The tremendous changes in networking is added advantage to networking users. The networking facilitates many services and it also open source for intrusions. In networking we have different type of intrusions. These intrusions are very effect in the performance of networking. In the existing system many algorithms are introduced for intrusion detection. Even though the existing algorithms are well in detection of intrusions. But, the performance time is very high. To overcome these problems and accurate the intrusion detection we proposed a novel Genetic Algorithm(GA). The proposed GA efficiently detect the various types of network intrusions. The GA is very useful to optimize the intrusion detection system. The Novel Genetic Algorithm achieved better performance. The performance of GA is test with the NSL-KDD99 benchmark dataset, it reduced performance time. The IDS with GA is recommended to use detect network intrusions in real time applications.
In modern days the network based computer systems plays vital role. The increasing use of network based computer systems is had open security issues. So there is a need of urgency to protect the computer network system. The intrusion detection system is a technology to defend from intrusions. The IDS is most prominent security management for computer networks. In computer networks the intrusion detection system helps to monitoring, detecting and all the activities of intrusions. The IDS is a potential dynamic defense system throughout the network. The IDS is very efficient in intrusion detection[4][5].
The intrusions are effected when the computer system security is compromised. The intrusions are can dissolve the computer network activities such as system integrity, resource availability and confidentiality. In computer systems some of the intrusion detection techniques are ahead in giving the security of authentication, encryption and programming errors. The computer networks are heavy complex and explore the weakness in the design and development of system, so the intrusion detection system is not sufficient to denied the intrusions. In this paper, combine the Intrusion Detection System with a novel efficient Genetic Algorithm.
Proposed Methodology
Genetic Algorithm
The genetic algorithm for intrusion detection system is start with random generation of chromosome population, the initial process is solution for all problems. The algorithm also uses evolution of chromosomes using selection, recombination and mutation operators. The function of evolution is used to calculate the nature of each chromosome for particular solution. The three main factors of genetic algorithm is impact of network applications, such as fitness function, representation of individuality and GA parameters.
Data Pre-Process
In this application for intrusion detection system we use NSL-KDD dataset. This is the benchmark dataset for intrusion detection methods. In pre-processing eliminate the redundant records. The duplicate records are removed from the dataset and help to increase the performance of the algorithm. In this data pre-process extracts the features of the dataset such as connection type, protocol type and etc.
Initial Population
The initial population is very essential in intrusion detection system. There must be know the type of attacks . In NSL KDD dataset contains random type of attacks.
SYSTEM REQUIREMENTS
SOFTWARE REQUIREMENTS:
• Web Technologies : HTML, CSS, JS. JSP
• Programming Language : Java and J2EE
• Database Connectivity : JDBC
• Backend Database : MySQL
• Operating System : Windows 08/10
HARDWARE REQUIREMENTS:
- Processor : Core I3
- RAM Capacity : 2 GB
- Hard Disk : 250 GB
- Monitor : 15″ Color
- Mouse : Two or Three Button Mouse
- Key Board : Windows 08/10
