improved.

Aiming at the problems of unbalanced energy and low coverage efficiency of the existing WSN, the invention provides a WSN energy efficient coverage method and device for improving a bacterial foraging algorithm (EABFO). According to the method, an adaptive variation method is adopted during position updating, before a predicted peak value is reached, the variation probability is increased along with the increase of the number of iterations, the randomness of bacteria is increased, global large-range search is realized, omission of an optimal solution is avoided, local optimum is tended, after the predicted peak value is reached, the variation probability is reduced, and local detail search is realized; time and space waste is reduced, and algorithm efficiency is improved. In addition, an elite breeding strategy is innovatively provided, the optimal fitness value is reserved, reverse growth of the fitness value is avoided, and the whole flora is evolved in a better direction. Compared with a genetic algorithm, the method has the advantages that the service life of the sensor matrix is prolonged by 200%, the method is suitable for energy-limited scenes such as large-scale static monitoring and dynamic target tracking, and the life cycle of the WSN is effectively prolonged.

The invention discloses a safety optimization method for an STAR-RIS auxiliary integrated common inductance and energy transfer system. The method mainly comprises the steps that S1, the STAR-RIS auxiliary multi-input single-output integrated common inductance and energy transfer system is constructed; s2, according to the communication scene, one IOT device is located in the STAR-RIS reflection area, and the base station transmits information to the IOT device through the straight-through link and the STAR-RIS reflection link. The sensing scene is that a target is located in a building, the signal intensity of a straight-through link is ignored, and the target is sensed only through an STAR-RIS transmission link; and S3, under the condition that the total emission energy of the base station is limited, constructing an optimization equation for maximizing the safety rate of the system. Performing decoupling optimization solution on active beam forming, artificial noise and STAR-RIS phase shift parameters of the base station based on an alternating optimization algorithm; s4, optimizing active beam forming and artificial noise of the base station; and S5, the STAR-RIS phase shift parameter is solved. According to the invention, the security of a communication system can be obviously improved.

The application is particularly a dual-power switching black start method of a liquid flow battery system, relates to the technical field of liquid flow battery energy storage system control, and comprises the following steps: collecting parameters such as the health degree of series-connected stacks, the residual state of charge, the local electrolyte temperature and the hydraulic resistance in real time, optimizing the pilot unit through normalization and a scoring model, combining multi-stage control such as dual-power switching and a feedback loop, and realizing the smooth construction of a microgrid. In the application, a dynamic optimization strategy of a pilot energy storage unit based on multidimensional constraints is constructed, and the complex technical pain points existing in the power supply reliability, energy storage unit optimization, pumping network cooperation and electrical safety of the all-vanadium liquid flow battery in the black start process under the conditions of limited standby energy and extreme environment are solved. The application has the beneficial effects of high start-up precision, strong system stability and wide adaptability.

The present application relates to a kind of cable health on-line monitoring system of wind energy harvesting, belong to self-powered cable condition monitoring field.It includes the acquisition layer, back end layer and front end layer connected in turn;Acquisition layer includes several acquisition points, each acquisition point includes communication module, MCU microcontroller, temperature and humidity sensor and wind speed and direction sensor, MCU microcontroller controls communication module to transmit the information of the environment where cable is located collected by temperature and humidity sensor and wind speed and direction sensor to back end layer;Each acquisition node also includes a wind energy harvesting module, which powers the load of acquisition layer;Back end layer forwards the received data to front end layer by wireless communication mode;The present application can effectively solve the problem of energy limitation, reduce the laying and operation cost of sensor network, monitor the environmental change of each node in real time, and provide important guarantee for the safe operation of cable health and power network.

