High sensitivity terahertz metasurface biosensor integrating plasmonic coupling and machine learning for wearable diabetes monitoring K. Vijayakumar, S. Karthikeyan, Harishchander Anandaram, K.V. Soumya, Jonas Muheki Sensing and Bio Sensing Research, 2026 This study presents a novel terahertz (THz) metasurface biosensor for highly sensitive and non-invasive glucose detection. Comprehensive finite element method (FEM) simulations performed using COMSOL Multiphysics demonstrate exceptional sensing performance, achieving a maximum sensitivity of 2000 GHz/RIU over a refractive index range of 1.335–1.347, corresponding to physiological glucose concentration variations. The proposed sensor exhibits a high figure of merit (FOM) of 44.444 RIU −1 , a quality factor (Q) of 6.244 and stable operation within the 0.27–0.31 THz frequency band with a tunability of 50 GHz. The sensing mechanism is driven by strong plasmonic coupling and enhanced electromagnetic field confinement arising from the hybrid integration of graphene, silver and gold resonators. To further validate the robustness of the sensing response, a machine learning–based predictive model is employed, yielding high prediction accuracy with R 2 values ranging from 84% to 100% across incident angles from 0° to 80°. Compared to conventional invasive glucose monitoring techniques, the proposed label-free optical biosensor offers superior sensitivity, real-time response and improved patient comfort. These results highlight the strong potential of the proposed THz metasurface platform for future wearable glucose monitoring and point-of-care diabetes management systems.
Real-Time Detection of Heavy Vehicles on Hairpin Bends using YOLO Algorithm S. Karthikeyan, U. Nithish, S. Sanjay, T. Sibiraj, J. Vishnu Proceedings of 5th International Conference on Soft Computing for Security Applications Icscsa 2025, 2025 Hairpin curves along hilly roads are accident-prone areas with less visibility, particularly where heavy transport vehicles such as buses and trucks come from the other direction. The Ministry of Road Transport and Highways (MoRTH) reports that more than 11% of road accidents in India take place on sharp turns and ghat roads, and a major percentage is due to blind spot collisions. This article suggests a real-time crash prevention system through image processing based on the YOLO (You Only Look Once) object detection model. A camera placed strategically in front of the upward slope senses vehicles approaching, while an LCD display on the downward slope warns drivers if heavy vehicles like buses or trucks are approaching. As opposed to cars or motorcycles, these vehicles tend to encroach on the other lane when taking sharp turns, making them more susceptible to collision. Apart from automatic warning systems, there is an emergency switch positioned at the curve to allow passengers or onlookers to alert nearby control rooms in case of an accident. The system facilitates a proactive safety mechanism that couples AI-driven vision and human-activated emergency response, substantially eliminating the possibility and consequences of such accidents in high-risk areas.
Design and Fabrication of Fire Extinguishing Drone S. Karthikeyan, U. Nithish, S. Sanjay, T. Sibiraj, J. Vishnu SAE Technical Papers, 2025 <div class="section abstract"><div class="htmlview paragraph">Nowadays, there are many technologies emerging like firefighting robots, quadcopters, and drones which are capable of operating in hazardous disaster scenarios. In recent years, fire emergencies have become an increasingly serious problem, leading to hundreds of deaths, thousands of injuries, and the destruction of property worth millions of dollars. According to the National Crime Records Bureau (NCRB), India recorded approximately 1,218 fire incidents resulting in 1,694 deaths in 2020 alone. Globally, the World Health Organization (WHO) estimates that fires account for around 265,000 deaths each year, with the majority occurring in low- and middle-income countries.</div><div class="htmlview paragraph">The existing fire-extinguishing systems are often inefficient and lack proper testing, causing significant delays in firefighting efforts. These delays become even more critical in situations involving high-rise buildings or bushfires, where reaching the affected areas is particularly challenging. The leading causes of fires include electrical faults, open flames, and negligence, with a notable percentage stemming from cooking-related incidents. This research introduces a prototype of a quadcopter designed to extinguish fires, which is equipped with a mechanism for releasing balls. The prototype utilizes experimental items to mimic fire-suppressing balls, establishing a basis for potential applications in firefighting technology.</div></div>
FPGA Based Integrated Control of Brushless DC Motor for Renewable Energy Storage System Karthikeyan S., Lakshmi K. Cybernetics and Systems, 2024 To reduce air pollution and global warming, renewable energy technologies may generate power. Wind, solar PV, and fuel cell energy are the primary sources. Solar PV system-powered brushless direct current motor (BLDC) drives are used in the automobile industry due to their importance. In this study, Sheppard–Taylor (S-T) converter and Pulse Width Modulated (PWM) Inverter-fed BLDC provide steady voltage across the BLDC motor drive independent of solar PV system power output. When renewable energy is scarce, the proposed battery-supercapacitor hybrid energy storage system (BS-HESS) provides electricity. S-T converters may be used for load matching and power processing to create energy-efficient systems and stabilize PV panel output voltage. The variable step size open circuit voltage-Maximum Power Point Tracking (VSSOCV-MPPT) technique in S-T converter switching pulses extracts maximum power from the solar PV system. This study considers the PVSWPS control function as a Port-Controlled Hamiltonian (PCH) system to continue rural growth and reduce the greatest demand and load. MATLAB/Simulink software simulates the system’s performance, and FPGA controllers validate the controller’s real-time performance.
