← Back to Archives
VOLUME 13, ISSUE 7, JULY 2025
MACHINE LEARNING APPROACHES FOR ACCURATE HEART DISEASE CLASSIFICATION
SANDI SUNANDA, RAYALA SURESH BABU, A.YASHASWI
REAL TIME EMOTION RECOGNITION AND GENDER CLASSIFICATION
SANDI SUNANDA, Dr.N.SRIHARI RAO, N. Nikhil
SWEET DREAMS POWERED BY MACHINES: MACHINE LEARNING IN SLEEP DISORDER ANALYSIS
SANDI SUNANDA, RAYALA SURESH BABU, BANDARI CHAITANYA KRISHNA
SMART AGRICULTURAL MONITORING SYSTEM USING ESP8266, ANDROID APP AND POWER BI
SANDI SUNANDA, Dr.N.SRIHARI RAO, BANOTHU RAJESH
Automatic Medical Dispatcher with Dynamic Tele Monitoring System Using IoT in Rural Areas
Savithri G R, Neha D, Rakshitha Pawar, Shalini C, Kruthika Rani DS
A touch-sensing system for precise robotic arm control, using force-sensing resistors (FSR), VHDL and FPGAs
Dr Evangelos I. Dimitriadis, Leonidas Dimitriadis, Aristeidis Grigoriadis
HOME AUTOMATION USING RASBERRY PI
Sneha Karre, Komal Kudal, Amruta Sutar, Satyawa Dindure, Shweta Vhankamble, Prof. S. A. Malvekar
A Comprehensive Review on Solar-Powered Lawn Mowers: Design Evolution, Automation Integration, and Sustainability
Shrihari T M Bhat, Shrishail, Shashikumar V, Siddartha L
Development of a Machine Learning-Based Crop Recommendation System for Sustainable Agriculture
Aarya Dhaygude, Aryan Karve, Prof. Dipali Pawar
Solar Tree
Ms. Priyanka Shrishail Chivadshetti, Prof. J. A. Patil
Image Compression Using DPCM Quantization
Igal Dudchenko, Yotam Tsfira, Mr. Ozi Chester
Hardware Accelerators for FFT Optimization: A Review of Genetic Algorithm-Based Approaches
Maenas Kirubahar E, Kishan S, Aaditya Pandey, Mukund Woodi
Deepfake Detection System using ResNet50 (based lightweight for static images) – Deep Dect
Mrs. Vedhapriya P, Vyshali M, Yashoda N, V Lavanya, Kusuma J M
Comparative Performance Analysis of Orthogonal and Non-Orthogonal Multiple Access Schemes in 5G Systems
Subhana Ayesha Siddiqui, Dr. P. Sreesudha
AI and ML-Driven Smart Attendance System: Real-Time Face Recognition Using DNN and NVIDIA Jetson Nano
V Mohamed Yousuf Hasan, Majid Rashid Alshammakhi, Nama Nasser Alhinai, Bushra Said Alsheriyani, Nada Yaqoob Alyahmadi
A Comparative Study of Traditional PAPR Reduction Techniques in MIMO-OFDM Systems
C.Shruthi, Dr.Rajkumar L Biradar, Dr.G. Krishna Reddy
Abstract
MACHINE LEARNING APPROACHES FOR ACCURATE HEART DISEASE CLASSIFICATION
SANDI SUNANDA, RAYALA SURESH BABU, A.YASHASWI
DOI: 10.17148/IJIREEICE.2025.13701
Abstract: Heart disease (HD), including heart attacks, is a leading cause of death worldwide, making accurate determination of a patient's risk a significant challenge in medical data analysis. Early detection and continuous monitoring by physicians can significantly reduce mortality rates, but heart disease is not always easily detectable, and physicians cannot monitor patients around the clock. Machine learning (ML) offers a promising solution to enhance diagnostics through more accurate predictions based on data from healthcare sectors globally. This study aims to employ various feature selection methods to develop an effective ML technique for early-stage heart disease prediction. The feature selection process utilized three distinct methods: chi-square, analysis of variance (ANOVA), and mutual information (MI), leading to three selected feature groups designated as SF-1, SF-2, and SF-3. We then evaluated ten different ML classifiers, including Naive Bayes, support vector machine (SVM), voting, XGBoost, AdaBoost, bagging, decision tree (DT), K-nearest neighbor (KNN), random forest (RF), and logistic regression (LR), to identify the best approach and feature subset. The proposed prediction method was validated using a private dataset, a publicly available dataset, and multiple cross-validation techniques. To address the challenge of unbalanced data, the Synthetic Minority Oversampling Technique (SMOTE) was applied. Experimental results showed that the AdaBoost classifier achieved optimal performance with the combined datasets and the SF-2 feature subset, yielding rates of 96.84% for accuracy, 95.32% for sensitivity, 91.12% for specificity, 94.67% for precision, 92.36% for F1 score, and 98.50% for AUC. Additionally, an explainable artificial intelligence approach utilizing SHAP methodologies is being developed to provide insights into the system's prediction process. The proposed technique demonstrates significant promise for the healthcare sector, facilitating early-stage heart disease prediction with reduced costs and minimal time.
