Training neural encoders requires a differentiable channel model for backpropagation. This can be bypassed by approximating the channel distribution using pilot signals. A common method for this is the use of generative adversarial networks (GANs). In this paper, we introduce diffusion models (DMs) for channel generation and propose an efficient training algorithm. Our DMs provide a solution that achieves near-optimal end-to-end symbol error rates (SERs). Importantly, DMs outperform GANs in high signal-to-noise ratio regions. Here, in particular, we explore the trade-off between sample quality and speed. We also show that the right noise scheduling can significantly reduce sampling time with a minor increase in SER.

Integrated sensing and communications (ISAC) is envisioned to be a key to advanced applications in future wireless networks. In this paper, we study the coded modulation designs for ISAC transmissions with short block lengths over correlated Rayleigh fading channels. In line with the short block length transmission, we consider the non-coherent communication detection and coherent radar sensing, where a neural network (NN)-assisted frame-wise constellation design is proposed. Specifically, we first derive the optimal communication and radar receivers. Then, we present some heuristic understandings of the code designs by considering special cases, based on which a conjecture on the optimal codes for the considered ISAC transmissions is developed. The constellation obtained from the proposed NN agrees with our conjecture and shows an important conclusion that the optimal codes of the considered problem may be a combination of the “on-off keying” and phase-shifted keying signalings. Our numerical results show that the proposed code exhibits promising communication and sensing performance simultaneously and outperforms the transmissions with a standard channel code and symbol-wise modulation.

Federated learning (FL) enables large-scale machine learning with user data privacy due to its decentralized structure. However, the user data can still be inferred via the shared model updates. To strengthen the privacy, we consider FL with local differential privacy (LDP). One of the challenges in FL is its huge communication cost caused by iterative transmissions of model updates. It has been relieved by quantization in the literature, however, there have been not many works that consider its effect on LDP and the unboundedness of the randomized model updates. We propose a communication-efficient FL algorithm with LDP that uses a Gaussian mechanism followed by quantization and the Elias-gamma coding. A novel design of the algorithm guarantees LDP even after the quantization. Under the proposed algorithm, we provide a trade-off analysis of privacy and communication costs theoretically: quantization reduces the communication costs but requires a larger perturbation to enable LDP. Experimental results show that the accuracy is mostly affected by the noise from LDP mechanisms, and it becomes enhanced when the quantization error is larger. Nonetheless, our experimental results enabled LDP with a significant compression ratio and only a slight reduction of accuracy in return. Furthermore, the proposed algorithm outperforms another algorithm with a discrete Gaussian mechanism under the same privacy budget and communication costs constraints in the experiments.

We consider a set of security and privacy problems under reliability and storage constraints that can be tackled by using codes and particularly focus on the secret-key agreement problem. Polar subcodes (PSCs) are polar codes (PCs) with dynamically-frozen symbols and have a larger code minimum distance than PCs with only statically-frozen symbols. A randomized nested PSC construction, where the low-rate code is a PSC and the high-rate code is a PC, is proposed for successive cancellation list (SCL) and sequential decoders. This code construction aims to perform lossy compression with side information, i.e., Wyner-Ziv (WZ) coding. Nested PSCs are used in the key agreement problem with physical identifiers and two terminals since WZ-coding constructions significantly improve on Slepian-Wolf coding constructions such as fuzzy extractors. Significant gains in terms of the secret-key vs. storage rate ratio as compared to nested PCs with the same list sizes are illustrated to show that nested PSCs significantly improve on all existing code constructions. The performance of the nested PSCs is shown to improve with larger list sizes, unlike the nested PCs considered. A design procedure to efficiently construct nested PSCs and possible improvements to the nested PSC designs are also provided.

Federated learning (FL) allows to train a massive amount of data privately due to its decentralized structure. Stochastic gradient descent (SGD) is commonly used for FL due to its good empirical performance, but sensitive user information can still be inferred from weight updates shared during FL iterations. We consider Gaussian mechanisms to preserve local differential privacy (LDP) of user data in the FL model with SGD. The trade-offs between user privacy, global utility, and transmission rate are proved by defining appropriate metrics for FL with LDP. Compared to existing results, the query sensitivity used in LDP is defined as a variable, and a tighter privacy accounting method is applied. The proposed utility bound allows heterogeneous parameters over all users. Our bounds characterize how much utility decreases and transmission rate increases if a stronger privacy regime is targeted. Furthermore, given a target privacy level, our results guarantee a significantly larger utility and a smaller transmission rate as compared to existing privacy accounting methods.

Coded distributed computing has been considered as a promising technique which makes large-scale systems robust to the "straggler" workers. Yet, practical system models for distributed computing have not been available that reflect the clustered or grouped structure of real-world computing servers. Also, the large variations in the computing power and bandwidth capabilities across different servers have not been properly modeled. We suggest a group-based model to reflect practical conditions and develop an appropriate coding scheme for this model. The suggested code, called group code, employs parallel encoding for each group. We show that the suggested coding scheme can asymptotically achieve optimal computing time in the regime of infinite n, the number of workers. While theoretical analysis is conducted in the asymptotic regime, numerical results also show that the suggested scheme achieves near-optimal computing time for any finite but reasonably large n. Moreover, we demonstrate that decoding complexity of the suggested scheme is significantly reduced by the virtue of parallel decoding.