Title: PM-LPDR: a prediction model for lost packets based on data reconstruction on lossy links in sensor networks

Authors: Zeyu Sun; Guozeng Zhao; Xiaoyan Pan

Addresses: School of Computer Science and Engineering, Luoyang Institute of Science and Technology, Luoyang, China; Department of Computer Science and Technology, Xi'an Jiaotong University, Xi'an, China ' School of Computer Science and Engineering, Luoyang Institute of Science and Technology, Luoyang, China ' School of Computer Science and Engineering, Luoyang Institute of Science and Technology, Luoyang, China

Abstract: During the data gathering process in sensor networks, lots of transmitted data packets can be lost due to the limiting node energy and the influence of data redundancy. In order to solve this problem, a prediction model is proposed for lost packets based on data reconstruction on lossy links in sensor networks. In this model, retransmission is adopted for data recovery when a random packet loss is predicted while prediction algorithms based on time sequences can be employed for data recovery when a random packet loss cannot be predicted. The operation of the whole system can still be guaranteed with this model when the packet loss probability of the network is lower than 15% while the relative error for data reconstruction remains between 0.17%~0.22% when the packet loss probability is higher than 22%. It is thus proven that this prediction model exhibits a strong stability and flexibility.

Keywords: sensor networks; lossy links; matching for lost packets; data redundancy; data reconstruction.

DOI: 10.1504/IJCSE.2019.100238

International Journal of Computational Science and Engineering, 2019 Vol.19 No.2, pp.177 - 188

Received: 06 Apr 2018
Accepted: 11 Jun 2018

Published online: 17 Jun 2019 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article