In many IIoT architectures,various devices connect to the edge cloud via gateway systems.For data processing,numerous data are delivered to the edge cloud.Delivering data to an appropriate edge cloud is critical to im...In many IIoT architectures,various devices connect to the edge cloud via gateway systems.For data processing,numerous data are delivered to the edge cloud.Delivering data to an appropriate edge cloud is critical to improve IIoT service efficiency.There are two types of costs for this kind of IoT network:a communication cost and a computing cost.For service efficiency,the communication cost of data transmission should be minimized,and the computing cost in the edge cloud should be also minimized.Therefore,in this paper,the communication cost for data transmission is defined as the delay factor,and the computing cost in the edge cloud is defined as the waiting time of the computing intensity.The proposed method selects an edge cloud that minimizes the total cost of the communication and computing costs.That is,a device chooses a routing path to the selected edge cloud based on the costs.The proposed method controls the data flows in a mesh-structured network and appropriately distributes the data processing load.The performance of the proposed method is validated through extensive computer simulation.When the transition probability from good to bad is 0.3 and the transition probability from bad to good is 0.7 in wireless and edge cloud states,the proposed method reduced both the average delay and the service pause counts to about 25%of the existing method.展开更多
With the arrival of 5G,latency-sensitive applications are becoming increasingly diverse.Mobile Edge Computing(MEC)technology has the characteristics of high bandwidth,low latency and low energy consumption,and has att...With the arrival of 5G,latency-sensitive applications are becoming increasingly diverse.Mobile Edge Computing(MEC)technology has the characteristics of high bandwidth,low latency and low energy consumption,and has attracted much attention among researchers.To improve the Quality of Service(QoS),this study focuses on computation offloading in MEC.We consider the QoS from the perspective of computational cost,dimensional disaster,user privacy and catastrophic forgetting of new users.The QoS model is established based on the delay and energy consumption and is based on DDQN and a Federated Learning(FL)adaptive task offloading algorithm in MEC.The proposed algorithm combines the QoS model and deep reinforcement learning algorithm to obtain an optimal offloading policy according to the local link and node state information in the channel coherence time to address the problem of time-varying transmission channels and reduce the computing energy consumption and task processing delay.To solve the problems of privacy and catastrophic forgetting,we use FL to make distributed use of multiple users’data to obtain the decision model,protect data privacy and improve the model universality.In the process of FL iteration,the communication delay of individual devices is too large,which affects the overall delay cost.Therefore,we adopt a communication delay optimization algorithm based on the unary outlier detection mechanism to reduce the communication delay of FL.The simulation results indicate that compared with existing schemes,the proposed method significantly reduces the computation cost on a device and improves the QoS when handling complex tasks.展开更多
The Internet of Medical Things(Io MT) is regarded as a critical technology for intelligent healthcare in the foreseeable 6G era. Nevertheless, due to the limited computing power capability of edge devices and task-rel...The Internet of Medical Things(Io MT) is regarded as a critical technology for intelligent healthcare in the foreseeable 6G era. Nevertheless, due to the limited computing power capability of edge devices and task-related coupling relationships, Io MT faces unprecedented challenges. Considering the associative connections among tasks, this paper proposes a computing offloading policy for multiple-user devices(UDs) considering device-to-device(D2D) communication and a multi-access edge computing(MEC)technique under the scenario of Io MT. Specifically,to minimize the total delay and energy consumption concerning the requirement of Io MT, we first analyze and model the detailed local execution, MEC execution, D2D execution, and associated tasks offloading exchange model. Consequently, the associated tasks’ offloading scheme of multi-UDs is formulated as a mixed-integer nonconvex optimization problem. Considering the advantages of deep reinforcement learning(DRL) in processing tasks related to coupling relationships, a Double DQN based associative tasks computing offloading(DDATO) algorithm is then proposed to obtain the optimal solution, which can make the best offloading decision under the condition that tasks of UDs are associative. Furthermore, to reduce the complexity of the DDATO algorithm, the cacheaided procedure is intentionally introduced before the data training process. This avoids redundant offloading and computing procedures concerning tasks that previously have already been cached by other UDs. In addition, we use a dynamic ε-greedy strategy in the action selection section of the algorithm, thus preventing the algorithm from falling into a locally optimal solution. Simulation results demonstrate that compared with other existing methods for associative task models concerning different structures in the Io MT network, the proposed algorithm can lower the total cost more effectively and efficiently while also providing a tradeoff between delay and energy consumption tolerance.展开更多
