边缘计算
Mobile edge computing（MEC）makes it possible to deploy and provide service locally,which is close to the users,by enabling the IT and cloud computation capacity at the radio access network (RAN).Thus,MEC can improve the quality of experience (QoE) by reducing the latency,and decrease the network operation cost through reducing backhaul capacity requirement.Meanwhile,based on the network context information (location,network traffic load,radio information,etc) exposure to applications and services,MEC can further improve the QoE of user and provide the platform to third-party partners for application and service innovation.Besides the introduction of MEC,the detailed MEC platform was presented.Furthermore,the traffic offloading solution based on MEC was proposed and compared with the solution of local IP access and selected IP traffic offload (LIPA/SIPTO).In addition,the problems and challenges of the MEC were also discussed in detail.
As a cost-efficient way to meet the wide range of use cases that the fifth generation wireless network will provide,network slicing and edge computing have been advocated by both academia and industry.In the concept of network slicing,the network entities are sliced into several logical networks to provide the requested services for different use cases.The edge computing pushes the frontier of computing applications,data,and services away from centralized nodes to the logical extremes of a network and even the user equipments,which can improve the traditional mobile broadband service capability,and handle with the emerging machine type service.The edge computing based radio access network slicing was proposed as a combination of edge computing and network slicing.The radio access network slicing can meet the diverse use cases and business models in 5G,and enable operators to flexibly provide personalized network services based on third-party need and network situation in a cost-efficient way.
Mobile edge computing (MEC) technology integrates IT service environment and cloud computing technology at the edge of the network,improving the capacities of computing and storage of the edge of network,reducing network operation and service delivery delay and further enhancing the quality of experience (QoE) of users.Thus,MEC has attracted extensive attention and has been added in the 5G standard as a key technology.Firstly,the basic concept,reference architecture and technical advantages of MEC were introduced,and the current research state of MEC were summarized from three perspectives:academia,industry and standard setting.Then,the key enablers,application scenarios and specific deployment methods of MEC were introduced in detail.Finally,the existing problems of MEC,such as mobility,security and billing,as well as privacy protection,were discussed.
A new multi-mode node was designed for the emergency communication with the demand of electricity.The node had the multi-mode communication capability of the private network and the public network technology,which could switch smoothly in three modes:the power private network,the wireless Ad Hoc network and the wireless public network.Data were automatically uploaded to the multi-mode terminal,multi-mode terminal through the mode of rapid switching,with the public network resources to achieve rapid network recovery and continuous transmission of data to achieve off-line emergency communications.Meanwhile,an intelligent routing algorithm based on smoothness was proposed,and the three principles of edge network communication were defined.Through the edge calculation method,the calculation and updating of the smoothness were realized to ensure the intelligent,efficient and reliable operation of the edge network in the offline state.
With the rapid development and extensive application of the Internet of things (IoT),big data and 5G network architecture,the massive data generated by the edge equipment of the network and the real-time service requirements are far beyond the capacity if the traditional cloud computing.To solve such dilemma,the edge computing which deploys the cloud services in the edge network has envisioned to be the dominant cloud service paradigm in the era of IoT.Meanwhile,the unique features of edge computing,such as content perception,real-time computing,parallel processing and etc.,has also introduced new security problems especially the data security and privacy issues.Firstly,the background and challenges of data security and privacy-preserving in edge computing were described,and then the research architecture of data security and privacy-preserving was presented.Secondly,the key technologies of data security,access control,identity authentication and privacy-preserving were summarized.Thirdly,the recent research advancements on the data security and privacy issues that may be applied to edge computing were described in detail.Finally,some potential research points of edge computing data security and privacy-preserving were given,and the direction of future research work was pointed out.
With the rise of edge computing,more and more attentions have been paid to security issues of edge computing.The basic concepts,system architecture and the relationship between edge computing and other computing paradigms were introduced.Then the security threats to edge computing were analyzed,and the security technology of edge computing was discussed for these security threats.Finally,the technologies of intrusion detection,access control,defense strategy and key management in edge computing were summarized and the further research directions were pointed out.
