123 results about "Lyapunov optimization" patented technology

The invention discloses an adaptive streaming adaptation and resource optimization method applied to a cloud-game system. Based on a user QoE model (Quality of Experience) and a service and cost model, user distribution and resource allocation are carried out. The adaptive streaming adaptation and resource optimization method applied to the cloud-game system comprises the following steps: a cloud-game platform operational cycle is divided into multiple time periods; user information and a visual machine resource are updated in a cloud-game platform system in each time period; a user distribution and resource allocation plan is decided based on a current system state, the service cost of a cloud-game service provider is minimized by the plan in the system operational cycle and the user QoE sum is maximized; the user distribution and resource allocation plan is applied to each game user. By establishing the QoE model, namely, the QoE model, a user utility model and a cloud-platform costing model, and based on a Lyapunov optimizing framework, the optimizing method of the user distribution and resource allocation is designed based on a game type and the adaptive streaming adaptation technology.

The invention belongs to the technical field of wireless communication, and relates to a resource allocation method in a distributed heterogeneous environment in mobile edge computing. The method comprises the steps of establishing diversified task unloading models according to unloading delays corresponding to different service types in an MEC environment; establishing a buying and selling game model between the user and the MEC server; respectively establishing maximized income models of the user and the MEC server; adopting a Lyapunov optimization algorithm to improve the user model, and solving an optimal purchase strategy of the user through a Lagrange multiplier method and KKT conditions; solving an optimal quotation strategy of the MEC server based on the strategy; and if the optimal purchase strategy and the optimal quotation strategy satisfy a Stackleberg equilibrium solution, allocating computing resources of different users as required according to the optimal strategy by the MEC server. By means of the method, compromise of user unloading income and time delay can be achieved, and elastic control and on-demand distribution of task unloading and computing resource distribution can be achieved.

The invention relates to a resource allocation method for a low-delay high-reliability service in mobile edge computing, and belongs to the technical field of mobile communication. According to the method, in a multi-MEC multi-user environment, a user task queue model and an MEC task queue model are described respectively, a theoretical model of mobile service provider network utility maximizationis established with the task queue overflow probability as the constraint, and joint allocation is performed on power resources, bandwidth resources and computing resources. Considering that a constraint condition in the optimization model comprises a limit constraint of a task queue overflow probability; a random optimization problem of time averaging is converted and decomposed into three sub-problems of single-time-slot solving through a Lyapunov optimization theory, including calculation resource allocation of users, bandwidth and power allocation and calculation resource allocation of MEC, and then the three sub-problems are solved respectively. According to the method, the average total income of the MSP is improved while the requirements of users for low time delay and high reliability are met.

The invention provides a Lyapunov optimization-based multi-server mobile edge computing unloading method and device, and the method comprises: building a task cache queue for a task generated by eachmobile device, and building a virtual task cache queue for a task which needs to be unloaded and is not processed by each mobile device; establishing a terminal power consumption model, a bandwidth model and a QoE model; collecting task cache queue information and virtual cache queue information of the mobile device and the bandwidth condition of the server at each preset time slice; constructinga Lyapunov drift and penalty function according to the collected information; and based on Lyapunov drift and a penalty function, a calculation unloading decision in the current time slice is solved.The technical effects of calculating and unloading based on the multi-server MEC, effectively reducing the task completion time and reducing the energy consumption of the mobile device are achieved.

The invention provides a dynamic unloading method of fog calculation based on optimization, belonging to the field of wireless network communication. The present invention divides the calculation request into a local calculation portion and a fog calculation portion. By offloading computationally intensive requests to fog nodes, application performance can be significantly improved. An online joint radio and computational resource algorithm based on Lyapunov optimization is used to derive the upper bound of Lyapunov migration penalty function. By minimizing the upper bound from the perspectiveof different decision variables, , the optimal CPU cycle frequency of local processor is obtained with a convex optimization method. By using the predefined offload priority function, the optimal transmission power of the optimal subchannel is obtained. On the fog node, the optimal request scheduling decision is obtained by absurdity proof.

