Research Of Cache Pollution Attack In Content Centric Network

As a typical example of Information-Centric Networking,Content-Centric Networking(CCN)replaces the IP address of the middle layer with the content name as the core element of the network.With the advantage of its nodes being cached generally,achieving the efficient distribution and response of content and becoming the hot spots of next generation of network architecture research.However,the universal caching also brings new security risks while improving the efficiency of content distribution.One of the most representative security threats is cache pollution attack.The feature of cache pollution attack in CCN is that the attacker sends a large number of malicious interest requests for certain low popularity content or illegal content,and increases the popularity of these content,so that it stays in the cache of nodes along the way for a long time,so as to achieve a reduction of the legitimate user interest requests hit rate at cache and increase the network delay for obtaining the content.Aiming at the problem of cache pollution attack in CCN,this paper studies from the following two aspects:1.Cache pollution attack detection.Aiming at the cache pollution attack detection in CCN and considering that the cache pollution attack will have different degrees of influence on different parameters of the node,based on this,this paper presents a multi-parameter detection algorithm based on fuzzy analytic hierarchy process(MPDA).The main idea of the algorithm is to comprehensively consider the multiple network parameters under attack for joint detection to improve the accuracy of attack detection and extend the scope of attack detection.2.Cache pollution attack defense.Aiming at the cache pollution attack defense in CCN,A Cache Pollution Defence Policy based on Content Request Locality(CPDP-CRL)strategy is proposed.The main idea of this strategy is based on the CacheShield strategy and consider the distribution of the content request arrival on each port of the node.At the same time,the division of content popularity is introduced to set different cache probabilities for different popular content.Theoretically,this strategy can effectively defend against two typical cache pollution attack in CCN.Finally,the ndnSIM simulation platform based on NS-3 is used to simulate and analyze the proposed MPDA algorithm and CPDP-CRL strategy.The simulation results showed that compared with the existing typical cache pollution detection and defense algorithms,the MPDA algorithm and the CPDP-CRL strategy showed better network performance.