A lightweight data transmission reduction method based on a dual prediction technique for sensor networks

Abstract

An essential design concern in a resource‐constraint sensor network is optimizing data transmission for each sensor node (SN) to prolong the network lifetime. Many research works cited that the dual prediction technique remains the most efficient technique for data reduction. A large amount of redundant data is usually transmitted across the network, leading to collisions, loss of data, and energy dissipation. This article proposes a data transmission reduction method (DTRM) to solve these problems, implemented on the cluster heads and operates in rounds. DTRM is lightweight in processing, has low complexity costs, and needs a limited memory footprint, but it is robust and effective. It can be combined with any form of cluster‐based data aggregation. We have incorporated the proposed DTRM with the data aggregation‐adaptive frame method (DA‐AFM), implemented on the SNs within the clusters. DA‐AFM can eliminate temporal redundancies and correlations in the sensor's time‐series readings. This helps the SN take fewer readings, which improves the efficiency of reducing data transmission and decreases the amount of energy spent during sensing. The proposed DTRM approach decreases the average transmission rates of data while maintaining data quality. This study is evaluated on real data obtained from the Intel Berkeley Lab and compared with three recent data reduction techniques focused on prediction. The results show that DTRM consumes up to 70% less energy while preserving the expected quality of data and reducing transmission.

Data transmission reduction techniques for improving network lifetime in wireless sensor networks: An up‐to‐date survey from 2017 to 2022

“…GO‐LMS and GA‐PCA provide higher data transmission reduction ratio and high data recovery accuracy at BS but do not depict estimations of energy consumption and network lifetime. Jain and Kumar 116 in 2021 proposed a data transmission reduction method (DTRM) to reduce redundant data transmission which leads to collisions, data loss, and energy consumption in WSN. DTRM is implemented on the CHs and BS and it can be integrated with several cluster‐based data aggregation techniques.…”

“…Energy‐efficient redundant data transmission reduction is required for scaling such WSN applications. As a result, achieving a longer network lifetime is still considered the most difficult issue in the research 12,13,57–65,67,79,83–94,109–112,114–117,121 …”

“…The aggregator or the CH combines the data collected by multiple SNs to reduce redundant data transmission and thus reduce data size and maintain accuracy before transmitting it to the BS. However, data aggregation requires processing as it implements aggregation function and aggregation schedule 57,60,79,83,84,86,92–93,112,116–117 …”

A two‐vectordata‐predictionmodel forenergy‐efficient data‐aggregationin wireless sensor network

“…Jain et al 24 propose a data transmission reduction method based on a dual prediction technique and is implemented on the cluster heads (CHs)…”

“…Jain et al 24 propose a data transmission reduction method based on a dual prediction technique and is implemented on the cluster heads (CHs) to operate in rounds. DTRM is lightweight in processing, has low complexity costs, and needs a limited memory footprint, but also it is robust and effective.…”