Gracia Nirmala Rani Duraisamy
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Gracia Nirmala Rani Duraisamy has a total of 39 co-authors. Here are the top co-authors by number of shared publications:
- Rajaram Sivasubramanian
Department of ECE, Thiagarajar College of Engineering, Madurai, India
8 shared publications - A. Andrew Roobert
Department of Electronics and Communication Engineering Francis Xavier Engineering College Tirunelveli Tamil Nadu India
7 shared publications - G. Gifta
Department of ECE, Thiagarajar College of Engineering, Madurai, India
5 shared publications - Arun Samuel Thankamony Sarasam
Department of ECE, National Engineering College, Kovilpatti, India
2 shared publications G. Gifta
Thiagarajar Coll Engn, Madurai, Tamil Nadu, India
2 shared publications- Nirmala Natarajan
Department of Electronics and Communication Engineering, Thiagarajar College of Engineering, Madurai, India
2 shared publications
Latest Publications
- 23 October 2026
- Published by Springer Nature in Sādhanā
- Vol. 51 (4) , 411
An energy-efficient deep learning framework for lung cancer detection and segmentation leveraging FPGA acceleration
- 23 October 2026
- Published by Springer Nature in Sādhanā
- Vol. 51 (4) , 411
An energy-efficient deep learning framework for lung cancer detection and segmentation leveraging FPGA acceleration
- 14 July 2026
- Published by Taylor & Francis in Assistive Technology
- p. 1-17
Toward equity: TSLYOLOv8s - transcribe for Tamil Sign Language recognition.
Sign language relies on hand gestures, facial expressions, and body postures. Vision-based systems are crucial for recognizing these gestures to convert sign language into text or speech. Tamil Sign Language Recognition (TSLR) remains underdeveloped, with limited studies on Tamil vowels and consonants. Many Tamil words can be represented with one hand, but no benchmark dataset exists for Tamil Sign Language. To address this, we constructed a dataset with 56 classes, including vowels, consonants, and Tamil Single-Hand Alphabets (TSHA), incorporating variations in scale, background, pose, resolution, and lighting. We propose Tamil Sign Language - You Only Look Once-small (TSL-YOLOv8s), an enhanced YOLOv8s model integrating Space to Depth - Convolution, Batch Normalization, and Sigmoid Activation (SPD-CBS) with Dynamic Head (DyHead) to improve recognition accuracy without increasing computational cost. The proposed model achieves 94% accuracy, outperforming state-of-the-art methods and demonstrating its effectiveness in real-world applications.
- 14 July 2026
- Published by Taylor & Francis in Assistive Technology
- p. 1-17
Toward equity: TSLYOLOv8s - transcribe for Tamil Sign Language recognition.
Sign language relies on hand gestures, facial expressions, and body postures. Vision-based systems are crucial for recognizing these gestures to convert sign language into text or speech. Tamil Sign Language Recognition (TSLR) remains underdeveloped, with limited studies on Tamil vowels and consonants. Many Tamil words can be represented with one hand, but no benchmark dataset exists for Tamil Sign Language. To address this, we constructed a dataset with 56 classes, including vowels, consonants, and Tamil Single-Hand Alphabets (TSHA), incorporating variations in scale, background, pose, resolution, and lighting. We propose Tamil Sign Language - You Only Look Once-small (TSL-YOLOv8s), an enhanced YOLOv8s model integrating Space to Depth - Convolution, Batch Normalization, and Sigmoid Activation (SPD-CBS) with Dynamic Head (DyHead) to improve recognition accuracy without increasing computational cost. The proposed model achieves 94% accuracy, outperforming state-of-the-art methods and demonstrating its effectiveness in real-world applications.
- 8 July 2025
- Published by Springer Nature in Microfluidics and Nanofluidics
- Vol. 29 (8) , 1-19
Optimization techniques in digital microfluidic biochips: a survey of sample preparation algorithmic solutions and challenges
- 8 July 2025
- Published by Springer Nature in Microfluidics and Nanofluidics
- Vol. 29 (8) , 1-19
Optimization techniques in digital microfluidic biochips: a survey of sample preparation algorithmic solutions and challenges
- 28 June 2024
- Published by IGI Global Scientific Publishing
- p. 344-364
Implementing Secure Machine Learning
This chapter investigates integrating a decision tree algorithm to boost smart grid stability within the 5G and fiber optics security framework. It stresses robust cyber defense for critical infrastructure, using machine learning, especially decision trees, to analyze smart grid stability data. Steps involve thorough data preprocessing, algorithm justification, model training, and evaluation. Visual aids aid interpretation. The chapter ends with future research suggestions, underlining machine learning's evolving role in enhancing security.
