arXiv cs.AI· Thanh Le, Hai Duong, Takeshi Matsumura, ThanhVu Nguyen·· 9 天前AI 评分22
DeepJSCC 神经网络验证：首个有界传播框架
Neural Network Verification for Deep Joint Source-Channel Coding
AI 导读
研究者提出首个针对 DeepJSCC 解码器的有界传播验证框架，可在给定无线信道噪声区域内界定最坏情况重建误差。该框架扩展了 PReLU、转置卷积与 Rayleigh 衰落的验证技术，并结合 GloRo 全局鲁棒性训练，在图像传输模型上将中位认证界降低最多 41%，认证安全案例数从 19 增至 192。OFDM 软件无线电实测中，认证界为 0.128，链路最差观测误差为 0.082。
正文 · 原文
Abstract:Deep joint source-channel coding (DeepJSCC) transmits data end-to-end over wireless channels using a neural encoder-decoder, but reconstruction quality can degrade sharply under adversarial perturbations and channel disturbances; no method formally bounds this degradation for DeepJSCC. We present the first bound-propagation framework for verifying DeepJSCC's decoder, bounding worst-case reconstruction error over a given wireless channel's noise region. Current deep neural network (DNN) verifiers do not support three DeepJSCC decoder components: parametric rectified linear activations (PReLU), transposed convolutions, and Rayleigh fading. We extend state-of-the-art techniques for optimization of linear relaxation in DNN verification for PReLU, replace the transposed convolution with its restricted upsample-then-convolution form, and formulate Rayleigh fading as a structural perturbation prepended directly into the decoder, thereby reducing the dimensionality of the verification problem. We also instantiate Lipschitz-regularized global robustness training, denoted GloRo, improving global robustness and enabling tight certification of DeepJSCC models for the first time. On DeepJSCC model for image transmission, this global robustness training procedure combined with structural encoding lowers the median certified bound by up to 41% and certifies about ten times more safe cases (192 against 19) than GloRo with interval encoding at a 10-degree error in channel estimation. Over-the-air validation with an orthogonal frequency-division multiplexing (OFDM) implementation on software-defined radio devices confirm the certificate holds on real hardware, with a worst observed error on radio link at 0.082 against a certified bound of 0.128.
| Subjects: | Software Engineering (cs.SE); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.11994 [cs.SE] |
| (or arXiv:2610.11994v1 [cs.SE] for this version) | |
| https://doi.org/10.48550/arXiv.2610.11994 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Hai Duong [view email]
[v1]
Thu, 8 Oct 2026 14:02:49 UTC (1,052 KB)
来源：arXiv cs.AI · arxiv.org