The invention relates to a near field communication-sensing-energy charging fusion method of a non-intrusive wireless body area network, which belongs to the technical field of near field communication and wireless body area networks, and comprises the following steps of: constructing a system model consisting of a multi-antenna access node and a user wearing a physical sign information sensor; the sensor realizes wireless energy charging through an excitation signal transmitted by the access node, transmits a sensing signal into a human body to collect physical sign information, and establishes low-power-consumption connection with the access node based on backscatter communication to realize data transmission. Aiming at the current situations that energy in the wireless body area network is limited and the penetration depth of far-field electromagnetic waves to a human body is relatively small, the invention provides a wireless body area network integrated communication, perception and energy charging system architecture, and designs a joint optimization method of an emission strategy, time slot allocation and beam forming, so that the cooperation of near-field wireless energy transmission and physical sign information perception is realized, and the energy utilization rate of the wireless body area network is improved. And meanwhile, an efficient and low-energy-consumption communication means is provided for transmission of the physical sign information.

This invention relates to a two-layer structured multi-UAV swarm anti-DoS attack formation coordination method based on energy-constrained hybrid control. First, a two-layer UAV swarm architecture integrating physical and virtual layers is proposed, which can uniformly divide a large number of UAV nodes into groups and establish biomimetic communication between these nodes. Second, a hybrid control protocol based on energy constraints is designed by combining continuous-time control and pulse control mechanisms and introducing a saturation function. When applied to formation coordination, this protocol simultaneously solves the problems of control efficiency, operating cost, and system robustness. Finally, a novel DoS attack model is introduced, which integrates "multi-interval periods" and "recovery probabilities." These multi-interval periods include a buffer period during which the recovery probability of the attacked communication channel follows a Bernoulli distribution. This method effectively addresses the limitations of existing DoS attack models in practical applications.

The invention provides a temperature and energy constrained task scheduling and computing power distribution method in satellite calculation. The method comprises the following steps: determining a satellite type according to a satellite task arrival rate and a satellite operation state; the satellites comprise overload satellites, adaptive-load satellites and light-load satellites; determining a dynamic task scheduling and computing power distribution target strategy; the dynamic task scheduling and computing power allocation target strategy comprises task dynamic allocation to local processing or task dynamic scheduling to other satellites or task dynamic unloading to a ground cloud center; establishing a satellite calculation time delay model; formulating a task scheduling execution strategy and minimum time delay for preventing computing power overload when the task is completed; formulating a computing power distribution execution strategy; and determining the optimal balance of the task scheduling execution strategy and the computing power distribution execution strategy through the game model, and balancing and optimizing the satellite computing service quality and the satellite energy efficiency. The balance of task delay, energy consumption and system income is considered, and the efficient operation efficiency of the satellite network is improved.

The invention discloses a Mars unmanned aerial vehicle task self-organizing scheduling method and device based on hierarchical reinforcement learning and credit distribution, relates to the technical field of deep space exploration, and aims to solve the problem that autonomous cooperation of an unmanned aerial vehicle cluster is difficult to realize due to long time delay of ground fire, discontinuous communication and limited energy in Mars exploration. According to the method, a high-level strategy is constructed to generate a team-level task target by obtaining a terrain grid map, an energy distribution map and an interest point confidence map; the task targets are input into a middle-level strategy for distributed task distribution combining competition and game; inputting a distribution result into a bottom strategy to complete path planning and task execution; dynamic environment mapping is updated through observation data, and an interest point confidence map is formed to be recalled by a high-level strategy; unmanned aerial vehicle marginal contribution is calculated through differential credit distribution and is used for strategy updating; and maintaining a scheduling link by using a disconnection tolerant communication mechanism when the communication is unstable. The method is suitable for unmanned aerial vehicle cluster autonomous planning and cooperative scheduling work in a Mars exploration task.

The invention relates to a routing device and a routing method, in particular to a wireless sensor network multi-path routing method and a wireless sensor network multi-path routing device based on energy perception and double-layer trust. The device comprises an energy sensing module for generating energy state information. And the trust evaluation module is used for updating the comprehensive trust value of the node according to the energy state information and the data packet interaction behavior of the node. And the path construction module is used for forming a plurality of candidate paths containing energy and trust attributes. And the path selection module is used for evaluating to obtain the relative proximity of the path, and determining the transmission path by combining a selection mechanism based on probability deviation and penalty factors. And the data transmission optimization module is used for monitoring the performance of the transmission path in a data transmission stage and judging whether to apply the path penalty factor or not according to a monitoring result. According to the invention, efficient and reliable data transmission can be realized in an industrial Internet of Things environment in which energy limitation and security threats coexist.

