76 results about "Computation offloading" patented technology

A new multi-edge smart community system consisting of communication, computing, and energy harvesting models, where the task execution delay and energy consumption are formalized as the optimization objectives under multiple constraints. For single-edge scenarios, we propose an improved twin-delayed DRL-based algorithm. For multi-edge scenarios, we develop a novel personalized FL-based training framework for DRL. Using the real-world settings and testbed, extensive experiments are conducted to validate the effectiveness of the proposed PFR-OA. The results show that the PFR-OA achieves better trade-offs between delay and energy consumption and exhibits higher task execution success rates than benchmark methods under different scenarios. Notably, the PFR-OA reaches a faster convergence speed compared to advanced DRL-based and FRL-based methods. Moreover, we further verify the practicality and superiority of the PFR-OA via real-world testbed experiments.

The invention belongs to the technical field of vehicle-mounted edge computing (VEC), and particularly relates to an unloading and resource allocation method for a DAG task in a vehicle-mounted edge computing environment. The method comprises the following steps: firstly, constructing a VEC unloading system in a two-way lane scene, and collecting task and equipment information and establishing an optimization model in combination with a vehicle moving model, a communication model and a calculation model; and secondly, aiming at a task in a directed acyclic graph (DAG), a task priority scheduling method based on a DAG topological structure is designed, so that a task scheduling strategy is optimized. On the basis, the problem is modeled as a Markov decision process by taking minimization of task completion time delay and system energy consumption as optimization objectives. The invention relates to the field of resource allocation, in particular to a task-dependent computing unloading and resource allocation algorithm based on a depth deterministic policy gradient (DDPG), and further provides a task-dependent computing unloading and resource allocation algorithm based on the depth deterministic policy gradient (DDPG) so as to solve the optimization problem.

Provided is a traffic-aware lightweight layered offloading framework for an adaptive slicing-enabled space-air-ground integrated network (SAGIN), where the adaptive slicing-enabled SAGIN is divided into a communication access platform (CAP) and a computation offloading platform (COP), and resources on each of the CAP and the COP are managed by network slicing; an edge service provider (ESP) provides computation offloading while performing resource allocation; for the resource allocation, a dynamic traffic change is captured by using ProbSparse self-attention, and adaptive network slicing is executed in accordance with predicted traffic and a system load; and for the computation offloading, a communication process and a computation process are separated to allocate a sub-channel as required in accordance with a channel state, then a virtual machine is allocated to a task through a lightweight computation offloading algorithm, and a converged policy is extracted as a lightweight neural network for online inference.

The embodiment of the invention provides a calculation unloading and resource allocation joint optimization method and device, and the method comprises the steps: constructing an edge calculation system model, determining a long-term optimization problem of calculation unloading and resource allocation based on the model, and carrying out the calculation unloading and resource allocation joint optimization through employing a preset optimization method. A long-term optimization problem is converted into an online optimization problem for determining a calculation unloading strategy and a resource allocation strategy of each time slot, the calculation unloading strategy of each time slot is determined based on a reinforcement learning method for the online optimization problem, and an optimal resource allocation strategy is determined by adopting a preset resource allocation method based on the calculation unloading strategy. According to the method, factors influencing performance, such as resources, energy consumption and time delay, are comprehensively considered to carry out joint optimization of calculation unloading and resource allocation, and the overall performance and service quality of the system can be improved.

The invention discloses a calculation unloading and resource allocation method and system based on unmanned aerial vehicle trajectory optimization, and relates to the field of industrial Internet of Things and air-ground integrated networks. Firstly, a multi-unmanned aerial vehicle assisted cloud side end computing unloading architecture is constructed, and unmanned aerial vehicles serve as edge computing nodes and relay nodes to assist computing unloading. Then, an optimization target is defined to minimize the average information age of all users under constraint conditions of calculation and communication resources, flight speed, observation range and the like; and finally, optimizing a hybrid strategy of trajectory planning, bandwidth allocation and task processing of the unmanned aerial vehicle by adopting a heterogeneous multi-agent dominant strategy-value algorithm, and performing parameter updating through federated learning to promote cooperation between agents. According to the method provided by the invention, the average information age can be effectively reduced, and reasonable bandwidth allocation and trajectory optimization are realized.