Aiming at introducing edge computing into operator’s network,a bearing and deployment thinking based on NFV was presented.Firstly,the industry development trends of edge computing were introduced.Then the concept,service scenarios and special requirements of edge computing were described,as well as the relationship between edge computing and NFV.Then,a bearing thinking of edge computing based on NFV was provided.Furthermore,some preliminary discussions about deploying edge computing platform on operator’s network were offered.Techni cal reference foRthe future deployment of edge computing was provided.
Computation offloading in mobile edge computing would transfer the resource intensive computational tasks to the edge network.It can not only solve the shortage of mobile user equipment in resource storage,computation performance and energy efficiency,but also deal with the problem of resource occupation,high latency and network load compared to cloud computing.Firstly the architecture of MEC was introduce and a comparative analysis was made according to various deployment schemes.Then the key technologies of computation offloading was studied from three aspects of decision on computation offloading,allocation of computing resource within MEC and system implement of MEC.Based on the analysis of MEC deployment scheme in 5G,two optimization schemes on computation offloading was proposed in 5G MEC.Finally,the current challenges in the mobility management was summarized,interference management and security of computation offloading in MEC.
In order to cope with the challenges brought by the “internet of everything”,mobile edge computing has been highly valued by academics and industry for its unique advantages.However,traditional IP networks are unable to provide network services dynamically in an efficient way for mobile edge computing,due to deficiencies from their original design.To this end,based on emerging network technologies,a solution for flexible adaptation of network services in mobile edge computing scenarios was proposed,aiming at realizing on-demand and effective deployment of network services to meet the diversified and personalized needs of different users.
Mobile edge computing (MEC) is an innovative computing paradigm.Offloading computing tasks to MEC server can effectively solve the contradiction between resource-constrained of mobile user and computing-intensive applications.Moreover,it can meet the delay of computing task and reduce energy consumption of mobile user.The current research results and achievements of computation offloading scheme in MEC system were summarized and the existing problems and challenges were analyzed.Furthermore,the development directions of MEC in the future were explored.
Edge computing is a new type of computing models that performs computing tasks at the edge of the network.Compared with cloud computing,it can respond to user’s needs more quickly and reliably.Starting from the shortcomings of cloud computing,the concept and general architecture of edge computing were illustrated,and then two reference frameworks proposed by industry alliances were elaborated.Four challenges of edge computing and their latest research progress were summarized.With the development of theory and technology related to edge computing,it will become a key technology to promote the upgrade of Internet of things (IoT) services.For this reason,two applications of edge computing in manufacturing and security monitoring were introduced.
The traditional centralized architecture,known as cloud computing,cannot accommodate such user demands in an efficient and timely manner.To cope with this problem,edge computing architectures have been proposed with the core concept of that“data processing should be close to the data source”.Firstly,paradigms of edge computing was introduced,including micro data center,cloudlet,fog computing,and mobile edge computing,and the advantages of edge computing from the perspective of resource integration was discussed.Then,related works of resource optimization in edge computing was reviewed and summarized,and these works was discussed via three directions,i.e.,computation offloading,distributed caching and high performance transmission,corresponding to core resources as computing,storage and communication.Finally,trends of development and future directions were presented as well.
With the continuous development of Internet of things applications,a large number of mobile terminal devices participate in service computing.Traditional cloud computing models cannot adapt to the rapid growth of data generated by network edge devices,and edge computing models emerge and become a research hotspot in recent years.The concept of edge computing and the reference model of edge computing in the Internet of things were introduced,the vulnerabilities of edge devices and the main research results of cryptographic security technology in edge computing were summarized,and that symmetric cryptography technology is not suitable for communication between edge devices,and identity-based cryptography technology is more suitable for communication between edge devices and edge devices was pointed out.Paired-based cryptography is more suitable for the communication between edge devices and base stations.The application of two post-quantum cryptography technologies in edge devices was discussed.Finally,some suggestions on the research of edge computing security technology were put forward.