The invention discloses a multi-sensor fusion filtering method for a 2-D nonlinear system under a measurement deletion and energy harvesting mechanism. The method comprises the following steps: firstly, establishing a Tobit model for a censored measurement value; then, under the condition that the energy of the sensor is limited, the probability that the sensor can successfully transmit the censored measurement value and the conditional expectation and conditional variance of the censored measurement value relative to the system state are calculated; on the basis, a local outlier suppression Tobit Kalman filter is constructed and is used for realizing reliable estimation of the system state under the conditions of measurement deletion and energy limitation. Furthermore, a stochastic analysis method and a 2-D mathematical induction method are utilized to deduce the upper bound of the filtering error variance, and the upper bound is minimized by optimizing the filter gain. And finally, based on the local estimation result of each sensor, a more accurate global fusion estimation value is obtained by adopting a covariance cross fusion method.

This invention relates to the field of wireless communication technology and discloses a multi-node self-organizing wireless coverage method for photovoltaic power plants. The method includes: collecting and tracking the physical tilt angle and link quality of the support structure; mapping the tilt angle to a mechanical sector index; constructing a mechanical spectrum fingerprint database and marking poorly performing channels as forbidden states; calculating negatively correlated energy potential weights based on real-time power generation current; fusing and constructing an uplink data transmission path pointing to the aggregation gateway; predicting the mechanical phase trend based on the tilt angle; and, when the trend falls into a communication blind zone and the real-time power generation current meets the maintenance threshold, activating a data retention mode and writing service data into a priority circular buffer. During transmission, forbidden channels are eliminated, and the operating frequency is selected for transmission. This invention effectively solves the problems of dynamic physical shading and energy limitation caused by support structure movement in photovoltaic power plants through coordinated scheduling of mechanical phase mapping and energy channels, achieving adaptive and reliable communication in complex environments.

This application discloses a method, apparatus, device, and storage medium for controlling the duty cycle of backscatter communication. The method includes: acquiring the voltage values ​​of the energy storage capacitor of a passive tag node at different sampling times; calculating the voltage change rate based on the voltage values; calculating the energy harvesting power of the passive tag node based on the voltage change rate; obtaining the tag transmission power consumption of the passive tag node performing backscatter communication; determining a target duty cycle based on the energy harvesting power and the tag transmission power consumption; and controlling the on-time and off-time of the backscatter modulator of the passive tag node during the communication cycle based on the target duty cycle. This scheme uses the voltage change of the energy storage capacitor to estimate the environmental energy harvesting capability in real time, and combines this with the tag transmission power consumption to adaptively adjust the communication duty cycle, enabling the tag node to dynamically match the balance between energy input and communication consumption under energy-constrained conditions.

The invention belongs to the field of self-powered wireless sensing, and relates to a self-powered wireless sensing system and method based on surface acoustic wave resonator modulation, and the system comprises a self-powered wireless transmitting module and a wireless receiving module. The transmitting module comprises a high-entropy energy collector, an energy management module and a radio frequency oscillation circuit based on acoustic surface wave resonator modulation. The energy management module comprises a half-wave rectification circuit, a switched capacitor converter and a thyristor switch circuit, and is used for carrying out rectification, energy conversion and storage on an electric pulse signal output by the high-entropy energy collector, and releasing energy to the radio frequency oscillation circuit when a trigger threshold value is reached; and the radio frequency oscillation circuit outputs a radio frequency signal carrying sensing information under the action of the surface acoustic wave resonator and wirelessly sends the radio frequency signal to the receiving module through the antenna to extract the sensing information. According to the invention, the energy utilization efficiency of the system can be greatly improved while the working frequency band of long-distance wireless transmission is provided, and the system is suitable for energy-limited or maintenance-free long-distance sensing application scenarios.

