Random Node Sampling for Energy Efficient Data Collection in Wireless Sensor Networks

Wireless Sensor networks are envisaged to be the building blocks for future smart applications, where tiny battery operated sensor nodes are deployed in thousands in the environment to provide real-time sensory data of their surroundings. Thousands of sensor nodes will be installed in our homes to collect and communicate sensory readings to feed information-hungry smart applications. Sensory applications need to be designed to conserve energy to prolong the lifetime of the network. This paper shows that employing a random sampling approach with sensory applications to provides accurate event status whilst prolonging the sensor network's lifetime. The proposed approach conserves energy by randomly selecting a subset of sensor nodes to provide the status of the environment, while setting the rest of the network in the sleep mode. Moreover, the randomness of the approach provides means to utilizing the sensor nodes in a fair manner, thus avoiding utilization hot spots and unbalanced utilization of sensor nodes.