Deep Joint Source-Channel Coding for Multi-Task Network

Multi-task learning (MTL) is an efficient way to improve the performance of related tasks by sharing knowledge. However, most existing MTL networks run on a single end and are not suitable for collaborative intelligence (CI) scenarios. In this work, we propose an MTL network with a deep joint source...

Full description

Saved in:
Bibliographic Details
Published in:IEEE signal processing letters Vol. 28; pp. 1973 - 1977
Main Authors: Wang, Mengyang, Zhang, Zhicong, Li, Jiahui, Ma, Mengyao, Fan, Xiaopeng
Format: Journal Article
Language:English
Published: New York IEEE 2021
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Coding
Collaborative intelligence
deep joint source-channel coding
Edge computing
Electronic devices
Feature extraction
Mobile handsets
multi-task learning
Networks
Object detection
Object recognition
Performance enhancement
Semantic segmentation
Semantics
Servers
Task analysis
Training
ISSN:1070-9908, 1558-2361
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Multi-task learning (MTL) is an efficient way to improve the performance of related tasks by sharing knowledge. However, most existing MTL networks run on a single end and are not suitable for collaborative intelligence (CI) scenarios. In this work, we propose an MTL network with a deep joint source-channel coding (JSCC) framework, which allows operating under CI scenarios. We first propose a feature fusion based MTL network (FFMNet) for joint object detection and semantic segmentation. Compared with other MTL networks, FFMNet gets higher performance with fewer parameters. Then FFMNet is split into two parts, which run on a mobile device and an edge server respectively. The feature generated by the mobile device is transmitted through the wireless channel to the edge server. To reduce the transmission overhead of the intermediate feature, a deep JSCC network is designed. By combining two networks together, the whole model achieves 512× compression for the intermediate feature and a performance loss within 2% on both tasks. At last, by training with noise, the FFMNet with JSCC is robust to various channel conditions and outperforms the separate source and channel coding scheme.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1070-9908
1558-2361
DOI:10.1109/LSP.2021.3113827