FuRL: fuzzy RBM learning framework to detect and mitigate network anomalies in Information Centric Network

Abstract

Information Centric Network (ICN) is a promising next-generation internet architecture in which the network focuses on retrieving the content by employing open in-network caching scheme to provide an efficient content distribution to users. However, such open in-network caching is vulnerable to network anomalies. In particular, cache pollution attack disrupts the smooth working of in-network caching by flooding unpopular contents. Hence, the in-network caching malfunctions and legitimate consumer requests are dropped. To address this problem, a secure framework based on Fuzzy Restricted Boltzmann Machine has been proposed to detect the anomalies and defend against such pollution attacks in ICN. Further, a reward-based cache replacement (ReBac) algorithm that is capable of avoiding cache pollution attack has also been proposed. The experimental results obtained while testing the proposed framework show better detection rate compared with the state-of-art solution and the proposed framework shows better cache rate.