Artificial intelligence (AI) and edge computing (EC) represent two of today’s most popular technologies.There is a great potential to coordinate these two emerging techniques to facilitate the further advent of both sides.Through three research cases,the profound benefits were demonstrated when AI and EC synergize.Specifically,from the perspective of EC for AI,to efficiently run deep learning at the network edge,a collaborative and on-demand deep neural network (DNN) co-inference framework with device-edge synergy was proposed.By applying DNN partitioning and right-sizing,it minimizes the inference latency under target accuracy.On the other hand,from the perspective of AI for EC,for the dynamical placement of edge computing services,two methods were proposed:an online-learning based adaptive service migration strategy and a factor graph model driven predictive service migration technique.
How to mitigate such pressure has become a critical challenge for global grid enterprises.Empowered by the development of IoT techniques,high-speed communications,and AI-chips,traditional grid enterprises are trying to adopt these new techniques to solve the challenge,especially in heterogeneous data collection,energy scheduling,and AI-enabled monitoring.Firstly,the background of smart grid and the history of edge computing were introduced.Then,edge computing enabled solutions were illustrated in three typical grid scenarios.Finally,two edge computing platforms at home and abroad were introduced.The main technologies and application scenarios of them were also analyzed.
The application of cloud computing and image processing technology has greatly changed the way of forest fire monitoring.The forest fire monitoring system based on cloud architecture has shortcomings in real-time detection and image algorithm configuration.In order to solve these problems,a forest fire monitoring system based on edge computing was designed.This system adopts the edge computing method and uses edge computers or servers in forest monitoring stations to perform image processing tasks of fire detection,which significantly improves the real-time performance of fire detection and early warning.On the other hand,the system introduces a function module for algorithm reconfiguration,which facilitates the iteration and update of the algorithms,reduces the cost of system redevelopment,and further improves the practicability of the system.
In order to solve the problem that traditional mobile edge computing network can’t be straightforwardly applied to the Internet of vehicles (IoV) due to high speed mobility and dynamic network topology,a vehicular edge multi-access computing network (VE-MACN) was introduced to realize collaborative computing offloading between roadside units and smart vehicles.In this context,the collaborative computation offloading was formulated as a joint multi-access model selection and task assignment problem to realize the good balance between long-term system utility,diverse needs of IoV applications and energy consumption.Considering the complex joint optimization problem,a deep reinforcement learning-based collaborative computing offloading scheme was designed to overcome the curse of dimensionality for Q-learning algorithm.The simulation results demonstrate that the feasibility and effectiveness of proposed offloading scheme.
In mobile edge computing system,the quality of computing experience can be improved greatly by offloading computing tasks from mobile devices to mobile edge computing servers.Consider incorporating renewable energy into a multi-user mobile edge system.Moreover,a battery as an energy harvesting device was added to the model to harvest energy and storage.The task allocation strategy in mobile edge computing system was formulated through the resource management algorithm based on reinforcement learning,which achieved the cost minimization of mobile devices (including delay cost and computing cost).The simulation results show that the proposed algorithm significantly minimizes the cost of mobile devices compared with other algorithms.
In order to meet the requirements of higher computing speed,lower communication delay and higher access density in the future Internet of things,a computational framework of EdgeFlow for mobile edge computing and a task unloading algorithm considering blocking and non-blocking state were proposed.The simulation experiments on the universal software radio platform show that the computational efficiency can be improved significantly and the communication delay can be reduced by EdgeFlow.It points out a new development direction,lays a theoretical and practical foundation for solving the problem of higher computational requirements of the Internet of things.And it will be widely used in the Internet of things scenarios.
Based on the advantages of high-bandwidth and low-latency brought by mobile edge computing (MEC),which could provide IT service environment and cloud computing capability,combined with the long-term evolution unlicensed (LTE-U) technology,the task offloading decision and resource allocation issues in vehicle heterogeneous network were studied.Considering the link differentiation requirements,which were the high capacity of vehicle-to-roadside unit (V2I) links and the super reliability of vehicle-to-vehicle (V2V) links,quality of service (QoS) was modeled as the combination of capacity and latency.Firstly,the improved K-means algorithm was used to cluster the request vehicles according to different QoS to determine the communication mode.Secondly,the LTE-U technology based on non-competition period (CFP) which was combined with carrier aggregation (CA) technology,and the distribution Q-Learning algorithm were adopted to allocate the channel and power.The simulation results show that the proposed mechanism can maximize the V2I link traversal capacity while ensuring the reliability of the V2I link.