The invention relates to the technical field of wireless communication, in particular to a task unloading method for heterogeneous network edge computing, which comprises the following steps of: requesting a computing task by a mobile user, establishing a task queue model, and obtaining an unloading task quantity of the mobile user; setting a constraint condition according to the maximum unloadingtask amount of the mobile user; establishing a system model, and constructing a task unloading optimization problem by taking maximization of time average unloading income as a target; converting a task unloading optimization problem into an optimization problem in each time slot by utilizing a Lyapunov optimization theory, and obtaining a new optimization equation by minimizing the sum of Lyapunov drift and penalty terms; providing a selection criterion based on an extreme value and on the basis, deciding an optimal target server unloaded by a user and an optimal task load unloaded by the user. The optimal target server for unloading and the optimal task unloading amount can be determined, and the time average unloading income of the system is maximized while the stability of the systemis ensured.

The invention discloses a method for reducing task unloading delay of a 6G digital twin edge computing network. The method mainly provides a new digital twin edge computing network, uses digital twinto estimate the state of an edge server so as to provide training data, formalizes a series of unloading decision optimization problems in a user moving process, uses a Lyapunov optimization method tosimplify a long-term migration cost constraint into a multi-target dynamic optimization problem, and finally, a deep learning framework based on Actor-Critic is used for solving the calculation unloading optimization problem, and the training Agent is realized by a digital twin edge calculation network. According to the method, the unloading delay, the task failure rate and the migration rate arereduced to the maximum extent on the premise of keeping relatively low system cost.

The invention relates to the field of resource optimization in an Internet of Vehicles, in particular to an Internet-of-Vehicles resource optimization method based on a non-orthogonal multiple accesstechnology, which comprises the following steps of: when a vehicle task is processed in an NOMA-assisted vehicle edge computing system, taking minimization of total energy consumption of the vehicle edge computing system as a principle; determining an unloading and caching decision of the system, calculating and allocating caching resources, namely considering random flow arrival and queue stability of a vehicle user, and defining as a random optimization problem through joint optimization calculation of the unloading decision and the content caching decision and calculation and allocation ofthe caching resources; and utilizing a Lyapunov optimization theory to propose a dynamic joint calculation unloading, content caching and resource allocation algorithm for solving the problem, decoupling the algorithm into two independent sub-problems, and utilizing 0-1 integer programming and linear programming to solve the two sub-problems. According to the invention, the computing resources ofthe mobile edge computing server can be effectively processed, and the energy consumption of the system is reduced.

In the contemporary data center industry, energy consumption of air conditioning systems accounts for a large proportion of total energy consumption, so how to reduce the total energy consumption of air conditioners and servers is what the industry and the academia focus on. The present invention provides a data center energy consumption optimization resource control algorithm under a server average temperature constraint. With quality of service and a server average temperature constraint ensured, an average temperature-aware power minimization model (called ATPM for short) is established, and a Lyapunov optimization theory is used in an algorithm to approximately solve an ATPM problem. The algorithm does not need to measure statistical information of a workload in advance, and the algorithm is low in complexity and easy for implementation.

The invention relates to a distributed energy management and optimization method in a multi-user mobile edge computing system. The distributed energy management and optimization method in the multi-user mobile edge computing system comprises the following steps: modeling a system according to the characteristics of the mobile edge computing technology and the energy harvesting technology, and an energy consumption problem of a mobile device to plan a system energy management and optimization problem, and using the target of minimizing the system energy consumption as an objective function; andperforming reasonable relaxation on building the system energy management and optimization problem, using a Lyapunov optimization method to deal with the converted problem to obtain an online optimization algorithm, and further obtaining a distributed optimization algorithm by using an ADMM distributed method. The distributed energy management and optimization method in the multi-user mobile edgecomputing system can effectively reduce the energy consumption in the multi-user mobile edge computing system.

