Nature Inspired Computing for Wireless Sensor Networks

Wireless sensor network (WSN) is the primary environment monitoring infrastructure of IoT system, where environmental information about hazard locations is collected through the collaborative functioning of sensor nodes. Considering the energy constraint of sensor nodes, energy efficiency is the primary requisite of protocols designed for WSN. Cluster-based routing protocols have been widely used to conserve sensors’ energy in WSN. Although, an extensive research has been done on cluster-based routing, but fault-aware routing is still an open research issue. In this chapter, we present a fault-aware routing algorithm called FAR for WSN-based on genetic algorithm (GA) approach. FAR is developed with a novel chromosome generation scheme which ensures that each CH in the network has a routing path to the remote station (RS). In FAR, we have derived a fitness function where the objective is to balance the load of CHs during data routing. The proposed algorithm has been extensively analyzed with some existing related algorithms and compared their performance in terms of different metrics like energy efficiency, number of alive nodes, and packet delivery ratio.

A wireless sensor network (WSN) is a collection of more than one sensor nodes which is used both collecting as well as sensing data from its environment (Rongbo in Int J Distrib Sens Netw 2010(1155):1–7, [1], Herbert and Donald in Schilling principles of communication systems. McGraw-Hill, New York, [2]). The main aim of this process is to achieve several operations efficiently in terms of different applications such as intelligent building, precise agriculture, medicine and health care, preventive maintenance, machine surveillance, disaster relief operation and biodiversity mapping. The stated applications are optimized efficiently in terms of cost, scalability and readiness. Although, there are so many fruitful advantages of WSN, but it consists of limited capacity of batteries which is insufficient during any operation. The sensor nodes are directly or indirectly connected with base station as well as sink node. Sometimes, due to network variation or failure of hardware sensor nodes fail to transmit the data packet. Moreover, due to limited energy, sometimes sensor node exhausts before the delivery of the data packet and gets converted into faulty node (Heinzelman, Chandrakasan and Balakrishnan in Energy efficient communication protocol for wireless micro sensor networks, pp. 8020–8030, [3], Bhajantri, Nalini in Int J Comput Netw Inf Secur 6(12):37–46, [4]). This faulty node treated as dead node during operation. So, there is need to design an effective algorithm for detecting as well as calculating total dead nodes and provide optimum solution. In this paper, an efficient technique is proposed based on the direct diffusion technique that aims to find optimum path by recovering dead nodes. The proposed algorithm enhances the network lifetime by reducing data packet loss as well as energy consumption.

In the modern era, the applications of the wireless network increase rapidly in the forms of several variations. Wireless Body Area Sensor Network (WBASN) is one of the variations of the wireless network. The purpose of this network is to monitor and detect several characteristics of the body and transmit into the proper destination. This is an intelligent wearable electronic component that consists of several inherent elements to achieve the main goal. It provides real-time diagnosis and treatment to the patients. This network contains several conflicting elements that help to raise traffic. Moreover, each node of this network consists of limited capacity of battery which is crucial point of the traffic. In this paper, an intelligent Genetic Algorithm (GA) based traffic management technique is proposed for WBASNs. The intelligent technique GA is used to enhance the network lifetime efficiently by maximizing green signal of the network. The proposed method is compared with some existing techniques in terms of some features. The final comparison shows that the proposed method outperformed the existing methods.

Sensor nodes in wireless sensor networks (WSNs) are randomly deployed in hostile environments. Real-time experience shows that sensor nodes are prone to faulty. Different faults of sensor nodes are inevitable due to internal and external influences such as adverse environmental conditions, low battery, calibration and sensor ageing effect. Since WSNs applications rely on the fidelity of data reported by the sensor nodes, it is important to detect a faulty sensor and isolate them. Most of the existing fault detection techniques in literature are statistical based which demands sensor domain knowledge and the data from the neighbouring sensors. There may be a problem of detecting a sensor fault by analyzing the sensor data in distributed approach is non-trivial since a faulty sensor reading could mimic non-faulty sensor data. Currently, machine learning algorithms have been successfully used to identify and classify various types of faults in WSNs to avoid such kind of problems. However, the application of deep learning (DL) methods has sparked great interest in both the industry and academia in the last few years. In this chapter, neural network methods will be used in fault diagnosis in WSN with DL algorithms. The focus on diagnosis of fault includes hard, soft, intermittent and transient types.