In this paper,we investigate the energy efficiency maximization for mobile edge computing(MEC)in intelligent reflecting surface(IRS)assisted unmanned aerial vehicle(UAV)communications.In particular,UAVcan collect the ...In this paper,we investigate the energy efficiency maximization for mobile edge computing(MEC)in intelligent reflecting surface(IRS)assisted unmanned aerial vehicle(UAV)communications.In particular,UAVcan collect the computing tasks of the terrestrial users and transmit the results back to them after computing.We jointly optimize the users’transmitted beamforming and uploading ratios,the phase shift matrix of IRS,and the UAV trajectory to improve the energy efficiency.The formulated optimization problem is highly non-convex and difficult to be solved directly.Therefore,we decompose the original problem into three sub-problems.We first propose the successive convex approximation(SCA)based method to design the beamforming of the users and the phase shift matrix of IRS,and apply the Lagrange dual method to obtain a closed-form expression of the uploading ratios.For the trajectory optimization,we propose a block coordinate descent(BCD)based method to obtain a local optimal solution.Finally,we propose the alternating optimization(AO)based overall algorithmand analyzed its complexity to be equivalent or lower than existing algorithms.Simulation results show the superiority of the proposedmethod compared with existing schemes in energy efficiency.展开更多
In this paper,we consider mobile edge computing(MEC)networks against proactive eavesdropping.To maximize the transmission rate,IRS assisted UAV communications are applied.We take the joint design of the trajectory of ...In this paper,we consider mobile edge computing(MEC)networks against proactive eavesdropping.To maximize the transmission rate,IRS assisted UAV communications are applied.We take the joint design of the trajectory of UAV,the transmitting beamforming of users,and the phase shift matrix of IRS.The original problem is strong non-convex and difficult to solve.We first propose two basic modes of the proactive eavesdropper,and obtain the closed-form solution for the boundary conditions of the two modes.Then we transform the original problem into an equivalent one and propose an alternating optimization(AO)based method to obtain a local optimal solution.The convergence of the algorithm is illustrated by numerical results.Further,we propose a zero forcing(ZF)based method as sub-optimal solution,and the simulation section shows that the proposed two schemes could obtain better performance compared with traditional schemes.展开更多
This paper investigates the age of information(AoI)-based multi-user mobile edge computing(MEC)network with partial offloading mode.The weighted sum AoI(WSA)is first analyzed and derived,and then a WSA minimization pr...This paper investigates the age of information(AoI)-based multi-user mobile edge computing(MEC)network with partial offloading mode.The weighted sum AoI(WSA)is first analyzed and derived,and then a WSA minimization problem is formulated by jointly optimizing the user scheduling and data assignment.Due to the non-analytic expression of the WSA w.r.t.the optimization variables and the unknowability of future network information,the problem cannot be solved with known solution methods.Therefore,an online Joint Partial Offloading and User Scheduling Optimization(JPOUSO)algorithm is proposed by transforming the original problem into a single-slot data assignment subproblem and a single-slot user scheduling sub-problem and solving the two sub-problems separately.We analyze the computational complexity of the presented JPO-USO algorithm,which is of O(N),with N being the number of users.Simulation results show that the proposed JPO-USO algorithm is able to achieve better AoI performance compared with various baseline methods.It is shown that both the user’s data assignment and the user’s AoI should be jointly taken into account to decrease the system WSA when scheduling users.展开更多
In edge computing,a reasonable edge resource bidding mechanism can enable edge providers and users to obtain benefits in a relatively fair fashion.To maximize such benefits,this paper proposes a dynamic multiattribute...In edge computing,a reasonable edge resource bidding mechanism can enable edge providers and users to obtain benefits in a relatively fair fashion.To maximize such benefits,this paper proposes a dynamic multiattribute resource bidding mechanism(DMRBM).Most of the previous work mainly relies on a third-party agent to exchange information to gain optimal benefits.It isworth noting thatwhen edge providers and users trade with thirdparty agents which are not entirely reliable and trustworthy,their sensitive information is prone to be leaked.Moreover,the privacy protection of edge providers and users must be considered in the dynamic pricing/transaction process,which is also very challenging.Therefore,this paper first adopts a privacy protection algorithm to