Raghavendra V. Kulkarni

Computational Intelligence in Wireless Sensor Networks: A Survey

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 data aggregation and fusion, energy aware routing, task scheduling, security, optimal deployment and localization. CI provides adaptive mechanisms that exhibit intelligent behavior in complex and dynamic environments like WSNs. CI brings about flexibility, autonomous behavior, and robustness against topology changes, communication failures and scenario changes. However, WSN developers are usually not or not completely aware of the potential CI algorithms offer. On the other side, CI researchers are not familiar with all real problems and subtle requirements of WSNs. This mismatch makes collaboration and development difficult. This paper intends to close this gap and foster collaboration by offering a detailed introduction to WSNs and their properties. An extensive survey of CI applications to various problems in WSNs from various research areas and publication venues is presented in the paper. Besides, a discussion on advantages and disadvantages of CI algorithms over traditional WSN solutions is offered. In addition, a general evaluation of CI algorithms is presented, which will serve as a guide for using CI algorithms for WSNs.

Particle Swarm Optimization in Wireless-Sensor Networks: A Brief Survey

Wireless-sensor networks (WSNs) are networks of autonomous nodes used for monitoring an environment. Developers of WSNs face challenges that arise from communication link failures, memory and computational constraints, and limited energy. Many issues in WSNs are formulated as multidimensional optimization problems, and approached through bioinspired techniques. Particle swarm optimization (PSO) is a simple, effective, and computationally efficient optimization algorithm. It has been applied to address WSN issues such as optimal deployment, node localization, clustering, and data aggregation. This paper outlines issues in WSNs, introduces PSO, and discusses its suitability for WSN applications. It also presents a brief survey of how PSO is tailored to address these issues.

Bio-inspired Algorithms for Autonomous Deployment and Localization of Sensor Nodes

Optimal deployment and accurate localization of sensor nodes have a strong influence on the performance of a wireless sensor network (WSN). This paper considers real-time autonomous deployment of sensor nodes from an unmanned aerial vehicle (UAV). Such a deployment has importance, particularly in ad hoc WSNs, for emergency applications, such as disaster monitoring and battlefield surveillance. The objective is to deploy the nodes only in the terrains of interest, which are identified by segmentation of the images captured by a camera on board the UAV. Bioinspired algorithms, particle swarm optimization (PSO) and bacterial foraging algorithm (BFA), are presented in this paper for image segmentation. In addition, PSO and BFA are presented for distributed localization of the deployed nodes. Image segmentation for autonomous deployment and distributed localization are formulated as multidimensional optimization problems, and PSO and BFA are used as optimization tools. Comparisons of the results of PSO and BFA for autonomous deployment and distributed localization are presented. Simulation results show that both the algorithms perform multilevel image segmentation faster than the exhaustive search for optimal thresholds. Besides, PSO-based localization is observed to be faster, and BFA-based localization is more accurate.

Bio-inspired node localization in wireless sensor networks

Many applications of wireless sensor networks (WSNs) require location information of the randomly deployed nodes. A common solution to the localization problem is to deploy a few special beacon nodes having location awareness, which help the ordinary nodes to localize. In this approach, non-beacon nodes estimate their locations using noisy distance measurements from three or more non-collinear beacons they can receive signals from. In this paper, the ranging-based localization task is formulated as a multidimensional optimization problem, and addressed using bio-inspired algorithms, exploiting their quick convergence to quality solutions. An investigation on distributed iterative localization is presented in this paper. Here, the nodes that get localized in an iteration act as references for remaining nodes to localize. The problem has been addressed using particle swarm optimization (PSO) and bacterial foraging algorithm (BFA). A comparison of the performances of PSO and BFA in terms of the number of nodes localized, localization accuracy and computation time is presented.

Neural network based secure media access control protocol for wireless sensor networks

This paper discusses an application of a neural network in wireless sensor network security. It presents a multilayer perceptron (MLP) based media access control protocol (MAC) to secure a CSMA-based wireless sensor network against the denial-of-service attacks launched by adversaries. The MLP enhances the security of a WSN by constantly monitoring the parameters that exhibit unusual variations in case of an attack. The MLP shuts down the MAC layer and the physical layer of the sensor node when the suspicion factor, the output of the MLP, exceeds a preset threshold level. Backpropagation and particle swarm optimization algorithms are used for training the MLP. The MLP-guarded secure WSN is implemented using the Vanderbilt Prowler simulator. Simulation results show that the MLP helps in extending the lifetime of the WSN.

Neural networks letter: Adaptive critics for dynamic optimization

A novel action-dependent adaptive critic design (ACD) is developed for dynamic optimization. The proposed combination of a particle swarm optimization-based actor and a neural network critic is demonstrated through dynamic sleep scheduling of wireless sensor motes for wildlife monitoring. The objective of the sleep scheduler is to dynamically adapt the sleep duration to nodes battery capacity and movement pattern of animals in its environment in order to obtain snapshots of the animal on its trajectory uniformly. Simulation results show that the sleep time of the node determined by the actor critic yields superior quality of sensory data acquisition and enhanced node longevity.

Generalized neuron: Feedforward and recurrent architectures

Feedforward neural networks such as multilayer perceptrons (MLP) and recurrent neural networks are widely used for pattern classification, nonlinear function approximation, density estimation and time series prediction. A large number of neurons are usually required to perform these tasks accurately, which makes the MLPs less attractive for computational implementations on resource constrained hardware platforms. This paper highlights the benefits of feedforward and recurrent forms of a compact neural architecture called generalized neuron (GN). This paper demonstrates that GN and recurrent GN (RGN) can perform good classification, nonlinear function approximation, density estimation and chaotic time series prediction. Due to two aggregation functions and two activation functions, GN exhibits resilience to the nonlinearities of complex problems. Particle swarm optimization (PSO) is proposed as the training algorithm for GN and RGN. Due to a small number of trainable parameters, GN and RGN require less memory and computational resources. Thus, these structures are attractive choices for fast implementations on resource constrained hardware platforms.