Browsing by Author "Panda G."

Now showing 1 - 20 of 129

Accurate partitional clustering algorithm based on immunized PSO
(2012) Nanda S.J.; Panda G.
Hybrid evolutionary algorithms are created by suitably combining the good features of two parent evolutionary algorithms, normally provide better solutions than the individual ones. In this paper we have formulated the partitional clustering as an optimization problem and solved it by a newly developed hybrid evolutionary algorithm Immunized PSO. Simulation studies on four benchmark UCI datasets demonstrate the superior performance of the proposed algorithm compared to the standard K-means, Correlation, PSO and CLONAL clustering algorithms in terms of percentage of accuracy, convergence characteristics, stability and computational efficiency achieved over fifty independent runs. � 2012 Pillay Engineering College.
Active control of nonlinear noise processes using cascaded adaptive nonlinear filter
(2013) George N.V.; Panda G.
A novel nonlinear adaptive filter based on a cascade combination of a functional link artificial neural network (FLANN) and a Legendre polynomial has been proposed in this paper for nonlinear active noise control (ANC). The performance of the new controller has been compared with that obtained by a FLANN based ANC system trained using a filtered-s least mean square (FsLMS) algorithm as well as with a Legendre neural network (LeNN) based ANC system trained using a filtered-l LMS (FlLMS) algorithm. The training of the cascaded controller has been achieved using a filtered-sl LMS (FslLMS) algorithm, which simultaneously adapts the weights of both the component adaptive controllers. The new controller has been shown to achieve improved noise mitigation capability in comparison to its constituent filters. � 2012 Elsevier Ltd. All rights reserved.
Advances in active noise control: A survey, with emphasis on recent nonlinear techniques
(2013) George N.V.; Panda G.
This paper discusses the evolution of active noise control systems over the past 75 years. The focus of this study is on the use of signal processing and some recent soft computing tools on the development of active noise control systems. Special attention has been paid to the advances in nonlinear active noise control achieved during the past decade. � 2012 Elsevier B.V.
Analysis and design of 1 GHz PLL for fast phase and frequency acquisition
(2014) Rout P.K.; Panda B.P.; Acharya D.P.; Panda G.
Phase locked loop (PLL) being a mixed signal circuit involves design challenges at high frequencies. In this work a mixed signal PLL for faster phase and frequency locking is designed. The PLL is designed and synthesized using GPDK090 library of CMOS 90 nm process in CADENCE Virtuoso Analog Design Environment for an operating frequency of 1 GHz. Its locking time is 280.6 ns and observed to consume a power of 11.9 mW with a 1.8 V supply voltage. The complete layout of the PLL is drawn in CADENCE Virtuoso XL and its behaviour and performance is observed in Spectre. Copyright � 2014 Inderscience Enterprises Ltd.
Automated retinal nerve fiber layer defect detection using fundus imaging in glaucoma
(2018) Panda R.; Puhan N.B.; Rao A.; Padhy D.; Panda G.
Retinal nerve fiber layer defect (RNFLD) provides an early objective evidence of structural changes in glaucoma. RNFLD detection is currently carried out using imaging modalities like OCT and GDx which are expensive for routine practice. In this regard, we propose a novel automatic method for RNFLD detection and angular width quantification using cost effective redfree fundus images to be practically useful for computer-assisted glaucoma risk assessment. After blood vessel inpainting and CLAHE based contrast enhancement, the initial boundary pixels are identified by local minima analysis of the 1-D intensity profiles on concentric circles. The true boundary pixels are classified using random forest trained by newly proposed cumulative zero count local binary pattern (CZC-LBP) and directional differential energy (DDE) along with Shannon, Tsallis entropy and intensity features. Finally, the RNFLD angular width is obtained by random sample consensus (RANSAC) line fitting on the detected set of boundary pixels. The proposed method is found to achieve high RNFLD detection performance on a newly created dataset with sensitivity (SN) of 0.7821 at 0.2727 false positives per image (FPI) and the area under curve (AUC) value is obtained as 0.8733. � 2018 Elsevier Ltd
Automatic clustering algorithm based on multi-objective Immunized PSO to classify actions of 3D human models
(2013) Nanda S.J.; Panda G.
