Rabindra Kishore Mishra
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Rabindra Kishore Mishra has a total of 44 co-authors. Here are the top co-authors by number of shared publications:
- Dhurba Charana Panda
Berhampur University
8 shared publications - Priyadarshini Raiguru
ITER, SOA Deemed to be University,Department of ECE,Bhubaneswar,Odisha,India
5 shared publications - Rajanikanta Swain
Berhampur University, Berhampur, India
4 shared publications D. R. Poddar
Department of E &TCE, Jadavpur University, Kolkata, West Bengal, India
3 shared publications- Rajeev Kumar Parida
Electronic Science and Technology, Berhampur University, Berhampur, India
3 shared publications - Rowdra Ghatak
Department of Electronics and Communication Engineering, National Institute of Technology Durgapur, Durgapur, India
3 shared publications
Latest Publications
- 12 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Rotation-Dependent RCS Signature Analysis and Machine-Learning-Based Identification of a Chipless RFID Tag for Passive Defence Sensing Applications
Defence, logistics and security services are using low-cost and passive Chipless Radio Frequency Identification (CRFID) technology for identification applications. Rotationdependent spectral variation degrades tag readability when using bistatic radar-style interrogation due to polarization mismatch which reduces the measurable Radar Cross Section (RCS). In this work the possibility of employing AI to negate such effect is the centre of investigation when analyzing backscattered spectral signature of frequency-coded CRFID tag. It employs a compact CRFID tag on 1.6 mm thick FR4 substrate ($7 \text{cm} \times 7 \text{cm}$) containing 18 corner slotted concentric square-ring resonators. The tag characterisation uses EM simulations of RCS (for rotation angles of 0°, 1 5°, 3 0°, and 4 5°) and laboratory bistatic $S_{21}$ measurement (for 0°) on fabricated prototype. Baseline correction via median filtering gave equivalent RCS of measured $S_{21}$ for comparison with simulation results. Application of controlled frequency warping, noise injection, and amplitude scaling generated synthetic experimental dataset from four simulated orientations expanding them to 800 unique signatures. These spectral traces trained and evaluated two machine-learning (ML) pipelines: a dip-feature Support Vector Machine (SVM) classifier and a 1D Convolutional Neural Network (CNN). The accuracy in the former is between 72-75% while in the later it is around 80%. Results show possibility of high accuracy orientation classification through data-driven methods confirming availability of sufficient encoded information within the rotation-induced spectral variation. So, in future passive defence and security systems can employ AI (MLassistance) for identification and authentication.- 12 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Rotation-Dependent RCS Signature Analysis and Machine-Learning-Based Identification of a Chipless RFID Tag for Passive Defence Sensing Applications
Defence, logistics and security services are using low-cost and passive Chipless Radio Frequency Identification (CRFID) technology for identification applications. Rotationdependent spectral variation degrades tag readability when using bistatic radar-style interrogation due to polarization mismatch which reduces the measurable Radar Cross Section (RCS). In this work the possibility of employing AI to negate such effect is the centre of investigation when analyzing backscattered spectral signature of frequency-coded CRFID tag. It employs a compact CRFID tag on 1.6 mm thick FR4 substrate ($7 \text{cm} \times 7 \text{cm}$) containing 18 corner slotted concentric square-ring resonators. The tag characterisation uses EM simulations of RCS (for rotation angles of 0°, 1 5°, 3 0°, and 4 5°) and laboratory bistatic $S_{21}$ measurement (for 0°) on fabricated prototype. Baseline correction via median filtering gave equivalent RCS of measured $S_{21}$ for comparison with simulation results. Application of controlled frequency warping, noise injection, and amplitude scaling generated synthetic experimental dataset from four simulated orientations expanding them to 800 unique signatures. These spectral traces trained and evaluated two machine-learning (ML) pipelines: a dip-feature Support Vector Machine (SVM) classifier and a 1D Convolutional Neural Network (CNN). The accuracy in the former is between 72-75% while in the later it is around 80%. Results show possibility of high accuracy orientation classification through data-driven methods confirming availability of sufficient encoded information within the rotation-induced spectral variation. So, in future passive defence and security systems can employ AI (MLassistance) for identification and authentication.- 12 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (22) , 38125-38134
Machine Learning-Aided Sparse Direction of Arrival Estimation
This paper introduces a design and method to address the challenge of decreased accuracy in direction-of-arrival (DOA) estimation when utilizing planar sparse arrays with closely spaced sensors. It proposes a design leveraging Koch fractal geometry, which effectively reduces mutual coupling and enhances the detection of a greater number of sources compared to the number of sensors in the array. Additionally, the paper presents a technique that integrates machine learning, specifically combining a deep stacked online sequential random vector functional link network with autoencoder (DSOSRVFLN-AE) and a robust multikernel random vector functional link network (RMKRVFLN), for DOA estimation. This technique involves feature extraction through unsupervised learning and DOA estimation via supervised learning. Results demonstrate superior performance, with a root mean square of 0.02 in 0 dB noisy environment, surpassing other prominent methods in terms of computational efficiency, learning speed, accuracy, and robustness. Similarly, figures depict that for minimum resolution grid interval of 0.5°, the proposed DSOSRVFLN-AE-RMKRVFLN method estimates DOA with MAE less than 0.02. The method effectively identifies multiple closely spaced targets (with a minimum angular separation of 0.5°) and is validated through digital architecture implemented on a Xilinx Virtex-5 field programmable gate array (FPGA), indicating its simplicity and practicality.