Given the advantages of low cost and easy deployment,large-scale Internet of things (IoT) has been deployed for environment monitoring pervasively.Within such systems,cloud platform is typically utilized as a remote data and control center.However,tremendous amount of data uploading and processing induce huge challenges on bandwidth load and real-time data gathering.In order to overcome these challenges,edge computing enabled IoT system architecture was proposed for environmental monitoring.As the intermediate layer,local processing could be supported for end devices with low latency and assist with preliminary analysis to offload computational tasks from cloud and the amount of data uploading could be reduced.Based on this system architecture,an autoencoder neural network-based abnormal data detection scheme was developed newly.Performance evaluation has been conducted based on the practical oceanic atmospheric data.Simulation results indicate that the proposed scheme can accurately detect the abnormal data by fully exploiting the spatial data correlation.
Edge computing integrate the computing,storage and other resources on the edge of the network into a unified user service platform.According to the principle of nearest service,the network edge node task request is timely and correspondingly processed.Computational migration is a key issue in edge computing due to limited edge node capabilities,resources,bandwidth,and energy.The causes,evolution and development trends of mobile edge computing were reviewed from the distributed computing,pervasive computing,cloud computing to edge computing as the main line.The characteristics of computing migration at each stage were compared and analyzed,and the classical models were reviewed.The latest research progress and application fields of edge computing were focused on,mobile edge computing models based on energy optimization management were researched and compared.Finally,a time division multiple access based multi-user edge mobile terminal computing migration strategy system for LTE applications was proposed.
Edge computing is a method of processing data physically close to where data is generated,and it is one of the key technologies of 5G.The development of network technology and the demand of business applications have jointly promoted the maturity of 5G edge computing.The 5G edge computing technology system was elaborated through the perspective of edge infrastructure,edge computing network and edge platform,and the 5G edge computing application was prospected.
As a new architecture,mobile edge computing gives edge users stronger capabilities of computing,storage and communication,but it needs reasonable incentives mechanism to motivate edge users to provide resources.In terms of the three typical scenarios of mobile intelligent edge computing:computation offloading,edge caching and data collection,the incentive mechanism in the above scenarios was studied at first,then the core scientific problems were proposed that need to be solved in the incentive mechanism design of mobile intelligent edge computing from three perspectives of service quality,network quality and data quality.Finally,the technical challenges in the process of solving the above problems were analysed deeply and the corresponding feasible solutions were given.
The demand is getting higher and higher for data processing due to big data volume, thus, data-intensive service shave emerged. When solving complex problems, multiple data-intensive services are often united as a service portfolio. Due to the huge data transmission between service components, a great transmission delay will affect the overall performance of the system. In the edge computing environment, an optimized deployment strategy based on the negative selection algorithm was proposed to reduce the data transmission time in the service composition. Firstly, the definition of such a data-intensive service component deployment problem was given, and the deployment problem was modeled as an optimization model; then, a negative selection algorithm was designed to obtain the best deployment solution. In order to evaluate the applicability and convergence of the algorithm, it was compared with genetic algorithm and simulated annealing algorithm. The results show that proposed algorithm outperforms other schemes in this data-intensive service deployment problem.
Edge computing has become an important innovative business model in the 5G era,especially its low latency characteristics,which are considered to be unavailable in traditional solutions.Therefore,edge computing can provide more service capabilities and more application scenarios.However,the synergy of computing resources between edge computing and cloud computing has become a new technical problem,so it is necessary to realize cloud network collaboration,cloud edge collaboration and even edge collaboration between edge computing,cloud computing and network,so as to achieve the optimization of resource utilization.A computing network solution based on cloud,network and edge depth fusion was introduced,and a typical system for AI application was proposed,which could effectively cope with the future.
The highly efficient network resources are provided by multi-access edge computing at the edge of the network,but high security capability is required also due to its distributed position and organization.Based on mimic defense theory,mimic defense structure for distributed multi-access edge computing was proposed.By segmenting data,padding check data and processing data at multiple edge node,dynamic scheduling and decision-making functions according to checksum were implemented.The simulation results show that with the increase of delay cost,the data manipulation and leak rates can be reduced effectively by the proposed structure.The edge node scheduling strategy based on trust and cost is proposed to improve the efficiency and security of the system.