Multi-objective clustering algorithms are preferred over its conventional single objective counterparts as they incorporate additional knowledge on properties of data in the from of objectives to extract the underlying clusters present in many datasets. Researchers have recently proposed some standardized multi-objective evolutionary clustering algorithms based on genetic operations, particle swarm optimization, clonal selection principles, differential evolution and simulated annealing, etc. In many cases it is observed that hybrid evolutionary algorithms provide improved performance compared to that of individual algorithm. In this paper an automatic clustering algorithm MOIMPSO (Multi-objective Immunized Particle Swarm Optimization) is proposed, which is based on a recently developed hybrid evolutionary algorithm Immunized PSO. The proposed algorithm provides suitable Pareto optimal archive for unsupervised problems by automatically evolving the cluster centers and simultaneously optimizing two objective functions. In addition the algorithm provides a single best solution from the Pareto optimal archive which mostly satisfy the users' requirement. Rigorous simulation studies on 11 benchmark datasets demonstrate the superior performance of the proposed algorithm compared to that of the standardized automatic clustering algorithms such as MOCK, MOPSO and MOCLONAL. An interesting application of the proposed algorithm has also been demonstrated to classify the normal and aggressive actions of 3D human models.� 2012 Elsevier Ltd. All rights reserved.
Automatic clustering using MOCLONAL for classifying actions of 3D human models
(2012) Nanda S.J.; Panda G.
Conventional clustering algorithms use a single objective function optimization criterion for classification which may not provide satisfactory results to determine the underlying clusters in many datasets. In such scenario multi-objective algorithms are preferred which improve the clustering performance due to additional knowledge of data properties in the form of objective functions. In this paper we have proposed an automatic multi-objective clustering algorithm based on clonal selection principle of artificial immune system (AIS) and is termed as MOCLONAL. The proposed algorithm is capable of providing a single best solution from the Pareto optimal archive which mostly satisfy the user requirement. Simulation studies on synthetic and real life datasets demonstrate the superior performance of the proposed algorithm compared to benchmark multi-objective clustering algorithm MOCK. An interesting application of the proposed algorithm have been demonstrated to classify the normal and aggressive actions of 3D human models. � 2012 IEEE.
BESAC: Binary External Symmetry Axis Constellation for unconstrained handwritten character recognition
(2016) Dash K.S.; Puhan N.B.; Panda G.
We propose a novel perception driven feature extraction called binary external symmetry axis constellation (BESAC) and a fast Boolean matching character recognition technique. A constellation model using a set of external symmetry axes which are perceptually significant can uniquely represent a handwritten character pattern. This model generates two histograms of orientations that are binary coded and concatenated to produce the proposed BESAC feature. A two stage classification strategy is adopted where a parallel Hamming Distance dissimilarity matching is performed on the extracted BESAC feature to achieve fast recognition along with perceptual closure part detection on look-alike characters. We adopt a 10-fold cross validation strategy to evaluate the performance of our algorithm on two major Indian languages, Bangla and Odia with four benchmark databases (ISI Kolkata Bangla numeral, ISI Kolkata Odia and IITBBS Odia numeral, and a newly created IITBBS Odia character database). The average time for classifying an unknown handwritten character is reported to be significantly less than the existing methods. The average recognition accuracy using the proposed approach is proved to outperform the state-of-the-art accuracy results on ISI Kolkata Odia numeral database (99.35%), IITBBS Odia numeral (98.9%), ISI Kolkata Bangla numeral database (99.48%) and IITBBS Odia character (95.01%) database. � 2016 Elsevier B.V.
Block least mean squares algorithm over distributed wireless sensor network
(2012) Panigrahi T.; Pradhan P.M.; Panda G.; Mulgrew B.
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. � Copyright 2012 T. Panigrahi et al.
A clustering model based on colliding bodies optimization for analysis of seismic catalog
(2016) Nanda S.J.; Panda G.
Nature has been the key source of inspiration for development of many heuristic optimization algorithm. In this manuscript a new clustering model is developed based on a recently developed nature inspired algorithm 'Colliding Bodies Optimization (CBO)'. The CBO is based on the phenomenon of collision between bodies where each body try to occupy a convenient position in the search space. The proposed clustering model is applied to analyze the seismic activities of Japan catalog. If the number of clusters are known aprori with the help of seismologist then the proposed model provide accurate clustering performance with lower computation. Comparison with recently developed 'fast density based clustering' the proposed model provide equivalent clustering output for Japan catalog with lower computational time. � 2015 IEEE.
Comparative performance analysis of evolutionary algorithm based parameter optimization in cognitive radio engine: A survey
(2014) Pradhan P.M.; Panda G.