Scientist and researchers have shown higher interest in the development of biologically inspired algorithms in recent years, to solve multiple complex computational problems. Different solutions were proposed by various authors using artificial immune system (AIS), ant colony optimization (ACO), particle swarm optimization (PSO), artificial bee colony (ABC) algorithm, and genetic algorithm (GA). Fault diagnosis in wireless sensor network (WSN) is very crucial because of the application where it is used. The issue of fault diagnosis in wireless sensor network can be comparable in many aspects with an artificial immune system. Different approaches to artificial immune system have been discussed in this chapter that can be applied to fault diagnosis of wireless sensor network. An overall view of the biological immune system is explained in detail. Different artificial immune system’s applications are also discussed.

Applications of wireless sensor network (WSN) have experienced a rapid growth recently due to the heterogeneous nature of network topology. At different levels, different entities such as source node, sink node, hop nodes and base station (BS) in WSN are positioned at remote locations to perform specific assigned operations. Since each sensor node in WSN employs battery having limited capacity, it is imperative to determine optimal routing which otherwise may lead to network transmission failure. This present work aims to introduce a new approach based on the African buffalo optimization (ABO) routing in the WSN. ABO is a nature-inspired combinatorial optimization technique based on the behavior of African buffalos. Here, ABO acts as the main controller of the WSN and manages all the sensor nodes in correspondence with the BS. It also helps to transfer data packets from the source node to the sink node efficiently. Further, it enhances the network lifetime and improves other performance metrics of the WSN.

Source localization and tracking is an important application in wireless sensor networks. Sources are localized by measuring the direction of arrival from the signal impinging on the array of sensors. Most of the existing array signals processing algorithms to estimate the source directions of arrival are centralized-based which need more communication in the network. Therefore, a distributed maximum-likelihood-based direction of arrival estimation strategy is developed by following the diffusion cooperation among the nodes in sensor network to minimize the communication overhead. Modified particle swarm optimization is proposed to optimize the multimodal maximum-likelihood function in distributed scenario. The experimental results exhibit improved performance for the distributed method over non-cooperative algorithm. Further, clustering-based approach is proposed where the nodes are clustered and then act as random arrays. Then, each cluster estimates the source direction of arrival by optimizing the maximum-likelihood function locally with cooperation of other clusters. The distributed in-clustering approach offers low communication overheads and better estimation accuracy compared to other methods.

Wireless communication technologies are under the process of rapid development. In the past few years, this field has experienced a steep growth in teaching and research activity, especially, in the area of wireless ad hoc and sensor network. In recent studies, continuously research work is going on for the optimum design of the metering and low-power devices for daily life usages. Wireless sensor networks (WSNs) may be one of the options to meet the above requirement. It deals with the major issues of energy limitations, challenges of handling traffic and lifetime of the battery. Network lifetime, energy efficiency, energy consumption and solving routing problems are some of the problems that need to be discussed with WSN. To achieve these constraints, an effective optimization method may be used as an effective and useful tool. Most of the optimization techniques are inspired by some phenomenon found in nature. One of them is quasi-oppositional harmony search algorithm. Although it is under developing stage, it is still a powerful optimization technique. It has the potential ability to solve various engineering optimization problems. It has many advantages that make it applicable to use in WSN and its related work.

Routing protocols are responsible for discovering and maintaining energy-efficient routes in wireless sensor networks (WSNs) to make reliable and efficient communication. The main aim of the routing protocol design is collecting data of the sensor field efficiently. In general, routing in WSNs can be classified into three groups: flat routing, hierarchical routing, and location routing. According to the literature, hierarchical routing has more advantages compared to other types, for example, hierarchical routing reduces the redundant data transmission and balances the load among the sensor nodes in an efficient way. Recently, many intelligent-based hierarchical routing protocols are developed for controlling the consumption power of WSNs. Selecting an appropriate routing protocol for specific applications is an important and difficult task for the designer of WSNs. Therefore, this chapter presents a comprehensive survey of the recently intelligent-based hierarchical routing protocols that are developed based on Particle Swarm Optimization, Ant Colony Optimization, Fuzzy Logic, Genetic Algorithm, and Artificial Immune Algorithm. These protocols will review in detail according to different metrics such as WSN type, node deployment, control manner, network architecture, clustering attributes, protocol operation, path establishment, communication paradigm, energy model, protocol objectives, and applications. Moreover, a comparison between the reviewed protocols is investigated here depending on delay, network size, energy efficiency, and scalability with mentioning the advantages and drawbacks of each protocol.

