Affiliations: [a] Computer Science Department, Faculty of Computers and Artificial Intelligence, University of Benha, Egypt
| [b] School of Computer and System Science, Jawaharlal Nehru University, India, New Delhi, India
| [c] Computer Science Department, University of Sharjah, Sharjah, UAE
| [d] Hanoi University of Home Affairs, Hanoi, Vietnam
| [e] VNU Information Technology Institute, Vietnam National University, Hanoi, Vietnam
| [f] Division of Computational Mathematics and Engineering, Institute for Computational Science, Ton Duc Thang University, Ho Chi Minh City, Vietnam
| [g] Faculty of Mathematics and Statistics, Ton Duc Thang University, Ho Chi Minh City, Vietnam
| [h] Tashkent State University of Economics, Tashkent, Uzbekistan
| [i] Mathematics Department, Zagazig University, Zagazig, Egypt
Correspondence:
[*]
Corresponding author. Le Minh Tuan, E-mail: letuan104@gmail.com.
Abstract: Data acquisition problem on large distributed wireless sensor networks (WSNs) is considered as a challenge in the growth of Internet of Things (IoT). Recently, the combination of compressive sensing (CS) and routing techniques has attracted much attention of researchers. An open question of this combination is how to integrate these techniques effectively for specific tasks. On the other hand, CS data reconstruction process is considered as one of the CS challenges because it requires to recover N data from only M measurement where M< <N. Through this paper, we propose a new scheme for data gathering in IoT based heterogeneous WSN that includes a new effective Deterministic Clustering using CS technique (DCCS) to handle the data acquisition problem. DCCS reduces the total overhead computational cost needed to self-organize WSN using a simple approach and then uses CS at each sensor node to decrease the overall energy consumption and increase the network lifetime. The proposed scheme includes also an effective CS reconstruction algorithm called Random Selection Matching Pursuit (RSMP) to improve the recovery process at the base station (BS). RSMP adds a random selection process during the forward step to give the opportunity for more columns to be selected as an estimated solution in each iteration. The simulation results show that the proposed scheme succeeds to minimize the overall network power consumption and prolong the network lifetime beside provide better performance in CS data reconstruction.
Keywords: Internet of things, clustering based
wireless sensor networks, compressive sensing, routing
techniques, data reconstruction, network
lifetime