One of the important features of the cognitive radio engine is to adapt the parameters of radio to fulfill certain objectives in a time varying wireless environment. In order to achieve this adaptation, six evolutionary algorithms are employed for optimizing the predefined fitness functions in the radio environment. The performance of genetic algorithm, particle swarm optimization, differential evolution, bacterial foraging optimization, artificial bee colony optimization and cat swarm optimization algorithm in different modes of operation are studied in detail. Each algorithm is tested in single and multicarrier communication system in order to acknowledge the advantage of multicarrier communication systems in wireless environment. The spectral interference introduced by the cognitive user into the primary user's band and that introduced by the primary user into the cognitive user's band are also investigated. The performance of different algorithms are compared using convergence characteristics and four statistical metrics. � 2014 Elsevier B.V. All rights reserved.
A comparative performance assessment of a set of multiobjective algorithms for constrained portfolio assets selection
(2014) Mishra S.K.; Panda G.; Majhi R.
This paper addresses a realistic portfolio assets selection problem as a multiobjective optimization one, considering the budget, floor, ceiling and cardinality as constraints. A novel multiobjective optimization algorithm, namely the non-dominated sorting multiobjective particle swarm optimization (NS-MOPSO), has been proposed and employed efficiently to solve this important problem. The performance of the proposed algorithm is compared with four multiobjective evolution algorithms (MOEAs), based on non-dominated sorting, and one MOEA algorithm based on decomposition (MOEA/D). The performance results obtained from the study are also compared with those of single objective evolutionary algorithms, such as the genetic algorithm (GA), tabu search (TS), simulated annealing (SA) and particle swarm optimization (PSO). The comparisons of the performance include three error measures, four performance metrics, the Pareto front and computational time. A nonparametric statistical analysis, using the Sign test and Wilcoxon signed rank test, is also performed, to demonstrate the superiority of the NS-MOPSO algorithm. On examining the performance metrics, it is observed that the proposed NS-MOPSO approach is capable of identifying good Pareto solutions, maintaining adequate diversity. The proposed algorithm is also applied to different cardinality constraint conditions, for six different market indices, such as the Hang-Seng in Hong Kong, DAX 100 in Germany, FTSE 100 in UK, S&P 100 in USA, Nikkei 225 in Japan, and BSE-500 in India. � 2014 Elsevier B.V.
Comparative performance evaluation of multiobjective optimization algorithms for portfolio management
(2009) Mishra S.K.; Meher S.; Panda G.; Panda A.
The objective of portfolio optimization is to find an optimal set of assets to invest on
Comparative performance study of wore segmentation techniques for handwritten Odia documents
(2017) Pradhan A.; Behera S.; Panda G.; Majhi B.
Word segmentation of handwritten documents is a vital step in the Optical Character Recognition system as its accuracy greatly influences the overall recognition performance. In the literature, various methods have been proposed for word segmentation of handwritten documents of various languages. However, it is observed that for Odia, which is an important Indian language, very little work has been reported on word segmentation. Hence, the objective of this paper is to employ two standard existing methods to segment words of Odia handwritten documents and compare the segmentation performance of these methods with the lone Water Reservoir Algorithm available in the literature and finally rank those methods based on their segmentation performance. It is observed that out of three methods, the Tree Structure method performs the best comparing four different performance measures. � 2016 IEEE.
Computational analysis of the GI/G/1 risk process using roots
(2018) Panda G.; Banik A.D.; Chaudhry M.L.
In this paper, we analyze an insurance risk model wherein the arrival of claims and their sizes occur as renewal processes. Using the duality relation in queueing theory and roots method, we derive closed-form expressions for the ultimate ruin probability, the distribution of the deficit at the time of ruin, and the expected time to ruin in terms of the roots of the characteristic equation. Finally, some numerical computations are portrayed with the help of tables. � 2018, Springer Nature Singapore Pte Ltd.
Computationally efficient algorithm for high sampling-frequency operation of active noise control
(2015) Rout N.K.; Das D.P.; Panda G.
In high sampling-frequency operation of active noise control (ANC) system the length of the secondary path estimate and the ANC filter are very long. This increases the computational complexity of the conventional filtered-x least mean square (FXLMS) algorithm. To reduce the computational complexity of long order ANC system using FXLMS algorithm, frequency domain block ANC algorithms have been proposed in past. These full block frequency domain ANC algorithms are associated with some disadvantages such as large block delay, quantization error due to computation of large size transforms and implementation difficulties in existing low-end DSP hardware. To overcome these shortcomings, the partitioned block ANC algorithm is newly proposed where the long length filters in ANC are divided into a number of equal partitions and suitably assembled to perform the FXLMS algorithm in the frequency domain. The complexity of this proposed frequency domain partitioned block FXLMS (FPBFXLMS) algorithm is quite reduced compared to the conventional FXLMS algorithm. It is further reduced by merging one fast Fourier transform (FFT)-inverse fast Fourier transform (IFFT) combination to derive the reduced structure FPBFXLMS (RFPBFXLMS) algorithm. Computational complexity analysis for different orders of filter and partition size are presented. Systematic computer simulations are carried out for both the proposed partitioned block ANC algorithms to show its accuracy compared to the time domain FXLMS algorithm. � 2014 Elsevier Ltd. All rights reserved.
