Energy-efficient cache node placement using genetic algorithm in wireless sensor networks

Wireless sensor network (WSN) applications are required to report events and service queries with minimum delay and minimal energy consumption. The network lifetime of a WSN can be extended if the amount of communication in the network is reduced. We can achieve this by caching useful data closer to the requesting node. Caching successfully reduces data access latency and also the number of packet transmissions in the network, thereby increasing network lifetime. However, the important aspect of caching schemes is to identify nodes that can implement caching decisions and also place such cache nodes in a way that they can provide services to as many sensor nodes as possible in their vicinity. This has led to the study of optimal deployment of these cache nodes in a WSN. We carried out experiments to demonstrate the use of a multi-objective genetic algorithm (GA) for cache node placement in a WSN. In this paper, GA optimization aims to increase two parameters: sensors per cache in charge and field coverage. We also show that the GA successfully helps in selecting sensor nodes to implement caching and request forwarding decisions. Finally, we run the Scaled Power Community Index Cooperative Caching scheme (scaPCICC) on the optimized network and compare the delay and total number of overhead messages in the network. We conclude that by reducing the number of messages in the network and reducing the data access latency, the energy consumption of the network is reduced and network lifetime is increased. The experiments were run on MATLAB and ns2.