Arup Lal Chakraborty
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
Arup Lal Chakraborty has a total of 42 co-authors. Here are the top co-authors by number of shared publications:
- Chandan Kumar Jha
Indian Institute of Technology Gandhinagar, Gandhinagar, Gujarat, India
6 shared publications Anirban Roy
Indian Institute of Information Technology, Surat, Gujarat, India
5 shared publications- Anirban Roy
Indian Institute of Technology Gandhinagar
5 shared publications - Abhishek Upadhyay
Department of Electronic and Electrical Engineering, University of Strathclyde, Glasgow G1 1XW, UK
3 shared publications - Walter Johnstone
Department of Electronic and Electrical Engineering, University of Strathclyde, Glasgow, U.K.
3 shared publications - A. S. Zarin
Indian Inst Technol Gandhinagar, Discipline Elect Engn, Gandhinagar 382355, India
2 shared publications
Latest Publications
- 23 May 2025
- Published by SPIE-Intl Soc Optical Eng
- Vol. 13527, 55-182
Speech enhancement in FBG-based throat microphones: a tailored long short-term memory recurrent neural network approach
Fiber Bragg Grating (FBG)-based throat microphone’s superior background noise suppression makes them ideal for wearable Automatic Speech Recognition (ASR) devices. However, achieving naturalness and intelligibility remains challenging due to the low-pass filtering effects of tissue and bones. This study presents a deep learning framework using a Long Short-Term Memory (LSTM) recurrent neural network for speech enhancement in FBG microphones and explores the impact of microphone placement and sex on ASR performance. The microphone, designed with a 1530.12nm prestrained FBG, captured vocal vibrations from six participants reciting Harvard sentences. An LSTM model trained with spectral mapping restored high-frequency components, reducing the log-spectral distance by 26% and improving the Non-Intrusive Short-Time Objective Intelligibility (NI-STOI) score by 2%. Character Error Rate (CER) and NI-STOI score showed significantly better performance at the lower throat position, emphasizing the importance of optimal microphone placement. Speaker sex, however, had no significant effect on CER or intelligibility.
- 23 May 2025
- Published by SPIE-Intl Soc Optical Eng
- Vol. 13527, 55-182
Speech enhancement in FBG-based throat microphones: a tailored long short-term memory recurrent neural network approach
Fiber Bragg Grating (FBG)-based throat microphone’s superior background noise suppression makes them ideal for wearable Automatic Speech Recognition (ASR) devices. However, achieving naturalness and intelligibility remains challenging due to the low-pass filtering effects of tissue and bones. This study presents a deep learning framework using a Long Short-Term Memory (LSTM) recurrent neural network for speech enhancement in FBG microphones and explores the impact of microphone placement and sex on ASR performance. The microphone, designed with a 1530.12nm prestrained FBG, captured vocal vibrations from six participants reciting Harvard sentences. An LSTM model trained with spectral mapping restored high-frequency components, reducing the log-spectral distance by 26% and improving the Non-Intrusive Short-Time Objective Intelligibility (NI-STOI) score by 2%. Character Error Rate (CER) and NI-STOI score showed significantly better performance at the lower throat position, emphasizing the importance of optimal microphone placement. Speaker sex, however, had no significant effect on CER or intelligibility.
- 17 April 2023
- Published by Optica Publishing Group in Applied Optics
- Vol. 62 (12) , 3160-3168
Detailed analysis of the R1f/ΔI1 WMS technique and demonstration of significantly higher detection sensitivity compared to 2f WMS for calibration-free trace gas sensing
Recognizing that wavelength modulation spectroscopy (WMS) is particularly important in the development of high-sensitivity gas sensing systems, this paper presents a detailed analysis of the R1f/ΔI1 WMS technique that has recently been successfully demonstrated for calibration-free measurements of the parameters that support detecting multiple gases under challenging conditions. In this approach, the magnitude of the 1f WMS signal (R1f) was normalized by using the laser’s linear intensity modulation (ΔI1) to obtain the quantity R1f/ΔI1 that is shown to be unaffected by large variations in R1f itself due to the variations in the intensity of the received light. In this paper, different simulations have been used to explain the approach taken and the advantages that it shows. A 40 mW, 1531.52 nm near-infrared distributed feedback (DFB) semiconductor laser was used to extract the mole fraction of acetylene in a single-pass configuration. The work has shown a detection sensitivity of 0.32 ppm for 28 cm (0.089 ppm-m) with an optimum integration time of 58 s. The detection limit achieved has been shown to be better than the value of 1.53 ppm (0.428 ppm-m) for R2f WMS by a factor of 4.7, which is a significant improvement.