Abstract
REAL TIME EMOTION RECOGNITION AND GENDER CLASSIFICATION
SANDI SUNANDA, Dr.N.SRIHARI RAO, N. Nikhil
DOI: 10.17148/IJIREEICE.2025.13702
Abstract: The proposed prototype aims to classify the gender and the emotions of a person in real time or using the image of the person in live cam or hard copy of the photograph. The gender classification would be implemented by simple yet robust real time convolutional neural network. Whereas for the emotion recognition, instead of fully connected layers, this model boasts and consists of depth wise separable convolution. It not only reduces the number of parameters and computation utilized in convolution but further increasing the efficiency. It has proved to achieve success in image classification models in terms of both, i) in obtaining better models than previously possible for a given parameter count required to perform at a given level and ii) acquire state-of-the-art results. The output is provided in the form of classes and these are seven classes of emotion recognition (angry, fear, sad, happy, surprise, neutral, disgust) and two classes of gender classification (Male and Female). The present accuracy for gender classification is 95%, whereas the accuracy for facial emotion recognition is around 67%. Also, a large reduction of hyper parameters is the main goal to reduce the model size.
Abstract
SWEET DREAMS POWERED BY MACHINES: MACHINE LEARNING IN SLEEP DISORDER ANALYSIS
SANDI SUNANDA, RAYALA SURESH BABU, BANDARI CHAITANYA KRISHNA
DOI: 10.17148/IJIREEICE.2025.13703
Abstract: Sleep disorders, such as Insomnia, Sleep Apnea, and other conditions, significantly impact individuals' health and well-being. Accurate and efficient classification of these disorders can aid in early diagnosis and effective treatment, enhancing the quality of life for affected individuals. The existing systems predominantly rely on Artificial Neural Networks (ANN) for classification, which, while effective, can be computationally intensive and less interpretable. This study proposes a Random Forest-based approach for classifying sleep disorders, utilizing a dataset consisting of 400 samples with 13 relevant features. Random Forest model was selected for its robustness, interpretability, and superior ability to handle complex, non-linear relationships within the data. By employing this algorithm, the study aims to classify sleep disorders into three classes: Insomnia, None, and Sleep Apnea, demonstrating improved performance compared to traditional ANN-based systems. The evaluation of the Random Forest model is conducted using standard performance metrics, including accuracy, precision, recall, and F1-score, which show that the proposed approach outperforms existing models, offering enhanced accuracy and reliability in the classification of sleep disorders.
Abstract
SMART AGRICULTURAL MONITORING SYSTEM USING ESP8266, ANDROID APP AND POWER BI
SANDI SUNANDA, Dr.N.SRIHARI RAO, BANOTHU RAJESH
DOI: 10.17148/IJIREEICE.2025.13704
Abstract: This project introduces a Smart Agricultural Monitoring System that integrates IoT, mobile technology, and data analytics to enhance irrigation efficiency and environmental monitoring in agriculture. Utilizing the ESP8266 NodeMCU microcontroller, the system gathers real-time data from a DHT22 sensor for temperature and humidity and a soil moisture sensor. Based on the soil moisture levels, it automatically controls a water pump via a relay module to maintain optimal soil conditions. The collected data is transmitted to the ThingSpeak cloud platform and visualized through a custom Android application, enabling farmers to monitor field conditions remotely. Additionally, the system connects to Power BI for advanced data analysis and visualization, offering insights into environmental trends and supporting precision farming. This cost-effective and scalable solution aims to reduce water wastage, minimize manual labor, and promote sustainable farming practices through smart automation and real-time decision-making.
