+ All Categories
Home > Presentations & Public Speaking > Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Date post: 09-Jun-2015
Category:
Upload: mabualsh
View: 941 times
Download: 0 times
Share this document with a friend
Popular Tags:
25
Machine Learning (ML) in Wireless Sensor Networks (WSNs) Mohammad Abu Alsheikh School of Computer Engineering Nanyang Technological University June 2014 1 / 25
Transcript
Page 1: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Mohammad Abu Alsheikh

School of Computer EngineeringNanyang Technological University

June 2014

1 / 25

Page 2: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Survey and summary of the field

This material is based on our recently published survey article:Alsheikh, M.A.; Lin, S.; Niyato, D.; Tan, Hwee-Pink, "Machine Learning in Wireless SensorNetworks: Algorithms, Strategies, and Applications," IEEE Communications Surveys &Tutorials, DOI: 10.1109/COMST.2014.2320099

2 / 25

Page 3: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Outline

Outline

1 Introduction

2 Machine learning (ML) algorithms

3 Functional challenges

4 Non-functional challenges

5 Some open research directions

6 Summary

3 / 25

Page 4: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Introduction Motivation

Motivation

A wireless sensor network (WSN) is composed of multiple autonomous, tiny, low cost andlow power sensor nodes that gather data about their environment and collaborate toforward sensed data to centralized backend unitsMachine learning (ML) is the adoption of computational methods for improving machineperformance by detecting and describing consistencies and patterns in training data [LS95]ML was introduced in the late 1950’s as a technique for artificial intelligence (AI) [Ayo10]

4 / 25

Page 5: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Introduction Motivation

Why machine learning is important

Wireless sensor networks monitor dynamic environments that change rapidly over time.This dynamic behavior is either caused by external factors or initiated by the systemdesigners themselvesTo adapt to such conditions, sensor networks often adopt machine learning techniques toeliminate the need for unnecessary redesign. Machine learning also inspires manypractical solutions that maximize resource utilization and prolong the lifespan of thenetwork

5 / 25

Page 6: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Introduction Motivation

Summary of benefits (1)

Machine learning is important in WSN applications for the following main reasons:

Sensor networks usually monitor dynamic environments that change rapidly over time,and it is desirable to develop sensor networks that can adapt and operate efficiently insuch environments.In some applications, due to the unexpected behavior patterns that may arise in suchscenarios, system designers may develop solutions that initially may not operate asexpected. System designers would rather have robust ML algorithms that are able tocalibrate itself to newly acquired knowledge.WSNs are usually deployed in complicated environments where researchers cannot buildaccurate mathematical models to describe the system behavior. Meanwhile, some tasks inWSNs can be prescribed using simple mathematical models but may still need complexalgorithms to solve them (e.g., the routing problem [KEW02, AKK04]). Under similarcircumstances, ML provides low-complexity estimates for the system model.

6 / 25

Page 7: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Introduction Motivation

Summary of benefits (2)

Sensor network designers often have access to large amounts of data but may be unable toextract important correlations in them, e.g., in node connectivity and energy sustainability.ML methods can then be used to discover important correlations in the sensor dataNew uses and integrations of WSNs, such as in cyber-physical systems (CPS),machine-to-machine (M2M) communications, and Internet of things (IoT) technologies,have been introduced with a motivation of supporting more intelligent decision-makingand autonomous control [WCX+13]. Here, Machine learning methods can then be usedto discover important correlations in the sensor data with limited human intervention[Ben09].

7 / 25

Page 8: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Introduction Motivation

Drawbacks and limitations

However, there are a few drawbacks and limitations that should be considered:

1 As a resource limited framework, WSN drains a considerable percentage of its energybudget to predict the accurate hypothesis and extract the consensus relationship amongdata samples. Thus, the designers should consider the trade-off between thealgorithm’s computational requirements and the learned model’s accuracy

2 Generally speaking, learning by examples requires a large data set of samples to achievethe intended generalization capabilities (i.e., fairly small error bounds), and thealgorithm’s designer will not have the full control over the knowledge formulation process[Hof90]

8 / 25

Page 9: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Machine learning (ML) algorithms

Overview

ML experts recognize it as a rich field with very large themes and patterns. Understandingsuch themes will be beneficial to those who wish to apply machine learning to WSNs.