A server (200) includes a userland APL (230) that cooperates with an accelerator (212) while bypassing an OS (220). The userland APL (230) includes an ACC-NIC common data parsing part (232) that parses reception data in which an input data format of an ACC utilizing function and an NIC reception data format are made common.

The invention discloses a production line batching conveying control method based on the Internet of Things, which comprises the following steps: deploying a hyperspectral camera at a key position of a batching production line of a glass electric melting furnace, and identifying the types and positions of pollutants in materials in real time; environment sensors are arranged in the workshops, and digital twin bodies equal to the physical workshops in proportion are constructed; when pollutants are detected, the digital twin starts a simulation engine, a pollution source diffusion path is predicted, and a feature library is established; the system tracks the pollution track in real time, accurately calculates the shunting time, and automatically triggers the shunting valve to guide the polluted material into the waste pool; meanwhile, the air pressure gradient and the cleaning time of the clean room are dynamically adjusted to form an airflow barrier to block diffusion; and after cleaning, executing secondary monitoring and generating a traceability report. The method realizes full-process automatic management and control of pollution, improves production safety and efficiency, and reduces material loss.

The present invention discloses a computing offloading method based on a hybrid genetic algorithm in mobile edge computing, including: S1. Establishing a system model to obtain the computing delay of subtask sets on each processor and the transmission delay between processors, and determining the task layer values of each task in the subtask set according to the constraint relationship of the subtask set; S2. Initializing the population according to the determined task layer values and random strategy to obtain the initial population individuals of the subtask set, and performing symbolic encoding to obtain a task scheduling sequence, and optimizing the individuals in the initial population; S3. Constructing a fitness evaluation function and performing a selection operation on the individuals in the optimized initial population; S4. Constructing a crossover mechanism and performing crossover on the individuals in the new population using a crossover operation based on a tabu list search algorithm; S5. Performing a mutation operation on the individuals in the new population using a mutation operation based on a simulated annealing algorithm; S6. Judging whether the iteration step size is reached. If not, repeat steps S3-S5; if so, output the global optimal solution.

The present invention discloses a multi-agent reinforcement learning method for offloading computation in an Internet of Vehicles (IoV), comprising the following steps: step S1, constructing a system model; step S2, constructing a communication model; step S3, constructing a task and computation model; step S4, constructing an energy consumption model; step S5, constructing an optimization model and establishing a Markov decision process; step S6, using the MADDPG algorithm to perform neural network training on the optimization model; and step S7, deploying the trained network to each agent. The present invention uses the service vehicles and MEC servers in the IoV system as agents in reinforcement learning to make resource allocation decisions, and uses multi-agent reinforcement learning to allow each agent to obtain its own strategy network. After training, each agent can quickly output its own actions based on current local information without excessive communication, thus being able to cope with the rapidly changing IoV environment and reduce the system's energy consumption.

This invention discloses a task offloading and resource allocation method under D2D-assisted MEC, belonging to the field of wireless communication and computing resource management technology. The method introduces service-oriented devices assisted by terminal direct transmission communication technology and rationally selects task processing modes, achieving efficient resource utilization and avoiding resource waste. By jointly optimizing computation offloading, D2D selection, computing resource allocation, and spectrum resource allocation, it can both guarantee service quality and reduce network congestion, thereby improving user experience. The two-stage iterative algorithm of this invention combines block coordinate descent, reconstruction linearization technology, and convex optimization methods, effectively solving non-convex optimization problems, and has a fast convergence speed and low computational complexity. Furthermore, the D2D-assisted MEC architecture and optimization method are applicable to different types of intelligent devices and diverse application scenarios, possessing broad applicability and flexibility, and can meet the needs of future mobile edge computing development.