The invention discloses a cloud-side-end collaborative resource allocation method for a three-dimensional heterogeneous power internet of things, namely, cloud-side-end collaborative task unloading and computing resource allocation in the space-air-ground integrated three-dimensional heterogeneous power internet of things are jointly optimized. The method comprises the following steps: (1) constructing a system model to establish a three-dimensional heterogeneous power Internet of Things scene formed by a satellite, an unmanned aerial vehicle and a terminal; (2) refining the model; (3) proposing and converting a queuing time delay constraint and joint optimization problem; (4) based on a Lyapunov optimization principle, combining decomposition and solution of an optimization problem, and minimizing an upper bound of drift plus penalty in each time slot; (5) a cloud edge-end collaborative task unloading decision algorithm based on deep reinforcement learning. According to the method, the high-dimensional task unloading problem is solved based on deep reinforcement learning, and the problem of curse of dimensionality under information uncertainty is effectively solved in combination with the complex function approximation capability provided by the neural network and the decision-making capability provided by the act-critic algorithm.

The invention discloses an online power control method based on a Lyapunov framework in an energy collection safety transmission system. A sending end is a source node powered by an energy collectiondevice, a source node transmits information to the two destination nodes, each time slot selects to send information to one of the two destination nodes according to a channel state, and the information needs to be confidential for the other destination node. In a transmission process, the data, the energy arrival amount and the channel state change randomly, and under the condition that the statistical information is unknown, a Lyapunov optimization framework is utilized to convert a sending power control problem which takes the energy use efficiency in the optimization security transmissionas a target into an optimization problem which minimizes the drift and punishment upper bound, and finally, a solution of the optimal sending power is obtained. According to the control method, statistical characteristics of channels and collected energy do not need to be known in advance, transmission fairness of two destination nodes is considered while collected energy is efficiently utilized to transmit confidential information, and the control method is a practical method.

The invention discloses an electric power Internet of Things low-delay security cloud side-end cooperation method, which comprises the following steps: constructing a system model consisting of PIOT equipment, a ground base station, an unmanned aerial vehicle, an edge server, a cloud server and a satellite; constructing a task unloading model; constructing a task processing model; constructing a queuing time delay model; constructing an alliance block chain model; the total queuing time delay of the PIOT equipment in the SAG-PIOT network under the long-term security constraint is minimized; and long-term constraint and short-term decision optimization are decoupled by using Lyapunov optimization. The method has the advantages that the system cost is reduced, the learning performance of model training is improved, and the allocation of computing resources is optimized; and the anti-EMI capability is improved, EMI perception is realized, and the adverse effect of EMI is reduced.

The invention discloses a distributed dynamic service deployment method based on mobile edge computing, which considers the high mobility of a user and the limited storage capacity of an edge server,maximizes the long-term system utility while ensuring the stability of a storage queue of the server, and achieves efficient unloading in a mobile edge network environment. According to the method, the long-term system utility maximization problem is decomposed into the online Lyapunov drift and penalty function minimization problem by utilizing Lyapunov optimization. Under the condition of no priori knowledge of the future moving track of the user, the future system utility is obtained through a sampling average approximation algorithm. In addition, service deployment probability distributionis introduced, and a Markov approximation model is utilized to dynamically deploy service request data. Theoretical analysis shows that the hybrid strategy service deployment decision provided by theinvention is a progressive optimal solution. The invention provides a new method for distributed dynamic service data deployment.

The invention relates to a multi-user calculation unloading method based on Lyapunov optimization, and the method specifically comprises the steps: system modeling: multi-user edge calculation is composed of one base station and N mobile devices, in an edge calculation system, each mobile device has the energy acquisition capability, each base station is provided with one server, the mobile devices can communicate with the edge server through wireless communication technologies such as 5G; a local calculation model and an edge server calculation model are constructed to obtain time delay and energy consumption of the application in local calculation and time delay and energy consumption required for transmitting the application to an edge server, and an energy model is constructed to obtain an energy queue; the execution cost of the mobile device in a single time slot is obtained, and a problem of minimizing the average execution cost of the mobile equipment is constructed; energy causal constraints are eliminated through a Lyapunov method, and then an optimal calculation unloading decision and a resource allocation scheme are obtained through a variable replacement method. According to the invention, the execution cost of the mobile device can be reduced, and the application rejection rate is reduced.

