VOLUME 14, ISSUE 6, JUNE 2026

Abstract: Rapid urbanization has greatly increased the need for efficient and hygienic municipal solid waste management systems. Traditional waste collection at the door still depends largely on manual work, fixed schedules, and there is no segregation of waste at the source which leads to operational inefficiencies and environmental problems. This paper discusses the design and development of a GPS, guided autonomous trash collection robot with automatic waste segregation. The proposed system makes use of an Arduino Uno microcontroller as the main control unit, coupled with a GPS module and a digital compass to enable accurate navigation outdoors. The robot receives user, defined pickup locations through a Bluetooth, enabled mobile application. Once the robot arrives at the location, it notifies the user and collects the waste which is automatically divided into wet and dry categories by a moisture, based sensing system. Waste segregation is further facilitated by a servo, actuated unit that directs the different wastes into the appropriate compartments. Testing the prototype confirms the excellent navigation of the robot, accurate waste separation, and no, fuss collecting of the garbage by the drone. The proposed system provides an economically viable and scalable alternative to conventional methods of waste management and is very likely to be adopted by both residential and institutional environments.

Keywords: Autonomous robot, Waste management, GPS navigation, Automatic waste segregation, Arduino Uno, Smart cities, Mobile robotics.

Abstract: The current energy meter reading system relies on human labor, which has drawbacks such as computation errors, customer absences during billing, and additional costs associated with the billing procedure. The proposed system overcomes these limitations by implementing an automatic wireless energy monitoring and tariff calculation system. The system uses a digital energy meter, PIC16F72 microcontroller, ZigBee wireless communication modules, LCD display, optocoupler, relay circuit, and regulated power supply. The energy meter generates pulses proportional to electrical energy consumption. These pulses are detected using an IR sensor and processed by the microcontroller to calculate consumed energy units. The measured data is transmitted wirelessly to the Electricity Board (EB) section through ZigBee communication. At the EB section, automatic tariff calculation is performed based on predefined electricity rates. The billing information is transmitted back to the consumer side and displayed on the LCD screen. The system also includes an automatic load disconnection feature using a relay in case the electricity bill is not paid within the due date. The proposed system provides accurate billing, reduced manpower, real-time monitoring, improved efficiency, and secure wireless communication. The project can be effectively used in smart homes, industries, and modern automated energy management systems.

Keywords: ZigBee, Energy Meter, Wireless Communication, Automatic Billing, PIC16F72, Smart Metering, Tariff Calculation.

Abstract: Adaptive Terahertz (THz) beam steering is a promising and energy efficient technology which has the potential to enhance signal strength, link reliability and signal rate in deep space communication systems where conventional fixed-beam THz links experience severe path loss, atmospheric attenuation and degradation of the pointing error. Under deep space long-distance propagation conditions and a stable high-gain communication link, the problem of maintaining a stable high-gain communication link becomes one of the most critical issues. Conventional fixed-beam THz systems, based on fixed radiation patterns and mechanical steering, have limited pointing error rejection and high misalignment sensitivity, and are infeasible with spacecraft vibration and relative motion. This piece of work proposes a low-complexity adaptive beam steering scheme of the THz deep space communication links where closed-loop beamforming algorithm dynamically adapts the phased array radiation pattern [13], [15] in response to received signal strength feedback with no full channel state information required. The steering issue is formulated as an optimization of beamforming weights in real-time and is solved with the assistance of an adaptive algorithm which is scalable in nature and is less burdensome in terms of processing. To model more realistic deep space conditions, the system model is further developed to include distance-varying SNR degradation and dynamics of the pointing error. The characteristics of the channel capacity [1], [5] and the bit error rate of the adaptive THz beam steering is analyzed as the distance and pointing offset increase and displays the resilience of adaptive THz beam steering against non- adaptive fixed-beam systems. The simulation results, obtained by a full implementation in MATLAB, confirm the proposed strategy leads to the reduction of the sensitivity to pointing errors by a significant margin, high integrity of link maintenance, and spectral efficiency at the deep space propagation conditions. The presented framework offers a viable and scalable next generation deep space communication system solution to future lunar, Mars and interplanetary missions. The simulation results in MATLAB confirm that adaptive THz beam steering is always better than conventional fixed-beam steering in all the measures considered, such as cumulative distribution of channel capacity [1], [5], signal-to-noise ratio over long distance and the ability to tolerate normalized pointing errors. This work provides a solid base on which it is possible to implement intelligent and self-aligning THz communication terminals onboard deep space probes, eliminating the need to rely on bulky mechanical gimbals and allowing autonomous maintenance of links without necessarily ground-based intervention. The further development of this framework can include predictive beam steering with orbital dynamics and channel prediction made by machine learning to achieve even greater efficiency in deep space communication networks.

Keywords: Terahertz communication, deep space, beam steering, adaptive beamforming, phased array, MATLAB simulation, high-data-rate links.

