Buchumschlag

Pulsar Search on Mobile Devices

Gespeichert in:
Veröffentlicht in:The Institute of Electrical and Electronics Engineers, Inc. (IEEE) Conference Proceedings (2025), p. 104-109
1. Verfasser: Peng, Shilin
Weitere Verfasser: Hu, Qianchen, Wang, Layue, Lu, Jiang, Wang, Jian, Peng, Zhicheng
Veröffentlicht:
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Energy consumption
Search process
Data processing
Big Data
Servers
Multilayers
Smartphones
Electronic devices
Edge computing
Nodes
Artificial intelligence
Power consumption
Pulsars
Distributed processing
Economic
Online-Zugang:Citation/Abstract
Tags: Tag hinzufügen
Keine Tags, Fügen Sie das erste Tag hinzu!
Abstract:Conference Title: 2025 8th International Conference on Artificial Intelligence and Big Data (ICAIBD)Conference Start Date: 2025 May 23Conference End Date: 2025 May 26Conference Location: Chengdu, ChinaThis paper present an edge computing model for pulsar search on mobile devices, to reduce the computing load of server nodes, make full use of distributed computing nodes. Using edge devices for dedispersion processing and acceleration searching, which makes the previously time-consuming and challenging to achieve parallel computing dedispersion algorithm divided into multiple small computing tasks in time domain and frequency domain, and a large number of computing tasks can be distributed into various edge devices. Based on that, a multilayer directed connection distributed search model is designed to reduce the amount of computation on the server side in the search process. Raspberry Pi 3B and 4B and smart phones are used as distributed computing nodes to test the data processing capability of mobile devices with different performance and the energy consumption in computing. With the characteristics of volunteer computing nodes, the network bandwidth requirements of different performance mobile computing nodes are discussed according to the test results. Experiments have shown that those devices can afford such computing tasks, especially mobile phones with GPGPU computing, which can carry such computing tasks well under the low power consumption.
DOI:10.1109/ICAIBD64986.2025.11082002
Quelle:Science Database