Article Preview
Top1. Introduction
Wireless sensor networks are proliferating into our everyday life in the form of different applications, such as eHealth (Ho, ym., 2009; Honeine;Mourad;Kallas;Snoussi;Amoud;& Francis, 2011; Singh;Tiwary;Hoon-Jae;& Chung, 2009), environment and infrastructure monitoring (Yunseop;Evans;& Iversen, 2008), home automation(Langhammer & Kays, 2012; Tudose, ym., 2011; Zhang;Song;Wang;& Meng, 2011), traffic control (Laisheng;Xiaohong;Zhengxia;Bing;& Pengzhi, 2009; Zou;Yang;& Cao, 2009), agriculture (Vijayakumar & Rosario, 2011; Zhang Y., 2011)[29-30], and manufacturing(Bertocco;Gamba;Sona;& Vitturi, 2008; Yamaji;Ishii;Shimamura;& Yamamoto, 2008). Most of wireless sensor networks have unique characteristics, such as the ability to withstand unfavorable environmental conditions, dynamic network topology, communication failures, large scale deployment, node mobility, unattended operation and limited energy resource. As energy is a limited resource in most cases, efforts need to be made to save energy in all operating conditions and network layers(Dargie & Pollebauer, 2010). The type of protocols and the selected parameters at different network layers play a vital role in energy consumption. Different strategies have been proposed in order to achieve better energy efficiency and organizing sensor nodes into clusters is among such solutions (Mamalis;Gavalas;Konstantopoulos;& Pantziou, 2009). The high density of the network may lead to multiple adjacent sensors generating redundant data. In order to solve this issue, data aggregation within clusters can be used to eliminate data redundancy and reduce communication load. Clustering can also reduce the number of nodes responsible for long distance transmission which helps in saving the network’s power consumption.
Wireless sensor networks are vulnerable to attacks and the possible security attacks have different degree of impact to the system, such as jamming(Xu;Ma;Trappe;& Zhang, 2006), denial-of-service(Raymond & Midkiff, 2008), node compromise(Lin, 2009), impersonation attacks, and protocol specific attacks. Security can be considered as a non-functional requirement that maintains the overall system usable and reliable by protecting the information and information systems. Thus, security is of paramount importance, the network must be adequately protected against malicious threats that can affect its functionality. Due to the role of wireless sensor networks as sensing and actuating systems, any disturbances in such a network may have consequences in the real world. The communication between the sensor nodes can be secured by implementing energy efficient cryptography.
The main focus of this work is developing a secure communication channel between cluster heads and base station for transmission of aggregated data and analyzing its energy efficiency. In order to accomplish this, first clustering algorithm is implemented in Contiki operating system and the clustered network is simulated in Cooja network emulator to evaluate its energy efficiency. The next step involves developing of a secure architecture for the communication of aggregated data between cluster heads and base station by incorporating Advanced Encryption Standard (AES) protocol on the clustering algorithm. The energy overhead is analyzed after the implementation of encryption protocol and detailed comparisons are carried out.