TY - GEN
T1 - Low-Power LiDAR Signal Processor with Point-of-Cloud Transformation Accelerator
AU - Park, Seunghyun
AU - Park, Daejin
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - 3D image processing is an important technology that needs to be fundamentally studied in the current, constantly evolving artificial intelligence market. Among this technology, in the field of autonomous driving, the Lidar sensor provides a 3D spatial image of the surrounding. A high-performance processor is required to process a large amount of information quickly. However, because the power consumption increases as the performance increases, a low-power design is also an essential consideration [1]. In general, large amounts of computation can be sped up by parallelized data processing using the GPU. In this paper, to determine advantages parallelization has in terms of performance in processing 3D images, we designed an MCU equipped with parallelized memory and a multicore processor with Verilog as well as compared the processing speed and power with the existing single-core processor. We conducted an experiment to check the 3D image processed through the MCU. The data were imported into CloudCompare software, and the result data was visible. The experimental results confirmed that the designed quad-core MCU consumes 52% less energy than the single-core MCU as the number of operations increases.
AB - 3D image processing is an important technology that needs to be fundamentally studied in the current, constantly evolving artificial intelligence market. Among this technology, in the field of autonomous driving, the Lidar sensor provides a 3D spatial image of the surrounding. A high-performance processor is required to process a large amount of information quickly. However, because the power consumption increases as the performance increases, a low-power design is also an essential consideration [1]. In general, large amounts of computation can be sped up by parallelized data processing using the GPU. In this paper, to determine advantages parallelization has in terms of performance in processing 3D images, we designed an MCU equipped with parallelized memory and a multicore processor with Verilog as well as compared the processing speed and power with the existing single-core processor. We conducted an experiment to check the 3D image processed through the MCU. The data were imported into CloudCompare software, and the result data was visible. The experimental results confirmed that the designed quad-core MCU consumes 52% less energy than the single-core MCU as the number of operations increases.
KW - 3D image processing
KW - CloudCompare
KW - multicore processor
KW - parallelized processing
KW - Verilog
UR - http://www.scopus.com/inward/record.url?scp=85138727442&partnerID=8YFLogxK
U2 - 10.1109/ICCE-Taiwan55306.2022.9869189
DO - 10.1109/ICCE-Taiwan55306.2022.9869189
M3 - Conference contribution
AN - SCOPUS:85138727442
T3 - Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022
SP - 57
EP - 58
BT - Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022
Y2 - 6 July 2022 through 8 July 2022
ER -