Connectivity constrained wireless sensor deployment using multiobjective evolutionary algorithms and fuzzy decision making
(2012) Pradhan P.M.; Panda G.
Deployment of sensor nodes is an important issue in designing sensor networks. The sensor nodes communicate with each other to transmit their data to a high energy communication node which acts as an interface between data processing unit and sensor nodes. Optimization of sensor node locations is essential to provide communication for a longer duration. An energy efficient sensor deployment based on multiobjective particle swarm optimization algorithm is proposed here and compared with that of non-dominated sorting genetic algorithm. During the process of optimization, sensor nodes move to form a fully connected network. The two objectives i.e. coverage and lifetime are taken into consideration. The optimization process results in a set of network layouts. A comparative study of the performance of the two algorithms is carried out using three performance metrics. The sensitivity analysis of different parameters is also carried out which shows that the multiobjective particle swarm optimization algorithm is a better candidate for solving the multiobjective problem of deploying the sensors. A fuzzy logic based strategy is also used to select the best compromised solution on the Pareto front. � 2012 Elsevier B.V. All rights reserved.
Constrained LMMSE-based object-specific reconstruction in compressive sensing
(2017) Tripathy S.R.; Panda G.; Majhi B.
In many applications like surveillance, it is essential to detect the presence of specific objects by seeing an image or video without being concerned about details of the scene. The reconstruction algorithms proposed in the compressive sensing literature try to iteratively reconstruct the full image. Hence, those are computationally expensive. If some prior knowledge of the object is available, then a closed-form reconstruction algorithm can be formulated. The goal is to reconstruct the object of interest efficiently without being bothered about the quality of reconstruction of the scene. To address this situation, a constraint is formulated and incorporated into linear minimum mean square error (LMMSE) estimator to form a closed-form solution. This compact solution is capable of meaningfully reconstructing the object relative to the scene. In the proposed method, an illconditioned matrix inversion problem has been faced and overcome by regularization method. To boost the speed of the algorithm, a modified Euler method is proposed for finding the regularization parameter. To further speedup the reconstruction process, larger images are divided into several pieces and each piece is reconstructed separately using constrained LMMSE. For a given number of measurements, the simulation-based results demonstrate acceptable quality of reconstruction with minimal computational effort. � The Institution of Engineering and Technology 2017.
Constrained multiobjective optimization based design of CMOS ring oscillator
(2014) Rout P.K.; Acharya D.P.; Panda G.
In this paper a popular multiobjective optimization Non-dominated Sorting Genetic Algorithm (NSGA-II) based integrated circuit design methodology using simple equation models is presented. The method is applied to CMOS ring oscillator circuit where the design parameters are estimated so that the circuit offers optimal performance. The circuit is designed using these parameters in Cadence Virtuoso Analog Design Environment (ADE) with GPDK 90nm process to test the predicted performance. The proposed method saves the design cycle time ensuring the optimal performance of the CMOS ring oscillator in a constrained environment. � 2014 IEEE.
Constrained portfolio asset selection using multiobjective bacteria foraging optimization
(2014) Mishra S.K.; Panda G.; Majhi R.
Portfolio asset selection (PAS) is a challenging and interesting multiobjective task in the field of computational finance, and is receiving the increasing attention of researchers, fund management companies and individual investors in the last few decades. Selecting a subset of assets and corresponding optimal weights from a set of available assets, is a key issue in the PAS problem. A Markowitz model is generally used to solve this optimization problem, where the total profit is maximized, while the total risk is to be minimized. However, this model does not consider the practical constraints, such as the minimum buy in threshold, maximum limit, cardinality etc. The Practical constraints are incorporated in this study to meet a real world financial scenario. In the proposed work, the PAS problem is formulated in a multiobjective framework, and solved using the multiobjective bacteria foraging optimization (MOBFO) algorithm. The performance of the proposed approach is compared with a set of competitive multiobjective evolutionary algorithms using six performance metrics, the Pareto front and computational time. On examining the performance metrics, it is concluded that the proposed MOBFO algorithm is capable of identifying a good Pareto solution, maintaining adequate diversity. The proposed algorithm is also successfully applied to different cardinality constraint conditions, for six different market indices. � 2013 Springer-Verlag Berlin Heidelberg.