Deep Architecture for DOA Trajectory Localization

Jaiswal, Shreyas; Pandey, Ruchi; Nannuru, Santosh

doi:10.1109/icassp49357.2023.10095493

Cited by 1 publication

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To model complex trajectories, polynomial and bandlimited models are used. The linear trajectory model, proposed in our earlier work 30,31 , can be recovered as a special case of the polynomial model.…”

Section: Signal Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…In our previous work 30,31 , we introduced a parametric trajectory model incorporating linear source motion within a block and performed trajectory localization (TL). We developed TL-CBF and TL-SBL algorithms 30 and deep U-Net architecture 31 for estimating the linear parameters, which perform better than traditional methods for moving sources. While TL-SBL has a higher resolution than TL-CBF, the computational cost is far too high for real-time applications, and the performance is compromised when the source trajectories don't lie on the predefined grid.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Parametric Models for Doa Trajectory Localization

Pandey

Nannuru

2022

ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Directions of arrival (DOA) estimation or localization of sources is an important problem in many applications for which numerous algorithms have been proposed. Most localization methods use block-level processing that combines multiple data snapshots to estimate DOA within a block. The DOAs are assumed to be constant within the block duration. However, these assumptions are often violated due to source motion. In this paper, we propose a signal model that captures the linear variations in DOA within a block. We applied conventional beamforming (CBF) algorithm to this model to estimate linear DOA trajectories. Further, we formulate the proposed signal model as a block sparse model and subsequently derive sparse Bayesian learning (SBL) algorithm. Our simulation results show that this linear parametric DOA model and corresponding algorithms capture the DOA trajectories for moving sources more accurately than traditional signal models and methods.

show abstract

Section: Signal Modelmentioning

confidence: 99%