Computational Intelligence in Wireless Sensor Networks

Abstract

Wireless sensor networks (WSNs) are networks of distributed autonomous devices that can sense or monitor physical or environmental conditions cooperatively. WSNs face many challenges, mainly caused by communication failures, storage and computational constraints and limited power supply.Paradigms of Computational Intelligence (CI) have been successfully used in recent years to address various challenges such as optimal deployment, data aggregation and fusion, energy aware routing, task scheduling, security, and localization

CI provides adaptive mechanisms that exhibit intelligent behaviour in complex and dynamic environments like WSNs. CI brings about flexibility, autonomous behaviour, and robustness against topology changes, communication failures and scenario changes. However, WSN developers can make use of potential CI algorithms to overcome the challenges in Wireless Sensor Network. The seminar includes some of the WSN challenges and their solutions using CI paradigms.

Typically, sensor nodes are grouped in clusters, and each cluster has a node that acts as the cluster head. All nodes forward their sensor data to the cluster head, which in turn routes it to a specialized node called sink node (or base station) through a multi-hop wireless communication. However, very often the sensor network is rather small and consists of a single cluster with a single base station .Other scenarios such as multiple base stations or mobile nodes are also possible. Resource constraints and dynamic topology pose technical challenges in network discovery, network control and routing, collaborative information processing, querying, and tasking . CI combines elements of learning, adaptation, evolution and fuzzy logic to create intelligent machines. In addition to paradigms like neuro-computing, reinforcement learning, evolutionary computing and fuzzy computing, CI encompasses techniques that use swarm intelligence, artificial immune systems and hybrids of two or more of the above.

Introduction

Paradigms of CI have found practical applications in areas such as product design, robotics, intelligent control, biometrics and sensor networks. Researchers have successfully used CI techniques to address many challenges in WSNs. However, various research communities are developing these applications concurrently, and a single overview thereof does not exist. Their aim is to bridge the gap between CI approaches and applications, which provide the WSN researchers with new ideas and incentives. A discussion on yet-unexplored challenges in WSNs, and a projection on potential CI applications in WSN are presented with an objective of encouraging researchers to use CI techniques in WSN applications.Paradigms of CI have found practical applications in areas such as product design, robotics, intelligent control, biometrics and sensor networks. Researchers have successfully used CI techniques to address many challenges in WSNs. However, various research communities are developing these applications concurrently, and a single overview thereof does not exist. Their aim is to bridge the gap between CI approaches and applications, which provide the WSN researchers with new ideas and incentives.

A discussion on yet-unexplored challenges in WSNs, and a projection on potential CI applications in WSN are presented with an objective of encouraging researchers to use CI techniques in WSN applications.

Fuzzy Logic:

Classical set theory allows elements to be either included in a set or not. This is in contrast with human reasoning, which includes a measure of imprecision or uncertainty, which is marked by the use of linguistic variables such as most, many, frequently, seldom etc. This approximate reasoning is modeled by fuzzy logic, which is a multi-valued logic that allows intermediate values to be defined between conventional threshold values. Fuzzy systems allow the use of fuzzy sets to draw conclusions and to make decisions. Fuzzy sets differ from classical sets in that they allow an object to be a partial member of a set. For example, a person may be a member of the set tall to a degree of 0.8 . In fuzzy systems, the dynamic behavior of a system is characterized by a set of linguistic fuzzy rules based on the knowledge of a human expert.

Fuzzy rules are of the general form: if antecedent(s) then consequent(s), where antecedents and consequents are propositions containing linguistic variables. Antecedents of a fuzzy rule form a combination of fuzzy sets through the use of logic operations. Thus, fuzzy sets and fuzzy rules together form the knowledge base of a rule-based inference system as shown in Figure

Evolutionary algorithms model the natural evolution, which is the process of adaptation with the aim of improving survival capabilities through processes such as natural selection, survival-of-the-fittest, reproduction, mutation, competition and symbiosis. EC encompasses a variety of EAs that share a common underlying idea of survival-of-the-fittest. EAs use a population of solution candidates called chromosomes. Chromosomes are composed of genes, which represent a distinct characteristic. A fitness function, which the EA seeks to Computational Intelligence in WSNs maximize over the generations, quantifies the fitness of an individual chromosome.

Process of reproduction is used to mix characteristics of two or more chromosomes (called parents) into the new ones (called offspring). Offspring chromosomes are mutated through small, random genetic changes in order to increase diversity. Some fittest chromosomes are selected to go into the next generation, and the rest are eliminated. The process is repeated generation after generation until either a fit-enough solution is found or a previously set computational limit is reached.

Swarm Intelligence

Swarm Intelligence (SI) originated from the study of collective behavior of societies of biological species such as flocks of birds, shoals of fish and colonies of ants. SI is the property of a system whereby collective behaviors of unsophisticated agents interacting locally with their environment cause coherent functional global patterns to emerge. While graceful but unpredictable bird-flock choreography inspired the development of particle swarm optimization , impressive ability of a colony of ants to find shortest path to their food inspired the development of ant colony optimization . The honey bee algorithm mimics foraging behavior of swarms of honey bees

Fitness of a particle is determined from its position. The fitness is defined in such a way that a particle closer to the solution has higher fitness value than a particle that is far away. In each iteration, velocities and positions of all particles are updated to persuade them to achieve better fitness. The process of updating is repeated iteratively either until a particle reaches the global solution within permissible tolerance limits, or until a sufficiently large number of iterations is reached. Magnitude and direction of movement of a particle is influenced by its previous velocity, its experience and the knowledge it acquires from the swarm through social interaction.