Keywords: soil moisture, temperature, humidity, and light intensity.
Keywords: soil moisture, temperature, humidity, and light intensity.
Abstract
Automatic Medical Dispatcher with Dynamic Tele Monitoring System Using IoT in Rural Areas
Savithri G R, Neha D, Rakshitha Pawar, Shalini C, Kruthika Rani DS
DOI: 10.17148/IJIREEICE.2025.13705
Abstract: The AI-Powered Medicine Vending Machine is a smart healthcare solution for dispensing over-the-counter medicines. It integrates AI, cloud platforms, and payment gateways to improve access to medication in places like hospitals, clinics, and public areas. Users interact with an AI chatbot to input symptoms and receive suitable medicine suggestions. Once selected, they can make payments through an integrated gateway. The machine uses a servo motor- driven spiral mechanism for accurate dispensing and displays updates on an LED screen. A cloud-based platform like Firebase monitors inventory, automates reports, and provides real-time analytics. It ensures synchronization between the vending machine and the cloud. The system supports multi-user authentication for admins, operators, and customers, enhancing security and role-based access. In conclusion, the system addresses key challenges like accessibility, inventory management, and secure transactions, while offering AI-based medicine recommendations and a streamlined, user- friendly interface for better healthcare delivery.
Keyword: AI Powered, medication, cloud-based platform, healthcare delivery
Keyword: AI Powered, medication, cloud-based platform, healthcare delivery
Abstract
A touch-sensing system for precise robotic arm control, using force-sensing resistors (FSR), VHDL and FPGAs
Dr Evangelos I. Dimitriadis, Leonidas Dimitriadis, Aristeidis Grigoriadis
DOI: 10.17148/IJIREEICE.2025.13706
Abstract: A touch-sensing system, based on force-sensing resistances, FPGAs and VHDL, is presented here, capable of providing precise control of a robotic arm model, thus making it move up or down depending on which of two sen- sors is pressed. The robotic arm is connected to a step motor able to rotate clockwise or counterclockwise, in order to provide lowering or rising of the arm, respectively. In case that both sensors are stopped being pressed, robotic arm remains to its last obtained position. Both sensors use voltage divider circuit and their analog voltage values act as input to FPGA’s ADC converter and both values are presented to seven-segment displays. Three different rotation speed scales are used for both sensors, depending on the exerted pressure and respective LED lights up to present the scale. Additionally, buzzer alarm and half of FPGA’s board LEDs are activated if one of the sensors is pressed over a critical value which is dangerous for damaging sensors. Another control implemented in our system, makes the other half of FPGA’s board LEDs light up, when a critical time value of continuous step motor operation is exceeded, in order to protect motor from overheating. The system uses DE10-Lite FPGA board with two FSR 402 sensors connected to it. Our system can work with a variety of force sensors and critical input voltage limits can be set by the programmer de- pending on the application that the system is implemented.
Keywords: Force-sensing resistances, Step motor, FPGA, VHDL, Buzzer, LEDs, robotic arm.
Keywords: Force-sensing resistances, Step motor, FPGA, VHDL, Buzzer, LEDs, robotic arm.
Abstract
HOME AUTOMATION USING RASBERRY PI
Sneha Karre, Komal Kudal, Amruta Sutar, Satyawa Dindure, Shweta Vhankamble, Prof. S. A. Malvekar
DOI: 10.17148/IJIREEICE.2025.13707
Abstract: A home automation system using a Raspberry Pi enables users to control and monitor various devices within their homes remotely via the internet. This system typically involves integrating sensors, actuators, and a user interface (often web-based) with the Raspberry Pi, which acts as the central control unit. Users can manage lighting, appliances, and other systems from anywhere, enhancing convenience, comfort, and potentially energy efficiency. This paper proposes the design of Inter of Things (IoT) based home automation system using Raspberry pi. This project presents the design and implementation of an IoT-based home automation system using the Raspberry Pi as the central control unit. The system enables users to monitor and control household appliances such as lights, fans, and security systems remotely through a web or mobile interface, ensuring convenience, energy efficiency, and enhanced security. Using the Raspberry Pi's built-in Wi-Fi and GPIO capabilities, various sensors (e.g., temperature, motion, light) and actuators are integrated to create a smart environment. Communication protocols like MQTT or HTTP are employed to facilitate real-time data exchange between the devices and the user interface. The system supports both manual and automated control modes, making it adaptable to user preferences. This IoT-based solution offers a scalable, cost-effective, and user-friendly alternative to traditional automation systems, and it demonstrates the potential of Raspberry Pi in developing smart home technologies. User required to use different mobile devices like smart phones, Laptops, Tablets to operate the home appliances with the help of UI created on web page. Home automation plays an important role in establishing a smart home and is an ever exciting field that have grown largely over the last few years. New innovations has made the homes progressively appropriate, efficient and much increasingly secured. The low-cost devices used for joining the various electronic equipments and different sensors over an internet connection are supported and connected with the help of the Raspberry pi. Creating a smart home which can be effectively controlled and observed by the Raspberry pi by the means of Internet of Things is the principle target of present work.