Abstract: The Intelligent Reflecting Surfaces (IRS) have become a promising technology in the enhancement of wireless communication by providing programmable control of the propagation environment. Most of currently available IRS-assisted systems are however designed under quasi-static conditions of the channel, which may not be suitable for high-mobility scenarios envisioned in future 6G networks. This paper proposes a Doppler-aware IRS performance evaluation framework for high-mobility scenarios.A channel model which includes Doppler frequency is developed to accurately represent the impact of user mobility on signal propagation. Under high mobility, the proposed scheme is able to effectively recover from the channel degradation caused by Doppler effect.The behavior of the proposed scheme is analyzed based on the large-scale Monte Carlo simulation in the parameters of Bit Error Rate (BER) and achievable rate under different signal-to-noise ratios and user velocities up to 250 km/h. The simulation results have shown that the dynamic IRS scheme performs significantly better than both the conventional static IRS and non-IRS system, with significantly lower BER and higher achievable rates in all the tested conditions. These results indicate the suitability of the proposed approach in ensuring stable and highly efficient communication in high-mobility settings, thus making it an excellent choice in next-generation 6G wireless systems.

Keywords: Intelligent Reflecting Surface (IRS), 6G, High Mobility, Doppler Effect, Dynamic IRS, Phase Adaptation, Bit Error Rate (BER), Achievable Rate.

Abstract: Classic IoT-based smart agriculture is built on static sensor networks and a cloud-dominated architecture. While effective, such architectures tend to suffer from high latency, inefficient communication, and heavy power consumption due to continuous data transfer. To address this issue, we introduced Agro-Visconic – an IoT solution based on mobile computing that transfers intelligent data processing from the cloud to the edge of the network using Very Large-Scale Integration (VLSI). Instead of static architecture, our design features an autonomous robot as a mobile gateway. Thanks to AI-based visual processing performed on the FPGA platform through Verilog HDL programming, our solution is now completely cloudindependent and operates reliably even in places with no Internet access.

Abstract: The emergence of sixth generation (6G) wireless communication networks requires a huge amount of data rate, low latency, massive connectivity, and high reliability in communication. The use of millimeter-wave (mmWave) communication is seen as one of the key technologies that will satisfy these requirements because of the availability of a wide bandwidth in the mmWave bands. In contrast, however, the path loss for mmWave systems is very high, signals can be blocked by objects and there is a high overhead of beam alignment that makes access to the systems difficult in dynamic wireless environments. Traditional exhaustive beam sweeping techniques are very time consuming and complex since all the beam directions have to be swept prior to establishing communication. To overcome these challenges, a Deep Learning-based Initial Access (DeepIA) framework for fast and reliable beam alignment in AI-powered 6G mmWave Networks is introduced in this paper. The proposed approach is based on a Deep Neural Network (DNN) designed to predict the best beam direction based on Received Signal Strength (RSS) measurements instead of doing a beam sweeping. A novel beam selection method called Sequential Feature Selection (SFS) is used to select the most informative beam combinations to achieve accuracy in prediction while minimizing beam sweeping delay. Moreover, to further improve the system performance under Non-Line-of-Sight (NLoS) channels, a technique called RSS averaging is introduced as an approach to reduce the fluctuation and shadow fading effects of the channel. The simulation results show that the proposed DeepIA framework can accurately predict the beams with a very small number of beam sweeps, which can significantly shorten the delay of initial access and enhance the efficiency of communication. The proposed approach is scalable and intelligent that can be used in the future 6G mmWave communication systems with the use of AI.

Keywords: Deep Learning, 6G, mmWave Communication, Beam Alignment, Beamforming, Initial Access, Deep Neural Network (DNN), Artificial Intelligence (AI)

Abstract: Solar energy is a fantastic clean power source, but its weather-dependent nature makes it highly unpredictable. To beat this volatility and keep the power flowing, we designed and tested a hybrid Photovoltaic-Fuel Cell (PV–FC) system in MATLAB/Simulink. A key part of the puzzle was finding the best way to lock onto maximum solar power, so we put three popular Maximum Power Point Tracking (MPPT) techniques to the test under identical conditions: Perturb and Observe (P&O), Incremental Conductance (INC), and Particle Swarm Optimization (PSO). Our simulations show that the intelligent PSO algorithm is the clear winner, reaching an impressive 97.65% tracking efficiency compared to 94.02% for INC and 90.96% for P&O. It responds much faster to environmental shifts, and when the sun dips, the integrated fuel cell seamlessly steps in to smooth out the supply. Ultimately, this hybrid setup proves to be a highly reliable blueprint for standalone grids, rural electrification, and a more resilient green energy infrastructure.

Keywords: Photovoltaic System, Fuel Cell, MPPT, Particle Swarm Optimization, Hybrid Energy System, MATLAB/Simulink.