Volume 6, Issue 4

2025 : Volume 6 Issue 4

Title FPGA-Based Implementation of IEEE 754 Single Precision Floating Point Multiplier
Authors Author Name(s) Here
Affiliation Department / University Info Here
Abstract This paper discusses the FPGA-based implementation of IEEE 754 single precision floating point multiplier...
Download Download Paper
Title Context-Aware Intelligent Smart Data Recorder for Real-Time Vehicle Tracking and Accident Response
Authors Rangaswamy L, Bharath Kumar Naidu D, Meghanath P, Rani Samyuktha Y, Karthik B, Likhitha S
Affiliation Department of Electronics and Communication Engineering (ECE), Sri Venkateswara Institute of Technology, Anantapur, India
Abstract This paper presents a context-aware intelligent smart data recorder designed for real-time vehicle tracking and accident response. The system integrates MEMS sensors, force sensors, GPS modules, and a microcontroller to detect accidents using multi-sensor data fusion and machine learning techniques. Upon detection of a confirmed accident, emergency alerts containing precise location and severity details are transmitted instantly to responders. The system also records pre-event and post-event data for accident analysis and investigation. Experimental results demonstrate high detection accuracy, reduced latency, and improved emergency response efficiency.
Download Download Paper
Title Automatic Plant Watering System Using Arduino UNO and Soil Moisture Sensor
Authors Rangaswamy L, Lavanya G, Sreeja B, Siva Sai Reddy G, Ashok Kumar R, Anil T, Dr. Venkateswara Reddy Vennapusa
Affiliation Department of Electronics and Communication Engineering (ECE), Sri Venkateswara Institute of Technology, Anantapur, India
DOI 10.64264/ijisea–STL0908
Abstract The design and execution of an autonomous smart pesticide spraying system for effective agricultural field operations are presented in this study. The ESP32 microprocessor at the center of the suggested system uses a four-wheel drive mechanism driven by an L293D motor driver to coordinate the sprayer’s navigation, allowing for precise movement in every direction. The spraying process is controlled by a specialized PSI motor, which enables exact control over pesticide discharge in accordance with operational requirements. An ESP32-CAM module is added to offer real-time video streaming and remote supervision in order to enhance monitoring and navigation. Solar energy powers the entire system, guaranteeing sustainable and environmentally friendly operation in outdoor farming settings. The technology lowers environmental effect, improves application precision, and decreases pesticide abuse by combining autonomous navigation, controlled spraying, and real-time visual input. Experimental testing demonstrates the suggested sprayer’s viability for use in contemporary precision agriculture by confirming its dependability and efficacy.
Download Download Paper
Title Underground Cable Fault Detection Using IoT
Authors Rangaswamy L¹, Pravallika V², Jaswanth Yadav J³, Sivaiah G⁴, Manoj S⁵, Prathyusha K⁶, Dr. Venkateswara Reddy Vennapusa⁷
DOI 10.64264/ijisea – STL0908
Abstract To precisely detect and locate short-circuit faults in metropolitan power distribution networks, this research proposes an Internet of Things (IoT)-based underground cable fault detection system. Modern cities use a large number of underground cables; therefore, accurate and timely fault detection is essential for maintaining a reliable power supply and efficient maintenance.

The proposed system monitors voltage fluctuations caused by faults using an Arduino microcontroller and current-sensing circuitry. The microcontroller’s built-in Analog-to-Digital Converter (ADC) converts these signals into digital values, enabling the calculation of the fault distance from the base station in kilometers. Switches are used to simulate fault conditions, while relay circuits control system operations.

The type and location of the fault are displayed in real time on a 16×2 Liquid Crystal Display (LCD). In addition, fault data is transmitted to a cloud platform through an ESP8266 Wi-Fi module for remote monitoring and analysis. A buzzer is also included to immediately alert field personnel when a fault occurs.

This system provides a reliable, efficient, and cost-effective solution for detecting underground cable faults, reducing repair time, and improving service continuity.

Keywords: Arduino, ADC, ESP8266, LCD display, fault detection, Internet of Things (IoT), underground cable, short-circuit fault, cloud monitoring, power distribution.
Download Download Paper
Title An Adaptive Multi-Source Hybrid Deep Learning Framework with Multi-Scale CNN, BiLSTM, and Transformer-Based Attention for Uncertainty-Aware Stock Market Growth Prediction
Authors Ramakrishna Kosuri¹*, Yogesh²
Author Affiliations ¹ Research Scholar, CSE Department, NIILM University, Kaithal, Haryana
² Associate Professor, CSE Department, NIILM University, Kaithal, Haryana
DOI 10.64264/ijisea – STL1109
Abstract Stock market growth prediction remains a highly complex task due to nonlinear price behaviour, market volatility, macroeconomic influences, and sudden event-driven fluctuations.

To address these challenges, this research proposes an enhanced intelligent hybrid deep learning framework that extends the CNN–BiLSTM–Transformer architecture by integrating multi-source data fusion, adaptive attention mechanisms, and uncertainty-aware prediction modelling.

The proposed system incorporates historical price data and technical indicators (RSI, MACD, SMA, EMA, and Bollinger Bands), along with sentiment scores derived from financial news and social media, and selected macroeconomic indicators.

A multi-scale Convolutional Neural Network is used to extract short-, medium-, and long-term temporal patterns. A stacked Bidirectional LSTM layer captures deep sequential dependencies, while an adaptive multi-head Transformer encoder enhances contextual learning by prioritizing significant market events.

A probabilistic output layer using Monte Carlo Dropout estimates prediction confidence intervals, improving reliability under volatile conditions. The model is optimized using a hybrid loss function combining Mean Squared Error and directional accuracy loss.

Experimental results demonstrate improved forecasting accuracy, better R² values, and higher robustness compared to baseline models.

Keywords: Stock Market Prediction, Deep Learning, CNN, BiLSTM, Transformer, Multi-Source Data Fusion, Sentiment Analysis, Monte Carlo Dropout, Uncertainty Modeling.
Download Download Paper