Streaming Convolutional Neural Network FPGA architecture for RFSoC data converters

Research output: Contribution to conferencePaperpeer-review

133 Downloads (Pure)

Abstract

This paper presents a novel Convolutional Neural Network (CNN) FPGA architecture designed to perform processing of radio data in a streaming manner without interruption. The proposed architecture is evaluated for radio modulation classification tasks implemented on an AMD RFSoC 2x2 development board and operating in real-time. The proposed architecture leverages optimisation such as the General Matrix-to-Matrix (GEMM) transform, on-chip weights, fixed-point arithmetic, and efficient utilisation of FPGA resources to achieve constant processing of a stream of samples. The performance of the proposed architecture is demonstrated through accuracy results obtained during live modulation classification, while operating at a sampling frequency of 128 MHz before decimation. The proposed architecture demonstrates promising results for real-time, time-critical CNN applications.
Original languageEnglish
Pages1-5
Number of pages5
Publication statusPublished - 27 Jun 2023
Event21st IEEE Interregional New Circuit and Systems (NEWCAS) Conference: An IEEE CAS Society Interregional Flagship Conference - John McIntyre Conference Centre, Edinburgh, United Kingdom
Duration: 26 Jun 202328 Jun 2023
https://2023.ieee-newcas.org/

Conference

Conference21st IEEE Interregional New Circuit and Systems (NEWCAS) Conference
Abbreviated titleIEEE NEWCAS 2023
Country/TerritoryUnited Kingdom
CityEdinburgh
Period26/06/2328/06/23
Internet address

Keywords

  • deep learning
  • wireless communications
  • FPGA
  • RFSoC
  • AMD
  • tensorflow
  • MATLAB
  • signal processing

Fingerprint

Dive into the research topics of 'Streaming Convolutional Neural Network FPGA architecture for RFSoC data converters'. Together they form a unique fingerprint.

Cite this