This invention discloses a neighborhood-constrained edge computing method for low-Earth orbit (LEO) satellites. The method includes: acquiring user task attributes, satellite-to-ground visibility relationships, inter-satellite topology relationships, and satellite computing resource status in the LEO satellite network; constructing a candidate access satellite set and a one-hop neighborhood set, and generating a local observation state vector; constructing a topology mask; constructing a multi-branch dual-depth Q-network to output user association decisions, offloading decisions, and resource allocation decisions; constructing a reward function based on task transmission delay, inter-satellite forwarding delay, computation delay, and system energy consumption, and training the multi-branch dual-depth Q-network; and using the trained model to perform collaborative computation offloading in LEO satellite edge computing. This method can reduce the dimensionality of collaborative offloading decisions, suppress Q-value overestimation, and improve the task success rate and energy consumption optimization performance in dynamic satellite networks.

This invention provides a trajectory optimization and computational unloading method and system based on hierarchical reinforcement learning, comprising: acquiring the position information of each UE and the UAV at the current time; employing a sub-target-based HRL framework, dividing the agent into an upper-layer network and a lower-layer network, wherein the upper-layer network generates k cluster centers through a clustering algorithm, and uses the k cluster centers as a sub-target space, representing the expected location of the UAV; the upper-layer network selects a sub-target non-repeatingly every time slot, using the state space and action features as input; based on the sub-targets selected by the upper-layer network, the lower-layer network adjusts the flight trajectory of the UAV in real time in each time slot to achieve rapid response and optimization to the dynamic environment, guiding the UAV to the sub-target; after the UAV moves for a preset duration in each time slot and remains stationary, the lower-layer network first determines the priority of each UE based on the distance between each UE and the UAV, and then makes a computational unloading decision. This invention can perform trajectory optimization and computational unloading.

The present invention discloses a multi-task domain adaptation method for computational offloading. The method comprises: constructing an objective function based on the scenario in which the user processes the target task on their own or offloads it to an edge server for processing, obtaining multiple sets of source domain data, each set of source domain data including sample data and labels of multiple users, the labels being initial classification values ​​and initial regression values; inputting the multiple sets of source domain data into a feedforward neural network model to obtain a trained feedforward neural network model; obtaining multiple sets of target domain data, enhancing the multiple sets of target domain data to obtain multiple sets of enhanced target domain data; wherein each set of target domain data includes sample data of multiple users. The present invention solves the technical problems of the existing task offloading model in different domain environments, such as low generalization ability, high privacy risk, large computational load, and long computational time.

The application provides a kind of based on NOMA's perception data acquisition and computing offloading method, by establishing the optimization problem model based on NOMA's perception data acquisition and computing offloading with the goal of maximizing offloading data volume;Simplify optimization problem model;The simplified optimization problem model is converted into two-layer optimization problem of inner and outer nesting;Get terminal scheduling variables and acquisition and offloading time allocation variables;According to terminal scheduling variables, schedule acquisition terminals in turn according to acquisition and offloading time allocation variables within the specified time based on NOMA for data offloading, and the acquisition terminals that have not been scheduled continue to carry out data acquisition;Base station calculates data according to the calculation resource allocation within the set time;Compared with the existing method, the method can effectively improve the data acquisition and computing efficiency, can maximize the offloading data volume, and the performance improvement is more obvious when the total computing resource is larger.

The present disclosure relates to a system and method for dynamic redundancy-aware blockchain-based partial computation offloading for a metaverse within a computing environment in a network, and according to the present disclosure, in order to improve the QoS of a metaverse service, it is possible to provide an environment that may perform existing traditional task offloading, perform optimal offloading through an in-network computing agent, and provide an expandable network and an ultra-low latency service. In addition, it is possible to maximize incentives while minimizing the overhead of computation execution costs incurred when a user performs a task, and satisfy the constraints on latency and blockchain offloading costs.