- 17 April 2023
- Published by Optica Publishing Group in Applied Optics
- Vol. 62 (12) , 3160-3168
Detailed analysis of the R1f/ΔI1 WMS technique and demonstration of significantly higher detection sensitivity compared to 2f WMS for calibration-free trace gas sensing
Recognizing that wavelength modulation spectroscopy (WMS) is particularly important in the development of high-sensitivity gas sensing systems, this paper presents a detailed analysis of the R1f/ΔI1 WMS technique that has recently been successfully demonstrated for calibration-free measurements of the parameters that support detecting multiple gases under challenging conditions. In this approach, the magnitude of the 1f WMS signal (R1f) was normalized by using the laser’s linear intensity modulation (ΔI1) to obtain the quantity R1f/ΔI1 that is shown to be unaffected by large variations in R1f itself due to the variations in the intensity of the received light. In this paper, different simulations have been used to explain the approach taken and the advantages that it shows. A 40 mW, 1531.52 nm near-infrared distributed feedback (DFB) semiconductor laser was used to extract the mole fraction of acetylene in a single-pass configuration. The work has shown a detection sensitivity of 0.32 ppm for 28 cm (0.089 ppm-m) with an optimum integration time of 58 s. The detection limit achieved has been shown to be better than the value of 1.53 ppm (0.428 ppm-m) for R2f WMS by a factor of 4.7, which is a significant improvement.
- 23 March 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
- Vol. 69 (5) , 1682-1690
A Fiber Bragg Grating-Based Sensor for Passive Cavitation Detection at MHz Frequencies
Fiber Bragg gratings (FBGs) are a potential alternative to piezoelectric ultrasound sensors for applications that demand high sensitivity and immunity to electromagnetic interference (EMI). However, limited data exist on the quantitative performance characterization of FBG sensors in the MHz frequency range relevant to biomedical ultrasound. In this work, we evaluated an FBG to detect MHz-frequency ultrasound and tested the feasibility of measuring passive cavitation signals nucleated using a commercial contrast agent (SonoVue). The sensitivity, repeatability, and linearity of the measurements were assessed for ultrasound measurements at 1, 5, and 10 MHz. The bandwidth of the FBG sensor was measured and compared to that of a calibrated needle hydrophone. The FBG showed a sensitivity of 0.99, 0.769, and 0.818 V/MPa for 1, 5, and 10 MHz ultrasound, respectively. The sensor also exhibited linear response ( $0.975\leq {R}$ -Squared ≤ 0.996) and good repeatability with a coefficient of variation (CV) less than 5.5%. A 2-MHz focused transducer was used to insonify SonoVue microbubbles at a peak negative pressure of 175 kPa and passive cavitation emissions were measured, in which subharmonic and ultraharmonic spectral peaks were observed. These results demonstrate the potential of FBGs for MHz-range ultrasound applications, including passive cavitation detection (PCD).
