Research On Mobile Terminals Charging Network Based On Blockchain And Auction

The long-endurance of unmanned aerial vehicles(UAVs),or drones based on mobile crowd sensing has always been one of the focuses of smart cities research.The set-up charging station can conduct electricity trading with UAVs,help the UAV to charge in real time and complete a series of sensing tasks such as data collection and sharing as soon as possible.However,the application of these electricity transactions in real scenarios faces many challenges.Not only does massive transaction information processing lack a trusted platform or technical support,but the establishment of charging stations is restricted by the geographical location of the city.Even the use of mobile unmanned charging stations faces the problem of insufficient resources and space.Therefore,to solve aforementioned problems,the main research work of this thesis is stated as follows:(1)Through investigation and analysis of the application of traditional blockchain technologies in the Industrial Internet of Things,a mobile terminal charging network model composed of mobile charging stations and UAVs was established using relevant blockchain technologies;this thesis proposed the charging and electricity transaction verification architecture from both the off-chain and on-chain levels,and designed the overall network operation mechanism;this thesis wrote smart contracts to simulate the communications between different network nodes,so as to test the rationality of the operation mechanism design.(2)Since mobile charging stations are limited in size,their power resources and charging space can only support charging a small number of UAVs at the same time,so scheduling must be performed to support the charging of a large number of UAVs.In response to this problem,this thesis transformed the charging scheduling into an auction process.Through the traditional second-price auction and Myerson’s lemma,the designed auction was evaluated to meet the ideal auction’s requirements and UAVs’ charging demands;this thesis also designed a new auction mechanism named Auction Mechanism based on Deep Reinforcement Learning(AM-DRL)by leveraging the state-of-the-art deep reinforcement learning method.(3)A series of network model and auction mechanism performance testing experiments were planned and implemented.Experimental results show that,compared with the traditional second-price auction,the auction mechanism designed in this thesis not only provided higher auction revenues for mobile charging stations,but also ensured the truthful bidding of UAVs.The trained learning network also improved the adaptability of AM-DRL in the real environment;analyzing the network model’s prevention against network attackers,and the deployment and test of smart contracts verified the utility of blockchain technologies in the application of electricity trading.