- 12 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (22) , 38125-38134
Machine Learning-Aided Sparse Direction of Arrival Estimation
This paper introduces a design and method to address the challenge of decreased accuracy in direction-of-arrival (DOA) estimation when utilizing planar sparse arrays with closely spaced sensors. It proposes a design leveraging Koch fractal geometry, which effectively reduces mutual coupling and enhances the detection of a greater number of sources compared to the number of sensors in the array. Additionally, the paper presents a technique that integrates machine learning, specifically combining a deep stacked online sequential random vector functional link network with autoencoder (DSOSRVFLN-AE) and a robust multikernel random vector functional link network (RMKRVFLN), for DOA estimation. This technique involves feature extraction through unsupervised learning and DOA estimation via supervised learning. Results demonstrate superior performance, with a root mean square of 0.02 in 0 dB noisy environment, surpassing other prominent methods in terms of computational efficiency, learning speed, accuracy, and robustness. Similarly, figures depict that for minimum resolution grid interval of 0.5°, the proposed DSOSRVFLN-AE-RMKRVFLN method estimates DOA with MAE less than 0.02. The method effectively identifies multiple closely spaced targets (with a minimum angular separation of 0.5°) and is validated through digital architecture implemented on a Xilinx Virtex-5 field programmable gate array (FPGA), indicating its simplicity and practicality.
- 1 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Antennas and Wireless Propagation Letters
- Vol. 23 (10) , 3038-3042
An Innovative Mapping Technique for Sinusoidal Frequency-Selective Surfaces in Shielding Applications
This letter presents an innovative mapping technique to design and demonstrate a frequency selective surface (FSS) based bandstop filter on a sinusoidal surface. The novelty of the method lies in the uniform placement of the unit cells on different non-planar surfaces using a hybrid method comprising an analytical technique and full-wave simulation that significantly improves the shielding characteristic in a wideband spectrum (having a -10 dB stopband from 7 to 13 GHz with a fractional bandwidth of 60%). A conventional metallic square loop is considered as the constituent unit cell topology, which is intelligibly positioned over a large surface using the proposed method. The structure is then simulated using the finite element boundary integral, as well as fabricated exploiting the threedimensional (3-D) printing technology, thereby validating the proposed concept. Such non-planar FSS geometries have potential in diverse applications, including shielding and aerospace sectors.
- 1 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Antennas and Wireless Propagation Letters
- Vol. 23 (10) , 3038-3042
An Innovative Mapping Technique for Sinusoidal Frequency-Selective Surfaces in Shielding Applications
This letter presents an innovative mapping technique to design and demonstrate a frequency selective surface (FSS) based bandstop filter on a sinusoidal surface. The novelty of the method lies in the uniform placement of the unit cells on different non-planar surfaces using a hybrid method comprising an analytical technique and full-wave simulation that significantly improves the shielding characteristic in a wideband spectrum (having a -10 dB stopband from 7 to 13 GHz with a fractional bandwidth of 60%). A conventional metallic square loop is considered as the constituent unit cell topology, which is intelligibly positioned over a large surface using the proposed method. The structure is then simulated using the finite element boundary integral, as well as fabricated exploiting the threedimensional (3-D) printing technology, thereby validating the proposed concept. Such non-planar FSS geometries have potential in diverse applications, including shielding and aerospace sectors.