Sensor Network is typically event-based systems. A wireless sensor network consists of many sink nodes to which it subscribes to and streams by expressing interest or queries submitted by various applications by users or organizations in general. As the sensors are battery-operated devices energy plays prime criteria in the sustainability of the network. If the size of the sensor tree pertaining to a network increases then the number of slots required for the scheduling transmission also increases. Sensors are deployed to cover various target points according to the application need; hence the proper deployment of the coverage or data gathering nodes is essential to increase the lifetime of the network. Proper deployment of coverage nodes plays a key role in load balancing and formation of the optimal subtree along with its respective base station and relay sensors. Various stochastic, deterministic, as well as heuristic-based algorithms incorporating optimization techniques to perform the node distribution has been developed and researched over the years. Researchers have also developed variegated models with bio-inspired algorithms like genetic algorithm, PSO algorithm, etc. to tackle some of the crucial problems of WSN. The paper provides a survey of some of the models and algorithms used for sensor node distribution, data aggregation, and discuss the various issues related to load balancing—advantages and disadvantages according to various applications.

In recent days bio-inspired computing is playing an important role in the area of research. Especially bio-inspired algorithms which are inspired by the behavior of nature are massively used to perform optimization. Wireless Sensor Networks (WSN) are playing vital role in all sectors. Some crucial issues of WSN are clustering, optimal routing, dynamic allocation of motes, energy and lifetime optimization. Researchers are working for several years to resolve issues of WSN for better quality of service. Bio-inspired algorithms like Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO) are playing important role in solving the issues of WSN. Still some algorithms are insufficiently studied. Bio-inspired computing is gradually gaining interest from researchers for its intelligence and adaptive nature. Although these algorithms have perceived a lot of attention from researchers in current years, the domain-specific understanding still needs to be improved for its establishment. In this chapter bio-inspired algorithms are discussed concisely with their importance in the field of wireless sensor networks.

Day by day the applications of Wireless Sensor Networks (WSNs) increases rapidly due to its flexibility and efficient functionalities in our life. In this network, multiple sensor nodes are connected together to achieve the purpose of the users. Here, the purpose of the user is anything related to sensing information from the environment. In WSN, Cluster-head (CH) is a node that plays the role of main controller within the network. It helps to manage and control all other sensor nodes of the network. The CH node is superior to other nodes with respect to energy capacity. Each node of the WSN is consists of the limited capacity of battery which is insufficient for any operation. During operation, battery cannot be charge or replace. So, energy is a crucial parameter of the network. Hence, CH selection becomes difficult task in WSN. In this paper, an intelligent method is proposed for CH selection in WSN using Teaching-Learning-Based-Optimization (TLBO). This optimization consists of two basic elements such as teacher and student based natural relation between both entities. The TLBO helps to optimize several conflicting objectives of the network efficiently in terms of learning methods. Finally, it helps to select CH efficiently and dynamically in each iteration of the network.

With the growing popularity of IoT paradigms, the number of connected medical equipment is on the rise and includes a variety of products ranging from simple devices such as thermometers to more complex machinery like smart infusion pumps, patient monitoring systems, and MRI scanners, to name a few. The advantages of such interconnected architectures are manifold, enabling better access to patients’ electronic health records, and improving the quality of care through real-time monitoring systems. However, a healthcare facility packed with medical appliances could face issues such as dropped network connections, power loss, and failure in the transmission of critical alarms. In pervasive healthcare applications, such incidents could result in disastrous consequences and are non-negotiable. Wireless sensor networks for healthcare have emerged in the recent years due to advances in sensor technology and the pressing need for the design of reliable, low-power networks. The most important key in WSN design is the task of finding the optimal path for transmission of sensor data to achieve energy efficiency and reduce costs, while catering to the needs of the application. Nature is arguably the best coach and extracting eccentric elements of biological network design proves useful in solving complex problems. Computations using nature-inspired algorithms have emerged as a new epoch in computing, serving a varied range of applications. This paper explores the need of optimization in WSN design for pervasive healthcare systems, while studying the existing class of bio-inspired algorithms and routing protocols for reliable communication in such architectures. Further, a detailed study is conducted to identify the challenges in routing sensitive data over a cluster of light, portable sensing motes, and application-specific nature-inspired models are suggested based on the observations.