Acceleration of FPGA Based Convolutional Neural Network for Human Activity Classification Using Millimeter-Wave Radar

Peng Lei,Jiawei Liang,Tong Zheng,Zhenyu Guan,Jun Wang

doi:10.1109/access.2019.2926381

Peng Lei, Jiawei Liang + Show 3 more

Open Access

https://doi.org/10.1109/access.2019.2926381

Copy DOI

Journal: IEEE Access	Publication Date: Jan 1, 2019
Citations: 27	License type: CC BY 4.0

Affiliation: Beihang University

Abstract

Deep learning techniques have attracted much attention in the radar automatic target recognition. In this paper, we investigate an acceleration method of the convolutional neural network (CNN) on the field-programmable gate array (FPGA) for the embedded application of the millimeter-wave (mmW) radar-based human activity classification. Considering the micro-Doppler effect caused by a person's body movements, the spectrogram of mmW radar echoes is adopted as the CNN input. After that, according to the CNN architecture and the properties of the FPGA implementation, several parallel processing strategies are designed as well as data quantization and optimization of classification decision to accelerate the CNN execution. Finally, comparative experiments and discussions are carried out based on a measured dataset of nine individuals with four different actions by using a 77-GHz mmW radar. The results show that the proposed method not only maintains the high classification accuracy but also improves its execution speed, memory requirement, and power consumption. Specifically, compared with the implementation of the same network model on a graphics processing unit, it could achieve the speedup of about 30.42% at the cost of the classification accuracy loss of only 0.27%.

Highlights

Millimeter wave radars are an emerging and promising technique in sensing systems
Since the proposed acceleration method of field-programmable gate array (FPGA) based convolutional neural network (CNN) focus on the test phase, we briefly introduce the computation workload theoretically involved in its forward propagation as follows
In this paper, we present an acceleration method of CNN on FPGA for the mmW radar based human activity classification

Summary

INTRODUCTION

Millimeter wave (mmW) radars are an emerging and promising technique in sensing systems. With the success of deep learning techniques in computer vision, they have become much popular in the radar mD based classification of human targets over past few years Such methods are able to automatically find useful features instead of the handcrafted feature extraction in traditional approaches. In [17], a Bayesian learning method is employed to optimize the CNN architecture in the human activity classification with mD radar signals. In this paper, considering the characterization of time-frequency representation of human mD signals, a FPGA based CNN acceleration method is proposed for the human activity classification by using mmW mD signals. The baseband echoes from the complete human body micro-motion could be approximated as ybase (t) ≈ T

Ai exp j

Lj Kj Kj

Findings

CONCLUSION