In this study, the authors analyse the downlink (DL) performance of dynamic‐time division duplexing (D‐TDD) enabled millimetre‐wave (mmWave) network with different base station (BS) heights using a channel model that incorporates the effect of shadowed fading. Modelling the combined effect of small and large (shadowing) scale fadings with shadowed fading, they provide a general framework based on stochastic geometry that comprehensively includes the smallest path loss BS association with either a line‐of‐sight (LoS) path or a non‐LoS (NLoS) path. For the precise analysis of the network, they further consider different PL distributions for NLoS and LoS links which are associated with the 3D distance, blockage model that includes outage, and directional beamforming at the transmitter and the receiver. Specifically, they assume the directional beamforming at the transmitter and the receiver using scanning angle and number of antennas. Through simulation results, they first illustrate the performance efficiency of D‐TDD over the static‐TDD for DL mmWave networks. Later, they show the impact of shadowed fading parameters on the network performance under the both LoS and NLoS propagations, and only NLoS propagation. Next, the effect of BS heights on the coverage of the network is investigated to provide more insights on the performance in more realistic scenarios. Finally, the coverage of the network is analysed using the varying number of transmitting antennas, scanning angle and BS heights.

We have proposed a new M-estimator based structure for noncoherent demodulation of differentially phase-shift keyed (DPSK) signals transmitted simultaneously via direct sequence code division multiple access (CDMA) flat-fading channel and embedded in impulsive noise. The approach is also corroborated with simulation results to show the efficacy of the proposed detector with the proposed M-estimator in synchronous flat-fading non-Gaussian channels. The gross error sensitivity for the proposed estimator is analyzed. Further, lower bounds for the explosion and implosion breakdown points of the proposed estimator are derived.

This paper presents a robust multiuser detector for combating multiple access interference and impulsive noise in direct-sequence code-division multiple-access (DS-CDMA) communication systems. A new M-estimator is proposed for robustifying the detector. The lower bounds for the explosion and implosion breakdown points of the proposed estimator are derived. For appropriate tuning constants, it is shown that the breakdown points attain the maximum possible value. The approach is also corroborated with simulation results to evaluate the performance of the proposed robust multiuser detector in comparison with the linear decorrelating detector, Huber and Hampel M-estimator based detectors. Simulation results show that, in highly impulsive noise, the proposed detector with the proposed M-estimator with significant performance gain outperforms the linear decorrelating detector, Huber and Hampel M-estimator based detectors in both synchronous and asynchronous non-Gaussian channels.

In this paper, an hybrid filter is presented for restoration of color images in a mixed noise environment, where both impulsive and correlation noise may be present. The proposed hybrid filter is composed of two stages, the first stage to remove the impulsive noise and the second to remove the correlated noise. The median filters and their variants are the most popular filter types used for impulsive noise suppression. However, median filters and their variants tend to remove fine image details and destroy fine texture in the mixed noise environment or when the signal to noise ratio is low. In order to overcome the limitations of the median filters, an adaptive simplified-model Kalman filter (ASMKF) is propsed for the suppression of impulsive noise in the first stage of the hybrid filter. In the second stage of the proposed hybrid filter, to remove correlated noise, discrete Wavelet Transform (DWT) filter is applied on the impulsive-noise-free image obtained from the first stage. The efficacy of the proposed hybrid filter is illustrated through implementation results obtained on the restoration of a color image.

Various approaches based on Kalman filtering exist in the literature for image restoration. The reduced order model Kalman filter (ROMKF) has comparable performance with less computational complexity. For further reduction in computational complexity, we suggest a simplified model Kalman filter (SMKF) for image restoration. Furthermore, a hybrid approach based on discrete wavelet transform (DWT) and SMKF is proposed for image restoration with better SNRs especially when the observed image signal-to-noise ratio is low. In the first step, the approach uses the DWT with few resolution levels and a moderate threshold value for denoising the image. The denoised image will provide a better data to SMKF in the second step. The proposed approach is implemented on a visual image to evaluate its performance in comparison with the SMKF and DWT approaches.

Neural networks with real valued inputs have been proposed in the literature for adaptive equalization and have been used to improve performance of communication channel equalizers. However, neural networks with complex valued inputs and fast convergence are lacking for adaptive equalization. Therefore, in this paper, complex extended Kalman filter (CEKF)-based neural network with complex valued inputs for adaptive equalization of a communication channel is suggested. Performance comparison of the CEKF and complex backpropagation (CBP) neural networks is made through simulation results.