Traffic and Security using Randomized Dispersive Routes in Heterogeneous Sensor Network

Abstract: Generally traffic and the sensor network security have many challenges in the transmission of data in the network. The existing schemes consider homogeneous sensor networks which have poor performance and scalability. Due to many-to-one traffic pattern, sensors may communicate with small portion of its neighbours. Key management is the critical process in sensor nodes to secure the data. Most existing schemes establish shared keys for all the sensors no matter whether they are communicating or not. Hence it leads to large storage overhead. Another problem in sensor network is compromised node attack and denial of service attack which occurs because of its wireless nature. Existing multi path routing algorithms are vulnerable to these attacks. So once an adversary acquires the routing algorithm, it can compute the same routes known to the source, and hence endanger all information sent over these routes. If an adversary performs node compromise attack, they can easily get the encryption/ decryption keys used by that node and hence they can intercept the information easily. In this paper we are proposing a key management scheme which only establishes shared keys with their communicating neighbour and a mechanism to generate randomized multipath routes for secure transmission of data to the sink. Here we are adopting heterogeneous sensor networks and we are utilizing elliptic curve cryptography for efficient key management which is more efficient, scalable, and highly secure and reduces communication overhead. The routes generated by our mechanism are highly dispersive, energy efficient and making them quite capable of bypassing the back holes at low energy cost.