In order to guarantee the quality of service of edge computing,a trusted cooperative task distribution strategy for edge computing under multi-dimensional constraints was proposed.The strategy was based on the task requirements,and the edge computing collaborative service leader node organizes the coordinated service alliance members.The priority of cooperative service alliance was determined based on the K-dimension weight index of user task migration.Using the load balancing of the members as the adaptive function,the task allocation and dispatch of alliance members were performed by greedy algorithm.The backup node was selected based on routing piggyback,and the priority of migration was evaluated.The scheduling and migration of collaborative services in case of abnormal collaborative services were realized,which improved the quality of service of edge computing task migration and ensured the reliability of task migration.The simulation results show that the mechanism can effectively complete the task distribution and migration scheduling,improve the collaborative efficiency of edge computing,and guarantee the quality of network service.
With the development of the Internet of things,multiple real-time applications will emerge in future 6G communication network.Driven by demands on low-latency services of multiple real-time applications,mobile edge computing (MEC) will become an important technology to improve the user experience and reduce the network cost.However,the computing capacity of a single MEC server is limited,which induces that it is difficult to meet the low-latency requirement of data processing for the computation-intensive applications.A heterogeneous multi-layer mobile edge computing (HetMEC) network architecture was proposed to jointly utilize the computing and transmission resources of both the cloud computing center and multi-layer MEC servers.By reasonably dividing and offloading computing tasks,reliable and efficient computing services were provided for edge applications.Simulation results show that the proposed algorithm could effectively reduce the processing latency,improve the network processing rate and robustness.
Aiming at the fact that the existing encrypted image retrieval schemes do not consider different keys to encrypt images,a multi-key encrypted image retrieval system based on edge computing (including basic scheme and enhanced scheme) based on local sensitive hashing,secure nearest neighbor and proxy re-encryption technologies was proposed.The retrieval efficiency and accuracy were improved and the extra computational cost of query users was reduced.Security analysis shows that the basic scheme can only resist the known ciphertext attack,while the enhanced scheme can resist the known background attack.The experimental performance test based on the real-world dataset shows that the proposed schemes are feasible in practical application scenarios.
Aiming at maximizing the cache revenue,a joint user association and power allocation algorithm was proposed for the matching of base stations with users and power allocation in a NOMA heterogeneous network with caches,a NOMA joint optimization algorithm which was achieved by combing with message passing and DC programming was proposed.First,the constraints were incorporated into the objective function,and the user association result was obtained by calculating the marginal of the message passing between the function node and the variable node in the new optimization problem.Then,the original optimization problem was transformed into the form of the difference between two convex functions,and allocate power resources through DC programming.Finally,the final user association and power allocation results are obtained through iterative calculations.Simulation results prove that the proposed algorithm effectively improves network performance.
In order to solve the problem of the high requirements of data transmission rate and sensitivity to transmission delay in virtual reality (VR) based on cloud services,a Cloud VR system with MEC (mobile edge computing) was proposed,mainly including viewpoint-based VR processing and HDA (hybrid digital-analog) transmission optimization.Firstly,a dynamic streaming method based on user viewpoint and pyramid projection was used to implement a complete edge cloud VR system.Then,HDA transmission was introduced to optimize the transmission,and a heuristic algorithm for resource allocation was given.Finally,the base station protocol stack was transformed,and the MEC was integrated into the LTE (long term evolution) system to implement a complete mobile edge cloud VR system.Experimental results demonstrate that the proposed scheme has good robustness and efficient transmission by comparing with the existing schemes.
The popularity of smart devices has driven the development of the application of Internet of things (IoT) technology,and the resulting massive amount of IoT data has brought challenges to traditional centralized data management methods,such as performance,privacy,and security.Therefore,a data management framework of IoT based on blockchain and edge computing was proposed to support the distributed IoT data management.The distributed storage and access control could be provided by the framework for the IoT data.At the same time,a set of the built-in encryption scheme was designed to protect the data security and privacy and safeguard the data ownership.By introducing edge computing,the scalability bottleneck of the blockchain the system was relieved.The processes of data storage and data access based on this framework were given,and the algorithm of the system implementation based on the smart contract technology was explained in detail.Experiments show that the IoT data management system based on this framework outperforms the traditional cloud-based data management systems.