The invention provides a request allocation method oriented to a cloud video service. The method comprises the following steps: acquiring a system parameter from a cloud video service system; acquiring a minimum cost function and an optimization constraint according the system parameter; acquiring a displacement-penalty function of the minimum cost function and the upper bound of the displacement-penalty function with a Lyapunov optimization framework; and extracting a part which is relevant to request allocation from the upper bound of the displacement-penalty function and calculating a request allocation scheme. Through adoption of the request allocation method, the problem of resource supply is considered while the problem of request allocation is solved, and gaining of an optimal strategy is facilitated. Meanwhile, scenes occurring in practical applications such as randomly-arriving loads, different QoE (Quality of Experience) requests, synchronous service of a plurality of CSPs (Cloud Service Providers) and the like are considered at the same time; a solution under the complex scenes is provided; and a practical application value is achieved.

The invention discloses a ubiquitous power Internet of Things access method based on context awareness learning. The method comprises the following steps: S1, constructing a system model; S2, refiningthe constructed system model to obtain a task data/transmission model, an energy consumption model and a time delay model; S3, constructing an optimization problem based on maximization of equipmentenergy efficiency; S4, based on the theory of the virtual queue and the Lyapunov optimization theory, converting the optimization problem; S5, realizing the optimization of channel selection through learning, and further realizing the maximization of energy efficiency. Based on the Lyapunov optimization theory, energy perception and service reliability perception are combined with learning, so themaximum and optimal energy utilization rate and service reliability can be achieved through learning, and the utilization efficiency of network energy and the reliability of network service are improved.

The invention discloses a resource scheduling implementation method based on energy consumption and QoS collaborative optimization. The method comprises the following steps: constructing a cloud task arrival queuing model of multiple virtual machines in a cloud computing data center environment; extracting QoS (Quality of Service) features of a data center by utilizing a stacked noise reduction automatic encoder technology to obtain a matrix for describing QoS feature information after dimension reduction, and solving the maximum response time of a current virtual machine to perfect constraint conditions of a collaborative optimization objective function; and obtaining a resource scheduling algorithm based on a Lyapunov optimization theory by combining a cloud task arrival queuing model, a collaborative optimization objective function and a Lyapunov optimization method, and realizing resource scheduling based on energy consumption and QoS collaborative optimization by adopting the resource scheduling algorithm. According to the method, energy consumption of the data center is effectively reduced while QoS is guaranteed, and interference of fluctuation of cloud task arrival in a real scene of the cloud computing data center on optimization problem solving is overcome.

The invention discloses a ubiquitous power Internet-of-Things access method based on matching learning. The method comprises the following steps: S1, constructing a system model; S2, refining the model to obtain a task/data transmission model, an energy consumption model, a time delay model and a service reliability model; S3, maximizing long-term throughput, and determining an optimization problem; S4, based on the theory of the virtual queue and the Lyapunov optimization theory, converting the optimization problem; and S5, realizing the optimization of channel selection through learning andmatching, and further realizing the maximization of throughput. According to the invention, the optimal channel selection is realized through learning and matching, so that the maximization of throughput is realized, based on an MAB theory, a lyapunov optimization theory and a matching theory, energy perception and service reliability perception are combined with machine learning, so that the maximum and optimal energy utilization rate and service reliability are achieved.

According to the embodiment of the invention, the method comprises the steps: obtaining the state information of a current time slot of mobile equipment connected with an MEC server, and constructinga weighted sum function model based on the obtained state information and the operation energy consumption of the current time slot; simplifying the weighted sum function model by using a preset Lyapunov function model to obtain a first target optimization function model, and performing optimal solution calculation on the first target optimization function model by using a preset Lagrange multiplier method to obtain a resource allocation strategy; and allocating resources to each mobile device in the MEC system based on the resource allocation strategy. Compared with the prior art, according to the embodiment of the invention, the weighted sum function model is constructed by utilizing the obtained state information and the current operation energy consumption; by simplifying the weightedsum function model, the complexity of the weighted sum function model is reduced, the Lagrange multiplier method is used for solving, the solving process is irrelevant to feasible actions, and therefore the speed of solving the weighted sum function model can be increased, and the real-time performance of resource allocation is improved.

