| 111 | 0 | 2363 |
| 下载次数 | 被引频次 | 阅读次数 |
作为实现语义无线通信的关键技术之一，端到端联合源信道编码(Joint Source-Channel Coding, JSCC)取得了显著进展。JSCC严重依赖于深度学习模型，这些复杂的神经网络训练通常需要大量的计算资源，对于资源受限的平台，如移动或嵌入式设备，提出了重大挑战。通过介绍一种轻量级的语义无线图像传输方法——MobileJSCC,旨在显著降低计算复杂度，同时不降低通信质量，便于在移动或嵌入式系统上部署。实验验证了MobileJSCC在各种图像分辨率、信噪比和信道条件下，实现了与先进的JSCC方法相当的图像恢复性能，大大减少了网络模型所需的计算量和参数。
Abstract：As one of the key technologies for realizing semantic wireless communication, end-to-end Joint Source-Channel Coding(JSCC) has made significant progress in recent years. JSCC relies heavily on deep learning models, and the training of these complex neural networks often demands substantial computational resources. This presents a significant challenge for resource constrained platforms, such as mobile or embedded devices. A lightweight semantic wireless communication network, namely MobileJSCC, is introduced to significantly reduce computational complexity without compromising communication quality, thereby facilitating deployment on mobile or embedded systems. Experiments have validated that MobileJSCC achieves image recovery performance comparable to that of state-of-the-art JSCC methods under various image resolutions, signal to noise ratios, and channel conditions, while greatly decreasing the computational operations and parameters required by the network model.
[1] 张平，牛凯，姚圣时，等.面向未来的语义通信：基本原理与实现方法[J].通信学报，2023,44(5):1-14.
[2] 高静，冯金顺，董少然，等.小样本深度学习在目标识别分类领域的应用前景研究[J].无线电工程，2024,54(11):2594-2601.
[3] 秦志金，赵菼菼，李凡，等.多模态语义通信研究综述[J].通信学报，2023,44(5):28-41.
[4] YANG W T,DU H Y,LIEW Z Q,et al.Semantic Communications for Future Internet:Fundamentals,Applications,and Challenges[J].IEEE Communications Surveys & Tutorials,2023,25(1):213-250.
[5] ZENG C,WANG J B,XIAO M,et al.Task-oriented Semantic Communication over Rate Splitting Enabled Wireless Control Systems for URLLC Services[J].IEEE Transactions on Communications,2024,72(2):722-739.
[6] FU Q F,XIE H Q,QIN Z J,et al.Vector Quantized Semantic Communication System[J].IEEE Wireless Communications Letters,2023,12(6):982-986.
[7] FARSAD N,RAO M,GOLDSMITH A.Deep Learning for Joint Source-channel Coding of Text[C]//2018 IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Calgary:IEEE,2018:2326-2330.
[8] HU Q Y,ZHANG G Y,QIN Z J,et al.Robust Semantic Communications with Masked VQ-VAE Enabled Codebook[J].IEEE Transactions on Wireless Communications,2023,22(12):8707-8722.
[9] YILMAZ S F,OZYILKAN E,GüNDüZ D,et al.Distributed Deep Joint Source-channel Coding with Decoder-only Side Information[C]//2024 IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN).Stockholm:IEEE,2024:139-144.
[10] BOURTSOULATZE E,KURKA D B,GüNDüZ D.Deep Joint Source-channel Coding for Wireless Image Transmission[J].IEEE Transactions on Cognitive Communications & Networking,2019,5(3):567-579.
[11] WU H T,SHAO Y L,MIKOLAJCZYK K,et al.Channel-adaptive Wireless Image Transmission with OFDM[EB/OL].(2022-05-05)[2024-10-06].https://arxiv.org/abs/2205.02417.
[12] YANG K,WANG S X,DAI J C,et al.SwinJSCC:Taming Swin Transformer for Deep Joint Source-channel Coding[J].IEEE Transactions on Cognitive Communications and Networking,2024,11(1):90-104.
[13] YI P,CAO Y,KANG X,et al.Deep Learning-empowered Semantic Communication Systems with a Shared Knowledge Base[J].IEEE Transactions on Wireless Communications,2024,23(6):6174-6187.
[14] NAN G S,LI Z C,ZHAI J L,et al.Physical-layer Adversarial Robustness for Deep Learning-based Semantic Communications[J].IEEE Journal on Selected Areas in Communications,2023,41(8):2592-2608.
[15] ZHANG G Y,HU Q Y,CAI Y L,et al.Alleviating Distortion Accumulation in Multi-hop Semantic Communication[J].IEEE Communications Letters,2024,28(2):308-312.
[16] PENG J C,XING H L,LI Y,et al.Task-oriented Multi-user Semantic Communication with Lightweight Semantic Encoder and Fast Training for Resource-constrained Terminal Devices[J].IEEE Wireless Communications Letters,2024,13(9):2427-2431.
[17] JIA Y J,HUANG Z,LUO K,et al.Lightweight Joint Source-channel Coding for Semantic Communications[J].IEEE Communications Letters,2023,27(12):3161-3165.
[18] HE Y,XIAO L G.Structured Pruning for Deep Convolutional Neural Networks:A Survey[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2024,46(5):2900-2919.
[19] KRIZHEVSKY A.Learning Multiple Layers of Features from Tiny Images[EB/OL].(2009-04-08)[2024-11-20].http://www.cs.toronto.edu/～kriz/learning-features-2009-TR.pdf.
[20] SANDLER M,HOWARD A,ZHU M L,et al.MobileNetV2:Inverted Residuals and Linear Bottlenecks[C]//2018 IEEE Conference on Computer Vision and Pattern Recognition.Salt Lake City:IEEE,2018:4510-4520.
[21] DENG J,DONG W,SOCHER R,et al.ImageNet:A Large-scale Hierarchical Image Database[C]//2009 IEEE Conference on Computer Vision and Pattern Recognition.Miami:IEEE,2009:248-255.
[22] ZHAO H M,LIAO P Y.Kodak PhotoCD Dataset[Z/OL].[2024-11-20].https://service.tib.eu/ldmservice/dataset/kodak-photocd-dataset.
[23] LYU Z H,ZHU G X,XU J,et al.Semantic Communications for Image Recovery and Classification via Deep Joint Source and Channel Coding[J].IEEE Transactions on Wireless Communications,2024,23(8):8388-8404.
[24] WU H T,SHAO Y L,BIAN C H,et al.Deep Joint Source-channel Coding for Adaptive Image Transmission over MIMO Channels[J].IEEE Transactions on Wireless Communications,2024,23(10):15002-15017.
[25] ERDEMIR E,TUNG T Y,DRAGOTTI P L,et al.Generative Joint Source-channel Coding for Semantic Image Transmission[J].IEEE Journal on Selected Areas in Communications,2023,41(8):2645-2657.
[26] HUANG J H,LI D X,HUANG C,et al.Joint Task and Data-oriented Semantic Communications:A Deep Separate Source-channel Coding Scheme[J].IEEE Internet of Things Journal,2024,11(2):2255-2272.
基本信息:
中图分类号:TN919.8
引用信息:
[1]黄军韬,赵响,香晏.一种适用于资源受限平台的语义无线图像传输方法[J].无线电工程,2025,55(07):1422-1430.
基金信息:
国家自然科学基金地区科学基金项目(61961007)~~
2025-04-08
2025-04-08
2025-04-08