In this article, we present a novel localization and activity classification method for aerial vehicle using mmWave frequency modulated continuous wave (FMCW) Radar. The localization and activity classification for aerial vehicle enables the utilization of mmWave Radars in security surveillance and privacy monitoring applications. In the proposed method, Radar's antennas are oriented vertically to measure the elevation angle of arrival of the aerial vehicle from ground station. The height of the aerial vehicle and horizontal distance of the aerial vehicle from Radar station on ground are estimated using the measured radial range and the elevation angle of arrival. The aerial vehicle's activity is classified using machine learning methods on micro-Doppler signatures extracted from Radar measurements taken in an outdoor environment. To evaluate performance, various light weight classification models such as logistic regression, support vector machine (SVM), Light gradient boosting machine (GBM), and a custom lightweight convolutional neural network (CNN) are investigated. Based on the results, the logistic regression, SVM, and Light GBM achieve an accuracy of 93%. Furthermore, the custom lightweight CNN can achieve activity classification accuracy of 95%. The performance of the proposed lightweight CNN is also compared with the pre-trained models (VGG16, VGG19, ResNet50, ResNet101, and InceptionResNet). The proposed lightweight CNN suits best for embedded and/or edge computing devices.

In this letter, we obtain a generalized expression for the optimal number of cognitive users M for the K-out-of-M rule that minimizes the Bayes risk at the fusion center over noisy control channels. We show that many existing and new results for optimal M are special cases of the proposed solution. Numerical results are presented using energy detector. However, the expressions for optimal M obtained in this letter are applicable to any detector used in cooperative spectrum sensing.

This letter considers cooperative spectrum sensing (CSS) in a heterogeneous cognitive radio (CR) network, where $K$ non identical cognitive users (CUs) and a fusion center (FC) collaboratively detect the primary user's signal using the $N$ -out-of- $K$ rule in the presence of erroneous control channels. We formulate a generalized optimal fusion problem (GOFP) to optimize the $N$ -out-of- $K$ rule for the Bayesian test and then solve the GOFP to obtain the optimal value of $N$ . It follows that existing results on the optimal value of $N$ in the $N$ -out-of- $K$ rule are special cases of the proposed solution. We then propose a framework to obtain the smallest number of CUs required in CSS while satisfying a target error bound at the FC. Numerical results are presented using the energy detector (ED) as an example.

The design of the optimal non-randomized hard decision fusion rule under Neyman-Pearson (NP) criterion is known to be exponential in complexity. In this letter, we formulate a more generalized version of this problem called “Generalized Decision Fusion Problem (GDFP)” and relate it to the classical 0-1 Knapsack Problem. Consequently we show that the GDFP has a worst case polynomial time solution. Numerical results are presented to verify the effectiveness of the proposed solution.

The period after the COVID-19 wave is called the Echo-period. Estimation of crowd size in an outdoor environment is essential in the Echo-period. Making a simple and flexible working system for the same is the need of the hour. This article proposes and evaluates a non-intrusive, passive, and cost-effective solution for crowd size estimation in an outdoor environment. We call the proposed system as LTE communication infrastructure based environment sensing or LTE-CommSense. This system does not need any active signal transmission as it uses LTE transmitted signal. So, this is a power-efficient, simple low footprint device. Importantly, the personal identity of the people in the crowd can not be obtained using this method. First, the system uses practical data to determine whether the outdoor environment is empty or not. If not, it tries to estimate the number of people occupying the near range locality. Performance evaluation with practical data confirms the feasibility of this proposed approach.