- 23 March 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
- Vol. 69 (5) , 1682-1690
A Fiber Bragg Grating-Based Sensor for Passive Cavitation Detection at MHz Frequencies
Fiber Bragg gratings (FBGs) are a potential alternative to piezoelectric ultrasound sensors for applications that demand high sensitivity and immunity to electromagnetic interference (EMI). However, limited data exist on the quantitative performance characterization of FBG sensors in the MHz frequency range relevant to biomedical ultrasound. In this work, we evaluated an FBG to detect MHz-frequency ultrasound and tested the feasibility of measuring passive cavitation signals nucleated using a commercial contrast agent (SonoVue). The sensitivity, repeatability, and linearity of the measurements were assessed for ultrasound measurements at 1, 5, and 10 MHz. The bandwidth of the FBG sensor was measured and compared to that of a calibrated needle hydrophone. The FBG showed a sensitivity of 0.99, 0.769, and 0.818 V/MPa for 1, 5, and 10 MHz ultrasound, respectively. The sensor also exhibited linear response ( $0.975\leq {R}$ -Squared ≤ 0.996) and good repeatability with a coefficient of variation (CV) less than 5.5%. A 2-MHz focused transducer was used to insonify SonoVue microbubbles at a peak negative pressure of 175 kPa and passive cavitation emissions were measured, in which subharmonic and ultraharmonic spectral peaks were observed. These results demonstrate the potential of FBGs for MHz-range ultrasound applications, including passive cavitation detection (PCD).
- 1 January 2022
- Published by Optica Publishing Group
- Vol. 21, M3Z.1
A fibre Bragg grating sensor-based instrumented glove for virtual rehabilitation applications
A fibre Bragg grating sensor-based instrumented glove being developed for medical applications will be demonstrated that can be used to measure finger joint angles with a resolution of 0.1°.
- 1 January 2022
- Published by Optica Publishing Group
- Vol. 21, M3Z.1
A fibre Bragg grating sensor-based instrumented glove for virtual rehabilitation applications
A fibre Bragg grating sensor-based instrumented glove being developed for medical applications will be demonstrated that can be used to measure finger joint angles with a resolution of 0.1°.
- 10 September 2021
- Published by Optica Publishing Group in Optics Letters
- Vol. 46 (18) , 4562-4565
Light-induced in situ active tuning of the LSPR of gold nanorods over 90 nm
We demonstrate an easy and controllable method for light-induced active tuning of the longitudinal surface plasmon resonance (LSPR) of gold nanorods (AuNRs) over ∼ 94 n m . The red-shift of the LSPR can be controlled by varying the time of exposure to a 532 nm laser. The tuning is achieved by photo-induced dissolution of individual AuNRs by sodium dodecyl sulfate (SDS) under continuous illumination. The dissolution of the AuNRs increases the aspect ratio, and consequently the LSPR exhibits a gradual but large redshift. A key feature is that it is possible to selectively tune the LSPR of a specific AuNR in a group while leaving the others totally unaffected. Such controllable, light-induced, post-synthesis fine-tuning of the LSPR is useful for tailoring the plasmonic response of individual AuNRs for a wide range of applications.
- 10 September 2021
- Published by Optica Publishing Group in Optics Letters
- Vol. 46 (18) , 4562-4565
Light-induced in situ active tuning of the LSPR of gold nanorods over 90 nm
We demonstrate an easy and controllable method for light-induced active tuning of the longitudinal surface plasmon resonance (LSPR) of gold nanorods (AuNRs) over ∼ 94 n m . The red-shift of the LSPR can be controlled by varying the time of exposure to a 532 nm laser. The tuning is achieved by photo-induced dissolution of individual AuNRs by sodium dodecyl sulfate (SDS) under continuous illumination. The dissolution of the AuNRs increases the aspect ratio, and consequently the LSPR exhibits a gradual but large redshift. A key feature is that it is possible to selectively tune the LSPR of a specific AuNR in a group while leaving the others totally unaffected. Such controllable, light-induced, post-synthesis fine-tuning of the LSPR is useful for tailoring the plasmonic response of individual AuNRs for a wide range of applications.