This invention discloses a joint optimization method and system for task offloading and resource allocation based on SAGIN. It collects task and scheduling information through a SAG-IoT system; utilizes the task data and scheduling information to achieve optimal association between UAVs and IoT devices based on a matching game algorithm, and makes optimal decisions for task offloading to maximize the computational performance of the SAG-IoT system; based on optimal association, optimal task offloading decisions, and maximizing the computational performance of the SAG-IoT system, it constructs an online offloading framework based on deep reinforcement learning, constructing a path from input X(t) to optimal action x. * The low-complexity mapping strategy (t) is based on repeated interactions between different modules and a random environment, iterating and running sequentially to achieve joint optimization of task offloading and resource allocation. This invention achieves optimal computational performance while stabilizing the system queue. In addition to utilizing binary computation offloading, the optimization framework can also be extended to online partial computation offloading strategies consisting of multiple independent subtasks.

The invention discloses a calculation unloading method based on deep reinforcement learning in a cellular-free large-scale MIMO system, and the method comprises the following steps: inputting environment information into a task transmission MADDPG network in a time slot t, and outputting a current access point connection strategy through minimizing the task transmission energy consumption of a user; for the kth user side, all access points connected with the kth user side are clustered according to the access point connection strategy of the kth user side, a plurality of access point clusters are obtained, each access point cluster unloads the MADDPG network through calculation, energy consumption is calculated through minimization of tasks, and the access point cluster unloads the MADDPG network through calculation; selecting a main access point and allocating computing resources to perform computing unloading on a task of the kth user side, returning computing result data to the kth user side after computing unloading is completed, and ending the time slot; repeating the above process to realize calculation unloading; according to the method, the distributed cooperation characteristic of the access points in the non-cellular large-scale MIMO is fully utilized, and the total energy consumption and algorithm complexity of the system are reduced while the communication reliability requirement is met.

The application discloses a user scheduling-oriented calculation unloading method for an edge cell-free network, and relates to the technical field of edge computing (MEC) and cell-free network. After modeling an MEC-assisted cell-free communication model and a parallel calculation unloading model to construct an uplink network simulation environment, a joint optimization calculation unloading, user scheduling and calculation resource allocation strategy (JCSCS) model considering limited orthogonal pilot resources and edge calculation resources is constructed, deep reinforcement learning training based on a proximal policy optimization (PPO) is carried out in the uplink network simulation environment, and finally, the best calculation unloading strategy, user scheduling strategy and calculation resource allocation strategy are obtained through the PPO. The uplink channel is divided into multiple orthogonal sub-channels in the application, user equipment is dynamically scheduled to the sub-channels for unloading, and the influence of multi-user interference and pilot contamination on unloading delay can be significantly reduced.

A kind of computing offloading and resource allocation method based on deep reinforcement learning, aiming at the problem of high task computing delay in multi-user multi-edge server scenario, with the goal of minimizing the total task processing delay of users. By selecting the deep reinforcement learning method based on value function, the D3QN algorithm combining Double-DQN algorithm and Dueling-DQN algorithm based on DQN algorithm with GRU network is proposed for edge computing computing offloading and resource allocation processing method, so as to improve the resource utilization of edge server and effectively alleviate the overestimation and Q value uniqueness problem in DQN algorithm, and obtain a better offloading scheme. Finally, the algorithm is simulated and compared with various baseline algorithms, which proves that the proposed algorithm can achieve the purpose of optimizing the total task processing delay of users and reducing the task drop rate.

