In India, a large population depends on local water bodies such as ponds for daily use, yet the quality of these small, community-level sources is rarely monitored or made publicly accessible, even though contamination from waste disposal, human and animal faecal matter, plastic pollution, and poor sanitation makes timely monitoring essential for reducing public-health and environmental risks. This paper presents a smart environmental monitoring and management system that combines laboratory-based water quality testing with a web-based data-analysis platform. Twenty water samples were collected from ponds and public sources in Berhampur, Odisha, and analysed for pH, dissolved oxygen, chlorin... More >

Wiki-grounded LLM agents enforce output constraints through safety rules and scope boundaries derived from structured knowledge bases, yet selecting an appropriate constraint scorer threshold $\theta$ remains challenging. Since outputs are blocked when $s(o) \geq \theta$, a low threshold may over-restrict safe responses, whereas a high threshold may admit dangerous outputs. Existing approaches largely rely on manual, domain-agnostic tuning without systematically incorporating wiki content. This paper proposes WikiConstraintCalibration, a framework that extracts eight wiki-derived features spanning hazard indicators (dangerous assertion density, immediate-action fraction, domain risk prior, a... More >

India's stationary air-quality monitoring network remains sparse, particularly in underserved urban and peri-urban areas. EcoPulse-India addresses this limitation by using public transit buses as mobile sensing platforms within an integrated artificial intelligence of things (AIoT) framework. The framework integrates velocity-compensated aerodynamic correction (VCAC), a physics-informed convolutional graph neural network (PI-ConvGNN), multi-agent reinforcement learning (MARL)-based coverage optimisation, and privacy-respecting inference and masking architecture (PRIMA). VCAC compensates for motion-related measurement bias using vehicle kinematic and environmental information, while the quant... More >

This study introduces GenAI-FDT (GenFedTwin Resilience), a decentralized architecture designed to address the growing vulnerability of IoT-enabled smart infrastructures to climate change, particularly in the context of data silos, strict privacy regulations, and the scarcity of ground-truth data for rare catastrophic events such as floods in Maharashtra. Unlike traditional centralized digital twins, which struggle with data heterogeneity and excessive communication costs, the proposed framework integrates quantized diffusion-based Generative AI, semi-supervised federated learning (FedProx), and lightweight LSTM-CNN digital twins deployed on resource-constrained edge devices like Raspberry Pi... More >

Taking notes during meetings sounds easy, but in reality, people often miss key points—especially when multiple people are talking. Managing follow-ups in separate apps only adds to the hassle. MinuteMaster was built to simplify this. It's a web app that brings transcription, speaker identification, summarization, and scheduling into one place. It uses Whisper for multilingual speech-to-text, pyannote.audio to identify who's speaking, and BART to turn long transcripts into clear, short summaries. A built-in calendar helps users manage meetings without switching apps. We tested it on 30 recordings across English, Hindi, Telugu, Tamil, and mixed languages. Transcription accuracy ranged from... More >

Artificial Intelligence of Things (AIoT) is transforming linear waste management into intelligent, data-driven solutions that support circular economy (CE) principles. This narrative review synthesizes peer-reviewed studies from 2020–2025 on AIoT applications across intelligent waste sorting, data analytics for resource recovery, and sustainable infrastructure development. The review examines IoT-enabled smart bins, computer-vision robotic sorting, machine learning classifiers (VGG-16/19: 97.11–99.7% accuracy, ResNet: 91.5–98.16%), predictive analytics, and graph-based route optimization. Reported improvements include up to 50% reduction in overflow events, 15.5–30% fuel savings, and... More >

There is an increase in heart diseases at a very high rate in today's world, and therefore early detection becomes a necessity in the treatment of such cases. In this paper, we present a digital stethoscope with AI and ML that records heart sounds automatically. Phonocardiogram (PCG) signals are recorded by the system, and the recorded signals are then processed using filtering, normalization, and peak detection to obtain features including heart sounds S1 and S2. The extracted features are fed into the classification algorithm, which in this case is the Random Forest Classifier, and they are classified as either abnormal or normal heart conditions. IoT is implemented in the system to ensure... More >

The integration of artificial intelligence and the Internet of Things (AIoT) enables advanced edge computing but creates significant energy challenges for resource-constrained devices. To address these challenges, this paper introduces a novel comparative taxonomy and a systematic gap analysis matrix that directly maps hardware-aware TinyML paradigms to network-layer scheduling in sustainable AIoT systems. Synthesizing recent empirical studies (2023–2026) on power consumption models for IoT edge devices, machine learning techniques for energy optimization, and sustainable deployment strategies, we demonstrate that additive and regression-based models achieve low prediction error (MAPE 4–... More >