Trilochan Panigrahi

Detection of epileptic seizure using Kraskov entropy applied on tunable-Q wavelet transform of EEG signals

Abstract The electroencephalogram (EEG) signals are basically electrophysiological signals that are normally used to access the condition of brain. Epilepsy is one of the brains disorder. Automated diagnosis of epilepsy can be done by measuring and analyzing the nonlinear and non-stationary trends in EEG signals. This paper introduces a new diagnostic approach for analysis and classification of seizure and seizure-free EEG signals. In time-scale domain, the tunable-Q wavelet transform (TQWT) can reliably represent the sparsity in oscillatory signals. The proposed methodology begins with application of TQWT to efficiently characterize the non-stationary behavior and sparsity of EEG signals. TQWT decomposes the considered signals into a valuable set of band-limited signals termed as sub-bands for better feature extraction. Kraskov entropy is a nonlinear parameter to detect the presence of nonlinear trends in the signals. After decomposition, Kraskov entropy is computed from the specific sub-band as a decisive feature in order to discriminate seizure-free from the epileptic seizure EEG signals. Subsequently, obtained feature vectors are used for classifying the seizure and seizure-free EEG signals using the least square support vector machine (LS-SVM) classifier. While doing analysis, it has been observed that value of proposed Kraskov entropy based feature is significantly higher for seizure EEG signals as compared to that of seizure-free EEG signals. Furthermore, the experimental results of this work has demonstrated significant values of classification accuracy, sensitivity, specificity and Matthews correlation coefficient. It is noteworthy that proposed framework uses single feature to diagnose the epilepsy accurately. Also the application of the proposed work on EEG data from the University of Bonn, Germany highlights the consistency and in some cases superiority of the proposed method over other popular methods.

Distributed DOA estimation using clustering of sensor nodes and diffusion PSO algorithm

Abstract This paper proposes a distributed DOA estimation technique using clustering of sensor nodes and distributed PSO algorithm. The sensor nodes are suited by clustered to act as random arrays. Each cluster estimates the source bearing by optimizing the Maximum Likelihood (ML) function locally with cooperation of other clusters. During the estimation process each cluster shares its best information obtained by Diffusion Particle Swarm Optimization (DPSO) with other clusters so that the global estimation is achieved. The performance of the proposed technique has been evaluated through simulation study and is compared with that of obtained by the centralized and decentralized MUltiple SIgnal Classification (MUSIC) algorithms and distributed in-network algorithm. The results demonstrate improved performance of the proposed method compared to others. However, the new method exhibits slightly inferior performance compared to the centralized Particle Swarm Optimization-Maximum Likelihood (PSO-ML) algorithm. Further the proposed method offers low communication overheads compared to other methods.

Fault tolerant distributed estimation in wireless sensor networks

In distributed wireless sensor networks (WSNs), each sensor node estimates the global parameter from the local data in distributed manner. An iterative distributed estimation algorithm is used where the diffusion co-operation scheme is incorporated. Presence of faulty sensor node in the network leads to inaccurate estimation in the conventional error squared based distributed algorithms. Therefore, a fault tolerant distributed estimation in WSNs is proposed here when faulty sensor nodes are present in the network and the network is not aware of them. For this, a robust diffusion estimation algorithms using robust function like Hubers cost function and error saturation non linearity are proposed here in order to make the network fault tolerant. Further, to make the robust estimation algorithm energy efficient, the block adaptive diffusion adaptive algorithm is addressed. The proposed algorithms are validated by simulation and the result shows that the fault tolerant distributed estimation method is robust to node failure.

Error Saturation Nonlinearities for Robust Incremental LMS over Wireless Sensor Networks

The data collected by sensor nodes over a geographical region is contaminated with Gaussian and impulsive noise. The conventional gradient-based distributed adaptive estimation algorithms exhibit good performance in the presence of Gaussian noise but perform poorly in impulsive noise environments. Therefore, the objective of this article is to propose a robust distributed adaptive algorithm that alleviates the effect of impulsive noise. An error saturation nonlinearity-based robust distributed strategy is proposed in an incremental cooperative network to estimate the desired parameters in impulsive noise. The steady-state analysis of the proposed error saturation nonlinearity incremental least mean squares (SNILMS) algorithm is carried out by employing the spatial-temporal energy conservation principle. Both theoretical and simulation results show that the presence of the error nonlinearity has made the proposed SNILMS algorithm robust to impulsive noise.