Modeling and Control of Solar Powered BLDC Motor System with PID and Sheppard Taylor Converter S. Karthikeyan, K. Harish, S. Dhinesh Balaji, R. Harikrishnan, C. Anush Proceedings 2024 4th International Conference on Pervasive Computing and Social Networking Icpcsn 2024, 2024 The proposed system presents a comprehensive approach for designing, simulating, and implementing a solar-powered air-purification system. This system incorporates a Sheppard-Taylor converter for voltage regulation, a Brushless DC (BLDC) motor, and an MPPT (Maximum Power Point Tracking) algorithm. For analysis and simulation tasks involve MATLAB application, ensuring the system’s performance under diverse conditions. Sheppard-Taylor converter optimizes voltage regulation, while the BLDC motor drives the air purification unit efficiently. The MPPT algorithm maximizes power extraction from the solar panels. This integrated solution offers a sustainable approach to air purification, with MATLAB simulations validating its effectiveness for implementation. Suitable for off-grid and environmentally sensitive areas. Additionally, this system contributes to cleaner air and energy sustainability.
Implementation of IoT Based Hybrid Uninterrupted Power Supply S. Karthikeyan, S. Gobhinath, L. Sabarish, S. Sabeer, K. Vijay, V. K. Vijay Vignesh 10th International Conference on Advanced Computing and Communication Systems Icaccs 2024, 2024 The main aim of this research paper is to develop hardware along with the software system that will provide the data along with the control monitoring used in the Uninterrupted power supply. Our proposed system aims to improve the existing system by introducing the concept of hybrid integration technology based on IoT. This research work mainly focuses on solving the issues of energy availability stemming from the increasing growth of organizations and the population of individuals. The extraction of energy from the sun results in solar energy generation, energy from the flow of air in the form of wind. Nowadays, along with the power generation techniques, adding energy from different energy sources will be very difficult due to more challenges when we implement the integration process. Here our proposed system aims to develop an IoT-based Small Hybrid Uninterrupted Power Supply (HUPS). It is a small-scale power system that manages and optimizes energy output and consumption by combining energy sources and storage units. Appropriate real-time monitoring of HUPS is critical in providing reliable data that allows the users to get to know about the efficiency of the proposed system and energy loss data along with abnormalities. To improve the efficiency of solar cells, this proposal proposes using energy management of thermal dissipation. Along with the energy supplied by the Peltier Plate and Wind Turbine, the sun's radiated energy is stored as an alternate energy source. Peltier plates are designed to catch heat dissipation from the environment and transform it into electrical energy. Bismuth Telluride Semiconductor is the material used in Peltier plates. This hybrid technology improves the solar power system's efficiency by 30% when compared to the conventional UPS, the power utilization from the grid will be also reduced by 20% thereby allowing us to utilize green energy in both the presence and absence of sunlight. IoT is utilized to send data into the cloud, which makes data storage easier and allows for access at any moment during the energy generation process.
Design of Octamagnetocore Ceiling Fan S Karthikeyan, Asfak N, V P Jayasegar, Farhan Nazeer, R Balakumar Proceedings of the 9th International Conference on Electrical Energy Systems Icees 2023, 2023 The aim of this paper is to provide a sustainable Ceiling fan design that can generate a meaningful and significant output voltage from its rotatory motion.The paper explains the need for such a design all the while exploring the scopes of further improvement and implementation in real time applications. Apart from designing a basic generator set-up which is being mounted on the head of the fan, there are rectifiers, converters, inverters, all integrated into a circuit board designed to use the power generated from the fan efficiently. The circuit board is at zero-voltage state until the ceiling fan is switched ON. It also comprises a microcontroller for control schemes. In the present day scenario,electricity consumption is at an all time high, so energy saving is important in our modern day situation. This invention relates to electrically powered ceiling fans from which power is generated and is supplied to other light loads. Such a mechanism allows for conservation of energy and its sustainable consumption.
