Robust deep learning-based algorithm for automatic modulation classification
This dissertation provides a comprehensive analysis of deep learning-based Automatic Modulation Classification (AMC) algorithms. AMC is a method employed to determine the modulation types of unknown signals and is a crucial step in demodulation. In non-collaborative communication environments, many...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis-Master by Coursework |
| Language: | English |
| Published: |
Nanyang Technological University
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/181408 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | This dissertation provides a comprehensive analysis of deep learning-based Automatic Modulation Classification (AMC) algorithms. AMC is a method employed to determine the modulation types of unknown signals and is a crucial step in demodulation. In non-collaborative communication environments, many parameters of the received signals are uncertain and must be determined through AMC algorithms to ascertain the modulation scheme of the received signal. Consequently, accurately identifying modulation signals with limited parameters poses a significant challenge. Traditional AMC methods rely on manually extracted features, which not only entails considerable labor and computational complexity but also faces substantial limitations in accuracy. Recently, the continuous progress of deep learning, characterized by the elimination of manual feature extraction and the use of self-learning mechanisms within networks, has demonstrated exceptional performance. |
|---|