Fuzzy Logic-based Efficient Interest Forwarding (FLEIF) in Named Data Networking

Abstract

Content centric or Named Data Networking (NDN) is expected to replace IP-based communication in the future Internet architecture. Here, connections are established based on data instead of host addresses. This flexibility helps in decreasing the end-to-end delay, as the requested data may be available in any intermediate router's cache. In NDN, a small Interest packet sent on a link results in large sized incoming Data packets on the same link. Therefore, sending multiple Interest packets on a single outgoing face may cause congestion on the returning path. Hence, regulation of Interest packets based on link characteristics is important to avoid congestion. In this paper, an intelligent mechanism based on Neuro-Fuzzy Logic is proposed for calculating outgoing link prices of each intermediate node. The class of each outgoing link on a router is determined for Interest forwarding in order to avoid congestion. Comparison results show that the proposed method effectively improves throughput by decreasing packet loss ratio.