Increased Speed for Network Security through Multi-Character Processing

A clear trend that can be observed in the Internet is the increasing amount of packet data that is being inspected before a packet is delivered to its destination. More recently, Network Intrusion Detection Systems (NIDS), virus scanners, spam filters and other content-aware applications go one step further by also performing scans on the packet payload. Pattern matching algorithm is used in Network Intrusion Detection System (NIDS). The system is used to detect network attacks by identifying attack patterns. This paper proposes a memory-efficient pattern matching algorithm which can significantly reduce the number of states and transitions by merging pseudo-equivalent states while maintaining correctness of string matching. Pattern matching is achieved through Aho-Corasick (AC) algorithm. By comparison result we say our matching algorithm is memory efficient than previously proposed method. As an extension of our work, through Multi-character processing, sufficient speed has been increased. The reduction in clock cycle indirectly increases speed in operation.The architecture is coded in VHDL and simulated using Modelsim and Xilinx.