- 28 June 2024
- Published by IGI Global Scientific Publishing
- p. 344-364
Implementing Secure Machine Learning
This chapter investigates integrating a decision tree algorithm to boost smart grid stability within the 5G and fiber optics security framework. It stresses robust cyber defense for critical infrastructure, using machine learning, especially decision trees, to analyze smart grid stability data. Steps involve thorough data preprocessing, algorithm justification, model training, and evaluation. Visual aids aid interpretation. The chapter ends with future research suggestions, underlining machine learning's evolving role in enhancing security.
- Open Access
- 23 September 2023
- Published by MDPI AG in Micromachines
- Vol. 14 (10) , 1816
A 0.8 V, 14.76 nVrms, Multiplexer-Based AFE for Wearable Devices Using 45 nm CMOS Techniques
Wearable medical devices (WMDs) that continuously monitor health conditions enable people to stay healthy in everyday situations. A wristband is a monitoring format that can measure bioelectric signals. The main part of a wearable device is its analog front end (AFE). Wearables have issues such as low reliability, high power consumption, and large size. A conventional AFE device uses more analog-to-digital converters, amplifiers, and filters for individual electrodes. Our proposed MUX-based AFE design requires fewer components than a conventional AFE device, reducing power consumption and area. It includes a single-ended differential feedback operational transconductance amplifier (OTA) and n-pass MUX-based AFE circuits which are related to the emergence of low power, low area, and low cost AFE-integrated chips that are required for wearable biomedical applications. The proposed 6T n-pass multiplexer measures a gain of −68 dB across a frequency range of 100 kHz with a 136.5 nW power consumption and a delay of 0.07 ns. The design layout area is approximately 9.8 µm2 and uses 45 nm complementary metal oxide semiconductor (CMOS) technology. Additionally, the proposed single-ended differential OTA has an obtained input referred noise of 0.014 µVrms, and a gain of −5.5 dB, while the design layout area is about 2 µm2 and was designed with the help of the Cadence Virtuoso layout design tool.
- 23 September 2023
- Published by MDPI AG in Micromachines
- Vol. 14 (10) , 1816
Open AccessA 0.8 V, 14.76 nVrms, Multiplexer-Based AFE for Wearable Devices Using 45 nm CMOS Techniques
Wearable medical devices (WMDs) that continuously monitor health conditions enable people to stay healthy in everyday situations. A wristband is a monitoring format that can measure bioelectric signals. The main part of a wearable device is its analog front end (AFE). Wearables have issues such as low reliability, high power consumption, and large size. A conventional AFE device uses more analog-to-digital converters, amplifiers, and filters for individual electrodes. Our proposed MUX-based AFE design requires fewer components than a conventional AFE device, reducing power consumption and area. It includes a single-ended differential feedback operational transconductance amplifier (OTA) and n-pass MUX-based AFE circuits which are related to the emergence of low power, low area, and low cost AFE-integrated chips that are required for wearable biomedical applications. The proposed 6T n-pass multiplexer measures a gain of −68 dB across a frequency range of 100 kHz with a 136.5 nW power consumption and a delay of 0.07 ns. The design layout area is approximately 9.8 µm2 and uses 45 nm complementary metal oxide semiconductor (CMOS) technology. Additionally, the proposed single-ended differential OTA has an obtained input referred noise of 0.014 µVrms, and a gain of −5.5 dB, while the design layout area is about 2 µm2 and was designed with the help of the Cadence Virtuoso layout design tool.
- 2 July 2023
- Published by Springer Nature
- p. 69-78
Develop a Quantum Based Time Scheduling Algorithm for Digital Microfluidic Biochips
- 2 July 2023
- Published by Springer Nature
- p. 69-78
Develop a Quantum Based Time Scheduling Algorithm for Digital Microfluidic Biochips
- 1 May 2023
- Published by The Electrochemical Society in ECS Journal of Solid State Science and Technology
- Vol. 12 (5) , 057012
Analytical Modelling and Simulation of Graphene Based Biosensor to Detect SARS-COV-2 from Aerosal Particles
The health sector is focusing on the wellness of the society, is advancing in the phases of diagnosis and treatment. Biosensors based devices are used to diagnose a variety of human diseases. Recently, there was a sudden hike in the human mortality rate by chronic diseases caused by mutants of SARS-COV-2, on global scale. It is important to detect these kinds of diseases on an early stage to reduce the risk of spreading. For the analysis of Covid-19 influenza, tests such as Rapid Antigen Test (RAT), True NAT, CBNAAT and the commonly done RPT PCR were utilised. This proposal describes a non-invasive, quick and practical method for sensing the at-risk or infected persons with SARS-COV-2, aiming at controlling the epidemic. The proposed method employs a breath sensing device consisting of a Graphene Field Effect Transistor biosensor which can identify disease-specific biomarkers from exhaled sniff, hence allowing speedy and precise detection. This test aids screening of large populations as it is simple and quick and emerges as a promising candidate for SARS-COV-2 tests due to a high sensitivity. This work justifies the accurate diagnosis of Severe Acute Respiratory Syndrome COV 2 from aerosol particles by GFET Biosensor.
