Compressed sensing for different sensors: A real scenario for WSN and IoT

• Published in: 2016 IEEE 3rd World Forum on Internet of Things (WF-IoT) pp. 289 - 294
• Main Authors: Amarlingam, M., Mishra, Pradeep Kumar, Durga Prasad, K.V.V., Rajalakshmi, P.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.12.2016
• Subjects: Compressed sensing; Data aggregation; Internet of Things; Monitoring; Multiplexing; Temperature measurement; Wireless sensor networks
• DOI: 10.1109/WF-IoT.2016.7845487

Wireless Sensor Networks (WSN) are integrable basic elements of Internet of Things (IoT). WSN deployment is constrained with sensor node's energy, communication range, limited on-board resources etc. Optimization of the energy consumption over the network to improve network lifetime is a challenging problem. Compressed Sensing (CS) involvement in WSN brought a solution to energy efficient data aggregation. This article presents a method, which exploits compressed sensing and dictionary learning to achieve energy efficiency in the scenario of data aggregation in WSN, where sensor node measures different sensors data. We demonstrate performance analysis of multiple sensors method with metrics, probability of successful recovery and network transmission cost. Extensive simulations on practical data set shows that our data aggregation method for practical scenario can deliver data to sink with minimum transmission cost which inherently saves significant energy to prolong the network lifetime. The probability of successful recovery shows that our method can recover compressed data with maximum probability.