Malware is basically malicious software or programs which are a major challenge or major threats, for the computer and different computer applications in the field of IT and cyber security. Traditional anti-viral packages and their upgrades are typically released only after the malware’s key characteristics have been identified through infection. The most common detection method is the signature based detection that makes the core of every commercial anti-virus program. In this paper we improved the rate of malware detection using neural network classifier and compare with the other technique i.e. support vector machines, and show that the support vector machine classifier result better than the neural network classifier.

In today’s word internetworking change the concept of real life and provide various services and advantage based on the internet, due to increasing the popularity and usage of internet, security is a very challenging task. Computer security is two types one is host based and the other one is network based, sometime attack detection is also called intrusion detection. Intrusion detection system offers various tools and techniques for the ensuring to develop and remain more secure of our network. In this paper our aim to study various attacks and threats detection with their proposed techniques using for the network based detection system.

Cognitive radio-based wireless sensor network is the next-generation sensor network paradigm. Important to this emerging sensor network is the need to reduce energy consumption, paving way for ‘green’ communication among sensor nodes. Therefore, in this paper, we have proposed an energy-efficient, dynamic channel decision and access technique for cognitive radio-based wireless sensor networks. Using intelligent learning technique based on the previous experience, the cognitive radio-based wireless sensor network agent decides which available channel to access based on the energy-efficiency achievable by transmitting using the channel.

Vehicular Ad-hoc Networks (VANETs) have attracted a lot of attention in the research community in recent years due to their promising applications. VANETs help improve traffic safety and efficiency. Each vehicle can exchange information to inform other vehicles about the current status of the traffic flow or a dangerous situation such as an accident. Road safety and traffic management applications require a reliable communication scheme with minimal transmission collisions, which thus increase the need for an efficient Medium Access Control (MAC) protocol. In this paper we present the new modified MAC layer protocol and improve the results for the vehicular ad-hoc network.

Network lifetime is, arguably, the most important performance metric in wireless sensor networks. Since wireless sensor networks nodes are battery operated, in general, optimal utilization of the limited battery energy is vital for prolonging the network lifetime. Energy budget of WSNs is dominated by the energy dissipation on communication. So that the optimization of all aspects of WSN communication and networking is the overarching goal. In this paper we present the improve model for the smart grid and also improve the performance of smart grid for various applications such as packet delivery ratio and throughput.

Image Classification and retrieval is current research trend in computer vision. In concern of classification, the rate of classification depends on the feature attributes of image data and depends on behavior of classifier and image feature, Color is one of the most widely used low-level visual features and is invariant to image size and orientation. The easy-to-compute color histogram is a popular and widely used image feature, Texture is also one of the important types of classification and is being widely intensively used in pattern recognition. A supervised classification is the process of using samples of known identity to classify pixels of unknown identity. In this paper we present the content based image classification using the unsupervised and supervised learning techniques.