Keywords: Home Automation, Internet of Things, Raspberry pi, Server, Mobile Devices
Keywords: Home Automation, Internet of Things, Raspberry pi, Server, Mobile Devices
Abstract
A Comprehensive Review on Solar-Powered Lawn Mowers: Design Evolution, Automation Integration, and Sustainability
Shrihari T M Bhat, Shrishail, Shashikumar V, Siddartha L
DOI: 10.17148/IJIREEICE.2025.13708
Abstract: The solar-powered automatic lawn mower is designed to provide an eco-friendly alternative to traditional gasoline mowers, addressing environmental pollution and operational inefficiencies. It harnesses solar energy through a 100W photovoltaic panel to charge a 12V, 40Ah battery, which powers a DC motor connected to cutting blades. The mower operates in both manual and autonomous modes, the latter controlled by Arduino, Bluetooth, and ultrasonic sensors for obstacle detection. This dual-mode operation ensures user flexibility and improved safety. Fabricated using mild steel, the mower’s frame and blade are durable and cost- effective, with stress analysis confirming mechanical stability. The system offers quiet operation and zero emissions, making it suitable for gardens, parks, and institutions. Field testing showed an efficiency of 85%, with a 5-hour charge supporting 50 minutes of grass-cutting. Compared to conventional mowers, it requires less maintenance, emits no carbon gases, and reduces noise significantly. The project integrates prior research on blade design, sensor integration, and power management to optimize performance. Overall, it presents a sustainable, low-cost solution for modern lawn maintenance with scope for AI and IoT enhancements.
Keywords: Solar power,Ultrasonic sensor, AT mega 328P. Cutter, Solar Panel, DC Motor, lead acid battery
Keywords: Solar power,Ultrasonic sensor, AT mega 328P. Cutter, Solar Panel, DC Motor, lead acid battery
Abstract
Development of a Machine Learning-Based Crop Recommendation System for Sustainable Agriculture
Aarya Dhaygude, Aryan Karve, Prof. Dipali Pawar
DOI: 10.17148/IJIREEICE.2025.13709
Abstract: The agricultural sector has become data-driven in recent years to increase agricultural yields, improve resource use, and promote sustainable agriculture. This essay describes an overview of how a Crop Recommendation System using Machine Learning was developed to assist farmers in making the optimal crop choice in relation to climatic and soil conditions. The model draws upon an open-source database of soil nutrient concentrations (nitrogen, phosphorus, potassium), temperature, humidity, pH, and rain to predict the best crop for the input conditions. Preprocessing methods were used to clean and normalize the data so that the model could be trained to be reliable. Various machine learning algorithms, such as Decision Trees, Random Forests, Support Vector Machines, and K-nearest neighbors, were trained and cross-validated to identify the most appropriate model. With heavy training, testing, and tuning for hyperparameters, the top-performing model was a Random Forest classifier, with impressive performance in accuracy, precision, recall, and F1-score measures. The system suggests crops in real-time, enabling farmers to make decisions based on weather and soil type. This increases its efficiency and decreases the overuse of fertilizers and water, thereby saving the environment. The study illustrates the potential of machine learning in transforming traditional agriculture. In order to develop a holistic system, the future can witness additional variables such as market forces, pest presence, and satellite imagery. The method discussed here is essential to realizing artificial intelligence-based sustainable agriculture.