1 Supervised learning: algorithms are provided with a labeled training data set. This set isused to build the system model representing the learned relation between the input,output and system parameters

2 Unsupervised learning: algorithms are not provided with labels (i.e., there is no outputvector). Basically, the goal of an unsupervised learning algorithm is to classify the samplesets to different groups (i.e., clusters) by investigating the similarity between the inputsamples

3 Reinforcement learning (RL): the agent, e.g., sensor node, learns by interacting with itsenvironment (i.e., online learning)

9 / 25

Page 10: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Machine learning (ML) algorithms Supervised learning

Supervised learning

In supervised learning, a labeled training set (i.e., predefined inputs and known outputs) isused to build the system model. This model is used to represent the learned relation betweenthe input, output and system parameters. Widely known examples of such algorithms are:

1 K-nearest neighbor (k-NN)2 Decision tree (DT)3 Neural networks (NNs)4 Support vector machines (SVMs)5 Bayesian statistics

10 / 25

Page 11: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Machine learning (ML) algorithms Unsupervised learning

Unsupervised learning

Unsupervised learners are not provided with labels (i.e., there is no output vector). Basically,the goal of an unsupervised learning algorithm is to classify the sample set into differentgroups by investigating the similarity between them. Examples of such methods are:

1 K-means clustering2 Principal component analysis (PCA)3 Self-organizing maps (or Kohonen’s maps)

11 / 25

Page 12: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Machine learning (ML) algorithms Reinforcement Learning

Reinforcement learning (RL)

Reinforcement learning enables an agent (e.g., a sensor node) to learn by interacting with itsenvironment. The agent will learn to take the best actions that maximize its long-termrewards by using its own experience.

1 Q-learning

12 / 25

Page 13: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Functional challenges

Overview

In the design of WSNs, it is important to consider power and memory constraints of sensornodes, topology changes, communication link failures, and decentralized managementMachine learning paradigms have been successfully adopted to address various functionalchallenges of wireless sensor networks such as energy aware and real-time routing, queryprocessing and event detection, localization, node clustering and data aggregation

13 / 25

Page 14: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Functional challenges Routing in WSNs

Routing in WSNs

Machine learning allows a sensor network to learn from previous experiences, make optimalrouting actions and adapt to the dynamic environment. The benefits can be summarized asfollows:

1 Able to learn the optimal routing paths that will result in energy saving and prolongingthe lifetime of dynamically changing WSNs

2 Reduce the complexity of a typical routing problem by dividing it into simpler sub-routingproblems. In each sub-problem, nodes formulate the graph structures by considering onlytheir local neighbors, thus achieving low cost, efficient and real-time routing

3 Meet QoS requirements in routing problem using relatively simple computational methodsand classifiers

14 / 25

Page 15: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Functional challenges Clustering and data aggregation

Clustering & data aggregation

Principally, ML techniques improve the operation of node clustering and data aggregation asfollows:

1 Usage of machine learning to compress data locally at cluster heads by efficientlyextracting similarity and dissimilarity (e.g., from faulty nodes) in different sensors’ readings

2 Machine learning algorithms are employed to efficiently elect the cluster head, whereappropriate cluster head selection will significantly reduce energy consumption andenhance the network’s lifetime

15 / 25

Page 16: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Functional challenges Event detection and query processing

Event detection & query processing

Fundamentally, ML offers solutions to restrict query areas and assess event validity for efficientevent detection and query processing mechanisms. This adoption will result in the followingbenefits:

1 Learning algorithms enable the development of efficient event detection mechanisms withlimited requirements of storage and computing resources. Besides they are able to assessthe accuracy of such events using simple classifiers

2 Machine learning facilitates the development of effective query processing techniques forWSNs, that determine the search regions whenever a query is received without floodingthe whole network

16 / 25

Page 17: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Functional challenges Localization and targeting objects

Localization & targeting objects

Sensor nodes may encounter changes in their location after deployment (e.g., due tomovement). The benefits of using machine learning algorithms in sensor node localizationprocess can be summarized as follows:

1 Converting the relative locations of nodes to absolute ones using few anchor points. Thiswill eliminate the need for range measurement hardware to obtain distance estimations

2 In surveillance and object targeting systems, machine learning can be used to divide themonitored sites into a number of clusters, where each cluster represents specific locationindicator