The invention belongs to the technical field of Internet of Vehicles, and particularly relates to a task-driven vehicle calculation unloading method. The method comprises the following steps: firstly, training a precision neural network estimation model; establishing a time delay model and a calculation cost model of the vehicle target detection task in different unloading modes; secondly, according to the estimation model, estimating the precision of image acquisition of the vehicle end and the road end, according to the time delay model and the calculation cost model, calculating the time delay and the calculation cost corresponding to the service, and according to the estimation precision, the task completion time delay and the task completion cost, constructing an optimization problem; solving the optimization problem through a genetic algorithm to obtain an optimal solution; and finally, according to the obtained optimal solution, completing selection decision, resource allocation and calculation unloading of the vehicle-mounted service. According to the invention, a precision neural network prediction model is trained, and on the basis, a task-driven vehicle calculation unloading method is provided, so that the perception precision is maximized in an acceptable time delay and calculation cost range.

The application discloses a kind of unmanned plane assisted computing unloading method and device, belong to communication technical field, this method includes: constructing unmanned plane assisted computing unloading method corresponding unloading platform;Unmanned plane is constructed as leader, operator is follower, and spectrum transaction model based on Stackelberg game;Respectively, unmanned plane and operator optimization problem analysis is carried out;According to the analysis result, generate comprehensive optimization problem, find out game equilibrium point, obtain the optimal computing unloading method of computing unloading request user in the communication coverage range of unmanned plane, outside the communication coverage range of unmanned plane, request user passes through D2D relay technology, utilize idle user in the coverage range of unmanned plane as relay object, and the computing task is unloaded to unmanned plane and handled.The application can solve the problem that terminal in communication edge or blind area cannot directly communicate with unmanned plane, leading to computing task difficult to unload or local computing delay too long.

Systems and methods for offloading computation operations from a device to the network are provided. One or more network functions and corresponding connections can be configured in accordance with a request from the device for modifying a current data session between the device and network function, that is in communication with a data network that hosts an application service, for offloading computation operations associated with the data session to the network. Another network function is configured according to the request and has resources for processing the computation operations. Further systems and methods for releasing offloading computation operations and resources in the network associated therewith, associated with a data session between the device and the other network function, are provided.

Provided are systems, methods, and apparatuses for transferring computational tasks. In one or more examples, the systems, methods, and apparatuses include a first host configured to detect a trigger to offload instruction code from the first host to a second host; identify, based on the trigger, an address translation binding for the instruction code and an address translation binding for application data associated with the instruction code; copy the address translation binding for the instruction code and the address translation binding for the application data to a memory; and transfer control of execution of the instruction code to the second host based on the copying.

This invention provides a multi-satellite-assisted AI task computation offloading method and system for ocean-going vessels, belonging to the field of satellite communication and satellite computing technology. The method includes: before initiating an AI task computation offloading application, the vessel selects assisting satellites from a satellite system based on its environmental parameters and AI task requirements; after selecting an assisting satellite, if it determines that a satellite is currently assisting in executing another AI task and is not the selected assisting satellite, it enters a satellite switching process after the current task is completed; otherwise, it directly enters the access process for the selected assisting satellite; the vessel submits a computational task application to the satellite system to access the assisting satellite, and the satellite system generates a computation offloading plan based on the current satellite resource status and the vessel's AI task requirements. In the assisting satellite selection process, satellites with sufficient computing power are matched based on the vessel's own environment and AI task requirements, achieving efficient and accurate docking of satellite resources and vessel tasks, thus improving the efficiency of computation offloading.

The invention discloses a deep reinforcement learning calculation unloading method and system based on a multi-reviewer mechanism, and the method comprises the steps: obtaining a local state vector of an equipment user, carrying out the decision making through a multi-reviewer near-end strategy optimization algorithm, and obtaining an action decision of the equipment user in a current state; data interaction with the environment is carried out, the interaction time delay and the interaction energy consumption of the equipment user are obtained, calculation is carried out according to the completion condition of data interaction, and a time delay reward and an energy consumption reward are obtained; based on a centralized training distributed execution mechanism and a multi-commentator mechanism, performing training updating on the action decision of the device user to obtain a trained device user; and based on the trained equipment user, carrying out independent decision and executing a calculation unloading task. According to the invention, intelligent agents can be helped to coordinate with each other and make decisions independently. The deep reinforcement learning calculation unloading method and system based on the multi-commentator mechanism can be widely applied to the technical field of industrial Internet of Things edge calculation.