Keywords: Crop Recommendation System, Machine Learning, Random Forest Classifier, Precision Agriculture, Sustainable Farming
Keywords: Crop Recommendation System, Machine Learning, Random Forest Classifier, Precision Agriculture, Sustainable Farming
Abstract
Solar Tree
Ms. Priyanka Shrishail Chivadshetti, Prof. J. A. Patil
DOI: 10.17148/IJIREEICE.2025.13710
Abstract: Developing energy efficiencies solutions from sunlight to electricity is a crucial solution for the world’s energy shortage and reducing greenhouse gas emissions. However, the typical photovoltaic (PV) flat module has a poor sunlight energy collection capability without a solar tracking system. Despite its advantages, solar PV technology has difficulties with land demand, capturing effectiveness and public image, especially in metropolitan areas due to the lack of pleasant aesthetics. This article presents an overview about the recently modified solar tree technology that can address these challenges efficiently. The main technology configurations, operational aspects and types are deeply presented and discussed. Many innovations and technologies of solar trees are analyzed and several commercial prototypes are discussed. Moreover, the main challenges and limitations that restrict the technology commercialization also highlighted in comparison with the traditional PV systems along with some remarks for future incomers. Analyzed studies show that solar tree technology is a good energy conversion method as it need only 1% land compared with traditional PV systems to produce power as more as 10%. Besides, this technology could efficiently collect off-peak sunshine and reflect light, and thus, create greater solar fraction.
Keywords: PV treeSolar cellPV technologyPower generatingSolar energyPower per area
Keywords: PV treeSolar cellPV technologyPower generatingSolar energyPower per area
Abstract
Image Compression Using DPCM Quantization
Igal Dudchenko, Yotam Tsfira, Mr. Ozi Chester
DOI: 10.17148/IJIREEICE.2025.13711
Abstract: Image processing modifies pictures to improve, extract information, and change their structure (composition, image editing, image compression, etc.). Images can be processed by optical, photographic, and electronic means, but image processing using digital computers is the most common method due to its speed, flexibility, and precision. Compression involves reducing redundancy in the image data to optimize transmission and storage. Differential Pulse Code Modulation (DPCM) is a method that uses prediction and quantization techniques to efficiently compress images by removing unused bits. In this paper, we evaluate the results of image compression using 3-bit DPCM quantization, analysing the in this study quality through histograms, prediction mean square error, and distortion levels. The results demonstrate that DPCM with 3-bit quantization achieves a good balance between image reconstruction and file size reduction, providing a clear trade-off between compression ratio and image quality. This paper explores the effects of 3- bit quantization on image compression, focusing on prediction accuracy and distortion.
Keywords: Image Compression, Quantization, Prediction Error, Image Processing.
Keywords: Image Compression, Quantization, Prediction Error, Image Processing.
Abstract
Hardware Accelerators for FFT Optimization: A Review of Genetic Algorithm-Based Approaches
Maenas Kirubahar E, Kishan S, Aaditya Pandey, Mukund Woodi
DOI: 10.17148/IJIREEICE.2025.13712
Abstract: Fast Fourier Transform (FFT) algorithms are fundamental in a wide range of digital signal processing applications, including audio analysis, image processing, and telecommunications. However, optimizing FFT implementations for speed, efficiency, and hardware utilization remains a complex task due to the numerous tunable architectural parameters such as radix type, stage configuration, word length, and pipeline depth. This paper presents the design and implementation of a hardware accelerator optimized through a Genetic Algorithm (GA), which intelligently explores the design space to identify parameter sets that yield optimal performance. The GA operates by encoding FFT configuration parameters into chromosomes and evolving them based on a fitness function that considers throughput, latency, power efficiency, and resource usage. A hardware model of the FFT accelerator is developed and evaluated through simulations to measure performance improvements against traditional fixed-parameter designs. Results demonstrate that GA-optimized FFT configurations lead to notable gains in processing speed and computational efficiency, validating the effectiveness of evolutionary algorithms in hardware design optimization. This work showcases the synergy between artificial intelligence and hardware engineering for advanced digital signal processing systems.
