Research On Secure Receiver-forwarding Mechanism Based On Bayesian Theory For NDN-VANET

The current IP communication technology faces the dual challenges which are internal structure expansion problems brought by technologies such as mobile communication and changes in external user requirements.The significant advantages of the NDN network architecture in terms of mobile support and fast content retrieval make it is the most promising candidate architecture in the future network architecture.Due to its better support in mobile,using NDN architecture in VANET is a hot topic in this field.This thesis refers to this NDN-based VANET network architecture as the NDN-VANET architecture.In the NDN architecture,the breadcrumb routing mechanism based on interest packet flooding is used to retrieve content data.This approach will lead to interest packet broadcast storm problems.At the same time,due to the rapid change of VANET network topology,wireless media transmission and other characteristics,NDN-VANET interest packet broadcast problem has further increased,and the broadcast storm problem caused by the flood of interest packets has become a key factor limiting the network transmission performance.In addition,because NDN's interest packets do not use security technologies,an attacker can tamper with the content of the interest packets and forge interest packets.Broadcasting these false interest packets can overload the node's PIT table and cause DOS attacks.In order to solve the above problems,based on the Bayesian classifier and combined public key encryption mechanism,this thesis proposes a new Bayesian receiver-side secure forwarding mechanism named BSRF(Bayesian-based secure Receiver Forwarding Decision scheme for the interest packet)for NDN-VANET.The main research contents are as follows:1.Aiming at the interest packets broadcast storm problem of NDN-VANET,this thesis investigates the broadcast storm suppression schemes in various wireless network environments in recent years,and proposes a new classification method of broadcast storm suppression schemes.According to the proposed classification methods,a variety of broadcast storm suppression schemes are introduced,and the basic characteristics of each method are given.2.Through research on existing broadcast suppression technologies and machine learning algorithms,to address the broadcast storm problem in NDN-VANET,the BSRF mechanism requires that vehicle nodes in the network use the HELLO protocol to exchange current operating status information.Based on the current network conditions,the nodes use Bayesian classification algorithms to make forwarding decisions.At the same time,a probability backoff mechanism is introduced to further limit the occurrence of broadcast storms.3.The combined public key technology is used to authenticate the vehicle node identity and solve the confidentiality problem in the transmission of interest packets.The signature mechanism and hash function are introduced to provide authentication of the origin and integrity of the interest packets.Finally,the security analysis of the BSRF proposed in this thesis is carried out,and the experimental analysis of the new scheme is made under the simulation tool ndnSIM.Experiments show that BSRF can reduce the number of redundant interest packets in the network,can improve the data forwarding efficiency,and effectively suppress the broadcast storm problem.