The application discloses a kind of full duplex relay assisted double-hop covert communication system power distribution method.For a long-distance transmission based on relay and energy limited system's covert communication scene, the way of dividing system energy to sender and relay is often not optimal.In order to solve this problem, the application constructs the analysis framework of covert communication based on full duplex relay, analyzes the minimum false detection probability of detector and the covert constraint condition that system needs to meet.Further, the application solves the joint optimization problem of transmitter and full duplex relay transmission power distribution, and solves the closed expression of transmitter and full duplex relay transmission power.Further, the correctness of theoretical analysis is verified by using MATLAB platform simulation.The results show that the application solves the power distribution problem of double-hop covert communication system with total power limited well, and gives the influence of transmission power distribution on covert transmission rate, to guide the design of covert transmission strategy.

The invention relates to the technical field of intelligent monitoring and safety early warning, and discloses a rural water area intelligent drowning prevention system based on wind-solar complementary power supply and multi-mode sensing, which comprises a wind-solar complementary power supply module, a multi-mode sensing module, an intelligent analysis and control center and an early warning and execution module, according to the system, the frequency of alternating current output by a wind driven generator is used for determining vibration of a rod body so as to compensate video jitter, the temperature of a back plate of a photovoltaic module is used for estimating the junction temperature of an image sensor so as to compensate image thermal noise, and the intelligent analysis and control center is used for analyzing the state of charge of a battery and a future energy obtaining prediction result. And dynamically adjusting the network structure complexity of the neural network model to balance the calculation performance and the system energy consumption. The technical problem of a field independent power supply monitoring system under environment interference and energy limitation is solved, the operation parameters of the power supply system are used for image quality correction, closed-loop self-adaption of energy and calculation load is achieved, and multi-dimensional data fusion and an active verification strategy are combined.

The invention discloses a molecular communication probability constellation diagram shaping method and system, and the method comprises the steps: obtaining an input bit sequence, and generating a concentration signal after probability shaping through a high-energy symbol discontinuous mapping algorithm; transmitting the concentration signal through a liquid laminar flow channel, detecting the change of medium characteristics in the liquid laminar flow channel, and obtaining an observation signal corresponding to the concentration signal; performing detection and de-mapping processing on the observation signal, and recovering a bit sequence; and based on a comparison result of the recovered bit sequence and the original input bit sequence, optimizing a mapping strategy of the high-energy symbol discontinuous mapping algorithm so as to complete probability shaping of communication symbols. According to the invention, based on a high-energy symbol discontinuous mapping algorithm, the sending probability of high-energy-consumption symbols is optimized, so that the overall energy consumption of a molecular communication system is reduced, and the service time of equipment is effectively prolonged in an energy-limited application scene.

The invention provides a wireless energy supply NOMA system IRS precoding method based on OAM, and the method comprises the steps: firstly constructing a system architecture comprising a BS, K energy-limited IDs and an IRS, enabling the BS to be equipped with a transmission concentric transmission UCA to transmit an OAM signal, and enabling the IDs to achieve SWIPT through a PS protocol; then position and rotation parameters are defined, channel gains are calculated, and emission signals are constructed according to an NOMA power distribution strategy in which UCA channel gains are received from IRS to ID in an ascending order. Deducing a received signal, separating information from an energy signal, calculating a signal to interference plus noise ratio and capacity, and collecting energy; converting the collected energy into equivalent capacity, and performing weighted summation to obtain the total capacity of the system. And finally, constructing an optimization problem by taking maximization of the total capacity as a target, setting capacity and energy threshold constraints, alternately updating equalizer vectors, weight vectors and IRS reflection phase shifts through iteration by using a WMMSE algorithm until convergence, solving IRS optimal reflection phase shifts, and completing precoding design. According to the invention, by fusing OAM, NOMA and IRS technologies, the 6G spectrum efficiency and energy sustainability are improved.