Keywords: Genetic Algorithm (GA), Hardware Acceleration, Fast Fourier Transform (FFT), Parameter Optimization, Digital Signal Processing (DSP), High-Level Synthesis, Hardware Design Automation, Signal Processing Architecture
Keywords: Genetic Algorithm (GA), Hardware Acceleration, Fast Fourier Transform (FFT), Parameter Optimization, Digital Signal Processing (DSP), High-Level Synthesis, Hardware Design Automation, Signal Processing Architecture
Abstract
Deepfake Detection System using ResNet50 (based lightweight for static images) – Deep Dect
Mrs. Vedhapriya P, Vyshali M, Yashoda N, V Lavanya, Kusuma J M
DOI: 10.17148/IJIREEICE.2025.13713
Abstract: The rise of synthetic media, commonly known as deepfakes, has created serious threats to digital authenticity. This paper presents a simple yet effective deepfake detection system for static facial images. It uses transfer learning with the ResNet50 CNN to distinguish between real and manipulated images by detecting subtle differences in facial features. A user-friendly interface was created using Streamlit and deployed through Hugging Face Spaces, allowing for real-time classification. While the system achieves moderate accuracy (about 61%), it shows the potential for accessible and deployable AI tools in digital forensics.
Abstract
Comparative Performance Analysis of Orthogonal and Non-Orthogonal Multiple Access Schemes in 5G Systems
Subhana Ayesha Siddiqui, Dr. P. Sreesudha
DOI: 10.17148/IJIREEICE.2025.13714
Abstract: This paper presents a comparative performance analysis of Orthogonal Multiple Access (OMA) and Non- Orthogonal Multiple Access (NOMA) techniques for 5G wireless communication systems. The OMA performance is modelled using OFDM and MIMO-OFDM under single-user and multi-user scenarios, while NOMA is implemented as a two-user power-domain system with Successive Interference Cancellation (SIC). Bit Error Rate (BER) and system throughput (sum rate) are evaluated through MATLAB-based simulations against varying levels of transmit power under identical channel conditions. The results reveal that although OMA schemes provide simpler implementation and effective interference management, NOMA achieves significantly higher sum rate while maintaining acceptable BER performance, especially at higher transmit power levels. These findings underscore NOMA’s potential to meet the growing spectral efficiency and user connectivity requirements of future 5G networks.
Keywords: OMA, NOMA, OFDM, MIMO-OFDM, Bit Error Rate (BER), Sum Rate, Transmit Power, 5G, Spectral Efficiency, SIC (Successive Interference Cancellation)
Keywords: OMA, NOMA, OFDM, MIMO-OFDM, Bit Error Rate (BER), Sum Rate, Transmit Power, 5G, Spectral Efficiency, SIC (Successive Interference Cancellation)
Abstract
AI and ML-Driven Smart Attendance System: Real-Time Face Recognition Using DNN and NVIDIA Jetson Nano
V Mohamed Yousuf Hasan, Majid Rashid Alshammakhi, Nama Nasser Alhinai, Bushra Said Alsheriyani, Nada Yaqoob Alyahmadi
DOI: 10.17148/IJIREEICE.2025.13715
Abstract: Attendance management is an essential task in educational and organizational settings, often plagued by inefficiencies in traditional methods such as manual roll calls and sign-in sheets. This paper proposes an AI and ML- based Face Recognition Attendance System that leverages the NVIDIA Jetson Nano for real-time, automated attendance tracking. The system utilizes advanced facial recognition technology, employing the OpenCV Deep Neural Network (DNN) framework and Haar Cascade Classifier for accurate face detection and recognition. The NVIDIA Jetson Nano's GPU-accelerated capabilities ensure efficient processing, enabling the system to operate in real-time without reliance on external servers.
Key components of the system include capturing facial data through a webcam, detecting faces using Haar Cascade, and recognizing them through the OpenCV DNN framework. Attendance data is logged automatically and updated in real- time in an Excel spreadsheet, simplifying reporting and reducing administrative burdens. This fully automated system addresses challenges associated with traditional and biometric attendance methods, including time inefficiency, scalability limitations, and susceptibility to errors.
Experimental results demonstrate a recognition accuracy of 94% under standard conditions and 88% in low-light environments, with an average processing time of 1.5 seconds per recognition. These results highlight the system's reliability, scalability, and adaptability for diverse applications. The proposed solution aligns with the global trend toward digital transformation, showcasing the potential of AI and ML in addressing real-world challenges while maintaining data security and integrity.
