Associate Professor Kaushik Mahata

We consider the problem of interference management and resource allocation in a cognitive radio network (CRNs) where the licensed spectrum holders (primary users) share their spare capacity with the non-licensed spectrum holders (secondary users). Under such shared spectrum usage the transmissions of the secondary users should have a minimal impact on the quality of service (QoS) and the operating conditions of the primary users. Therefore, it is important to distinguish the two types of users, and formulate the problem of resource allocation considering hard restrictions on the user-perceived QoS (such as packet end-to-end delay and loss) and physical-layer channel characteristics (such as noise and interference) of the primary users. To achieve this goal, we propose to assign the bandwidth and transmission power to minimize the total buffer occupancy in the system subject to capacity constraints, queue stability constraints, and interference requirements of the primary users. We apply this approach for resource allocation in a CRN built upon a Third Generation Partnership Project (3GPP) long-term evolution (LTE) standard platform. Performance of the algorithm is evaluated using simulations in OPNET environment. The algorithm shows consistent performance improvement when compared with other relevant resource allocation techniques.

We investigate the problem of energy-efficient dynamic spectrum access (DSA) in a cognitive Third Generation Partnership Project (3GPP) long-term evolution (LTE) network based on IEEE802.22 architecture. In this architecture, the network resources are allocated to the end-users and the evolved NodeBs (eNBs) by the spectrum manager (SM) using some optimal resource allocation strategy. In particular, we propose to assign the bandwidth and transmission power to the uplink and downlink of LTE system so that the total transmission power is minimized subject to capacity constraints, queue stability constraints, and some integer restrictions on the bandwidth. To find the buffer occupancy in the system, we use modified Shannon expression which depends on signal-to-noise ratio (SNR) and modulation and coding scheme (MCS). Unlike previous works, the proposed resource allocation algorithm is derived based on realistic assumptions, such as discrete spectrum resources, relation between transmission rate, MCS and the channel quality. Performance of the algorithm is evaluated using simulations in OPNET environment. The algorithm shows consistent performance improvement when compared with other relevant resource allocation techniques. © 2014 Elsevier B.V. All rights reserved.

We present new insights into the modeling of the microcantilever in dynamic mode atomic force microscopy and outline a novel high-bandwidth tip-sample force estimation technique for the development of high-bandwidth z -axis control. Fundamental to the proposed technique is the assumption that in tapping mode atomic force microscopy, the tip-sample force takes the form of an impulse train. Formulating the estimation problem as a Kalman filter, the tip-sample force is estimated directly; thus, potentially enabling high-bandwidth z-axis control by eliminating the dependence of the control technique on microcantilever dynamics and the amplitude demodulation technique. Application of this technique requires accurate knowledge of the models of the microcantilever; a novel identification method is proposed. Experimental data are used in an offline analysis for verification.

An extended instrumental variable algorithm is proposed for identification of systems with arbitrary time delays. Our algorithm can handle time-delays which may be non-integer multiple of the sampling interval. It is further shown that the proposed approach can achieve the asymptotic Cramér-Rao bound by properly tuning the user's choices. In simulations the algorithm gives unbiased parameter estimates with good statistical accuracy. © 2007 IEEE.

In this paper, the problem of identifying stochastic linear continuous-time systems from noisy input/output data is addressed. The input of the system is assumed to be non-Gaussian, whereas the noises contaminating the data are assumed to be Gaussian. The fourth-order cumulants of the input/output data are then (asymptotically) insensitive to the noises, that can be coloured and/or mutually correlated. Two estimators based on this noise-cancellation property are proposed. The performance of the proposed algorithms are assessed through a numerical simulation. © 2007 IEEE.

Least-squares based non-parametric estimation of the wave propagation functions of a viscoelastic material is considered in this paper. Widely used nonlinear least-squares-based algorithms are often computationally expensive and suffer from numerical problems. In this paper, we propose a class of subspace estimators which assume equidistant sensor configuration. The proposed estimator is computationally economical and numerically robust. Analytical expressions for the estimation accuracy have been derived. It is also shown that the subspace estimator achieves the optimal accuracy under the optimal weighting. The algorithm is employed on simulated data as well as on real experimental data. The results therefrom are shown to confirm the analytical results. © 2003 Elsevier Ltd. All rights reserved.

In this paper we propose a parametric and a non-parametric identification algorithm for dynamic errors-in-variables model. We show that the two-dimensional process composed of the input-output data admits a finite order ARMA representation. The non-parametric method uses the ARMA structure to compute a consistent estimate of the joint spectrum of the input and the output. A Frisch scheme is then employed to extract an estimate of the joint spectrum of the noise free input-output data, which in turn is used to estimate the transfer function of the system. The parametric method exploits the ARMA structure to give estimates of the system parameters. The performances of the algorithms are illustrated using the results obtained from a numerical simulation study. © 2003 Published by Elsevier Science Ltd.