- 3 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Antennas and Wireless Propagation Letters
- Vol. 22 (3) , 586-590
CMA Approach for a DGS-Based Wideband Circular Patch Antenna With Low Cross Polarization
This paper presents a design approach for a defected ground structure (DGS) based wideband circular microstrip patch antenna using the characteristic mode analysis (CMA). It emphasizes improving two design aspects: cross-polarization reduction and bandwidth enhancement. The CMA unveils the intrinsic characteristic modes in the patch and ground plane, and then undesired modes are suppressed using a DGS, reducing cross-polarized radiations. Another defect, strategically incorporated into the same DGS, enhances a higher-order non-radiating mode to radiate and increases bandwidth. The proposed design fabricated on an FR-4 substrate achieves a $\simeq$ 16% impedance bandwidth (5.7–6.7 GHz) with dimensions $0.8\lambda _{0}\times 0.8\lambda _{0}\times 0.03\lambda _{0}$ ( $\lambda _{0}$ being the free-space wavelength at 6 GHz) and 10–15 dB average cross-polarization reduction at $\pm$ 60° angle of boresight. The measured results from a fabricated prototype agree well with the proposed design.
- 3 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Antennas and Wireless Propagation Letters
- Vol. 22 (3) , 586-590
CMA Approach for a DGS-Based Wideband Circular Patch Antenna With Low Cross Polarization
This paper presents a design approach for a defected ground structure (DGS) based wideband circular microstrip patch antenna using the characteristic mode analysis (CMA). It emphasizes improving two design aspects: cross-polarization reduction and bandwidth enhancement. The CMA unveils the intrinsic characteristic modes in the patch and ground plane, and then undesired modes are suppressed using a DGS, reducing cross-polarized radiations. Another defect, strategically incorporated into the same DGS, enhances a higher-order non-radiating mode to radiate and increases bandwidth. The proposed design fabricated on an FR-4 substrate achieves a $\simeq$ 16% impedance bandwidth (5.7–6.7 GHz) with dimensions $0.8\lambda _{0}\times 0.8\lambda _{0}\times 0.03\lambda _{0}$ ( $\lambda _{0}$ being the free-space wavelength at 6 GHz) and 10–15 dB average cross-polarization reduction at $\pm$ 60° angle of boresight. The measured results from a fabricated prototype agree well with the proposed design.
- 1 April 2022
- Published by Hindawi Limited in International Journal of RF and Microwave Computer-Aided Engineering
- Vol. 32 (7)
Compact reconfigurable antenna system for spectrum interweaved cognitive radio
- 1 April 2022
- Published by Hindawi Limited in International Journal of RF and Microwave Computer-Aided Engineering
- Vol. 32 (7)
Compact reconfigurable antenna system for spectrum interweaved cognitive radio
- 6 March 2022
- Published by Taylor & Francis in IETE Journal of Research
- Vol. 69 (11) , 7672-7683
Multi-Source Detection Performance of Some Linear Sparse Arrays
An investigation on fractal-based sub-array (FBSA), of isotropic elements, that evolves from two Non-uniform Linear arrays, i.e. the minimum redundancy array and fractal Array, with a hole-free difference co-array shows competitive performance in direction of arrival estimation using (spatial smoothing-) MUSIC algorithm. Numerical experiments demonstrate that the algorithm can resolve 2° separation between sources. Considering mutual coupling, in terms of RMSE, the FBSA performs better than other competing configurations.
- 6 March 2022
- Published by Taylor & Francis in IETE Journal of Research
- Vol. 69 (11) , 7672-7683
Multi-Source Detection Performance of Some Linear Sparse Arrays
An investigation on fractal-based sub-array (FBSA), of isotropic elements, that evolves from two Non-uniform Linear arrays, i.e. the minimum redundancy array and fractal Array, with a hole-free difference co-array shows competitive performance in direction of arrival estimation using (spatial smoothing-) MUSIC algorithm. Numerical experiments demonstrate that the algorithm can resolve 2° separation between sources. Considering mutual coupling, in terms of RMSE, the FBSA performs better than other competing configurations.
- 10 January 2022
- Published by Springer Nature in Circuits, Systems, and Signal Processing
- Vol. 41 (5) , 2977-2987
Hole-Free DCA for Augmented Co-Prime Array
- 10 January 2022
- Published by Springer Nature in Circuits, Systems, and Signal Processing
- Vol. 41 (5) , 2977-2987
Hole-Free DCA for Augmented Co-Prime Array
- 1 December 2021
- Published by Elsevier in AEU - International Journal of Electronics and Communications
- Vol. 142, 153986
A digital direction of arrival estimator based on fast on-line sequential random vector functional link network
- 1 December 2021
- Published by Elsevier in AEU - International Journal of Electronics and Communications
- Vol. 142, 153986
A digital direction of arrival estimator based on fast on-line sequential random vector functional link network