Keywords: Face Recognition, Deep Learning, OpenCV DNN, NVIDIA Jetson Nano, Haar Cascade.
Key components of the system include capturing facial data through a webcam, detecting faces using Haar Cascade, and recognizing them through the OpenCV DNN framework. Attendance data is logged automatically and updated in real- time in an Excel spreadsheet, simplifying reporting and reducing administrative burdens. This fully automated system addresses challenges associated with traditional and biometric attendance methods, including time inefficiency, scalability limitations, and susceptibility to errors.
Experimental results demonstrate a recognition accuracy of 94% under standard conditions and 88% in low-light environments, with an average processing time of 1.5 seconds per recognition. These results highlight the system's reliability, scalability, and adaptability for diverse applications. The proposed solution aligns with the global trend toward digital transformation, showcasing the potential of AI and ML in addressing real-world challenges while maintaining data security and integrity.
Keywords: Face Recognition, Deep Learning, OpenCV DNN, NVIDIA Jetson Nano, Haar Cascade.
Abstract
A Comparative Study of Traditional PAPR Reduction Techniques in MIMO-OFDM Systems
C.Shruthi, Dr.Rajkumar L Biradar, Dr.G. Krishna Reddy
DOI: 10.17148/IJIREEICE.2025.13716
Abstract: MIMO-OFDM (Multiple-Input Multiple-Output Orthogonal Frequency Division Multiplexing) has emerged as a key technology in modern wireless communication systems due to its high data rates and robustness against multipath fading [1]. However, a significant drawback of OFDM is its high Peak-to-Average Power Ratio (PAPR), which leads to power inefficiency and nonlinear distortion in power amplifiers [3]. This paper focuses on traditional PAPR reduction techniques that have been widely used to address this issue, particularly in MIMO-OFDM systems.
The study explores three well-known methods, Partial Transmit Sequence (PTS), Selected Mapping (SLM), and Clipping and Filtering (C&F) [18]. PTS divides the OFDM block into sub-blocks and optimally rotates them to minimize PAPR without distortion, though it requires side information and increased computational complexity [2]. SLM generates multiple OFDM candidate signals using different phase sequences and selects the one with the lowest PAPR for transmission, balancing complexity and performance [11]. Clipping and Filtering, a straightforward nonlinear technique, reduces PAPR by clipping the peaks of the signal followed by filtering to suppress out-of-band radiation, though it may introduce in-band distortion and BER degradation.
A comparative analysis of these techniques is presented in terms of PAPR reduction performance, computational cost, and impact on bit error rate (BER) [20]. This work provides a foundational understanding of traditional PAPR reduction strategies for MIMO-OFDM systems and offers insights for researchers aiming to improve energy efficiency and signal integrity in future wireless networks.
Keywords: PAPR reduction, MIMO-OFDM, Peak-to-Average Power Ratio, power amplifier nonlinearity, spectral efficiency, Bit Error Rate (BER), Side Information, OFDM Signal Distortion, Partial Transmit Sequence (PTS), Selective Mapping (SLM), Clipping and Filtering (C&F).
The study explores three well-known methods, Partial Transmit Sequence (PTS), Selected Mapping (SLM), and Clipping and Filtering (C&F) [18]. PTS divides the OFDM block into sub-blocks and optimally rotates them to minimize PAPR without distortion, though it requires side information and increased computational complexity [2]. SLM generates multiple OFDM candidate signals using different phase sequences and selects the one with the lowest PAPR for transmission, balancing complexity and performance [11]. Clipping and Filtering, a straightforward nonlinear technique, reduces PAPR by clipping the peaks of the signal followed by filtering to suppress out-of-band radiation, though it may introduce in-band distortion and BER degradation.
A comparative analysis of these techniques is presented in terms of PAPR reduction performance, computational cost, and impact on bit error rate (BER) [20]. This work provides a foundational understanding of traditional PAPR reduction strategies for MIMO-OFDM systems and offers insights for researchers aiming to improve energy efficiency and signal integrity in future wireless networks.
Keywords: PAPR reduction, MIMO-OFDM, Peak-to-Average Power Ratio, power amplifier nonlinearity, spectral efficiency, Bit Error Rate (BER), Side Information, OFDM Signal Distortion, Partial Transmit Sequence (PTS), Selective Mapping (SLM), Clipping and Filtering (C&F).