Multi Model Transmission Analysis Based Efficient Intrusion Detection System for Improved Performance
A. Anthony Paul Raj1, J. K. Kani Mozhi2
1A.Anthony Paul Raj *, Research Scholar, Periyar University, Salem, India. Dr.
2J. K. Kani Mozhi *, Professor, Department of Computer Applications, Sengunthar Arts & Science College, Salem Road, Tiruchengode, Namakkal, Tamil Nadu,
Manuscript received on July 30, 2019. | Revised Manuscript received on August 25, 2019. | Manuscript published on August 30, 2019. | PP: 4094-4103 | Volume-8 Issue-6, August 2019. | Retrieval Number: F8932088619/2019©BEIESP | DOI: 10.35940/ijeat.F8932.088619
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: The problem of intrusion detection in network systems has been well studied. There exist numerous techniques in the mitigation of intrusion attacks, but they struggle to produce expected performance. To solve this issue, an efficient multi model analysis based approach is described in this article. The network systems faces various challenges like modification, distributed denial of service, spoofing, eavesdrop and so on. The proposed multi model approach monitors the network packets in different level by analyzing the payload, path, host and frequency of incoming packets. The method considers the frequency of packets, path being used, and frequency of transmission, host details and payload features. For each features, the method computes the trust measure which has been used to classify the packets. The method estimates cumulative multi mode trust weight towards any packet being received. According to the weight measures of different analysis, the attack has been identified. The proposed method improves the performance of intrusion detection and increases the accuracy. 
Keywords: Intrusion Detection Systems, Multi Model Analysis, Payload Analysis, Path Analysis, Frequency Analysis, MMTW.