In the industrial Internet of things,limited by the computing capacity of field devices,the task offloading based on edge computing can effectively alleviate the computing burden of field devices and provide low-latency computing services.Moreover,because the load of edge servers are different in different areas of the network,it is necessary to reasonably arrange task offloading and allocate computing resources of edge servers,thereby reducing task completion delay and achieving load balance.Thus,the task offloading and resource allocation for edge computing in the industrial Internet of things was studied,a cross-domain offloading model for computing tasks in the industrial Internet of things was proposed,and a mixed integer nonlinear optimization problem that minimizes task completion time was formulated.The problem was decomposed it into two sub-problems of resource allocation and task offloading,based on the characteristics of the two sub-problems,the optimal solution of resource allocation and task offloading strategy were obtained through iterative and alternating solution.The experimental results show that compared with the non-cross-domain strategy,the load imbalance of the edge server and the task completion delay are reduced effectively by the proposed strategy.
The three application scenarios of 5G make forward higher requirements for network delay,bandwidth,connection and security,which drive 5G to sink the core capabilities of network,computing,storage and application to the edge of the network.The 5G base station which is closer to the terminal/user and has complete resources such as base station power,operation and maintenance will be the best choice for 5G edge computing deployment.Based on the multi-point 5G base station site resource and the architecture of edge computing deployment,the feasibility of multi-operator 5G edge computing shared deployment was studied,three feasible co-building and sharing schemes,such as base station infrastructure sharing,IaaS facility sharing and PaaS platform sharing were proposed.By coordinating the development needs of 5G edge computing and realizing the co-construction and sharing of edge computing,the construction cost and operation cost of edge computing can be greatly reduced,and the rapid scale development of edge computing can be realized,to build an intensive and efficient,economic and applicable,intelligent green,safe and reliable edge computing system.
With the rapid development of internet of things technology,in order to better develop modern medicine,avoid the “isolated” crisis of information,and meet the requirements of timeliness and computing performance of massive data generated by edge devices,it is a new trend of the times to combine edge computing with smart healthcare treatment.However,edge computing has a certain degree of openness,and it is prone to potential safety hazards.Therefore,people pay more and more attention to the security and privacy protection of edge computing system for smart healthcare treatment.The edge computing system and its architecture for smart healthcare were introduced,and the security risks encountered by smart healthcare in edge computing was explained,and some scholars’ solutions to the security risks were introduced.Finally,a security framework,and specific solutions for security and privacy protection under the framework were put forward,which providing some help for the trusted research of smart healthcare edge computing.
With the popularization of video surveillance systems and the diversification of their applications,traditional data processing methods using cloud computing to handle all the computations are more and more unsustainable.Edge computing pre-processes video data nearby,and performs well in terms of bandwidth,storage,and latency,but edge-cloud collaboration is still required to improve overall performance.The architecture of edge-cloud collaborative video surveillance system was proposed,then an open edge node design scheme and algorithm training and inferencing ideas were proposed.Finally,taking forest fire prevention and smart poles ＆amp; towers as examples,the application realization and value of edge computing in video surveillance services were illustrated.
Information security monitoring system is widely used in railway construction projects,but the existing safety supervision model has problems such as high server load,heavy network bandwidth pressure,monitoring data can’t be effectively shared,key monitoring data may be tampered,and security accident responsibility is difficult to determine.A credible sharing model of safety supervision data for railway construction projects based on edge computing was proposed.On one hand,the construction of a safety monitoring edge computing framework for construction projects was constructed,and the data processing flow of edge nodes was described; on the other hand,a trusted data sharing framework based on blockchain was established,and a safety supervision data sharing process was designed and trusted sharing data was implemented through smart contracts.The proposed solution combined with edge computing and blockchain technology,which not only could strengthen the safety supervision of railway construction projects by industry authorities,but also provide new opportunities for stakeholders to share safety monitoring data in real time.
Volumn Content
边缘计算
