The rapid growth of smart-city infrastructures has created an environment in which massive IoT deployments operated across dense, heterogeneous wireless networks. As device density increased, the communication channels have often experienced severe interference, unpredictable fading, and high noise levels that collectively limited estimation accuracy. Traditional estimation techniques relied on linear models that struggled to track the dynamic channel conditions of large-scale IoT environments. This scenario established the core problem: existing estimators have not maintained reliable performance when network density surged or when devices transmitted sporadic traffic. To address this, the study proposed an AI-driven channel estimation framework that has leveraged deep learning to extract latent channel characteristics from limited pilot signals. The method incorporated a hybrid convolutional–recurrent design that captured spatial variations while it tracked temporal fluctuations of each channel. The system also included an adaptive refinement block that has improved estimation accuracy when pilot contamination occurred. The architecture was trained with synthetic and real-world datasets that have represented typical smart-city IoT deployments, including traffic sensors, utility meters, and environmental monitoring nodes that operated under mixed mobility patterns. The evaluation demonstrates that the proposed framework consistently outperforms conventional estimators. The method achieves a 6.2% NMSE at 100 epochs compared with 10.4% for MMSE and 8.2% for CS, and reduces MAE to 4.0% compared with 7.2% for MMSE. Spectral efficiency increases to 6.9 bps/Hz, while pilot overhead is reduced by 25%, outperforming baseline methods. Computational time remains practical at 3.6 ms per batch, confirming that the AI-assisted estimation effectively enhances reliability and efficiency in large IoT smart-city deployments.

This study proposes the design and simulation of a low-power analog jammer that selectively targets LTE downlink signals based on real time uplink detection. The system architecture integrates a field strength detection unit, a PLL-controlled frequency sweeper, and a jamming signal generator using Zener-based noise injection and RF mixing via SA612A ICs. Simulations conducted in Proteus and MATLAB/Simulink validated the functional blocks, demonstrating accurate uplink detection, stable frequency synthesis, and effective jamming performance. Key results include spectral spreading between 21–33?dBJE, severe signal distortion, and bit-error rates exceeding 80% under interference conditions. While manual tuning and regulatory limitations constrain immediate deployment, the proposed solution offers a scalable foundation for controlled civilian use. The findings support future development of digitally enhanced, multi-band jamming systems tailored for educational or security-sensitive settings.

A low-profile dielectric resonator antenna backed by substrate integrated waveguide cavity is proposed in this paper. A novel feeding mechanism is used to excite a low-profile dielectric resonator antenna backed by substrate integrated waveguide cavity. Instead of a single slot, a 2x2 array of 2mm x 2mm slots is used to excite the dielectric resonator antenna. With this feeding technique, an antenna working in X band frequencies is designed by combining substrate integrated waveguide cavity and dielectric resonator antenna resonant modes. The paper compares the conventional single slot excited dielectric resonator antenna and the proposed 2x2 slot array excited dielectric resonator antenna. The proposed antenna is linearly polarized with a 2:1 VSWR bandwidth of 5.3% and a peak gain of 6.5dBi. Overall dimension of the antenna is 30mm x 30mm x 0.78mm.

Peer to peer (P2P) overlays dominate content distribution, collaborative applications, and edge services because they eliminate single points of failure and exploit aggregate bandwidth. Yet, heterogeneous node capacity, churn, and route redundancy often throttle end to end throughput. Classical P2P rate control and scheduling schemes (e.g., tit for tat, rarest first) optimise a single objective or operate on a single network layer, leaving cross layer interactions unexploited. This results in sub optimal bandwidth utilisation, especially under bursty traffic and high churn. We introduce E ThruEnsemble, an ensemble algorithm that fuses (i) adaptive chunk scheduling, (ii) topology aware path selection, and (iii) reinforcement learning guided rate control. The three weak learners each make local throughput estimates; a lightweight Bayesian combiner assigns dynamic weights based on recent prediction error. The final scheduling decision maximises a composite utility that jointly rewards link utilisation and delivery deadline satisfaction. We implement the scheme in NS 3 and instrument it with real trace latency variations. In 500 node overlays, E ThruEnsemble raises average throughput by 29?% over BitTorrent’s choking algorithm, 17?% over ML DOS, and 11?% over ChunkyStream while lowering 95 th percentile latency by 22?%. It converges within 25?seconds after a 20?% churn event and achieves a Jain fairness index of 0.93. Sensitivity studies confirm robustness to packet loss rates up to 5?%.

Vehicular Ad Hoc Networks (VANETs) have emerged as a critical component in intelligent transportation systems (ITS), enabling vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication. However, the highly dynamic topology, high mobility, and low latency requirements of VANETs present significant challenges for ensuring reliable and efficient data transmission. Traditional machine learning models often struggle to adapt to VANETs’ real-time data processing needs and variable network conditions. While deep learning offers promising capabilities in feature extraction and pattern recognition, standalone architectures may fall short due to overfitting, underfitting, or limited generalization in complex VANET environments. This study proposes an improvised ensemble deep learning framework that integrates Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and Transformer-based attention mechanisms. The ensemble model leverages the spatial-temporal feature extraction strength of CNN-RNN and the long-range dependency modeling capability of Transformers. A weighted majority voting and adaptive fusion layer are implemented to combine model outputs effectively. The framework is evaluated using real-time vehicular mobility datasets and simulated traffic scenarios to measure metrics such as packet delivery ratio (PDR), end-to-end delay, and throughput. The proposed ensemble framework achieved a 15–20% improvement in PDR, a 25% reduction in end-to-end delay, and a significant increase in throughput compared to existing deep learning baselines.