In this paper, we investigate the nonparametric estimation of the frequency dependent complex modulus of a viscoelastic material. The strains due to flexural wave propagation In a bar specimen are registered at different cross sections. The time domain data is transformed into frequency domain using discrete Fourier transform and a nonlinear least squares algorithm is then employed to estimate the complex modulus at each frequency. Inherent numerical problems due to associated ill-conditioned matrices are treated with special care. An analysis of the quality of the nonlinear least squares estimate is also carried out. The validity of the theoretical results are confirmed by numerical studies and experimental tests.

Many practical system identification problems can be formulated as linear regression problems. The parameter estimates can be computed using instrumental variables (IV) or total least squares (TLS) estimators, both of which have moderate computational complexity. In this work, explicit expressions for the asymptotic covariance matrix of the TLS estimates is derived and is shown to be same as that of the IV method. The accuracy of the parameter estimates for an errors-in-variables model using the above methods has been treated in particular, as standard analysis does not apply. The results obtained from the numerical simulations show that the practical behaviour of the estimators is well predicted by the theoretical results. We provide an explanation why for finite samples, the IV approach is found to be somewhat more robust than the TLS approach. On the other hand, the TLS approach has lower computational load than the IV method.

The paper gives an overview of various methods for identifying dynamic errors-in-variables systems. Several approaches are classified by how the original information in time-series data of the noisy input and output measurements is condensed before further processing. For some methods, such as instrumental variable estimators, the information is condensed into a nonsymmetric covariance matrix as a first step before further processing. In a second class of methods, where a symmetric covariance matrix is used instead, the Frisch scheme and other bias-compensation approaches appear. When dealing with the estimation problem in the frequency domain, a milder data reduction typically takes place by first computing spectral estimators of the noisy input-output data. Finally, it is also possible to apply maximum likelihood and prediction error approaches using the original time-domain data in a direct fashion. This alternative will often require quite high computational complexity but yield good statistical efficiency. The paper is also presenting various properties of parameter estimators for the errors-in-variables problem, and a few conjectures are included, as well as some perspectives and experiences by the authors. © 2002 Elsevier Science B.V. All rights reserved.

As the electricity market transits into deregulation, customers and generators are given the liberty of deciding on the bid functions they submit and the amount of transmission rights they purchase in the market. These determine the payoffs they achieve from the market. In this paper, we evaluate how generators maximize their payoffs achieved from the market by making strategic decisions about the bid functions they submit and the transmission rights they purchase in the network. Strategical interdependency between decisions for when there are 2 or more generators in the market is also considered and modelled using matrix game. A Nash equilibrium reflecting the strategies the generators should pick is then identified using the max-min theorem. The application of the max-min theorem on the matrix game is modelled as a mixed integer linear programming problem. The proposed matrix game is applied and analyzed on 2 bus and 3 bus systems.

This paper presents the application of bi-level programming to determine the profit maximization for transmission investor in deregulated power markets. We consider a deregulated market comprising merchant transmission investors responsible for transmission expansion and a central market operator responsible for market clearing. The problem is solved using two levels of optimization: Upper level problem, where the objective of the transmission merchant investor is to maximize profits and a Lower level problem, where the objective of the market operator is to maximize the social welfare of the generators and the customers. Transmission investment decisions made in the upper level adjust the transmission constraints in the lower level. The optimal solutions obtained in the lower level are used by transmission investors to calculate the transmission congestion price and profit. The bi-level model is applied on test systems and the results are consistent with the analytical method available in literature.

We explore the problem of resource allocation in a Third Generation Partnership Project (3GPP) long-term evolution (LTE) based cognitive radio network (CRN). The network model consists of a number of service providers (SPs) with fixed licensed spectrum bands. The network offers the wireless services to two types of users: primary and secondary. The primary users (PUs) get prioritized access to the licensed spectrum bands. The secondary users (SUs) are served on the best-effort (non-prioritized) basis. In this paper we consider the specific design features of LTE radio interface associated with the uplink spectrum access, scheduling process, and limited control channel capacity of the LTE system. We establish the relation between the number of users in the system, and the scheduling delay (which is the largest contributor to the packet end-to-end delay in LTE network). Using these results, we propose a simple algorithm to assign the spectrum for the SUs without violating the quality of service (QoS) requirements of the PU, and implement it in an LTE-based CRN. Consistent performance of the algorithm is verified using OPNET-based simulations.

Random access is being considered as one of the key bottlenecks for supporting machine-to-machine (M2M) communications over an IEEE 802.16-based WiMAX network. Apart from handling massive access attempts from a large number of devices, the random access plane requires service differentiation capability to meet the diverse QoS requirements of various M2M applications. To address these issues, in this paper, we propose an enhanced random access scheme, where the fixed/low-mobility M2M devices pre-equalize their random access codes using the estimated frequency response of the slowly-varying wireless channel. Consequently, the base station is able to detect a large number of codes as their mutual orthogonality remains preserved. Moreover, a differentiated random access strategy is proposed to provide QoS-aware access service to various M2M devices. The performance of the proposed scheme is demonstrated under two different code matrices using both theoretical analysis and simulation results.

