A Reliable Data Compression Scheme in Sensor-Cloud Systems Based on Edge Computing

Shaofei Lu,Yingping Lu,Qinhua Xia,Jingke She,Xiaolin Tang,Xuyang Zhang

doi:10.1109/access.2021.3068753

Shaofei Lu, Yingping Lu + Show 4 more

Open Access

https://doi.org/10.1109/access.2021.3068753

Copy DOI

Abstract

The rapid development of the IoT and cloud computing has spawned a new network structure - sensor-cloud system (SCS) where sensors, sensor networks, and cloud computing are integrated to perform data sensing, collection, transmission, and decision making. The large-scale deployment of sensors creates a massive amount of data, posing new challenges in data transmission and storage. As an intermediate platform between IoT and cloud platforms, edge computing provides IoT with data collection, processing, and scheduling services. This paper proposes a hybrid data compression scheme that incorporates lossy and lossless compression in SCS based on edge computing to address the increasing challenges. Moreover, we propose a new reliable lossy compression algorithm DFan, based on the simplified Fan algorithm with a high compression ratio (CR). By introducing the data tolerable deviation, DFan transforms single-factor decision-making into multi-factor decision-making, reducing the error of lossy compression. Through experiments on IntelLab and MIT-BIH datasets, the proposed hybrid data compression scheme achieves an overall CR of 4.21× and 3.88×, respectively. The lossy CR of DFan is 6.42× and 5.1×, respectively, and the Percentage RMS Difference (PRD) caused by lossy compression is 0.27% and 0.56%, respectively. The hybrid compression scheme, high compression ratio, and reliable data restoration make this scheme attractive to the data processing of sensors in SCS.

Highlights

With the rapid development and broad applications of wireless sensor networks (WSN) coupled with massive deployment and wide reception of cloud computing, the integration of WSNs and cloud computing, aka Sensor-Cloud System (SCS) has received considerable attention in both academic and industry communities [1]
In this paper, taking the actual application scenarios of data into consideration, we propose a novel hybrid data compression scheme based on edge computing to address the data transmission challenges in SCS
This paper proposes a reliable sensor data processing scheme for sensor-cloud systems based on edge computing

Summary

INTRODUCTION

With the rapid development and broad applications of wireless sensor networks (WSN) coupled with massive deployment and wide reception of cloud computing, the integration of WSNs and cloud computing, aka Sensor-Cloud System (SCS) has received considerable attention in both academic and industry communities [1]. S. Lu et al.: Reliable Data Compression Scheme in SCSs Based on Edge Computing only effective data are received, stored and processed in the cloud. Lu et al.: Reliable Data Compression Scheme in SCSs Based on Edge Computing only effective data are received, stored and processed in the cloud This can lead to lower storage requirements, faster data processing and better user experience. In this paper, taking the actual application scenarios of data into consideration, we propose a novel hybrid data compression scheme based on edge computing to address the data transmission challenges in SCS. This scheme combines both lossy and lossless compression algorithms, realizing mixed lossless transmission and transmission rate selection.

RELATED WORK

EVALUATION METRICS

PERFORMANCE IMPROVEMENT EVALUATION

Findings

CONCLUSION