prevent sensitive information from leakage.On the premise that the sensitive data of both edge providers and users are protected,the prices of providers fluctuate within a certain range.Then,users can choose appropriate edge providers by the price-performance ratio(PPR)standard and the reward of lower price(LPR)standard according to their demands.The two standards can be evolved by two evaluation functions.Furthermore,this paper employs an approximate computing method to get an approximate solution of DMRBM in polynomial time.Specifically,this paper models the bidding process as a non-cooperative game and obtains the approximate optimal solution based on two standards according to the game theory.Through the extensive experiments,this paper demonstrates that the DMRBM satisfies the individual rationality,budget balance,and privacy protection and it can also increase the task offloading rate and the system benefits.展开更多
With the rapid development of network and communication techniques,the teaching forms have become diversified.To enhance the education experience and improve the teaching environment,an increasing number of educationa...With the rapid development of network and communication techniques,the teaching forms have become diversified.To enhance the education experience and improve the teaching environment,an increasing number of educational institutions have adopted virtual simulation technology.A typical teaching mechanism is to exploit Virtual Reality(VR)technology,which affords participants an immersive experience.Unquestionably,such a VRbased mode is highly approved.However,the performance of this technology requires further optimization.On one hand,for VR 360video,the current intraframe decision cannot adapt to rapid response demands.On the other hand,the generated data size is considerably large and fast computation may not be realized,depending on the local VR device.Therefore,this study proposes an improved teaching mechanism empowered by edge computing–driven VR,called VE4T,that involves two parts.First,an intraframe decision algorithm for VR 360videos is devised to realize the rapid responses.Second,an edge computing framework is proposed to offload some tasks to an edge server for computation,where a task scheduling strategy is developed to check whether a task needs to be offloaded.Finally,experiments are performed using a practical teaching scenario with some VR devices.The obtained results demonstrate that VE4T is more efficient than existing mechanisms.展开更多
In a network environment composed of different types of computing centers that can be divided into different layers(clod,edge layer,and others),the interconnection between them offers the possibility of peer-to-peer t...In a network environment composed of different types of computing centers that can be divided into different layers(clod,edge layer,and others),the interconnection between them offers the possibility of peer-to-peer task offloading.For many resource-constrained devices,the computation of many types of tasks is not feasible because they cannot support such computations as they do not have enough available memory and processing capacity.In this scenario,it is worth considering transferring these tasks to resource-rich platforms,such as Edge Data Centers or remote cloud servers.For different reasons,it is more exciting and appropriate to download various tasks to specific download destinations depending on the properties and state of the environment and the nature of the functions.At the same time,establishing an optimal offloading policy,which ensures that all tasks are executed within the required latency and avoids excessive workload on specific computing centers is not easy.This study presents two alternatives to solve the offloading decision paradigm by introducing two well-known algorithms,Graph Neural Networks(GNN)and Deep Q-Network(DQN).It applies the alternatives on a well-known Edge Computing simulator called PureEdgeSimand compares them with the two defaultmethods,Trade-Off and Round Robin.Experiments showed that variants offer a slight improvement in task success rate and workload distribution.In terms of energy efficiency,they provided similar results.Finally,the success rates of different computing centers are tested,and the lack of capacity of remote cloud servers to respond to applications in real-time is demonstrated.These novel ways of finding a download strategy in a local networking environment are unique as they emulate the state and structure of the environment innovatively,considering the quality of its connections and constant updates.The download score defined in this research is a crucial feature for determining the quality of a download path in the GNN training process and has not previously been proposed.Simultaneously,the suitability of Reinforcement Learning(RL)techniques is demonstrated due to the dynamism of the network environment,considering all the key factors that affect the decision to offload a given task,including the actual state of all devices.展开更多
