MUSIC-lite: Efficient MUSIC using Approximate Computing: An OFDM Radar Case Study

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Bhattacharjya, Rajat, Sarkar, Arnab, Maity, Biswadip, Dutt, Nikil
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866929615585411072
author Bhattacharjya, Rajat
Sarkar, Arnab
Maity, Biswadip
Dutt, Nikil
author_facet Bhattacharjya, Rajat
Sarkar, Arnab
Maity, Biswadip
Dutt, Nikil
contents Multiple Signal Classification (MUSIC) is a widely used Direction of Arrival (DoA)/Angle of Arrival (AoA) estimation algorithm applied to various application domains such as autonomous driving, medical imaging, and astronomy. However, MUSIC is computationally expensive and challenging to implement in low-power hardware, requiring exploration of trade-offs between accuracy, cost, and power. We present MUSIC-lite, which exploits approximate computing to generate a design space exploring accuracy-area-power trade-offs. This is specifically applied to the computationally intensive singular value decomposition (SVD) component of the MUSIC algorithm in an orthogonal frequency-division multiplexing (OFDM) radar use case. MUSIC-lite incorporates approximate adders into the iterative CORDIC algorithm that is used for hardware implementation of MUSIC, generating interesting accuracy-area-power trade-offs. Our experiments demonstrate MUSIC-lite's ability to save an average of 17.25% on-chip area and 19.4% power with a minimal 0.14% error for efficient MUSIC implementations.
format Preprint
id arxiv_https___arxiv_org_abs_2407_04849
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MUSIC-lite: Efficient MUSIC using Approximate Computing: An OFDM Radar Case Study
Bhattacharjya, Rajat
Sarkar, Arnab
Maity, Biswadip
Dutt, Nikil
Hardware Architecture
Signal Processing
Multiple Signal Classification (MUSIC) is a widely used Direction of Arrival (DoA)/Angle of Arrival (AoA) estimation algorithm applied to various application domains such as autonomous driving, medical imaging, and astronomy. However, MUSIC is computationally expensive and challenging to implement in low-power hardware, requiring exploration of trade-offs between accuracy, cost, and power. We present MUSIC-lite, which exploits approximate computing to generate a design space exploring accuracy-area-power trade-offs. This is specifically applied to the computationally intensive singular value decomposition (SVD) component of the MUSIC algorithm in an orthogonal frequency-division multiplexing (OFDM) radar use case. MUSIC-lite incorporates approximate adders into the iterative CORDIC algorithm that is used for hardware implementation of MUSIC, generating interesting accuracy-area-power trade-offs. Our experiments demonstrate MUSIC-lite's ability to save an average of 17.25% on-chip area and 19.4% power with a minimal 0.14% error for efficient MUSIC implementations.
title MUSIC-lite: Efficient MUSIC using Approximate Computing: An OFDM Radar Case Study
topic Hardware Architecture
Signal Processing
url https://arxiv.org/abs/2407.04849