This article presents a miniaturized antenna operating on ultra-wideband (UWB) spectrum having two reconfigurable notch bands. Initially, an antenna is designed for UWB applications, offering a 3.3–14.5 GHz impedance bandwidth. Subsequently, a semicircular stub and an arc-shaped slot are etched to achieve two distinguish notch bands at 3.7 and 8.1 GHz. PIN diodes are loaded onto the antenna to enable on-demand switching between notch bands. The final geometry of the antenna offers compact size of 15 mm × 20 mm × 1.6 mm and is designed using commercially available substrate material FR-4. The structure of antenna consists of a rectangular stub loaded into a circular patch, while an additional semicircular stub and arc-shaped slot are utilized to achieve notch bands. The antenna provides a moderate value of gain surpassing 2.5 dBi in the operational region. The antenna has the capability to operate between the UWB mode or the UWB mode having single or dual notch bands using two PIN diodes. To validate the simulated results, a hardware prototype of the design is fabricated and tested, yielding similar results obtained from software-based antenna. Finally, the antenna performance is compared with recently reported similar work in literature, demonstrating advantages such as miniaturization, wideband, moderate gain, and high efficiency along with on-demand switching characteristics while maintaining simple geometrical configuration. These characteristics highlight the scientific contribution of the proposed work along with making it a potential candidate for targeted applications.

This study investigates the application of a Bidirectional Long Short-Term Memory (BiLSTM) model for Human Activity Recognition (HAR) using the PAMAP2 dataset. The aim was to enhance the accuracy and efficiency of recognizing daily activities captured by wearable sensors. The proposed BiLSTM-based model achieved outstanding performance, with 98.75% training accuracy and 99.27% validation accuracy. It also demonstrated high precision, recall, and F1 scores (all 0.99). Comparative analysis with state-of-the-art models, including Deep-HAR and CNN-BiLSTM-BiGRU, revealed that the proposed BiLSTM model surpassed their performance. These results highlight the potential of the proposed approach for real-time HAR applications in edge computing, particularly where accurate and efficient activity recognition is crucial.

The speed up of group operations on elliptic curves is proposed using a new type of projective coordinate representation. These operations are the most common computations in key exchange and encryption for both current and postquantum technology. The boost this improvement brings to computational efficiency impacts not only encryption efforts but also attacks. For maintaining security, the community needs to take note of this development as it may need to operate changes in the key size of various algorithms. Our proposed projective representation can be viewed as a warp on the Jacobian projective coordinates, or as a new operation replacing the addition in a Jacobian projective representation, basically yielding a new group with the same algebra elements and homomorphic to it. Efficient algorithms are introduced for computing the expression $Pk+Q$ where P and Q are points on the curve and k is an integer. They exploit optimized versions for particular k values. Measurements of the numbers of basic computer instructions needed for operations based on the new representation show clear improvements. The experiments are based on benchmarks selected using standard NIST elliptic curves.