Internet of Things (IoT) integrated with Wireless Sensor Networks (WSNs) plays a critical role in remote monitoring and intelligent decision-making. However, energy conservation remains a major concern due to the limited battery life of sensor nodes. Efficient cluster head (CH) selection directly influences network lifetime and energy consumption. LEACH and fuzzy-based clustering are examples of classic methods that usually can't deal with nodes that act in complicated ways, environments that change, and trying to reach many goals at the same time. This research points to a new Deep Genetic Mechanism (DGM) that could help people make good decisions about CHs. This method uses a genetic algorithm (GA) and a fitness assessor that is based on deep learning. This allows you pick CHs that are stable and use less energy right now. A deep neural network looks at the current state of the network, the energy levels, and the placement of the nodes to discover the best ways to organize them. This network then informs genetic algorithms what to perform, like crossover, selection, and mutation. We utilize MATLAB to do tests on the suggested DGM with real WSN properties. Some of the methods that are compared to it are LEACH, PSO-based CH selection, and fuzzy C-means clustering. DGM is wonderful in many respects, such as how much energy it needs, how many packets it transmits, how long the network lasts, and how stable the nodes are. The network's lifetime went up by 27.4% compared to LEACH, and the number of nodes that failed because of unbalanced residual energy was reduced.

The rapid evolution of Internet of Vehicles (IoV) systems has enabled smart transportation through Vehicle-to-Everything (V2X) communications. Cyber dangers include message manipulation, impersonation, and denial of service (DoS) attacks put both cars and data at risk. More and more cars are connecting to the internet, which makes these attacks happen more often. Traditional Intrusion Detection Systems (IDS) often lack the capability to process high-dimensional IoV traffic data efficiently and fail to generalize across evolving attack patterns. Lightweight machine learning methods underperform in feature representation and temporal correlation detection, especially in real-time vehicular environments. This study proposes a cyberattack detection model utilizing Residual Neural Networks (ResNet) to capture complex spatiotemporal patterns in IoV data. The ResNet architecture is trained on a benchmark vehicular network dataset to classify normal and malicious traffic efficiently. ResNet’s skip connections enable deeper networks to avoid vanishing gradients and improve learning efficiency, even with limited labeled data. The proposed ResNet-based IDS achieved superior detection accuracy compared to conventional models like CNN, LSTM, and SVM. It yielded a classification accuracy of 98.7%, precision of 98.9%, and a recall of 98.3%, outperforming benchmark systems by an average margin of 5–8% in all metrics. The framework shows potential for real-time deployment in smart vehicular ecosystems.

Sleep is vital in maintaining overall health and cognitive function. Sleep disorders, like insomnia, can significantly impact sleep quality and quantity, leading to adverse effects on humans. Around 30% of the adult population in the world has insomnia symptoms. Current treatments for insomnia include lifestyle changes, therapy, and medication, but they often have limitations and potential side effects. Non-invasive electronic methods have shown promise in addressing sleep problems, but they lack feedback on the patient’s sleep condition. The study utilised a system that monitored heart rate, blood oxygen saturation, body temperature, and body movement. A microcontroller-based circuitry integrated the sensor data and provided real-time feedback. The electromagnetic coil generated a sleep-inducing electromagnetic field in the delta frequency band. The developed sleep inducer successfully monitored all the selected vital parameters. The microcontroller processed the sensor data and triggered the electromagnetic coil when deviations from specified conditions were detected. The system operated effectively to induce sleep, maintaining the coil activation for at least 5 minutes per parameter above threshold levels. We also tested the device on one volunteer for two consecutive days. The non-invasive sleep inducer demonstrated potential as an effective tool for insomnia treatment. By incorporating monitoring and feedback techniques, it addressed the limitations of existing methods and provided personalised sleep-inducing effects. The system’s low cost, simplicity, and patient comfort make it a promising alternative for insomnia management. Further research and refinement are necessary to enhance the system’s efficacy and ensure long-term safety due to limited testing.

Energy management is a critical issue in Wireless Sensor Networks (WSNs) due to the limited battery capacity of nodes, which directly impacts the design of energy-efficient routing protocols aimed at extending the network's lifetime and conserving node energy. Each sensor node relies on its battery for vital functions in the network, making energy conservation essential for network sustainability. While the Low Energy Adaptive Clustering Hierarchy (LEACH) is a popular algorithm, it treats all nodes with varying energy levels uniformly, leading to premature node failure. To address this, the HLEACH-PSO (High Lifetime Energy-Aware Cluster Head Particle Swarm Optimization) is introduced as a modified version of LEACH. In this approach, the Cluster Head (CH) role is assigned to randomly distributed nodes across the network. The introduction of two new parameters—Base station connectivity and Average node lifetime—modifies the CH selection process to ensure that previously selected CHs are not re-elected. A Particle Swarm Optimization (PSO)-based clustering method, enhanced with an energy-constrained fitness function, is proposed to optimize both inter- and intra-cluster routing. Simulation results show that HLEACH-PSO outperforms traditional LEACH by reducing energy consumption, increasing network lifetime, and enhancing communication quality.

Software-Defined Networks (SDNs) coupled with Cognitive Radio (CR) systems offer dynamic spectrum access capabilities, enabling efficient spectrum utilization. However, the high dimensionality of network parameters and unpredictable spectrum availability pose critical challenges in achieving real-time adaptation and optimal throughput. Existing adaptive search and decision-making algorithms often fail to scale effectively in high-dimensional state spaces, leading to reduced convergence rates and suboptimal spectrum allocation. Traditional ensemble techniques lack dynamic interaction between learning agents and real-time feedback mechanisms. This work introduces an Improvised Ensemble Method built upon an Empowered Adaptive Dimensional Search (EADS) algorithm. The proposed system yields a 17% increase in throughput, 22% lower latency, and 19% improvement in spectral efficiency.