Machine Learning in Wireless Sensor Networks: Algorithms, Strategies, and Applications

Alsheikh, Mohammad Abu; Lin, Shaowei; Niyato, Dusit; Tan, Hwee-Pink

doi:10.1109/comst.2014.2320099

Cited by 783 publications

(458 citation statements)

References 144 publications

(249 reference statements)

Supporting

Mentioning

454

Contrasting

Unclassified

Order By: Relevance

“…Artificial neural networks (ANNs) have been successfully used in the development of novel solutions for WSNs as they can capture nonlinear structures in data [9]. For example, an ANN-based method for minimizing environmental influences on sensor responses was proposed in [28].…”

Section: Neural Autoencoders (Aes)mentioning

confidence: 99%

“…The error bound mechanism first computes the residual r = x − p as shown in Figure 6. Any entry of the residual vector exceeding the bound will be transmitted, using the residual code ε = residualCode(r, ) = 1 J , (r j ) j∈J (9) where J ⊂ {1, . .…”

Section: Error Bound Mechanismmentioning

confidence: 99%

“…The routines for compression and decompression are implemented using only linear and sigmoidal operations. The compressed data features can be fed to machine learning algorithms [9] to automatically predict human activities and environmental conditions. 2) Unlike many conventional methods, our unified method is easily customized for both temporal and spatial compression.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Rate-Distortion Balanced Data Compression for Wireless Sensor Networks

et al. 2016

Self Cite

View full text Add to dashboard Cite

Abstract-This paper presents a data compression algorithm with error bound guarantee for wireless sensor networks (WSNs) using compressing neural networks. The proposed algorithm minimizes data congestion and reduces energy consumption by exploring spatio-temporal correlations among data samples. The adaptive rate-distortion feature balances the compressed data size (data rate) with the required error bound guarantee (distortion level). This compression relieves the strain on energy and bandwidth resources while collecting WSN data within tolerable error margins, thereby increasing the scale of WSNs. The algorithm is evaluated using real-world datasets and compared with conventional methods for temporal and spatial data compression. The experimental validation reveals that the proposed algorithm outperforms several existing WSN data compression methods in terms of compression efficiency and signal reconstruction. Moreover, an energy analysis shows that compressing the data can reduce the energy expenditure, and hence expand the service lifespan by several folds.Index Terms-Lossy data compression, error bound guarantee, compressing neural networks, Internet of things.

show abstract

Section: Neural Autoencoders (Aes)mentioning

confidence: 99%

Section: Error Bound Mechanismmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Rate-Distortion Balanced Data Compression for Wireless Sensor Networks

et al. 2016

Self Cite

View full text Add to dashboard Cite

show abstract

“…Alsheikh et al [11] present an overview of existing machine learning techniques used in WSNs. They group them into the categories of supervised, unsupervised and reinforcement learning.…”

Section: Related Workmentioning

confidence: 99%

Influence of on-Device Measurement Analysis on Energy Efficiency in Machine-to-Machine Systems

Skočir

Kušek

Jezic

2015

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

Abstract-Machine-to-Machine Communication (M2M) enables communication between heterogeneous devices without human intervention. It is considered to be a key enabler technology for the concept of Internet of Things (IoT) and Cyber Physical Systems (CPS). With M2M's integration with Wireless Sensor Networks (WSN), information from different kinds sensors can be obtained. In order to discover useful knowledge from sensor data, various data mining techniques need to be applied. Due to the development of microprocessors on end devices in M2M system which collect data from sensors, data processing can also be executed on those devices. However, since end devices are often battery powered, energy consumption when running those algorithms needs to be taken into account. In this paper we implement an algorithm in M2M system, on Libelium Waspmote devices, which detects temperature plummeting in an indoor space. Afterwards, energy consumption of Waspmote devices is analyzed for two cases: when algorithm is executed on-device and when algorithm is executed on gateway or on back-end system.

show abstract

“…In our case, we used more than one prediction algorithm so that we do not rely on a specific learning technique. We applied four of the best well-known approaches [34]: Support Vector Machines (SVM), k-Nearest Neighbors (kNN), Regression Trees (RT) and Rule-Based Regression (RBR).…”

Section: A Experimental Frameworkmentioning

confidence: 99%

Tracking and Predicting End-to-End Quality in Wireless Community Networks

Millan

Molina

Dimogerontakis

et al. 2015

2015 3rd International Conference on Future Internet of Things and Cloud

View full text Add to dashboard Cite

Abstract-Community networks are an emergent model with mottos like "a free net for everyone is possible" or "don't buy the network, be the network". Their social impact is measurable, as the community is provided with the right and opportunity of communication. The combination of wired and wireless links in these networks, and the unreliable nature of the wireless medium, poses several challenges to the routing protocol. End-toEnd quality tracking helps the routing layer to select paths that maximize the delivery rate and minimize traffic congestion. We believe that End-to-End quality prediction can be a technique that surpasses End-to-End quality tracking by foreseeing which paths are more likely to change quality. In this work, we focus on End-to-End quality prediction by means of time-series analysis. We apply this prediction technique in the routing layer of largescale, distributed, and decentralized networks. We demonstrate that it is possible to accurately predict End-to-End Quality with an average Mean Absolute Error of just 2.4%. Particularly, we analyze the path properties and path ETX behavior to identify the best prediction algorithm. Moreover, we analyze the EtEQ prediction accuracy some steps ahead in the future and also its dependency of the time of the day.

show abstract

Machine Learning in Wireless Sensor Networks: Algorithms, Strategies, and Applications

Cited by 783 publications

References 144 publications

Rate-Distortion Balanced Data Compression for Wireless Sensor Networks

Rate-Distortion Balanced Data Compression for Wireless Sensor Networks

Influence of on-Device Measurement Analysis on Energy Efficiency in Machine-to-Machine Systems

Tracking and Predicting End-to-End Quality in Wireless Community Networks

Contact Info

Product

Resources

About