Research On The Framework Of Swarm Intelligence Network Based On Blockchain And Its Security

With the advent of artificial intelligence 2.0 era,crowd intelligence as an important field has been widely concerned by academia and industry.Crowd intelligence is a new distributed problem solving paradigm.Crowd intelligence has been applied in intelligent transportation perception,weather forecast and public safety.The organizational forms of Crowd intelligence mainly include centralized structure,distributed structure and decentralized structure based on blockchain.Compared with other architectures,the decentralized architecture based on blockchain can more effectively resist DDoS attacks and prevent single point of failure and other security risks.However,the existing framework of crowd intelligence network based on blockchain has some shortcomings,such as low universality and scalability,centralized data processing mode leading to large delay of crowd intelligence task,etc.And the lack of efficient and lightweight group intelligence task data authentication strategy can not better guarantee the reliability and verifiability of group intelligence data sources.According to these shortcomings,the research contents of this thesis are as follows:（1）To solve the problem of lack of scalability and universality of the current crowd intelligence network system based on blockchain,this thesis studies the framework of crowd intelligence network,and proposes a general crowd intelligence network framework based on blockchain.A special smart contract is designed for function management without functional level,and the control contract status of each layer is relatively independent,which improves the scalability of the system.Based on Ethereum smart contract development platform,the smart contract designed in this paper is simulated,and the experiment shows that the scheme is feasible.（2）This thesis analyzes the security risks that the framework of crowd intelligence network may encounter,and mainly analyzes the framework system security and services security.The results show that the crowd intelligence network framework can effectively resist common malicious attacks.（3）In order to solve the problem of lack of group intelligence task data authentication strategy,this thesis studies the two-party cooperative digital signature technology,and proposes a secure two-party cooperative Edwards curve digital signature scheme.In this system,each group intelligence task data is required to be jointly signed by task participants and task publishers using the two party signature scheme to ensure the reliability and verifiability of group intelligence data.The security analysis of the proposed scheme is completed,and the simulation results show that the scheme is feasible.