- 1 May 2023
- Published by The Electrochemical Society in ECS Journal of Solid State Science and Technology
- Vol. 12 (5) , 057012
Analytical Modelling and Simulation of Graphene Based Biosensor to Detect SARS-COV-2 from Aerosal Particles
The health sector is focusing on the wellness of the society, is advancing in the phases of diagnosis and treatment. Biosensors based devices are used to diagnose a variety of human diseases. Recently, there was a sudden hike in the human mortality rate by chronic diseases caused by mutants of SARS-COV-2, on global scale. It is important to detect these kinds of diseases on an early stage to reduce the risk of spreading. For the analysis of Covid-19 influenza, tests such as Rapid Antigen Test (RAT), True NAT, CBNAAT and the commonly done RPT PCR were utilised. This proposal describes a non-invasive, quick and practical method for sensing the at-risk or infected persons with SARS-COV-2, aiming at controlling the epidemic. The proposed method employs a breath sensing device consisting of a Graphene Field Effect Transistor biosensor which can identify disease-specific biomarkers from exhaled sniff, hence allowing speedy and precise detection. This test aids screening of large populations as it is simple and quick and emerges as a promising candidate for SARS-COV-2 tests due to a high sensitivity. This work justifies the accurate diagnosis of Severe Acute Respiratory Syndrome COV 2 from aerosol particles by GFET Biosensor.
- 10 March 2022
- Published by Wiley in Transactions on Emerging Telecommunications Technologies
- Vol. 33 (7)
Design and analysis of 28 GHz CMOS low power LNA with 6.4 dB gain variability for 5G applications
A 28 GHz two stage low noise amplifier (LNA) is proposed with envelope detection technique for power reduction (21.62%) and tunable negative feedback capacitor for gain variation in 40 nm CMOS technology. The envelope detection circuit turn‐on the second half of the LNA by the RF signal input received at the first stage. The default gain is increased (31.53%) by the tunable negative feedback capacitor circuit of the LNA with the control voltage from 0 to 1 V. In addition, 6.22 GHz of bandwidth is achieved with the tunable gain from 20.3 dB to 26.7 dB. The first stage of the LNA is designed with the inductive source degeneration for the noise reduction, and the multiple‐gate topology is involved in the second stage to improve the linearity. The third‐order input intercept point and the noise figure of the LNA are −7 dBm and 2.86 dB, respectively. When the second stage is turned‐on and turned‐off the LNA consumes 7.4 mW and 5.8 mW of power, respectively, from the 1 V supply. The proposed LNA requires 0.19 mm2 of core area. The performance of the LNA under process corner variation and temperature variation are analyzed.
- 10 March 2022
- Published by Wiley in Transactions on Emerging Telecommunications Technologies
- Vol. 33 (7)
Design and analysis of 28 GHz CMOS low power LNA with 6.4 dB gain variability for 5G applications
A 28 GHz two stage low noise amplifier (LNA) is proposed with envelope detection technique for power reduction (21.62%) and tunable negative feedback capacitor for gain variation in 40 nm CMOS technology. The envelope detection circuit turn‐on the second half of the LNA by the RF signal input received at the first stage. The default gain is increased (31.53%) by the tunable negative feedback capacitor circuit of the LNA with the control voltage from 0 to 1 V. In addition, 6.22 GHz of bandwidth is achieved with the tunable gain from 20.3 dB to 26.7 dB. The first stage of the LNA is designed with the inductive source degeneration for the noise reduction, and the multiple‐gate topology is involved in the second stage to improve the linearity. The third‐order input intercept point and the noise figure of the LNA are −7 dBm and 2.86 dB, respectively. When the second stage is turned‐on and turned‐off the LNA consumes 7.4 mW and 5.8 mW of power, respectively, from the 1 V supply. The proposed LNA requires 0.19 mm2 of core area. The performance of the LNA under process corner variation and temperature variation are analyzed.
- 1 June 2021
- Published by Springer Nature in Journal of Electrical Engineering & Technology
- Vol. 16 (5) , 2793-2807
Design of Low Noise, Low Power, Bulk-Driven CMOS Based Operational Transconductance Amplifier for Biosensor Applications
- 1 June 2021
- Published by Springer Nature in Journal of Electrical Engineering & Technology
- Vol. 16 (5) , 2793-2807
Design of Low Noise, Low Power, Bulk-Driven CMOS Based Operational Transconductance Amplifier for Biosensor Applications
- 4 May 2021
- Published by Springer Nature in Wireless Personal Communications
- Vol. 120 (2) , 1229-1249
Design of 5-bit Flash ADC Using 180 nm Technology for Medical Applications
- 4 May 2021
- Published by Springer Nature in Wireless Personal Communications
- Vol. 120 (2) , 1229-1249
Design of 5-bit Flash ADC Using 180 nm Technology for Medical Applications