AI development has brought great success to upgrading the information age.At the same time,the large-scale artificial neural network for building AI systems is thirsty for computing power,which is barely satisfied by ...AI development has brought great success to upgrading the information age.At the same time,the large-scale artificial neural network for building AI systems is thirsty for computing power,which is barely satisfied by the conventional computing hardware.In the post-Moore era,the increase in computing power brought about by the size reduction of CMOS in very large-scale integrated circuits(VLSIC)is challenging to meet the growing demand for AI computing power.To address the issue,technical approaches like neuromorphic computing attract great attention because of their feature of breaking Von-Neumann architecture,and dealing with AI algorithms much more parallelly and energy efficiently.Inspired by the human neural network architecture,neuromorphic computing hardware is brought to life based on novel artificial neurons constructed by new materials or devices.Although it is relatively difficult to deploy a training process in the neuromorphic architecture like spiking neural network(SNN),the development in this field has incubated promising technologies like in-sensor computing,which brings new opportunities for multidisciplinary research,including the field of optoelectronic materials and devices,artificial neural networks,and microelectronics integration technology.The vision chips based on the architectures could reduce unnecessary data transfer and realize fast and energy-efficient visual cognitive processing.This paper reviews firstly the architectures and algorithms of SNN,and artificial neuron devices supporting neuromorphic computing,then the recent progress of in-sensor computing vision chips,which all will promote the development of AI.展开更多
基金supported by the National Research Foundation of Korea (NRF) grant funded by the Korea Government (MSIT) (No.2021R1C1C1013133)supported by the Institute of Information and Communications Technology Planning and Evaluation (IITP)grant funded by the Korea Government (MSIT) (RS-2022-00167197,Development of Intelligent 5G/6G Infrastructure Technology for The Smart City)supported by the Soonchunhyang University Research Fund.
文摘In many IIoT architectures,various devices connect to the edge cloud via gateway systems.For data processing,numerous data are delivered to the edge cloud.Delivering data to an appropriate edge cloud is critical to improve IIoT service efficiency.There are two types of costs for this kind of IoT network:a communication cost and a computing cost.For service efficiency,the communication cost of data transmission should be minimized,and the computing cost in the edge cloud should be also minimized.Therefore,in this paper,the communication cost for data transmission is defined as the delay factor,and the computing cost in the edge cloud is defined as the waiting time of the computing intensity.The proposed method selects an edge cloud that minimizes the total cost of the communication and computing costs.That is,a device chooses a routing path to the selected edge cloud based on the costs.The proposed method controls the data flows in a mesh-structured network and appropriately distributes the data processing load.The performance of the proposed method is validated through extensive computer simulation.When the transition probability from good to bad is 0.3 and the transition probability from bad to good is 0.7 in wireless and edge cloud states,the proposed method reduced both the average delay and the service pause counts to about 25%of the existing method.
基金supported by the National Natural Science Foundation of China(62032013,62072094Liaoning Province Science and Technology Fund Project(2020MS086)+1 种基金Shenyang Science and Technology Plan Project(20206424)the Fundamental Research Funds for the Central Universities(N2116014,N180101028)CERNET Innovation Project(NGII20190504).
文摘With the arrival of 5G,latency-sensitive applications are becoming increasingly diverse.Mobile Edge Computing(MEC)technology has the characteristics of high bandwidth,low latency and low energy consumption,and has attracted much attention among researchers.To improve the Quality of Service(QoS),this study focuses on computation offloading in MEC.We consider the QoS from the perspective of computational cost,dimensional disaster,user privacy and catastrophic forgetting of new users.The QoS model is established based on the delay and energy consumption and is based on DDQN and a Federated Learning(FL)adaptive task offloading algorithm in MEC.The proposed algorithm combines the QoS model and deep reinforcement learning algorithm to obtain an optimal offloading policy according to the local link and node state information in the channel coherence time to address the problem of time-varying transmission channels and reduce the computing energy consumption and task processing delay.To solve the problems of privacy and catastrophic forgetting,we use FL to make distributed use of multiple users’data to obtain the decision model,protect data privacy and improve the model universality.In the process of FL iteration,the communication delay of individual devices is too large,which affects the overall delay cost.Therefore,we adopt a communication delay optimization algorithm based on the unary outlier detection mechanism to reduce the communication delay of FL.The simulation results indicate that compared with existing schemes,the proposed method significantly reduces the computation cost on a device and improves the QoS when handling complex tasks.
