Research On The Security Of Underwater Wireless Sensor Network Based On Blockchain

The ocean,as the most extensive water body on the earth,contains a large number of scarce resources such as minerals and oil that can be developed and utilized by human beings.The design of a reliable and safe Underwater Wireless Sensor Network(UWSN)has important research value for marine environmental monitoring,marine development and other research fields.UWSN uses underwater acoustic communication.The underwater acoustic channel has the characteristics of low data rate,long propagation delay,narrow bandwidth,high bit error rate,etc.In addition,the open and shared characteristics of underwater acoustic channels make UWSN face various threats and attacks.However,most research on UWSN aim at saving energy and prolonging the lifetime of the UWSN,and ignore the potential security problems in UWSN.UWSN has the characteristics of mobility and self-organization,and it is difficult to build an absolutely trusted third party in UWSN,so the security of UWSN needs to be decentralized.The blockchain is essentially a decentralized distributed database.In addition,blockchain has many other advantages such as transparency,anonymity,transaction traceability,as well as data cannot be tampered with at will,so blockchain provides a new view for the security of UWSN.This thesis studies the security technology of underwater wireless sensor network based on blockchain,and proposes the node authentication and intrusion detection mechanisms of underwater wireless sensor network based on blockchain.The main contributions of this dissertation are as following:(1)By analyzing the characteristics,security threats,security requirements and topology of UWSN,combining with the blockchain infrastructure,a blockchain model of underwater wireless sensor network based on cluster network topology(BUWSNC)is proposed.In the BUWSNC model,an alliance blockchain network,composed of cluster heads,base stations and intra-cluster nodes,is jointly responsible for the blockchain management.Based on the proposed BUWSNC model,the mechanisms of node authentication and intrusion detection for UWSN are analyzed.(2)Based on the BUWSNC model,a node authentication mechanism for UWSN is proposed.Traditional identity authentication adopts centralized mechanism and requires a trusted third party authentication authority.Underwater nodes move with water flow or other underwater activities,so the topology of an UWSN network changes dynamically.Additionally,the characteristics of mobility and self-organization also make the UWSN lack a trusted third party.So traditional centralized identity authentication mechanism is prone to single-point failure,which threatens the security of the UWSN network.Therefore,this dissertation proposes two authentication mechanisms of nodes,one is the authentication mechanism for nodes in UWSNs based on the BUWSNC model,and the other is the CBFT consensus authentication algorithm based on the credit mechanism.The authentication mechanism based on the BUWSNC model is used to authenticate the identifications of cluster-head nodes and intra-cluster nodes by the base station,and the authentication information is stored in the blockchain;The CBFT consensus authentication algorithm based on the credit mechanism is designed based on the optimization of the PBFT algorithm and used to authenticate the nodes newly joined.The new nodes which have passed the authentication can access the network legally.Finally,the formal analysis of the UWSN node authentication based on the BUWSNC model is carried out using BAN logic.Analysis shows that the authentication mechanism based on the BUWSNC model can achieve two-way authentication between a base station and a cluster head;Experimental results show that the CBFT consensus algorithm can reduce delay,resource consumption,and improve the system efficiency.(3)A detection mechanism of malicious nodes in UWSNs based on smart contract is proposed.Aiming at the problem that the existing methods of detecting malicious nodes in UWSNs cannot guarantee the fairness and traceability of the detection process,a blockchain model for malicious node detection in UWSNs based on clustering structure(called the BUW model)is proposed.In the BUW model,the data of sensor nodes are stored in the blockchain,and the smart contracts are used to detect malicious nodes in the network while the consensus algorithms are used to ensure data consistency,all these provide more secure and reliable solution for the detection of malicious nodes in UWSN as well as the traceability of the detection process.The feasibility and performance of the model are analyzed experimentally.Experimental results show that the proposed mechanism can not only prevent malicious nodes from interfering with the network,but also ensure the transparency and traceability of the detection process.