Electricity is widely used around 80% of the world. Electricity theft has dangerous effects on utilities in terms of power efficiency and costs billions of dollars per annum. The enhancement of the traditional grids gave rise to smart grids that enable one to resolve the dilemma of electricity theft detection (ETD) using an extensive amount of data formulated by smart meters. This data are used by power utilities to examine the consumption behaviors of consumers and to decide whether the consumer is an electricity thief or benign. However, the traditional data-driven methods for ETD have poor detection performances due to the high-dimensional imbalanced data and their limited ETD capability. In this paper, we present a new class balancing mechanism based on the interquartile minority oversampling technique and a combined ETD model to overcome the shortcomings of conventional approaches. The combined ETD model is composed of long short-term memory (LSTM), UNet and adaptive boosting (Adaboost), and termed LSTM–UNet–Adaboost. In this regard, LSTM–UNet–Adaboost combines the advantages of deep learning (LSTM-UNet) along with ensemble learning (Adaboost) for ETD. Moreover, the performance of the proposed LSTM–UNet–Adaboost scheme was simulated and evaluated over the real-time smart meter dataset given by the State Grid Corporation of China. The simulations were conducted using the most appropriate performance indicators, such as area under the curve, precision, recall and F1 measure. The proposed solution obtained the highest results as compared to the existing benchmark schemes in terms of selected performance measures. More specifically, it achieved the detection rate of 0.92, which was the highest among existing benchmark schemes, such as logistic regression, support vector machine and random under-sampling boosting technique. Therefore, the simulation outcomes validate that the proposed LSTM–UNet–Adaboost model surpasses other traditional methods in terms of ETD and is more acceptable for real-time practices.

Massive multiple-input multiple-output (massive-MIMO) is foreseen as a potential technology for future 5G cellular communication networks due to its substantial benefits in terms of increased spectral and energy efficiency. These advantages of massive-MIMO are a consequence of equipping the base station (BS) with quite a large number of antenna elements, thus resulting in an aggressive spatial multiplexing. In order to effectively reap the benefits of massive-MIMO, an adequate estimate of the channel impulse response (CIR) between each transmit–receive link is of utmost importance. It has been established in the literature that certain specific multipath propagation environments lead to a sparse structured CIR in spatial and/or delay domains. In this paper, implicit training and compressed sensing based CIR estimation techniques are proposed for the case of massive-MIMO sparse uplink channels. In the proposed superimposed training (SiT) based techniques, a periodic and low power training sequence is superimposed (arithmetically added) over the information sequence, thus avoiding any dedicated time/frequency slots for the training sequence. For the estimation of such massive-MIMO sparse uplink channels, two greedy pursuits based compressed sensing approaches are proposed, viz: SiT based stage-wise orthogonal matching pursuit (SiT-StOMP) and gradient pursuit (SiT-GP). In order to demonstrate the validity of proposed techniques, a performance comparison in terms of normalized mean square error (NCMSE) and bit error rate (BER) is performed with a notable SiT based least squares (SiT-LS) channel estimation technique. The effect of channels’ sparsity, training-to-information power ratio (TIR) and signal-to-noise ratio (SNR) on BER and NCMSE performance of proposed schemes is thoroughly studied. For a simulation scenario of: $4\times 64$ massive-MIMO with a channel sparsity level of $80\%$ and signal-to-noise ratio (SNR) of $10\mathrm{dB}$ , a performance gain of $18\mathrm{dB}$ and $13\mathrm{dB}$ in terms of NCMSE over SiT-LS is observed for the proposed SiT-StOMP and SiT-GP techniques, respectively. Moreover, a performance gain of about $3\mathrm{dB}$ and $2.5\mathrm{dB}$ in SNR is achieved by the proposed SiT-StOMP and SiT-GP, respectively, for a BER of ${10}^{-2}$ , as compared to SiT-LS. This performance gain NCME and BER is observed to further increase with an increase in channels’ sparsity.