Design and Implementation of IoT Based Accident Detection and Prevention System S. Karthikeyan, J. Kiruthik, S. Madumitha, R. Manikandan, V. Prakash Raj 7th International Conference on Trends in Electronics and Informatics Icoei 2023 Proceedings, 2023 In the modern world, an increase in the usage of automobiles for commercial purposes has also increased the number of accidents occurring in commercial vehicles, which leads to the loss of life of the people involved in the accident. To minimize the death rates involved in an accident, the people who are met with the accident must claim medical assistance at the correct time. This study is concerned with two set-ups. One set-up is associated with the vehicle, where the use of a MEMS or gyroscopic sensor, a vibration sensor, and a gas sensor integrated with Arduino helps to detect the accident. Here, the location is detected by the GPS module and updated in the cloud by using the ESP8266 Wi-Fi module. If any accident is detected, the RF transmitter circuit sends the signal to the RF receiver. The other configuration is related to the Ambulance which consists of an RF receiver circuit integrated with the NodeMCU microcontroller. When the signal reaches the receiver, NodeMCU retrieves the information from the cloud and displays it on the LCD. Integration of a tracking system with a Radio frequency transmitter and receiver helps build IoT services using embedded systems. The system of providing medical assistance to the people involved in the accident would help us reduce the death rates.
Detection And Analysis of Earlier Cognitive Disorder Using Ai Based Transcranial Magnetic Stimulator S. Gobhinath, S. Karhtikeyan, K. Janani, U. Arunkumar 8th International Conference on Advanced Computing and Communication Systems Icaccs 2022, 2022 For the past decade, the suggested Monotonous Transcranial Magnetic Stimulation system is employed as ainvasive treatment of a variety of nervous, psychotic illnesses such as depression, Parkinson's disease and schizophrenia. Early detection of cognitive illnesses such as Alzheimer's and dementias helps patients keep their memories and regain the information they need with better quality care by detecting cognitive impairment as soon as feasible. Early detection and management of Alzheimer's and dementias, like other chronic disorders, helps people recognise the severity of the disease and plan for future care. The rTMS device works by applying a changing magnetic field that is utilised to modify the firing pulse of neurons and so trigger brain communication. Within the range of focus across the magnetic coil area, the suggested stimulator can either stimulate or inhibit cortical areas. The high frequency rTMS is normally thought more excitatory, whereas the low frequency rTMS is thought to be inhibitory, though it can vary among people depending on the trial results. Because Alzheimer's disease tangles and associated neuronal death usually start in memory-related areas like the progress to, finding Alzheimer pathology in these aphasic patients is difficult. Furthermore, Alzheimer's disease is usually symmetrical. How can be caused by a primarily limbic and symmetric pathology Initial research exploring the idea that Alzheimer's disease manifests itself in an atypical manner in primary progressive aphasia produced inconclusive results. In every case, Alzheimer's disease was the primary diagnosis. The purpose was to see if the two phenotypes of Alzheimer's disease had clinically concordant, and thus separate, distributions. In primary progressive aphasia, stereological counts of tangles and plaques demonstrated increased leftward asymmetry for tangles, but not in amnestic Alzheimer's disease (P 0.06). Because Alzheimer's disease tangles and associated cell death usually start in memory-related areas like the finding Alzheimer pathology in these aphasic patients is difficult.. In every case, Alzheimer's disease was the primary diagnosis. The purpose was to see if the two phenotypes of Alzheimer's disease had clinically concordant, and thus separate, disseminations.
Dynamic Objects Detection and Tracking from Videos for Surveillance Applications S. Gobhinath, S. Sophia, S. Karthikeyan, K. Janani 8th International Conference on Advanced Computing and Communication Systems Icaccs 2022, 2022 Artificial intelligence is a recent topic in computer science study right now. A machine must be aware of its surroundings in order to function intelligently. Humans require visual information to function. The field of research known as “Computer Vision” focuses on automatically interpreting images and image sequences. In other words computer vision is the scientific discipline that aims for extracting information from visual data. In a video sequence, moving object tracking refers to constantly recognising an object's position and orientation despite camera or object movement. The task of following one or more things around a scene from when they first appear to when they leave is known as object tracking. An item can be anything of interest in the scene that can be detected, and it is determined by the scenario's requirements. Object tracking systems are generally used in surveillance applications where moving individuals and/or vehicles must be monitored. Moving object detection, object tracking, and event recognition are the three basic components of any surveillance system. The performance of such system considerably depends on its first step that is detection of the foreground objects. These are the objects which are not background objects. These foreground objects have a important role in subsequent processes, such as event detection. To work properly, every object detector requires an object model, which specifies the look of objects, or a set of rules to determine whether a specific patch (region) from an input image contains an object of interest.
Solar Based Fast Tag Charger for Electrical Vehicle S. Karthikeyan, H. Bragruthshibu, R. Logesh, K. Srinivasan, S. Tarjanbabu 2021 7th International Conference on Advanced Computing and Communication Systems Icaccs 2021, 2021
