This paper presents a novel entropy correlation model with less computation complexity that could be applied practically. Moreover, two energy efficient aggregation schemes including on-off scheme which offers an efficient way to choos representative nodes in a cluster with permitted distortion and compression scheme which reduces in-network message length suitable to high correlation data are also presented in this paper using the proposed correlation models. | Journal of Science & Technology 128 (2018) 041-047 Study on Entropy Correlation in Wireless Sensor Networks for Energy Efficiency Nguyen Thi Thanh Nga*, Nguyen Kim Khanh, Ngo Hong Son Hanoi University of Science and Technology, No. 1, Dai Co Viet, Hai Ba Trung, Hanoi, Viet Nam Received: January 25, 2018; Accepted: May 25, 2018 Abstract Correlation characteristic can bring many significant potential advantages for the development of efficient communication protocols for wireless sensor networks. To exploit the correlation in WSNs, it is necessary to build the correlation model. However, most of the present correlation models only consider the linear and distance dependence correlation or computation complexity. This paper presents a novel entropy correlation model with less computation complexity that could be applied practically. Moreover, two energy efficient aggregation schemes including on-off scheme which offers an efficient way to choose representative nodes in a cluster with permitted distortion and compression scheme which reduces in-network message length suitable to high correlation data are also presented in this paper using the proposed correlation models. Keywords: Entropy correlation coefficient, Correlation model, Compression, Representative node, Distortion 1. Introduction* Matérn. In [4], the correlation model is proposed such as correlation coefficient is a function of distance among nodes. Other research papers consider the correlation as the similarity of sensed data [5]. Some papers define the correlation model in different ways such as linear predictive model [6], node weight [7], data density correlation degree [8]. In recent years, the advanced development in micro-electro-mechanical systems (MEMS) and the wireless communications have enabled the wide deployment of wireless sensor networks (WSN) which expand sensing capabilities in space and time that can satisfy requests from various modern applications. Because of low-cost, small in .