基金supported by National Natural Science Foundation of China(Grant No.62071377,62101442,62201456)Natural Science Foundation of Shaanxi Province(Grant No.2023-YBGY-036,2022JQ-687)The Graduate Student Innovation Foundation Project of Xi’an University of Posts and Telecommunications under Grant CXJJDL2022003.
文摘The Internet of Medical Things(Io MT) is regarded as a critical technology for intelligent healthcare in the foreseeable 6G era. Nevertheless, due to the limited computing power capability of edge devices and task-related coupling relationships, Io MT faces unprecedented challenges. Considering the associative connections among tasks, this paper proposes a computing offloading policy for multiple-user devices(UDs) considering device-to-device(D2D) communication and a multi-access edge computing(MEC)technique under the scenario of Io MT. Specifically,to minimize the total delay and energy consumption concerning the requirement of Io MT, we first analyze and model the detailed local execution, MEC execution, D2D execution, and associated tasks offloading exchange model. Consequently, the associated tasks’ offloading scheme of multi-UDs is formulated as a mixed-integer nonconvex optimization problem. Considering the advantages of deep reinforcement learning(DRL) in processing tasks related to coupling relationships, a Double DQN based associative tasks computing offloading(DDATO) algorithm is then proposed to obtain the optimal solution, which can make the best offloading decision under the condition that tasks of UDs are associative. Furthermore, to reduce the complexity of the DDATO algorithm, the cacheaided procedure is intentionally introduced before the data training process. This avoids redundant offloading and computing procedures concerning tasks that previously have already been cached by other UDs. In addition, we use a dynamic ε-greedy strategy in the action selection section of the algorithm, thus preventing the algorithm from falling into a locally optimal solution. Simulation results demonstrate that compared with other existing methods for associative task models concerning different structures in the Io MT network, the proposed algorithm can lower the total cost more effectively and efficiently while also providing a tradeoff between delay and energy consumption tolerance.
基金the Key Scientific and Technological Project of Henan Province(Grant Number 222102210212)Doctoral Research Start Project of Henan Institute of Technology(Grant Number KQ2005)+1 种基金Doctoral Research Start Project of Henan Institute of Technology(Grant Number KQ2110)Key Research Projects of Colleges and Universities in Henan Province(Grant Number 23B510006).
文摘In this paper,we investigate the energy efficiency maximization for mobile edge computing(MEC)in intelligent reflecting surface(IRS)assisted unmanned aerial vehicle(UAV)communications.In particular,UAVcan collect the computing tasks of the terrestrial users and transmit the results back to them after computing.We jointly optimize the users’transmitted beamforming and uploading ratios,the phase shift matrix of IRS,and the UAV trajectory to improve the energy efficiency.The formulated optimization problem is highly non-convex and difficult to be solved directly.Therefore,we decompose the original problem into three sub-problems.We first propose the successive convex approximation(SCA)based method to design the beamforming of the users and the phase shift matrix of IRS,and apply the Lagrange dual method to obtain a closed-form expression of the uploading ratios.For the trajectory optimization,we propose a block coordinate descent(BCD)based method to obtain a local optimal solution.Finally,we propose the alternating optimization(AO)based overall algorithmand analyzed its complexity to be equivalent or lower than existing algorithms.Simulation results show the superiority of the proposedmethod compared with existing schemes in energy efficiency.
基金This work was supported by the Key Scientific and Technological Project of Henan Province(Grant Number 222102210212)Doctoral Research Start Project of Henan Institute of Technology(Grant Number KQ2005)Key Research Projects of Colleges and Universities in Henan Province(Grant Number 23B510006).