17 / 25

Page 18: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Functional challenges Medium access control (MAC)

Medium access control (MAC)

Recently, machine learning methods have been used to enhance the performance of MACprotocols in WSNs. Specifically, this is achieved through the following points:

1 Machine learning can be used to adaptively determine the duty cycle of a node using thetransmission history of the network. In particular, the nodes, which are able to predictwhen the other nodes’ transmissions will finish, can sleep in the meantime and wake up(to transmit data) just when the channel is expected to be idle (i.e., when no other nodeis transmitting)

2 Achieving secured data transmission by combining the concepts of machine learning andMAC protocols. Such MAC layer security schemes are independent of the proposedapplication and are able to iteratively learn sporadic attack patterns

18 / 25

Page 19: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Non-functional challenges

Overview

Non-functional requirements include specifications that are not related to the basic operationalbehavior of the system. For example, WSN designers may need to ensure that the proposedsolution is always capable of providing up-to-date information about the monitoredenvironment

19 / 25

Page 20: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Non-functional challenges Security and anomaly intrusion detection

Security & anomaly intrusion detection

Basically, WSN security enhancements by adopting machine learning techniques will result inthe following earnings:

1 Save node’s energy and significantly expand WSN lifetime by preventing the transmissionof the outlier, misleading data

2 Enhance network reliability by eliminating faulty and malicious readings. In the same way,avoiding the discovery of unexpected knowledge that will be converted to important, andoften critical actions

3 Online learning and prevention (without human intervention) of malicious attacks andvulnerabilities

20 / 25

Page 21: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Non-functional challenges Quality of service (QoS), data integrity and fault detection

QoS, data integrity & fault detection

In the following, we review the latest efforts of using machine learning techniques to achievespecific QoS and data integrity constraints. In brief, this adoption results in the followingadvantages:

1 Different machine learning classifiers are used to recognize different types of streams, thuseliminating the need for flow-aware management techniques

2 The requirements for QoS guarantee, data integrity and fault detection depend on thenetwork service and application. Machine learning methods are able to handle much ofthis while ensuring efficient resource utilization, mainly bandwidth and power utilization

21 / 25

Page 22: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Non-functional challenges Miscellaneous applications

Miscellaneous applications

Moreover, there are many other applications such as

1 Resource management2 Clock synchronization3 Air quality monitoring4 Intelligent lighting control

22 / 25

Page 23: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Some open research directions

Future applications

Although machine learning techniques have been applied to many applications in WSNs, manyissues are still open and need further research efforts.

1 Compressive sensing and sparse coding2 Distributed and adaptive machine learning techniques for WSNs3 Resource management using machine learning4 Detecting data spatial and temporal correlations using hierarchical clustering

23 / 25

Page 24: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

Summary

Conclusions

Wireless sensor networks require innovative solutions for energy aware and real-timerouting, security, scheduling, localization, node clustering, data aggregation, faultdetection and data integrityMachine learning provides a collection of techniques to enhance the ability of wirelesssensor network to adapt to the dynamic behavior of its surrounding environment

24 / 25

Page 25: Machine Learning (ML) in Wireless Sensor Networks (WSNs)

References

ReferencesJamal N Al-Karaki and Ahmed E Kamal.Routing techniques in wireless sensor networks: A survey.IEEE Wireless Communications, 11(6):6–28, 2004.

Taiwo Oladipupo Ayodele.Introduction to machine learning.In New Advances in Machine Learning. InTech, 2010.

Yoshua Bengio.Learning deep architectures for AI.Foundations and Trends in Machine Learning, 2(1):1–127, 2009.

Achim G Hoffmann.General limitations on machine learning, 1990.

B. Krishnamachari, D. Estrin, and S. Wicker.The impact of data aggregation in wireless sensor networks.In 22nd International Conference on Distributed Computing Systems Workshops, pages 575–578, 2002.

Pat Langley and Herbert A Simon.Applications of machine learning and rule induction.Communications of the ACM, 38(11):54–64, 1995.

Jiafu Wan, Min Chen, Feng Xia, Li Di, and Keliang Zhou.From machine-to-machine communications towards cyber-physical systems.Computer Science and Information Systems, 10:1105–1128, 2013.

25 / 25


Recommended