- 19 May 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Technology Letters
- Vol. 33 (18) , 982-985
QCL-Based Open-Path, Single-Pass Measurement of Ambient Carbon Monoxide Using R 1f /ΔI₁ WMS
This paper outlines a new robust wavelength modulation spectroscopy technique to achieve calibration-free measurements. The technique normalizes the magnitude of the 1st harmonic signal by the amplitude of the laser’s linear intensity modulation, and uses the fact that the ratio of the intensity to the amplitude of the linear intensity modulation remains invariant at every point within a system and over time despite significant and unpredictable variations in the individual quantities. The technique is demonstrated using a mid-infrared quantum cascade laser for open-path measurements of ambient carbon monoxide with a single-pass detection limit of 12 ppb-m. The method is simple to implement and performs robustly for long durations of outdoor measurements.
- 19 May 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Technology Letters
- Vol. 33 (18) , 982-985
QCL-Based Open-Path, Single-Pass Measurement of Ambient Carbon Monoxide Using R 1f /ΔI₁ WMS
This paper outlines a new robust wavelength modulation spectroscopy technique to achieve calibration-free measurements. The technique normalizes the magnitude of the 1st harmonic signal by the amplitude of the laser’s linear intensity modulation, and uses the fact that the ratio of the intensity to the amplitude of the linear intensity modulation remains invariant at every point within a system and over time despite significant and unpredictable variations in the individual quantities. The technique is demonstrated using a mid-infrared quantum cascade laser for open-path measurements of ambient carbon monoxide with a single-pass detection limit of 12 ppb-m. The method is simple to implement and performs robustly for long durations of outdoor measurements.
- 22 December 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 21 (6) , 7620-7630
Design and Evaluation of an FBG Sensor-Based Glove to Simultaneously Monitor Flexure of Ten Finger Joints
This paper demonstrates a fiber Bragg grating sensor-based instrumented glove that simultaneously measures the range of motion of the ten finger joints (five metacarpophalangeal, four proximal interphalangeal and one interphalangeal joint) with very high angular resolution (0.1°). The accuracy and repeatability of all the ten sensors of the glove are compared with a pre-calibrated inertial measurement unit (IMU) sensor. The glove outperforms many earlier reported sensing gloves with a mean error of 0.80°. The standard deviation (1.01°) and range (2.60°) obtained in the reliability test performed on five healthy subjects are also smaller compared to many other gloves. The feasibility of using the glove in real-world applications has been shown by demonstrating a virtual reality (VR) platform that uses a Raspberry Pi-based module to interface the glove with custom VR games running on an Android tablet or a VR headset. VR-based gaming platforms have become popular in the rehabilitation of the upper limb in stroke patients in recent years. The high accuracy and reliability of this glove will enable accurate tracking of the movement quality and recovery progress of stroke patients during the virtual rehabilitation therapy.
- 22 December 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 21 (6) , 7620-7630
Design and Evaluation of an FBG Sensor-Based Glove to Simultaneously Monitor Flexure of Ten Finger Joints
This paper demonstrates a fiber Bragg grating sensor-based instrumented glove that simultaneously measures the range of motion of the ten finger joints (five metacarpophalangeal, four proximal interphalangeal and one interphalangeal joint) with very high angular resolution (0.1°). The accuracy and repeatability of all the ten sensors of the glove are compared with a pre-calibrated inertial measurement unit (IMU) sensor. The glove outperforms many earlier reported sensing gloves with a mean error of 0.80°. The standard deviation (1.01°) and range (2.60°) obtained in the reliability test performed on five healthy subjects are also smaller compared to many other gloves. The feasibility of using the glove in real-world applications has been shown by demonstrating a virtual reality (VR) platform that uses a Raspberry Pi-based module to interface the glove with custom VR games running on an Android tablet or a VR headset. VR-based gaming platforms have become popular in the rehabilitation of the upper limb in stroke patients in recent years. The high accuracy and reliability of this glove will enable accurate tracking of the movement quality and recovery progress of stroke patients during the virtual rehabilitation therapy.
- 1 November 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 20 (21) , 12691-12701
Intensity Modulation-Normalized Calibration-Free 1f and 2f Wavelength Modulation Spectroscopy
- 1 November 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 20 (21) , 12691-12701
Intensity Modulation-Normalized Calibration-Free 1f and 2f Wavelength Modulation Spectroscopy