Random access is being considered as one of the key bottlenecks for supporting machine-to-machine (M2M) communications over an IEEE 802.16-based WiMAX network. This paper presents a comprehensive analytical model to evaluate the performance of the WiMAX CDMA-based random access (CRA) procedure under the bursty M2M communications traffic in the Smart Grid. Compared to the literature, the proposed model incorporates the unique features of the CRA procedure such as multi-user multi-code transmission and parallel code detection in the performance analysis matrix. The accuracy of the model is validated by an extensive set of simulation results based on an OPNET simulation model. The results provide some interesting insights into the performance of the WiMAX random access plane for a range of controllable parameters, such as the number of available codes, the number of detectable codes, and the backoff parameters.

This paper studies the problem of channel estimation for the downlink in broadband LTE communications. A Kalman Filter is used to combine channel soundings made with pilot tones that are deployed at different times and in different frequency bands on a pre-selected pattern. This Kalman filter is presented in a form directly suitable for implementation as a periodically time varying filter, in the case where periodic pilot tone data is available. Simulation results illustrate the efficacy of this approach.

Australia currently faces high attrition rates which significantly concern universities and government agencies due to the serious needs for skilled and proficient workforce. Attrition rate is significant factor as it is a performance indicator for fund allocating. To gain a better understanding of the attrition and retention situation in Australia, this paper provides a review of recent empirical studies. Attrition factors and gender bias in Australia are discussed. In addition, it is explained how to support students who are at risk of attrition. © 2012 IEEE.

A novel direct approach for modeling continuous-time stochastic processes is proposed in this paper. First the observed data is passed through an input-to-state filter and the covariance of the output state is computed. The properties of the state covariance matrix is then exploited to estimate the positive real spectrum of the observed data at a set of prescribed points on the right half plane. Finally, the continuous-time parameters are obtained from the positive real spectrum estimates by solving a Nevanlinna-Pick interpolation problem. The estimated model is stable. The analytical results are illustrated using numerical simulations. Copyright © 2005 IFAC.

Least squares based nonparametric estimation of the wave propagation functions of a viscoelastic material is considered in this paper. Widely used nonlinear least squares based algorithms are often computationally expensive and suffer from numerical problems. In this paper we propose a class of subspace estimators which assume equidistant sensor configuration. The proposed estimator is computationally economical and numerically robust. Analytical expressions for the estimation accuracy have been derived. It is also shown that the subspace estimator achieves optimal accuracy under optilnal weighting. The algorithm is employed on real experimental data.

A one-dimensional heat diffusion system with an associated partial differential equation (PDE) model is considered. The PDE model involves unknown parameters, which need to be estimated to model the physical diffusion system. A recursive algorithum in the frequency domain is deviced for an efficient estimation of the parameters involved.

In this paper we propose a parametric and a non-parametric identification algorithm for dynamic errors-in-variables model. We show that the two-dimensional process composed of the input-output data admits a finite order ARMA representation. The non-parametric method uses the ARMA structure to compute a consistent estimate of the joint spectrum of the input and the output. A Frisch scheme is then employed to extract an estimate of the joint spectrum of noise free input-output data from the ARMA spectrum, which is used to estimate the transfer function. The parametric method exploits the ARMA structure to give consistent estimates of the system parameters. The performances of the algorithms are illustrated using the results obtained from a numerical simulation study.

Computationally efficient identification of dynamic errors-in-variables model is considered in this paper. The instrumental variable (IV) method is computationally efficient but it suffers from poor small-sample properties of the estimated parameters. The method presented in this work uses the prefiltered data. The input-output data is passed through a pair of user defined prefilters and the output data from the prefilters is subjected to a least-squares like algorithm. Compared to the IV approach, the proposed method shows a significant improvement in the small-sample properties of the MA parameter estimates, without any increase in the computational load.

Rotation of the binary image of a document page for correcting the skew in the case of OCR or signature verification systems entails disfigurement of the shape of the characters, which brings down the performance of the pattern classifier. An elegant algorithm is presented in this paper for distortion-free rotation. Multirate signal processing principles are applied to solve this problem, which was hitherto handled using only heuristics. The resolution of the image is increased during rotation. The rotated image is decimated and thresholded after low pass filtering. Polyphase implementation of the filter is performed for efficient computation. The results show that it solves the problem of mutilations and occasional breaks obtained in simple rotation schemes and further, performs better than the scheme involving interpolation in the original or the rotated domains. Consequent on the better retention of their original shape, the recognition accuracy of originally skewed documents improves by more than 5%.