A geometry-based three-dimensional (3D) novel stochastic channel model for air-to-ground (A2G) and ground-to-air (G2A) radio propagation environments is proposed. The vicinity of a ground station (GS) is modelled as surrounded by effective scattering points; whereas the elevated air station’s (AS) vicinity is modelled as a scattering-free region. Characterization of the Doppler spectrum, dispersion in the angular domain and second order fading statistics of the A2G/G2A radio communication channels is presented. Closed-form analytical expressions for joint and marginal probability density functions (PDFs) of Doppler shift, power and angle of arrival (AoA) are derived. Next, the paper presents a comprehensive analysis on the characteristics of angular spread on the basis of shape factors (SFs) for A2G/G2A radio propagation environments independently in both the azimuth and elevation planes. The analysis is further extended to second order statistics of the fading channel; where the behaviour of the level crossing rate (LCR), average fade duration (AFD), auto-covariance and coherence distance for the A2G/G2A radio propagation environment is studied. Finally, the impact of physical channel parameters, such as the mobility of AS, the height of AS, the height of GS and the delay of the longest propagation path, on the distribution characteristics of Doppler shift, angular spread and second order statistics is thoroughly studied.

This paper presents LUVS-Net, which is a lightweight convolutional network for retinal vessel segmentation in fundus images that is designed for resource-constrained devices that are typically unable to meet the computational requirements of large neural networks. The computational challenges arise due to low-quality retinal images, wide variance in image acquisition conditions and disparities in intensity. Consequently, the training of existing segmentation methods requires a multitude of trainable parameters for the training of networks, resulting in computational complexity. The proposed Lightweight U-Net for Vessel Segmentation Network (LUVS-Net) can achieve high segmentation performance with only a few trainable parameters. This network uses an encoder–decoder framework in which edge data are transposed from the first layers of the encoder to the last layer of the decoder, massively improving the convergence latency. Additionally, LUVS-Net’s design allows for a dual-stream information flow both inside as well as outside of the encoder–decoder pair. The network width is enhanced using group convolutions, which allow the network to learn a larger number of low- and intermediate-level features. Spatial information loss is minimized using skip connections, and class imbalances are mitigated using dice loss for pixel-wise classification. The performance of the proposed network is evaluated on the publicly available retinal blood vessel datasets DRIVE, CHASE_DB1 and STARE. LUVS-Net proves to be quite competitive, outperforming alternative state-of-the-art segmentation methods and achieving comparable accuracy using trainable parameters that are reduced by two to three orders of magnitude compared with those of comparative state-of-the-art methods.

In this paper, a novel modified optimization algorithm is presented, which combines Nelder-Mead (NM) method with a gradient-based approach. The well-known Nelder Mead optimization technique is widely used but it suffers from convergence issues in higher dimensional complex problems. Unlike the NM, in this proposed technique we have focused on two issues of the NM approach, one is shape of the simplex which is reshaped at each iteration according to the objective function, so we used a fixed shape of the simplex and we regenerate the simplex at each iteration and the second issue is related to reflection and expansion steps of the NM technique in each iteration, NM used fixed value of $\alpha$ , that is, $\alpha$  = 1 for reflection and $\alpha$  = 2 for expansion and replace the worst point of the simplex with that new point in each iteration. In this way NM search the optimum point. In proposed algorithm the optimum value of the parameter $\alpha$ is computed and then centroid of new simplex is originated at this optimum point and regenerate the simplex with this centroid in each iteration that optimum value of $\alpha$ will ensure the fast convergence of the proposed technique. The proposed algorithm has been applied to the real time implementation of the transversal adaptive filter. The application used to demonstrate the performance of the proposed technique is a well-known convex optimization problem having quadratic cost function, and results show that the proposed technique shows fast convergence than the Nelder-Mead method for lower dimension problems and the proposed technique has also good convergence for higher dimensions, that is, for higher filter taps problem. The proposed technique has also been compared with stochastic techniques like LMS and NLMS (benchmark) techniques. The proposed technique shows good results against LMS. The comparison shows that the modified algorithm guarantees quite acceptable convergence with improved accuracy for higher dimensional identification problems.