Research On Resource Allocation Algorithms Based On Double Auction In Mobile Edge Computing

With the massive growth of computing-intensive and delay-sensitive applications on mobile terminals,the high-latency computing mode based on the central cloud can no longer meet the computing needs of users;meanwhile,the local computing power and battery capacity of mobile terminals are limited,therefore,servers with computing capability need to be deployed near the mobile terminal to reduce the computing delay and energy consumption of the mobile terminal.As a result,Mobile Edge Computing(MEC)has received widespread attention.Mobile edge computing platform deploys resources at the edge of the network so that users can migrate tasks to edge nodes for processing.The resources of mobile edge computing nodes are limited and there are entities with conflicting interests in the system,which makes task migration in mobile edge computing face complex resource allocation problems.How to efficiently allocate mobile edge computing resources,coordinate conflicts of interest between different interest entities,and achieve efficient allocation of resources is an urgent problem to be solved.To this end,many scholars have introduced the double auction mechanism into the research of mobile edge computing resource allocation,but the existing research has the following deficiencies:(1)The resource allocation algorithm based on the single-round double auction mechanism does not consider the game process when users and edge nodes offer prices,nor does it analyze the impact of different resource allocation strategies on the algorithm;(2)The resource allocation algorithm based on the multi-round double auction mechanism does not consider that users and mobile edge computing nodes participating in the next round of auctions can adjust the bid price at the same time,nor does it analyze the impact of different payment rules on the allocation algorithm.Based on the above deficiencies,the article has conducted research from the following two aspects:(1)A computing and communication resource allocation algorithm based on single round double auction is proposed,and the bidding strategies of users and edge nodes are analyzed from a game perspective,and analyzes the impact of the three resource allocation strategies,such as random allocation,priority allocation for high-unit-bid users and priority allocation for high-bid users,on the algorithm.Simulation experiments show that the algorithm can obtain higher total system revenue.(2)A computing and communication resource allocation algorithm based on multiround double auctions is proposed,combining a bidding strategy with fixed adjustment ratio and a dynamic bidding strategy based on the Hurwicz criterion to form a new price strategy,at the same time,the impact of the three payment rules such as VCG payment rule,k-value payment rule and the highest price of failed users(the lowest price of successful users if there are no failed users)payment rule on the resource allocation algorithm is analyzed.Simulation experiments show that the algorithm can obtain higher total system revenue and faster convergence rate than the existing algorithms.