Ashutosh Deshwal

Ashutosh Deshwal

Auto-generated profile, based on publicly available information

Ashutosh Deshwal

Computer Science and Engineering, Indian Institute of Technology Kanpur, Kanpur, India

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
30 April 2018 in International Journal for Research in Applied Science and Engineering Technology
Verified
Journal Article
Closed access
ADAshutosh Deshwal
Ashutosh Deshwal

30 April 2019 in International Journal for Research in Applied Science and Engineering Technology
Verified
Journal Article
Closed access
ADAshutosh Deshwal
Ashutosh Deshwal

18 May 2026 in IEEE Transactions on Computational Imaging
Verified
Journal Article
Closed access
ADAshutosh Deshwal
Ashutosh Deshwal
AAAlireza Azizi
Alireza Azizi
SSShubham Sinha
Shubham Sinha
DPDivyanshu Pandey
Divyanshu Pandey
NMNishant Mehrotra
Nishant Mehrotra
ASAshutosh Sabharwal
Ashutosh Sabharwal
Reliable detection and localization of humans through structural obstructions is critical for time-sensitive applications such as disaster response. This paper presents a novel ultra-wideband (UWB) radar-based framework for multi-person localization and breathing rate estimation in obstructed environments. To overcome the challenges posed by signal attenuation, multipath interference, and limited spatial resolution, we propose a synthetic aperture radar (SAR) system that combines linear translation and on-axis rotation of a single radar. By fusing intensity maps from multiple radar viewpoints, the system achieves enhanced localization of human targets. A signal quality assessment pipeline, including SNR-based filtering, dominance ratio analysis, and periodicity checks, is used to extract clean respiratory signals from selected range bins. The estimates are further integrated into a spatial breathing-rate matrix, which is fused with radar motion maps to suppress false detections and improve reliability. The proposed algorithm is validated experimentally under line-of-sight, through-wall, and emulated rubble conditions. Results demonstrate significant improvements in localization accuracy and robustness of respiration-based human presence detection. The proposed post-processing filtering algorithms reduce the 80th percentile Chamfer distance by 1.53×, while the maximum Chamfer distance is reduced by $\sim 3\times$. The estimated breathing rate correlates strongly with ground truth while achieving a Pearson correlation of 0.79.