A Real-Time Massive Data Processing Technique for Densely Distributed Sensor Networks

Harb, Hassan; Makhoul, Abdallah; Jaoude, Chady Abou

doi:10.1109/access.2018.2872687

Cited by 28 publications

(17 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Numerous aggregation and compression approaches that take advantage of the temporal correlation in the collected data have been proposed [10,11,12,13,14].…”

Section: Data Reduction In Iot Applicationsmentioning

confidence: 99%

An energy efficient IoT data compression approach for edge machine learning

Azar

Makhoul

Barhamgi

et al. 2019

Future Generation Computer Systems

Self Cite

195

102

View full text Add to dashboard Cite

Many IoT systems generate a huge and varied amount of data that need to be processed and responded to in a very short time. One of the major challenges is the high energy consumption due to the transmission of data to the cloud. Edge computing allows the workload to be offloaded from the cloud at a location closer to the source of data that need to be processed while saving time, improving privacy, and reducing network traffic. In this paper, we propose an energy efficient approach for IoT data collection and analysis. First of all, we apply a fast error-bounded lossy compressor on the collected data prior to transmission, that is considered to be the greatest consumer of energy in an IoT device. In a second phase, we rebuild the transmitted data on an edge node and process it using supervised deep learning techniques. To validate our approach, we consider the context of driving behavior monitoring in intelligent vehicle systems where vital signs data are collected from the driver using a Wireless Body Sensor Network (WBSN) and wearable devices and sent to an edge node for stress level detection. The experimentation results show that the amount of transmitted data has been reduced by up to 103 times without affecting the quality of medical data and driver stress level prediction accuracy.

show abstract

“…Numerous aggregation and compression approaches that take advantage of the temporal correlation in the collected data have been proposed [10,11,12,13,14].…”

Section: Data Reduction In Iot Applicationsmentioning

confidence: 99%

An energy efficient IoT data compression approach for edge machine learning

Azar

Makhoul

Barhamgi

et al. 2019

Future Generation Computer Systems

Self Cite

195

102

View full text Add to dashboard Cite

show abstract

“…Thus, the data fidelity in terms of structural similarity in the continuous sensing applications for WSNs can be accomplished by SFDC. In addition, Harb et al [30] proposed a method of compressing collected data at the first level as well as eliminating redundant data generated by the neighboring node at the second level. The proposed scheme adapts clustering methods (EKmeans and TopK) to decrease the communication costs and enhance data mining in WSN.…”

Section: Technical Approaches To Wsn-based Big Data Systemsmentioning

confidence: 99%

Wireless Sensor Networks for Big Data Systems

Kim

Shah

et al. 2019

Sensors

View full text Add to dashboard Cite

Before discovering meaningful knowledge from big data systems, it is first necessary to build a data-gathering infrastructure. Among many feasible data sources, wireless sensor networks (WSNs) are rich big data sources: a large amount of data is generated by various sensor nodes in large-scale networks. However, unlike typical wireless networks, WSNs have serious deficiencies in terms of data reliability and communication owing to the limited capabilities of the nodes. Moreover, a considerable amount of sensed data are of no interest, meaningless, and redundant when a large number of sensor nodes is densely deployed. Many studies address the existing problems and propose methods to overcome the limitations when constructing big data systems with WSN. However, a published paper that provides deep insight into this research area remains lacking. To address this gap in the literature, we present a comprehensive survey that investigates state-of-the-art research work on introducing WSN in big data systems. Potential applications and technical challenges of networks and infrastructure are presented and explained in accordance with the research areas and objectives. Finally, open issues are presented to discuss promising directions for further research.

show abstract

“…We take the sensor node 1 and we present its correlation with neighbouring nodes for temperature, humidity and light conditions. The results show that the temperature sensor in node 1 has correlation with the set of temperature sensors [2,13,24,26,27,30,31,33,43] while the humidity and light sensors in node 1 have correlation with [13,24,26,27,28,29,30,31,33] and [2,3,15,21,22,23,26,32,33,35,36,41,43,46] respectively. Thus, we can deduce the following: 1) the sensors in the same node do not have the same number of correlated sensors; the light sensor has more correlations than temperature and humidity sensors.…”

Section: Periodic Number Of Candidatesmentioning

confidence: 99%

An energy-efficient data prediction and processing approach for the internet of things and sensing based applications

Harb

Jaoude

Makhoul

2019

Peer-to-Peer Netw. Appl.

Self Cite

View full text Add to dashboard Cite

The Internet of Things (IoT) is a vision in which billions of smart objects are linked together. In the IoT, "things" are expected to become active and enabled to interact and communicate among themselves and with the environment by exchanging data and information sensed about the environment. In this future interconnected world, multiple sensors join the internet dynamically and use it to exchange information all over the world in semantically interoperable ways. Therefore, huge amounts of data are generated and transmitted over the network. Thus, these applications require massive storage, huge computation power to enable real-time processing, and high-speed network. In this paper, we propose a data prediction and processing approach aiming to reduce the size of data collected and transmitted over the network while guaranteeing data integrity. This approach is dedicated to devices/sensors with low energy and computing resources. Our proposed technique is composed of two stages: on-node prediction model and in-network aggregation algorithm. The first stage uses the Lagrange interpolation polynomial model to reduce the amount of data generated by sensor nodes while, the second stage uses a statistical test, i.e. Kolmogorov-Smirnov, and aims to reduce the redundancy between data generated by neighbouring nodes. Simulation on real sensed data reveals that the proposed approach significantly reduces the amount of data generated and transmitted over the network thus, conserving sensors' energies and extending the network lifetime.

show abstract

A Real-Time Massive Data Processing Technique for Densely Distributed Sensor Networks

Cited by 28 publications

References 30 publications

An energy efficient IoT data compression approach for edge machine learning

An energy efficient IoT data compression approach for edge machine learning

Wireless Sensor Networks for Big Data Systems

An energy-efficient data prediction and processing approach for the internet of things and sensing based applications

Contact Info

Product

Resources

About