This invention proposes a method for computing offloading and resource allocation in the Internet of Vehicles (IoV) based on an improved Black Kite optimization algorithm. The method comprises the following steps: obtaining the computational latency and computational energy consumption of tasks in various computing scenarios based on multiple computing scenarios; calculating the total computational latency and total computational energy consumption of the tasks based on the computational latency and computational energy consumption of the tasks in various computing scenarios; defining a system utility function based on the total computational latency and total computational energy consumption of the tasks; constructing a computational latency and energy consumption model for mobile edge computing scenarios with the goal of minimizing the system utility function; designing a Black Kite optimization algorithm based on an elite reverse learning strategy, a Gompertz model, and a Gaussian mutation strategy; and using the designed Black Kite optimization algorithm to solve the computational latency and energy consumption model for mobile edge computing scenarios to obtain a final computation offloading and resource optimization allocation solution. This invention can accelerate the convergence of the algorithm and prevent the algorithm from prematurely falling into a local optimal solution.

This invention relates to a satellite-ground collaborative edge computing method for 6G space-ground integrated networks, comprising: constructing a system model for a mobile edge computing scenario; based on the system model, constructing a computation offloading model, a low Earth orbit satellite communication time model, and a channel model; based on the computation offloading model, the low Earth orbit satellite communication time model, and the channel model, constructing an offloading optimization problem with the goal of minimizing the weighted sum of total system latency and energy consumption; and iteratively solving the offloading optimization problem using a block coordinate descent algorithm, wherein the offloading optimization problem is decomposed into two sub-problems: task offloading optimization and resource allocation, which are solved separately. This invention solves the task allocation problem between satellites and base stations for multiple users by establishing a refined satellite-ground geometric model and resource scheduling model, minimizing the total system overhead while satisfying task deadlines and satellite service window constraints.

Systems and techniques for distributed processing are provided. For example, a process may include sending a discovery message; receiving a response to the discovery message, the response including an identifier of the neighboring device; sending an unloading request to an unloading server, wherein the unloading request comprises the identifier of the adjacent equipment; receiving, from the offload server, an indication that the neighboring device is selected to perform an offload process; sending data for processing to the neighboring device; receiving processed data from the neighboring device; and transmitting an indication of the amount of completed offload processing.

The invention provides a multi-type task adaptive calculation unloading method and system for the satellite Internet of Things, and the method is characterized in that the method comprises the following steps: S1 to S2, resource use conditions of the local Internet of Things device, each edge server corresponding to the local Internet of Things device and the satellite cloud are respectively collected as a current resource use condition of the device and a current resource use condition of other non-self devices; s3, taking the task attribute of the calculation task, the resource use condition of the current equipment and the resource use condition of other current non-self equipment as an input state space; s4, inputting the input state space into the unloading object decision model to obtain a current task unloading target; and S5, adopting the current task unloading target to process the calculation task. In a word, the method can optimize the long-term system overhead of the calculation task in the satellite Internet of Things.

The present disclosure relates to a method, a device, a system, and a computer program for processing-in-memory computation offloading for improving the inference performance of an artificial intelligence model. More specifically, the present disclosure provides a method for performing processing-in-memory (PIM) offloading by using a computing device, the method including: collecting information about a first computation to be processed, the first computation including an operator and at least one operand; determining the usefulness of offloading the first computation, based on the information about the first computation and the optimal operand size of a processing-in-memory (PIM); and offloading the first computation, based on the determination.