Advance Research in Communication Engineering and its Innovations

IntelligentSense Bot: A Multi-model Conversational AI Framework for Real-time Context-aware Interaction

Ganesh Satish Nalawade — 2025-05-14

This work presents the IntelligentSense Bot, a state-of-the-art AI-powered conversational assistant engineered to deliver real-time, context-aware responses across multiple industries. Traditional chatbots are often restricted by static, rule-based systems that struggle to adapt to dynamic user needs. In contrast, IntelligentSense Bot integrates advanced natural language processing (NLP), machine learning (ML), and context retention techniques to produce dynamic interactions that closely mimic human conversation. In benchmark tests, the multi-model framework achieved a 23% improvement in response accuracy, reduced average latency to 1.2 seconds (vs. 2.8 seconds for baseline systems), and yielded an overall user satisfaction score of 87% in live deployments. The LLaMA Mixtral model demonstrated >85% accuracy and >90% context-awareness, while DistilGPT2 delivered sub-second response times (0.5 s) for simpler queries. Efficiency scores combining speed and accuracy reached 85% for LLaMA Mixtral, 72% for T5-small, and 65% for DistilGPT2. The unified architecture is defined in detail here, which adapts continuously through domain-specific customizations and compares this approach with existing methodologies. Furthermore, this bot has outlined practical applications in transportation, education, and corporate environments, and provides preliminary performance metrics that demonstrate improvements in response accuracy, latency, and overall user satisfaction. This work contributes to the evolving landscape of conversational AI by addressing key challenges in adaptability, integration, and real-time processing, thereby setting a new benchmark for future research and development in this field.

A Review on Intelligent Door Lock Systems: Biometrics, IoT, and AI Applications

Neha Janardan Naukudkar — 2025-06-06

This review paper explores a range of advanced door lock systems and presents a comprehensive understanding of modern technologies used to enhance security against criminal activities such as theft and robbery. By analysing various research papers and technological advancements, the paper aims to provide insights into affordable and efficient security solutions for personal and property protection.

The primary objective of this review is to investigate different technological approaches that can be implemented to create secure and user-friendly door lock systems. Emphasis is placed on using cost-effective technologies that are easy to integrate and implement in real-world scenarios. The review highlights how technologies like Raspberry Pi, image processing, radio frequency identification (RFID), and other smart systems can be leveraged to design intelligent door-locking mechanisms.

One of the key takeaways from this paper is the potential of combining multiple technologies to build versatile and reliable security systems. These solutions not only enhance protection but also open doors to further technological innovation. The paper encourages thinking beyond conventional security methods and inspires the development of new applications by adapting existing technologies in novel ways.

In summary, this review serves as a valuable resource for anyone interested in the development and implementation of advanced door lock systems. It emphasizes the importance of integrating low-cost yet effective technologies, offering both technical understanding and inspiration for future innovations in the field of home and building security.

Real-Time Object Detection for Autonomous Drone Navigation Using YOLOv8

Vaibhav Godase — 2025-07-16

This paper presents an original implementation and evaluation of YOLOv8-based real-time object detection for autonomous drone navigation. While previous works have primarily focused on earlier YOLO versions or simulation environments, this study deploys and benchmarks multiple YOLOv8 variants (YOLOv8n to YOLOv8x) on real-world, drone-compatible embedded platforms, such as the NVIDIA Jetson AGX Xavier, and Xavier NX. We propose an optimized object detection pipeline integrated with autonomous navigation algorithms, which have been tested in various environments and weather conditions. YOLOv8s demonstrated the best trade-off between detection accuracy (mAP@0.5 = 89.3%) and real-time inference speed (45.2 FPS) on Jetson AGX Xavier. The system consistently achieved over 90% successful autonomous flight performance in urban, rural, and mixed scenarios. These results confirm the feasibility of YOLOv8s for reliable real-time object detection in autonomous UAVs, paving the way for robust deployment in real-world applications.

The Future of Coastal Resilience: Harnessing Satellite Technology

Kazi Kutubuddin Sayyad Liyakat — 2025-08-05

As coastal towns continue to face the numerous threats posed by climate change, satellite sensing is becoming a more crucial tool for risk assessment and management. Environmental scientists, policymakers, and urban planners may be able to make better judgments by using this technology, which could ultimately result in safer and more resilient coastal cities. Significant progress has been made in lessening the effects of extreme weather, environmental degradation, and rising sea levels as a result of the integration of satellite data into urban planning and disaster preparedness initiatives. Data from satellites has been used to do this. Adaptability and resilience are not choices; they are requirements, and satellite sensing offers the means to effectively achieve both of these objectives. Sea level monitoring has undergone a revolution thanks to the use of satellite sensors. We can better understand how climate change affects sea levels globally because of the vital information this technology gives us. As technology continues to develop, satellite observations will become more and more important for supporting climate resilience planning and helping vulnerable individuals worldwide. A new era in environmental monitoring is being fostered by the combination of advanced sensor technology, teamwork, and easier access to data. For our world to survive in the future, each of these elements is necessary.

A Review of Propagation models for Wireless Communication

Nishi Singh — 2025-08-13

Wireless communication is a rapidly advancing field, playing a pivotal role in modern technologies such as mobile networks, the Internet of Things (IoT), and satellite communication systems. A fundamental aspect of wireless system design understands how radio signals propagate through diverse environments. Propagation models are critical in predicting signal behavior between transmitters and receivers, helping estimate path loss, signal strength, coverage area, and overall network performance. This review provides an in-depth analysis of various propagation models, classified into empirical, deterministic, and stochastic categories. The Free Space Path Loss (FSPL) model offers a basic yet essential approach for calculating signal attenuation over unobstructed, line-of-sight paths. For more realistic scenarios, the Two-Ray Ground Reflection model accounts for both direct and ground-reflected rays, making it suitable for open terrains. Widely adopted empirical models like the Hata, Okumura-Hata, and COST-231 Hata models are specifically designed for urban and suburban environments, incorporating parameters such as frequency, antenna height, and terrain characteristics. These models enable practical network planning, especially in cellular system design.