The invention provides an energy efficiency optimization and resource allocation method. The method comprises the following steps: calculating a signal-to-interference-noise ratio received by users ofan authorized spectrum in a small base station on a subchannel; calculating a signal-to-interference-noise ratio received by users of an unauthorized spectrum in the small base station on the subchannel; calculating the total capacity of all users in the small base station according to the capacities of the users of the small base station; calculating a utility function; rewriting an equivalent objective function; rewriting an optimization problem according to a defined throughput function and the average throughput of the users; rewriting the optimization problem again according to the arrival of traffic of the users in the small base station; and getting a solution of three sub-problems by means of Lyapunov optimization. The invention also provides an energy efficiency optimization andresource allocation system. Through the technical scheme provided by the invention, the enhancement of utility performance and the balance of energy efficiency delay can be realized under various constraints.

The invention discloses a high energy-efficiency downloading method based on context awareness. The method comprises the steps of constructing a context awareness downloading method which utilizes an improved Lyapunov optimization framework and a system thereof; and when an intelligent mobile device requests to download network data, according to network environment in which the device locates currently and different link states, size of a data file which needs to be downloaded, and queue length waiting for being downloaded, determining whether to delay downloading, and further selecting a link for downloading; the downloading system comprises one or more intelligent mobile devices requesting to download the data and includes one or more network communication environments; a downloading module is embedded into the system of each intelligent mobile device; the downloading modules are used for obtaining current network environment state information, receiving a downloading request, and determining whether to download and selecting which link for downloading according to a specific situation and the size of the data file which needs to be downloaded, thereby realizing optimization of the data downloading. The high-quality data downloading service is provided, the network data flow of a user is effectively saved, and the energy consumption for mobile data downloading is reduced.

The invention discloses a data migration method of a cross-data center cloud computing system. The data migration method comprises the steps of obtaining system parameters from the cloud computing system, obtaining an objective function and constraint according to the system parameters, obtaining a flicker-penalty factor of the objective function and an upper bound of the flicker-penalty factor through a Lyapunov optimization framework; extracting a data distribution item from the upper bound of the flicker-penalty factor and generating the data migration method. Due to the fact that a technical method of using the Lyapunov optimization framework to obtain the flicker-penalty factor of the objective function and the upper bond of the flicker-penalty factor and extract the data distribution item is utilized, the cost of cross-data centers is balanced, and the scheduling problem of the cross-data center cloud computing system is solved in high performance, high availability and least cost.

The invention provides a non-orthogonal multiple access network resource allocation method, which is used to solve the problem of the prior art of lack of non-orthogonal multiple access network resource allocation mechanisms having strong feasibility. The non-orthogonal multiple access network resource allocation method comprises steps that various parameters are initialized, and a two-sided matching algorithm is used to solve a problem of associating a user and a channel; by adopting a Lyapunov optimization algorithm, an optimal solution of power allocation is acquired; by adopting iterativejoint user channel matching and power allocation algorithm, energy efficiency and economic benefits can be improved to the greatest extent, and at the same time, individual rationality and authenticity attribute are guaranteed. The reasonable allocation of NOMA network resources can be realized, and at the same time, maximization of economic benefits and energy effectiveness of content providers can be guaranteed, and therefore the energy efficiency can be improved to the greatest extent, a system capability is improved, and under a precondition of not reducing service quality and consideringan energy problem, an increasing data requirement is satisfied, and sustainable development is realized.

The invention discloses an optimization method for a dynamic spectrum slice framework in a super-5G Internet of Vehicles, which comprises the following steps of: for a service request of a specific type, dispatching the service request arriving in a time slot to different downlink resource planes to deliver a service; enabling each resource plane to maintain a queue for each type of service request in a resource slicing mode; according to the unmanned aerial vehicle packet processing capacity of each resource plane in each time slot, determining the number of unmanned aerial vehicles needing to be dispatched in order to meet the dynamic packet processing requirements from all the service requests in each time slot; according to the packet processing capability of each resource plane for each service request in the time slot, by taking minimization of service supply cost and maximization of service utility of all resource planes as optimization objectives, decoupling optimization parameters based on a Lyapunov optimization technology so as to determine optimal control parameters. According to the invention, the user service performance and the resource utilization rate are effectively improved.