Showing papers by "Sarat Kumar Patra published in 2018"

On the PTS Optimization Using the Firefly Algorithm for PAPR Reduction in OFDM Systems

Abstract: Orthogonal frequency division multiplexing (OFDM) is currently the most used technique for the high-data-rate transmission. However, OFDM systems have an inherent drawback in that the transmitted signals suffer from a high peak-to-average power ratio (PAPR). Partial transmit sequence (PTS) is generally applied to reduce the PAPR of an OFDM signal in wireless communication systems. Here, the search complexity for optimal phase vector increases exponentially with the number of phase vectors, since it involves an extensive random search over all the combinations of the allowed phase vectors. In this paper, a swarm intelligence algorithm for phase optimization based on the Firefly algorithm (FF) is applied to search the optimal combination of phase vectors. The proposed algorithm provides a superior trade-off between the improved PAPR performance and computational complexity when compared to the PTS scheme for a large number of sub-blocks. Simulation results show that the Firefly-algorithm-based PTS (...

Blind Identification of Radio Access Techniques Based on Time-Frequency Analysis and Convolutional Neural Network

Abstract: In the Last decade, various machine learning schemes have been investigated to make the cognitive radio (CR) more adaptive. Blind identification of radio accesses technology (RAT) indirectly aid the CR to adapt according to the real-time wireless environment. In this paper, some of the various wireless standards like GSM, Bluetooth and Wi-Fi are blindly identified using deep neural networks. The present work proposes the combination of time-frequency distributions and Convolutional Neural Network (CNN) based Machine Learning technique to identify the RATs. Time-Frequency Analysis (TFA) is used to obtain the spectral content of the signal and Convolutional Neural Network is used for feature extraction and identification purpose. The accuracy of the network is analyzed with performance plots of correct identification and the confusion matrix. Also, Performance of the deep neural network classifier has been compared with the previously proposed Machine Learning techniques.

Large-Scale Antenna System Performance with Imperfect CSI in Cooperative Networks

Abstract: In this paper, we have considered a relay assisted cooperative network, where relay station (RS) and base station (BS) have very large but finite number of antenna. The data detection is done by linear zero forcing (ZF) technique assuming BS and RS have imperfect channel state information (CSI). We derive a new analytical expression for the uplink rate in different channel imperfection scenario. In single hop signal transmission from mobile users (MU) to BS via RS, large number of RS and BS antennas play a vital role over the fixed channel error variance. For dealing with large MIMO, we have used the property of random matrix theory (specially Wishart Matrix decomposition). We have drawn the relation, where uplink rate in single hop signal transmission is the function of number of RS and BS antenna, both link channel error variance and other parameter also. Keeping other parameters constant the uplink rate vs number of RS and BS antenna has been numerically validated for large MIMO perspective under suitable simulation parameter.