Informatics Publishing Limited

A. Edwin Robert ¹, M. Hemalatha ¹

Affiliations
1 Department of Computer Science, Karpagam University, Coimbatore, IN

Abstract

In present world applications of software in other domains have their own privileges and their own control over other application also fulfilling their own testing methods acting as a tool in solving the given problem. Application integrity is highlighted in existing work but in our proposed approach MCBA (Malicious Code Behavior Analysis). In our proposed study the method of MCBA approach error correcting codes in the kernel is incorporated. Our objective is to incorporate a protection mechanism that saves the application even though the system’s memory gets corrupted. In exploring the trusted MCBA Server to identify and monitor the portion of the system where corruption occurs and the server segregates the reason for various malicious impacts. Therefore, two approaches have been simulated: one is authenticated check and the next is unauthenticated check. In an authenticated check, a matching schema (e.g., the MCBA) applies dataset pattern recognition techniques to check malicious pattern by comparing it to the incoming applications during execution, if malicious packet is found it protects the system, in an unauthenticated check, the malicious packet from the guest OS for example ischolar_main kits enters into our local system and it securely stores a cloned image of the guest OS memory at boot time, this method incorporates a VMM (Virtual Memory Monitor) and it allows only the instructions to read from the cloned copy of memory but never execute the instruction, though it is so, sometimes the instructions are malicious and it is unsecured. This paper emphasizes the MCBA architecture, incorporates monitoring, detecting and healing process which are so feasible to apply in real time environment ,it is very economically used for the technical programmers who are designing source code for various domains in Software market.

Keywords

Dataset, Error Correcting, Malicious, Matching, Privileges

Full Text

   

S. Sampath ¹, A. Edwin Robert ²

Affiliations
1 Department of Computer Science, Karpagam University, Coimbatore - 641021, Tamil Nadu, IN
2 Department of Computer Applications, Karpagam University, Coimbatore - 641021,Tamil Nadu, IN

Abstract

Objective: To improve IDS strategy with high detection accuracy and reduce the power consumption of the nodes in MANET. Methods: There are several intrusion detection systems are developed. In order to increase the performance of IDS, Bayesian Correlated Equilibrium based IDS which incorporates two main processes namely, Cluster Head selection and Hybrid IDS for MANET is proposed. The game theory is also used to increase the detection accuracy of IDS. Findings: Mobile Ad-hoc Network (MANET) is an autonomous system that consists of battery powered mobile nodes. MANETs are prone to several attacks as they are continuously self configuring and infrastructure less. As the nodes are mobile, they are susceptible to intrusions. The Intrusion Detection System has issues of heavy traffic related to IDS in the network, which causes congestion. It also leads to high energy consumption among the nodes. So designing an efficient MANET should have certain number of goals such as effective intrusion detection, light traffic and low energy consumption and power loss. Many Intrusion detection schemes were proposed that normally incurs power loss in the node as there is a need for continuous monitoring. Application/Improvements: To increase detection accuracy, decrease power consumption and the IDS traffic Bayesian Correlated Equilibrium based IDS for MANET is presented.

Keywords

Bayesian Nash Equilibrium, Correlated Equilibrium, Game Theory, Intrusion Detection System (IDS), Mobile Ad-hoc Network (MANET).

Full Text

   

D. Sabareeswaran ¹, A. Edwin Robert ²

Affiliations
1 Dept. of Computer Science, Karpagam University, Coimbatore, IN
2 Dept. of Computer Application, Karpagam University, Coimbatore, IN

Abstract

Objective: To study about different data mining methods utilized for detecting plant diseases, soil moisture and crop growth monitoring.

Methods: Different data mining techniques are used in agriculture for detecting crop diseases, soil moisture and crop growth monitoring such as Support Vector Machine (SVM), Artificial Neural Network (ANN) and Regression model.

Findings: The inclusion of modern technologies can enhance the crop production and resolve major issues in traditional farming. The crop production is mainly depends on the availability of arable land and influenced by yields, macro-economic uncertainty and consumption patterns. The actual yield is mostly depends on crop’s genetic potential, amount of sunlight, water and nutrients absorbed by crop, presence of weeds and pests. In addition, the crop production is enhanced by combining crop models with data mining approaches.

Applications/Improvements: Finally, different data mining techniques used in agriculture are compared in order to prove their effectiveness. Hence, the agricultural monitoring system can be enhanced by using data mining techniques.

Keywords

Crop Production, Data Mining Techniques, Crop Diseases, Soil Moisture, Crop Growth Monitoring.

Full Text