FPGA-based Low Power Acceleration of HARK Sound Source Localization

被引：1

作者：

Lin, Zirui ^{[1
,1
]}

Itoyama, Katsutoshi ^{[1
,2
]}

Nakadai, Kazuhiro ^{[1
]}

Amano, Hideharu ^{[3
]}

机构：

[1] Tokyo Inst Technol, Sch Engn, Dept Syst & Control Engn, Tokyo, Japan

[2] Honda Res Inst Japan Co Ltd, Saitama, Japan

[3] Keio Univ, Dept Informat & Comp Sci, Keio, Kanagawa, Japan

来源：

2024 IEEE SYMPOSIUM IN LOW-POWER AND HIGH-SPEED CHIPS, COOL CHIPS 27 | 2024年

关键词：

FPGA; Robot Audition; Sound Source Localization; Acceleration; Low Power; INTERNET;

D O I：

10.1109/COOLCHIPS61292.2024.10531180

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper presents an FPGA-based low-power acceleration of sound source localization in HARK, open-source software for robot audition. Due to the massive matrix operations, sound source localization in HARK takes substantial processing time in edge computing devices. To balance processing time and low power consumption, two functions in sound source localization that include many matrix operations are targeted and migrated on an FPGA SoC board called M-KUBOS. Compared to CPU-based computing on ARM Cortex A53, our implementation achieved a 2.0x speedup and 1.7x lower energy consumption.

引用

页数：6

共 22 条

[1]

Bando Y, 2015, 2015 IEEE INTERNATIONAL SYMPOSIUM ON SAFETY, SECURITY, AND RESCUE ROBOTICS (SSRR)

[2]

Biookaghazadeh Saman., 2018, USENIX WORKSHOP HOT

[3] Edge Computing: A Survey On the Hardware Requirements in the Internet of Things World [J].

Capra, Maurizio ;

Peloso, Riccardo ;

Masera, Guido ;

Roch, Massimo Ruo ;

Martina, Maurizio .

FUTURE INTERNET, 2019, 11 (04)

[4]

HARK Development Team, 2024, Hark document, version 3.4.0

[5] The Role of Edge Computing in Internet of Things [J].

Hassan, Najmul ;

Gillani, Saira ;

Ahmed, Ejaz ;

Yaqoob, Ibrar ;

Imran, Muhammad .

IEEE COMMUNICATIONS MAGAZINE, 2018, 56 (11) :110-115

[6]

HAYES M.H., 1996, STAT DIGITAL SIGNAL

[7] An FPGA off-loading of HARK sound source localization [J].

Hou, Zhongyang ;

Wei, Kaijie ;

Amano, Hideharu ;

Nakadai, Kazuhiro .

2022 TENTH INTERNATIONAL SYMPOSIUM ON COMPUTING AND NETWORKING WORKSHOPS, CANDARW, 2022, :236-240

[8] M-KUBOS/PYNQ Cluster for multi-access edge computing [J].

Inage, Takumi ;

Hironaka, Kazuei ;

Iizuka, Kensuke ;

Ito, Kohei ;

Fukushima, Yasuyu ;

Namiki, Mitaro ;

Amano, Hideharu .

2021 NINTH INTERNATIONAL SYMPOSIUM ON COMPUTING AND NETWORKING (CANDAR 2021), 2021, :95-101

[9]

Kazuhiro Nakadai, 2023, P AS PAC C ROB IOT S, V2023, P65

[10]

Kojima R, 2018, IEEE INT C INT ROBOT, P2497, DOI 10.1109/IROS.2018.8594070

← 1 2 3 →