Showing papers by "Susmita Das published in 2018"
TL;DR: Qualitative and quantitative study and analysis indicate that the proposed technique can be used as an effective tool for denoising of ECG signals and hence can serve for better diagnostic in computer-based automated medical system.
144 citations
01 Jul 2018
TL;DR: The proposed algorithm avoids the need for a central controller and complex communication structure thereby reducing the computational burden and the risk of single-point-failure and is efficient to provide quick restoration of the voltage and frequency whilst accurate power-sharing is achieved despite disturbances.
Abstract: This study presents a fully distributed control paradigm for secondary control of islanded AC microgrid (MG). The proposed method addresses both voltage and frequency restoration for inverter-based distributed generators (DGs). The MG system has droop controlled DG units with predominantly inductive transmission lines and different communication topologies. The restoration scheme is fully distributed in nature, and the DGs need to communicate with their neighbours using a sparse communication network. The proposed control scheme is efficient to provide quick restoration of the voltage and frequency whilst accurate power-sharing is achieved despite disturbances. Further, convergence and stability analysis of the proposed control scheme is presented. The proposed algorithm avoids the need for a central controller and complex communication structure thereby reducing the computational burden and the risk of single-point-failure. The performance of the proposed control scheme has been verified considering variations in load and communication topologies and link delay by pursuing an extensive simulation study in MATLAB/SimPowerSystem toolbox. The proposed control scheme supports plug-and-play demand and scalability of MG network. The proposed control scheme is also compared with the neighbourhood tracking error based distributed control scheme and observed that the former exhibit faster convergence and accurate performance despite disturbances in MG network.
30 citations
TL;DR: It is observed that the absence of a hidden layer in FLANN makes it more efficient than the MLP model in predicting the occupancy faster and with less complexity.
Abstract: Recent advancements in artificial neural networks (ANNs) motivated us to design a simple and faster spectrum prediction model termed the functional link artificial neural network (FLANN). The main objective of this paper is to gather realistic data to obtain utilization statistics for the industrial, scientific and medical band of 2.4–2.5 GHz. To present the occupancy statistics, we conducted measurement in indoors at the Swearingen Engineering Center, University of South Carolina. Further, we introduce different threshold-based spectrum prediction schemes to show the impact of threshold on occupancy, and propose a spectrum prediction algorithm based on FLANN to forecast a future spectrum usage profile from historical occupancy statistics. Spectrum occupancy is estimated and predicted by employing different ANN models including the Feed-forward multilayer perceptron (MLP), Recurrent MLP, Chebyshev FLANN and Trigonometric FLANN. It is observed that the absence of a hidden layer in FLANN makes it more efficient than the MLP model in predicting the occupancy faster and with less complexity. A set of illustrative results are presented to validate the performance of our proposed learning scheme.
14 citations
TL;DR: Empirical channel models for body surface communication in the various sub-bands specified by the IEEE 802.15.6 are proposed as the combination of propagation through wireless media and losses due to absorption from body tissues make it challenging.
Abstract: Body area networks are being developed to serve a wide range of purposes ranging from providing health care to patients on the move to tracking patients and motion sensing for gaming controls. There has been significant and sizeable amount of research in the various areas and applications of body area networks. Ultra wideband which operates in the 3.1–10.6 GHz band is slowly being preferred for high data rate communication in body area networks. The development of suitable applications and techniques for communication depends significantly on the channel models. The wireless channel is a crucial parameter as it provides significant information about the propagation characteristics and losses involved in the transmission medium. The existing channel models proposed are mostly in the spectra involving the wideband 3.1–10.6 GHz bands or the 3.1–6 GHz bands. However, the IEEE 802.15.6 specifies operation in various sub-bands of 499.2 MHz width. And the channel characteristics are significantly different for wideband and narrowband channels. In this article, we propose empirical channel models for body surface communication in the various sub-bands specified by the IEEE 802.15.6. The body surface scenario is chosen as the combination of propagation through wireless media and losses due to absorption from body tissues make it challenging. The proposed path loss models are developed from more than 300,000 received power measurements collected over a span of hours.
6 citations
01 Jan 2018
TL;DR: Two efficient relay selection algorithms on the performance of an IEEE 802.16j MMR WiMAX network are investigated and it is observed that the DF relay along with MRC combination technique outperforms the other relay-diversity combining techniques in terms of providing improved SER and channel capacity.
Abstract: This paper investigates two efficient relay selection algorithms on the performance of an IEEE 802.16j MMR WiMAX network. The relay selection algorithms considered here are max-min relay selection and harmonic mean of SNR relay selection. The WiMAX transmitter comprises of OFDMA transmission technique and the receiver with maximum likelihood (ML) detection method is adopted for reliable detection of transmitted bits. The relays used here are the Amplify-Forward (AF) relay and Decode-Forward (DF) relay. Both selection combining (SC) and maximal ratio combining (MRC) techniques are applied and the performance metrics are symbol error rate (SER) and channel capacity. The analysis of the network is performed using the MATLAB software. It is observed that the DF relay along with MRC combination technique outperforms the other relay-diversity combining techniques in terms of providing improved SER and channel capacity. Again in providing less SER, the harmonic mean of SNR relay selection algorithm is superior in comparison to max-min relay selection algorithm. Further, the max-min based relay selection algorithm provides better channel capacity about the harmonic mean of SNR relay selection algorithm for the considered MMR WiMAX network.
4 citations
01 Dec 2018
TL;DR: The proposed control scheme synchronizes the voltage and frequency of an islanded MG to their nominal values and a permissible communication delay bound is derived using the solution of delay differential equation model of the dynamic consensus algorithm.
Abstract: Communication networks are extensively used in microgrid (MG) networks for information sharing under distributed secondary control. The shared control information experiences inherent time delay while traversing the communication channels. This paper presents a communication latency impact analysis on dynamic average consensus-based secondary control scheme in an islanded MG. The proposed control scheme synchronizes the voltage and frequency of an islanded MG to their nominal values. A permissible communication delay bound is derived using the solution of delay differential equation model of the dynamic consensus algorithm. Also, the effect of communication latency on the convergence rate of the control scheme is presented. The relationship between maximum degree of the communication graph and permissible delay bound is established. Furthermore, the effectiveness of the proposed control scheme is evaluated by simulating a MG test system using MATLAB/SimPowerSystem. The effect of communication latency is presented by simulating a set of fixed link delays and results show that the convergence time increases with an increase in time delay.
2 citations
TL;DR: An empirical path loss model is estimated for an IEEE 802.16e standardised WiMAX network operating on a carrier frequency of 2.65 GHz deployed in a sub-urban area and it is concluded that this model closely approximates to Stanford University Interim model with path loss exponent value of 3.45.
Abstract: Deployment of an efficient cellular network is considered as a challenging task as it affects the performance measures like data rate, bit error rate, spectrum efficiency and energy efficiency etc. For this, the foremost important step is developing an accurate path loss model for the network in the deployment region. In this paper, an empirical path loss model is estimated for an IEEE 802.16e standardised WiMAX network operating on a carrier frequency of 2.65 GHz deployed in a sub-urban area. An experimental setup is designed for collecting the parameters such as carrier to interference plus noise ratio (CINR), received signal strength indicator (RSSI) for the concerned network and with the help of regression technique, the path loss model is formulated. The relationships between CINR, RSSI, and the distance between base station and customer premise equipment are formulated. The distributions of RSSI, CINR and path loss for the concerned network are also found out. Then by using the proposed path loss model, link budget analysis is performed. From the analysis, it is concluded that the proposed path loss model closely approximates to Stanford University Interim model with path loss exponent value of 3.45.
2 citations