文摘In this paper,we consider mobile edge computing(MEC)networks against proactive eavesdropping.To maximize the transmission rate,IRS assisted UAV communications are applied.We take the joint design of the trajectory of UAV,the transmitting beamforming of users,and the phase shift matrix of IRS.The original problem is strong non-convex and difficult to solve.We first propose two basic modes of the proactive eavesdropper,and obtain the closed-form solution for the boundary conditions of the two modes.Then we transform the original problem into an equivalent one and propose an alternating optimization(AO)based method to obtain a local optimal solution.The convergence of the algorithm is illustrated by numerical results.Further,we propose a zero forcing(ZF)based method as sub-optimal solution,and the simulation section shows that the proposed two schemes could obtain better performance compared with traditional schemes.
基金supported in part by the Fundamental Research Funds for the Central Universities under Grant 2022JBGP003in part by the National Natural Science Foundation of China(NSFC)under Grant 62071033in part by ZTE IndustryUniversity-Institute Cooperation Funds under Grant No.IA20230217003。
文摘This paper investigates the age of information(AoI)-based multi-user mobile edge computing(MEC)network with partial offloading mode.The weighted sum AoI(WSA)is first analyzed and derived,and then a WSA minimization problem is formulated by jointly optimizing the user scheduling and data assignment.Due to the non-analytic expression of the WSA w.r.t.the optimization variables and the unknowability of future network information,the problem cannot be solved with known solution methods.Therefore,an online Joint Partial Offloading and User Scheduling Optimization(JPOUSO)algorithm is proposed by transforming the original problem into a single-slot data assignment subproblem and a single-slot user scheduling sub-problem and solving the two sub-problems separately.We analyze the computational complexity of the presented JPO-USO algorithm,which is of O(N),with N being the number of users.Simulation results show that the proposed JPO-USO algorithm is able to achieve better AoI performance compared with various baseline methods.It is shown that both the user’s data assignment and the user’s AoI should be jointly taken into account to decrease the system WSA when scheduling users.
基金supported in part by National Natural Science Foundation of China under Grant No.62172349,62032020,and 62172350the Research Foundation of Education Bureau of Hunan Province under Grant No.21B0139+1 种基金the National Key Research and Development Program of China under Grant 2021YFB3101200Hunan Science and Technology Planning Project under Grant No.2019RS3019.
文摘In edge computing,a reasonable edge resource bidding mechanism can enable edge providers and users to obtain benefits in a relatively fair fashion.To maximize such benefits,this paper proposes a dynamic multiattribute resource bidding mechanism(DMRBM).Most of the previous work mainly relies on a third-party agent to exchange information to gain optimal benefits.It isworth noting thatwhen edge providers and users trade with thirdparty agents which are not entirely reliable and trustworthy,their sensitive information is prone to be leaked.Moreover,the privacy protection of edge providers and users must be considered in the dynamic pricing/transaction process,which is also very challenging.Therefore,this paper first adopts a privacy protection algorithm to prevent sensitive information from leakage.On the premise that the sensitive data of both edge providers and users are protected,the prices of providers fluctuate within a certain range.Then,users can choose appropriate edge providers by the price-performance ratio(PPR)standard and the reward of lower price(LPR)standard according to their demands.The two standards can be evolved by two evaluation functions.Furthermore,this paper employs an approximate computing method to get an approximate solution of DMRBM in polynomial time.Specifically,this paper models the bidding process as a non-cooperative game and obtains the approximate optimal solution based on two standards according to the game theory.Through the extensive experiments,this paper demonstrates that the DMRBM satisfies the individual rationality,budget balance,and privacy protection and it can also increase the task offloading rate and the system benefits.
基金supported by the Approved Project of Jilin Undergraduate Higher Education and Teaching Reform 2020(General Project).
文摘With the rapid development of network and communication techniques,the teaching forms have become diversified.To enhance the education experience and improve the teaching environment,an increasing number of educational institutions have adopted virtual simulation technology.A typical teaching mechanism is to exploit Virtual Reality(VR)technology,which affords participants an immersive experience.Unquestionably,such a VRbased mode is highly approved.However,the performance of this technology requires further optimization.On one hand,for VR 360video,the current intraframe decision cannot adapt to rapid response demands.On the other hand,the generated data size is considerably large and fast computation may not be realized,depending on the local VR device.Therefore,this study proposes an improved teaching mechanism empowered by edge computing–driven VR,called VE4T,that involves two parts.First,an intraframe decision algorithm for VR 360videos is devised to realize the rapid responses.Second,an edge computing framework is proposed to offload some tasks to an edge server for computation,where a task scheduling strategy is developed to check whether a task needs to be offloaded.Finally,experiments are performed using a practical teaching scenario with some VR devices.The obtained results demonstrate that VE4T is more efficient than existing mechanisms.