Maximum likelihood DOA estimation in distributed wireless sensor network using adaptive particle swarm optimization

Source direction of arrival (DOA) estimation is one of the challenging problem in wireless sensor network. Several methods based on maximum likelihood (ML) criteria has been established in literature. Generally, to obtain the exact ML (EML) solutions, the DOAs must be estimated by optimizing a complicated nonlinear multimodal function over a high-dimensional problem space. An adaptive particle swarm optimization (APSO) based solution is proposed here to compute the ML functions and explore the potential of superior performances over traditional PSO algorithm. Simulation results confirms that the APSO-ML estimator is significantly giving better performance at lower SNR compared to conventional method like MUSIC in various scenarios at less computational costs.

Block Least Mean Squares Algorithm over Distributed Wireless Sensor Network

In a distributed parameter estimation problem, during each sampling instant, a typical sensor node communicates its estimate either by the diffusion algorithm or by the incremental algorithm. Both these conventional distributed algorithms involve significant communication overheads and, consequently, defeat the basic purpose of wireless sensor networks. In the present paper, we therefore propose two new distributed algorithms, namely, block diffusion least mean square (BDLMS) and block incremental least mean square (BILMS) by extending the concept of block adaptive filtering techniques to the distributed adaptation scenario. The performance analysis of the proposed BDLMS and BILMS algorithms has been carried out and found to have similar performances to those offered by conventional diffusion LMS and incremental LMS algorithms, respectively. The convergence analyses of the proposed algorithms obtained from the simulation study are also found to be in agreement with the theoretical analysis. The remarkable and interesting aspect of the proposed block-based algorithms is that their communication overheads per node and latencies are less than those of the conventional algorithms by a factor as high as the block size used in the algorithms.

Amplitude only compensation for failed antenna array using particle swarm optimization

Particle swarm optimization (PSO) procedure has been applied to find out a compensation for degraded pattern of a failed antenna array by amplitude only perturbation of few of the unfailed elements. The optimization technique finds the optimum amplitude levels of some of the unfailed elements in the array, so that the compensated pattern will match with the pattern of the original array, though not perfectly, but up to a satisfactory level. The reason for applying PSO is that it is simple to code and has small computational cost as compared to other soft-computing techniques. The developed formulation is tested for a 16-element linear broadside array.

Robust distributed linear parameter estimation in wireless sensor network

In a wireless sensor network each sensor node collects scalar measurements of some unknown parameters, corrupted by independent Gaussian noise. Then the objective is to estimate some parameters of interest from the data collected across the network. In this paper a simple iterative robust distributed linear parameter estimation algorithm is proposed where the diffusion co-operation scheme is incorporated. Each node updates its information by using the data collected by it and the information received from the neighbours. When any node fails to transmit correct information to the neighbours and/or the data collected by the node is noisy then the least mean square based diffusion estimation is not accurate. Hence a robust diffusion linear estimation algorithm is proposed here in order to improve the accuracy of the estimation in distributed wireless sensor network.

Robust Incremental LMS over Wireless Sensor Network in Impulsive Noise

Distributed wireless sensor networks have been proposed as a solution to environment sensing, target tracking, data collection and others. Energy efficiency, high estimation accuracy, and fast convergence are important goals in distributed estimation algorithms for WSN. This paper studies the problem of robust adaptive estimation in impulsive noise environment using robust cost function like Wilcox on norm and error saturation nonlinearity. The incremental cooperative scheme conventionally used in sensor network in which each node have local computing ability and share them with their predefined neighbors, is not robust to impulsive type of noise or outliers. In this paper the robust norm is introduced in incremental cooperative distributed network to estimate the desired parameters in presence of Gaussian contaminated impulsive noise.

Learning with distributed data in wireless sensor network

Wireless sensor networks (WSN) have been proposed as a solution to environment sensing, target tracking, data collection and others. WSN collect an enormous amount of data over space and time. The objective is to estimate of a parameter or function from these data. Learning is used in detection and estimation problems when no probablistic model relating an observation. This paper investigates a general class of distributed algorithms for data processing, eliminating the need to transmit raw data to a central processor. This can provide significant reductions in the amount of communication and energy required to obtain an accurate estimate. The estimation problems we consider are expressed as the optimization of a cost function involving data from all sensor nodes. Here the distributed algorithm is based on an incremental optimization process. A parameter estimate is circulated through the network, and along the way each node makes a small adjustment to the estimate based on its local data.