P.P. Kanjilal

Bio: P.P. Kanjilal is an academic researcher from Indian Institute of Technology Kharagpur. The author has contributed to research in topics: Singular value decomposition & Singular spectrum analysis. The author has an hindex of 12, co-authored 31 publications receiving 900 citations. Previous affiliations of P.P. Kanjilal include Charles River Laboratories & University of Oxford.

Revisiting the role of correlation coefficient to distinguish chaos from noise

Abstract: The correlation coefficient vs. prediction time profile has been widely used to distinguish chaos from noise. The correlation coefficient remains initially high, gradually decreasing as prediction time increases for chaos and remains low for all prediction time for noise. We here show that for some chaotic series with dominant embedded cyclical component(s), when modelled through a newly developed scheme of periodic decomposition, will yield high correlation coefficient even for long prediction time intervals, thus leading to a wrong assessment of inherent chaoticity. But if this profile of correlation coefficient vs. prediction horizon is compared with the profile obtained from the surrogate series, correct interpretations about the underlying dynamics are very much likely.

Fast succesive selection of variables in linear models using modified QR factorisation

Abstract: A new approach for succesively selecting regressor variables in the linear-in-the-parameter modelling problem is presented. It is based on QR with column pivoting factorisation, where the columns are pivoted based on the correlation with the corresponding rotated output vector. In contrast to other competing methods, no parameter estimation is necessary. The method is computationally fast and numerically robust.

Spatiotemporal Clustering for Aggregating Hostile Units in Cluttered Environments

Abstract: We describe a novel clustering approach for aggregating mobile (typically potentially hostile) units in cluttered urban environments. The approach consists of a suite of spatiotemporal clustering algorithms that leverage the wealth of military sensor data available to provide insight into "what is strange" about a given situation, without knowing beforehand what exactly we are looking for. The algorithms perform a space and time-series analysis of sensor messages independently of any contextual or semantic information. The algorithms can, for example, detect patterns and track for spatially correlated moving units over time within the environment. The patterns thus detected trigger follow-up assessment of the newly developed situations, resulting in invocations of various doctrine-based computational models to identify higher-level situations (e.g. attack, ambush, interdiction, insurgency). We provide some experimental results analyzing the performance of the clustering algorithms.

Regression Diagnostics: Identifying Influential Data and Sources of Collinearity

Abstract: 1. Introduction and Overview. 2. Detecting Influential Observations and Outliers. 3. Detecting and Assessing Collinearity. 4. Applications and Remedies. 5. Research Issues and Directions for Extensions. Bibliography. Author Index. Subject Index.

Photoplethysmography and its application in clinical physiological measurement.

Abstract: Photoplethysmography (PPG) is a simple and low-cost optical technique that can be used to detect blood volume changes in the microvascular bed of tissue. It is often used non-invasively to make measurements at the skin surface. The PPG waveform comprises a pulsatile ('AC') physiological waveform attributed to cardiac synchronous changes in the blood volume with each heart beat, and is superimposed on a slowly varying ('DC') baseline with various lower frequency components attributed to respiration, sympathetic nervous system activity and thermoregulation. Although the origins of the components of the PPG signal are not fully understood, it is generally accepted that they can provide valuable information about the cardiovascular system. There has been a resurgence of interest in the technique in recent years, driven by the demand for low cost, simple and portable technology for the primary care and community based clinical settings, the wide availability of low cost and small semiconductor components, and the advancement of computer-based pulse wave analysis techniques. The PPG technology has been used in a wide range of commercially available medical devices for measuring oxygen saturation, blood pressure and cardiac output, assessing autonomic function and also detecting peripheral vascular disease. The introductory sections of the topical review describe the basic principle of operation and interaction of light with tissue, early and recent history of PPG, instrumentation, measurement protocol, and pulse wave analysis. The review then focuses on the applications of PPG in clinical physiological measurements, including clinical physiological monitoring, vascular assessment and autonomic function.

Nonlinear Time Series Analysis.

Abstract: : This thesis applies neural network feature selection techniques to multivariate time series data to improve prediction of a target time series. Two approaches to feature selection are used. First, a subset enumeration method is used to determine which financial indicators are most useful for aiding in prediction of the S&P 500 futures daily price. The candidate indicators evaluated include RSI, Stochastics and several moving averages. Results indicate that the Stochastics and RSI indicators result in better prediction results than the moving averages. The second approach to feature selection is calculation of individual saliency metrics. A new decision boundary-based individual saliency metric, and a classifier independent saliency metric are developed and tested. Ruck's saliency metric, the decision boundary based saliency metric, and the classifier independent saliency metric are compared for a data set consisting of the RSI and Stochastics indicators as well as delayed closing price values. The decision based metric and the Ruck metric results are similar, but the classifier independent metric agrees with neither of the other metrics. The nine most salient features, determined by the decision boundary based metric, are used to train a neural network and the results are presented and compared to other published results. (AN)