Wireless sensor networks (WSN) can be referred to as a group of spatially dispersed and dedicated sensors that are self-capable of collecting environmental information, processing it as per requirement, and forwarding it to a base station for different application. In this chapter, the authors focus on secure data transformation in wireless sensor node. The intrusion detection system using sensor node is complicated as sensors lifetime depends upon power consumption and for which the network's lifetime also get affected. Security as well as network lifetime can be enhanced by relocating the computational load of the operation from the sensor nodes to the base station. At base station by encrypting and compressing the data using cryptography and artificial neural network (ANN), the authors increase the security and increase the throughput of the network. On the server side it is decompressed and then decrypted.
TopIntroduction
Wireless Sensor Networks (WSNs) consist of a huge range of tiny sensor/detector nodes that send the collected data using the wireless channels. Most sensor networks consist of many low powers, less cost nodes deployed to keep an eye on the environment. This sensor network is a mix of different variety of sensors and actuators with variable computational power. WSNs have many wide range applications ranging from military surveillance to under water monitoring, critical health supervising, urban traffic controlling etc. (Yick et al., 2008; Akyildiz et al., 2002). Most of the applications of WSNs require se- cure transmission of data from node to server and vice versa. As sensor nodes have limited storage capacity and energy for processing the data, security in WSNs becomes a vital concern. But it is common problem that Sensor nodes can face different unwanted attacks during deployment or functioning in any hostile environments (Chen et al., 2009; Patel & Aggarwal, 2003).
For securing that data or information from various types of attacks normally cryptography is used which is an art of hiding the data from intruder. It is a technique in which the security to the data is provided by encrypting the whole data which cannot be read by the attacker .Now a day’s attacker uses very complicated and modern technologies. To deal with those attacks it requires higher computational cryptographic algorithm but the sensor nodes have limited computational power to execute those processes (Kahate, 2014).
From the above context, a need of a new technology which is more se- cure than the existing technique is realized. In this Chapter, an efficient algorithm has proposed in which a hybrid technology of cryptography and Artificial Neural Network (ANN) are used for offering additional security with more secure exchange of information with less computational power.
Wireless Sensor Network (WSN) can be refer as a group of heterogeneous sensors with less energy, storage capacity and computation power. It has a wide range of applications in our day-to-day life (El-Bendary, 2015; Abdollahzadeh & Navimipour, 2016).
As wireless network is more prone to outer threats and attacks so security is an important concern of these types of networks. One of the ways to tackle this kind of situation is the use of cryptography. Cryptography encrypts the data before transmitting. Due to physical size and energy level of sensor nodes it has restricted processing power, storage capacity and communication bandwidth. These things have to be considered particularly while dealing with security issues (El-Bendary, 2015; Yang, 2014; Kodali & Chougule, 2013). The security model which are: confidentiality, integrity, and availability. Each of these goals is in brief explained below.
(a). Confidentiality: Confidentiality is to stop illegal access of the information in data communication
(b). Integrity: Integrity makes sure that data being sent to the server in a correct and complete way such that the rightness and completeness of the data is sustained
(c). Availability: The timely availability of data to authorized users (El-Bendary, 2015; Xiao et al., 2009).