基金funding from TECNALIA,Basque Research and Technology Alliance(BRTA)supported by the project aOptimization of Deep Learning algorithms for Edge IoT devices for sensorization and control in Buildings and Infrastructures(EMBED)funded by the Gipuzkoa Provincial Council and approved under the 2023 call of the Guipuzcoan Network of Science,Technology and Innovation Program with File Number 2023-CIEN-000051-01.
文摘In a network environment composed of different types of computing centers that can be divided into different layers(clod,edge layer,and others),the interconnection between them offers the possibility of peer-to-peer task offloading.For many resource-constrained devices,the computation of many types of tasks is not feasible because they cannot support such computations as they do not have enough available memory and processing capacity.In this scenario,it is worth considering transferring these tasks to resource-rich platforms,such as Edge Data Centers or remote cloud servers.For different reasons,it is more exciting and appropriate to download various tasks to specific download destinations depending on the properties and state of the environment and the nature of the functions.At the same time,establishing an optimal offloading policy,which ensures that all tasks are executed within the required latency and avoids excessive workload on specific computing centers is not easy.This study presents two alternatives to solve the offloading decision paradigm by introducing two well-known algorithms,Graph Neural Networks(GNN)and Deep Q-Network(DQN).It applies the alternatives on a well-known Edge Computing simulator called PureEdgeSimand compares them with the two defaultmethods,Trade-Off and Round Robin.Experiments showed that variants offer a slight improvement in task success rate and workload distribution.In terms of energy efficiency,they provided similar results.Finally,the success rates of different computing centers are tested,and the lack of capacity of remote cloud servers to respond to applications in real-time is demonstrated.These novel ways of finding a download strategy in a local networking environment are unique as they emulate the state and structure of the environment innovatively,considering the quality of its connections and constant updates.The download score defined in this research is a crucial feature for determining the quality of a download path in the GNN training process and has not previously been proposed.Simultaneously,the suitability of Reinforcement Learning(RL)techniques is demonstrated due to the dynamism of the network environment,considering all the key factors that affect the decision to offload a given task,including the actual state of all devices.
基金Project supported in part by the National Key Research and Development Program of China(Grant No.2021YFA0716400)the National Natural Science Foundation of China(Grant Nos.62225405,62150027,61974080,61991443,61975093,61927811,61875104,62175126,and 62235011)+2 种基金the Ministry of Science and Technology of China(Grant Nos.2021ZD0109900 and 2021ZD0109903)the Collaborative Innovation Center of Solid-State Lighting and Energy-Saving ElectronicsTsinghua University Initiative Scientific Research Program.
文摘AI development has brought great success to upgrading the information age.At the same time,the large-scale artificial neural network for building AI systems is thirsty for computing power,which is barely satisfied by the conventional computing hardware.In the post-Moore era,the increase in computing power brought about by the size reduction of CMOS in very large-scale integrated circuits(VLSIC)is challenging to meet the growing demand for AI computing power.To address the issue,technical approaches like neuromorphic computing attract great attention because of their feature of breaking Von-Neumann architecture,and dealing with AI algorithms much more parallelly and energy efficiently.Inspired by the human neural network architecture,neuromorphic computing hardware is brought to life based on novel artificial neurons constructed by new materials or devices.Although it is relatively difficult to deploy a training process in the neuromorphic architecture like spiking neural network(SNN),the development in this field has incubated promising technologies like in-sensor computing,which brings new opportunities for multidisciplinary research,including the field of optoelectronic materials and devices,artificial neural networks,and microelectronics integration technology.The vision chips based on the architectures could reduce unnecessary data transfer and realize fast and energy-efficient visual cognitive processing.This paper reviews firstly the architectures and algorithms of SNN,and artificial neuron devices supporting neuromorphic computing,then the recent progress of in-sensor computing vision chips,which all will promote the development of AI.