Abstract-To analyze the distributed attack actions in computer networks, different models are discussed. Firstly, the Hybrid P2P, Pure P2P and S uper P2P structures of attack are presented. The, the broadcast mechanisms of every structure are introduced. The attack methods mainly focus on the index poisoning and routing table poisoning on the general nodes and access point. By the presented model, high attack flexibility, no single point of failure, high freedom may be achieved. It is difficult to cluster, identify the attack traffic and trace the attacker. S o, security researchers of computer networks should catch the attention of P2P attacks.

At present, the active defense strategy based on game theory is based on the complete information game model. For such model cannot cope with attackers and defenders do not know each other's behavior problems, based on the cooperation, and incomplete information game theory is proposed to improve the existing strategy gains quantitative method, calculation and comprehensive analysis of the bayesian equilibrium.

Abstract-The existing anonymous region construction schemes based on continuous Location-based service (LBS) queries do not solve the temporal and spatial relationship of common user sets in adjacent anonymous regions. The attacker can reduce the anonymous region and reduce the common user set by the Maximum movement boundary attack, and reduce the privacy protection level of the user. Aiming at the above problems, considering the time accessibility of adjacent anonymous region and common user set, this paper proposes a location privacy enhancement scheme based on historical footprint for k-anonymity. The security analysis shows that the scheme does not increase the computational cost and communication cost of the system while guaranteeing the privacy queries of the users, and does not affect the user's quality of service. It has good validity and practicability.

Abstract-K-anonymity technology is one of the important technologies of location privacy protection based on LBS(Location-based service, LBS), it protects users' location privacy by constructing ASR(anonymous spatial region, ASR). Existing construction schemes for ASR based on continuous LBS queries giving rise to the problem of generating a large ASR, which increases query latency and server load, thus affect the QoS(quality of service, QoS) of users. Considering the shortages of existing ASR construction schemes for continuous LBS queries and the influence of aggregation property and moving directions of historical users in ASR, a new ASR construction scheme is proposed in this paper by selecting regions with high frequency of users and introducing the idea of ASR segmentation. The experimental results show that this scheme can further reduce the ASR and improve the QoS under the premise of satisfying users' location privacy demands.

Network protocol analysis can help security personnel analyze network vulnerabilities. However, the network protocol analysis is confronted with more and more kinds of protocols and more and more complex issues. Firstly, based on the basis function of the network protocol structure representation method, the improved k-means clustering algorithm based on the basis function is presented. Finally, a network protocol analysis process based on basis function is presented. Performance analysis shows that the proposed method is superior to traditional matching methods and statistical methods.