ABSTRACT
Soft computing measures are an understanding of the imprecision, defenselessness, ambiguity, and approximation.
Just as human beings can survey the likelihood of possibilities, comparatively, soft computing systems utilize some intelligence-based strategies to evaluate the ongoing issue with diagnostic models [1]. The fundamental segments of soft computing incorporate machine learning, probabilistic thinking, swarm intelligence (for example, ant colony optimization and particle swarm optimization [PSO]), and ANN (artificial neural network), perception, fuzzy logic, evolutionary computing, and genetic algorithms [2]. In Chapter 4, there is a broad analysis of intelligence-based soft computing strategies connected in different operational parts of wireless sensor networks.
Wireless sensor networks (WSNs) are an assembly of self-organized sensors which are grouped together to screen and record physical or natural conditions (i.e., those used to gauge temperature, sound, and weight) and pass accumulated data to the central region. WSNs make the connection between correct and virtual conditions, which makes them more useful for real-time applications. Mainly, WSNs are most suitable [24] for military applications, but nowadays they are used in various other sectors, like industrial applications, consumer applications, healthcare applications, and many more. Despite having many advantages, there are some challenging issues that also takes place in WSNs, like radio range problems, energy hole problems, routing issues, coverage problems, load balancing problems, and so on [25]. These issues affect different factors of WSNs, such as energy consumption, stability, quality, deployment time, and the lifetime of a network, which degrade the performance of the WSN. To solve these issues, various researchers develop different mechanisms. Considering all of the above critical reasons, Chapter 4 analyzes the important applications of soft computing, which is an emerging computational strategy in the current scenario of various improved and challenging solutions in WSNs. Different kinds of soft computing paradigms are discussed as well as various types of soft computing techniques (such as swarm intelligence, fuzzy logic, neural networks, reinforcement learning, and evolutionary algorithms [26]) with their implementation details with respect to various suitable applications in wireless sensor networks.
