Improved genetic feedback algorithm based network security policy framework contains some drawback for security. This has motivated the need of a strong network security policy framework. In this paper a strong network model for security function is presented. The fitness function is examined for defining the gene of a network packet and a method to calculate fitness function is explained. In this model passive attacks are more difficult to deal as compare to the active attack. The basic attacks can be categorized as buffer overflow, array index out of bound, etc. We have dealt with passive attack, active attack, its types and brute force attack. These attacks are analyzed and security is provided. We finally find the best policy using a comparator. The main aim of this paper is threat detection, time optimization, performance increase in terms of accuracy and policy automation. We are experimenting using the real data set from the internet and local network. Our work is carried out on client server environment providing data confidentiality and authentication.

1. “Improved Genetic Feedback Algorithm Based Network Security Policy Framework” by Atish Mishra, Arun Kumar Jhapate, Prakash Kumar Second International Conference on Future Networks on page 8-10. http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?armunber=5431893
2. Chen Xiao-su Wu Jin-hua Ni Jun this paper appears in Wireless Communications, Networking and Mobile Computing, 2007 WiCom 2007.Internatinal Conference on page 2278-2281.
3. “An Improved Interactive Genetic Algorithm Incorporating Relevant Feedback” by Hang-Fei Wang, Xu-Fa Wang and Jia Xue.
4. “Designing Rule base for Genetic Feedback Algorithm Based Network Security policy Framework using State Machine” by Atish Mishra, Arun Kumar Jhapate and Prakash Kumar on pp 415-417 International Conference on Signal Processing Systems,2009.
5. “A Policy-Based Management System with Automatic Policy Selection and creation capabilities by using a Singular Value Decomposition Technique” by Hoi Chan; Kwok, T.; IBM Thomas J. Watson Res. Center, Hawthorne, NY on June 2006.
6. “Storing Scheme for State Machine Based Rule Base of Genetic Feedback Algorithm Based Network Security Policy Framework Depending on Memory Consumption” by Atish Mishra, Prakash Kumar on 2009 at IACSIT vol, 3 Singapore.
7. x “An Artificial Intelligence Perspective on Automatic Computing Policies” by Jeffrey O. Kephart and William E.Walsh IBM Thomas J. Watson Research Center Yorktown Heights, New York 10598.
8. “Reinforcement Learning: A Survey” by L.P.Kaelbling, M. L. Littman, A. W. Moore.
9. “Improved Algorithms for Finding Gene Teams and Constructing Gene Team Trees” by Biing Feng Wang, Chien-Hsin Lin.
10. RFC 2573, “A Framework for Policy-based Admission Control”, http://www.faqs.org/rfcs/rfc2753.html.

genetic algorithm fitness function