The application discloses a kind of unmanned aerial vehicle self-organizing network double-layer Q learning dynamic routing method based on multi-dimensional state perception fusion, to solve the problem that unmanned aerial vehicle node high-speed movement leads to topology dramatic change in prior art, unmanned aerial vehicle battery energy is limited, global routing decision perception performance weak problem, realize the collaborative optimization of unmanned aerial vehicle self-organizing network dynamic routing reliability, energy efficiency and task adaptability.The method includes that node is based on two-hop neighbor perception mechanism to realize topology comprehensive perception, and the interval of Hello information package is sent by dynamic time slot allocation adjustment;Construct "link-topology-energy-task" multidimensional fusion state model;Based on the double-layer decision framework of Q (λ) learning, combine routing hole dynamic avoidance mechanism to optimize routing selection;Balancing multi-optimization target using multi-objective reward function.The application has good environmental adaptability and generalization ability, can adapt to different network size and node moving speed scene, significantly improve packet delivery rate, reduce energy consumption and shorten end-to-end delay.

The invention discloses a simultaneous information and energy transmission method and system based on a non-linear circuit model, and belongs to the technical field of wireless communication, the simultaneous information and energy transmission method comprises the steps that waveform parameters and indexes of the waveform parameters are optimized through centralized decision making by a transmitting end, and low-complexity circuit regulation based on a lookup table is executed by a receiving end; and a closed-loop adaptive triggering mechanism based on an energy acquisition state is introduced, so that the information demodulation performance is ensured, the energy acquisition efficiency is improved, the calculation complexity of a receiving end is reduced, and the method is suitable for wireless communication nodes with limited energy. When a channel state or a load condition changes, performance degradation caused by nonlinear coupling can be sensed and compensated in time, and the accuracy of energy collection efficiency and the stability of information demodulation performance are ensured. Furthermore, the information demodulation performance is ensured, collaborative optimization of information transmission and energy transmission is achieved, and the method has good engineering realizability and application and popularization value.

The invention discloses a wireless energy supply Internet of Things information age minimization scheduling method, and the method comprises the steps: building a wireless energy supply Internet of Things network which comprises a hybrid access point and a plurality of energy limited devices; energy collection, energy consumption and information age evolution processes are analyzed based on the wireless energy supply Internet of Things; modeling a target problem of searching an optimal resource scheduling strategy into an optimization problem with the goal of minimizing the information age expectation weighted sum of the network; based on a Lyapunov optimization framework, designing an adaptive scheduling strategy of information age, and determining mode selection; deducing a closed power control scheme under the condition that the number of non-orthogonal multiple access devices is given; solving a sub-channel allocation problem by using a matching algorithm based on a delay acceptance method and an exchange operation; according to the method, the information age and the energy state are jointly optimized through the multi-carrier non-orthogonal multiple access technology, a triple self-adaptive mechanism is provided, the average weighted information age of the network is remarkably reduced, and the calculation overhead is reduced.

The invention relates to the field of mobile edge computing, and discloses an unmanned aerial vehicle assisted asynchronous federated learning online scheduling method and system based on convergence perception. The method comprises the following steps: establishing a system model and defining model obsolessness; constructing a convergence cost penalty function containing data isomerism and old degree influence on the basis of theoretical derivation; under the condition that energy budget and bandwidth constraints are met, a joint optimization problem with the goal of minimizing long-term convergence cost is established; constructing a virtual queue by using a Lyapunov optimization technology, and converting a long-term problem into a single-slot deterministic optimization sub-problem; and finally, a deep reinforcement learning algorithm based on an encoder-decoder structure is adopted to solve a flight position and a client selection strategy online. The system comprises various modules and computing equipment which are configured to execute the steps of the method. According to the method, model deviation caused by asynchronous updating can be effectively relieved, and the convergence speed and precision of the model are remarkably improved in an energy-limited environment.