An Efficient Secured PIT Management and Attack Detection Strategy Enhanced by CSOA-DCNN Algorithm in a Named Data Networking (NDN)

Abstract

Named Data Networking (NDN) is a developing Internet design that utilizes a new network communication model dependent on the identity of Internet content. Its core component, the Pending Interest Table (PIT) serves an important role of recording Interest packet information. In managing PIT, the issue of flow PIT measuring has been very challenging because of the huge use of long Interest lifetime especially when there is no adaptable replacement strategy, subsequently affecting PIT performance. Named Data Networking (NDN) might experience some emerging threats such as Interest Flooding Attacks (IFA). In this paper, we focus on the IFA that can seriously devour the memory resource for the Pending Interest Table (PIT) of each included NDN router by flooding a huge amount of malicious Interests with spoofed names. To extricate the pressure of PIT attacked by IFA, we propose a methodology of efficient Secured PIT management and attack detection strategy by using a cuckoo search optimization algorithmDeep convolutional neural network (CSOA-DCNN) algorithm in Named Data Network. The CSO algorithm initially utilizes a learning technique and afterward considers improved search operators and deep convolutional neural network architecture (DCNN) for classification. The network simulation tool is utilized to design and calculate PIT management. The results of the study on a 20 Gbps gateway trace shows that the corresponding PIT contains 1.5 M entries, and the lookup, insert and delete frequencies are 1.4 M/s, 0.9 M/s and 0.9 M/s. The contribution of this study is significant for Interest packet management in NDN routing and forwarding systems.