A Study Of Timing Attacks Detection And Defense Strategies In Named Data Network

Named Data Network(NDN)is an alternative to host-centric networking exemplified by today's Internet.One key feature of NDN is in-network caching that reduces access delay and query overhead by caching popular contents at the source as well as at a few other nodes.But attackers can use cached content in a router to violate a user's privacy,such as getting a user's internet history.The attack is called timing attacks and is an attack that inferred whether a consumer has recently requested certain content based on the difference in round-trip time between the cached content and the uncached content.In order to prevent the privacy leakage and resist such kind of attacks,this paper does the following:(1)We propose a detection scheme.It uses cache hit rate,average request interval,request frequency,and request content type as eigenvalues for timing attacks,and in order to capture these eigenvalues in a timely manner,our scheme divides a constant-sized time window into several small time slices on which to extract features of the traffic.Finally,we trained an LSTM(Long Short-Term Memory)model to detect timing attacks.(2)We designed a groupbased privacy protection collaborative caching that ensures user privacy while improving network performance.First,the scheme use a grouping algorithm based on greed.Grouping intermediate nodes based on cache similarity and information entropy improves consumer anonymity.Then,consistent hashing ideas are used within the group to reduce the redundancy of cached content within the group and a heuristic algorithm is designed to address the load imbalance in consistent hashing.Finally,a routing forwarding of interest packets within the group is implemented so that interest packets can be transmitted securely within and outside the group.This paper evaluates the detection scheme in terms of classification accuracy,detection rate,false alarm rate and F-measure and evaluates the defense mechanism in terms of average hop,cache hit rate,anonymous set and information entropy based on the popular named data network simulation tool ndnSIM.The experimental results show that our proposed timing attacks detection scheme can detect timing attacks effectively and defensive policies that not only protect user privacy but also have good caching performance.