상세 보기
SR-DeepSC-Pro: Progressive Curriculum Learning with Adversarial Robustness for Satellite Semantic Communication
- Do, Tung Son;
- Thanh Phung Truong;
- Quang Tuan Do;
- Cho, Sungrae
WEB OF SCIENCE
0초록
We present SR-DeepSC-Pro, an advanced super-resolution semantic communication system for bandwidthconstrained satellite image transmission. Through a novel two-stage training algorithm combining curriculum learning and adversarial robustness, our system achieves 10.2% higher PSNR (30.33 dB) while reducing satellite-side parameters by 26% and processing latency by 28.9%. SR-DeepSCPro maintains robust performance across SNR conditions with strongest gains at challenging low-SNR levels. This work demonstrates that advanced training strategies can overcome traditional quality-efficiency trade-offs, producing models that simultaneously excel in reconstruction performance and computational efficiency-critical advances for next-generation satellite communication systems.
키워드
- 제목
- SR-DeepSC-Pro: Progressive Curriculum Learning with Adversarial Robustness for Satellite Semantic Communication
- 저자
- Do, Tung Son; Thanh Phung Truong; Quang Tuan Do; Cho, Sungrae
- 발행일
- 2025-11
- 유형
- Proceedings Paper
- 저널명
- 2025 30TH ASIA-PACIFIC CONFERENCE ON COMMUNICATIONS, APCC
- 언어
- ENG
- 출판사
- IEEE
- 발행국가
- 미국
- ISSN
- P 2163-0771