The past few years have witnessed the flourishing of large-scale deep neural network models with ever-growing parameter numbers. Training such large-scale models typically requires massive memory and computing resources, necessitating distributed training. As GPU performance has rapidly evolved in recent years, computation time has shrunk, making communication a larger portion of the overall training time. Consequently, optimizing communication for distributed training has become crucial. In this article, we briefly introduce the general architecture of distributed deep neural network training and analyze relationships among Parallelization Strategy, Collective Communication Library, and Network from the perspective of communication optimization, which forms a three-layer paradigm. We then review current representative research advances within this three-layer paradigm. We find that layers in the current three-layer paradigm are relatively independent and there is a rich design space for cross-layer collaborative optimization in distributed training scenarios. Therefore, we advocate “Vertical” and “Horizontal” co-designs which extend the three-layer paradigm to a five-layer paradigm. We also advocate “Intra-Inter” and “Host-Net” co-designs to further utilize the potential of heterogeneous resources. We hope this article can shed some light on future research on communication optimization for distributed training.

The advent of softwarization and disaggregated architectures has transformed modern communication networks, sparking innovation by separating network functionalities from the underlying hardware. Following this trend, in future quantum networks, the virtualization and softwarization of critical components will be essential to achieve global interoperability and enhance adaptability. Virtualization allows for the abstraction of hardware resources, making it easier to manage and scale networks dynamically, while softwarization promotes flexibility, reconfigurability, and interoperability by enabling the use of standardized, software-defined protocols that can be easily updated and modified to work across diverse and evolving network environments. In this paper, we introduce and examine an operational model for QKD networks that leverages the virtualization of control and key management functionalities. We detail its elements and procedures to optimize QKD services. The design paves the way for the integration of quantum network with functions already established in modern mobile networks while adhering to established QKD network standards. The proposed operational model has been validated at the 5G Telefonica Open Network Innovation Centre (5TONIC) using an environment that deploys digital twins of QKD networks.

In the software rich environment of 6G, systems will be surrounded by edge devices that support distributed software systems which are critical to operations. Such systems may also be subject to frequent updates or uploads of individual software components. Trust in such systems will therefore depend on our ability to rapidly ensure that such software is not vulnerable to cyberattacks or malicious compromises. Thus this paper presents a novel System-Wide Vulnerability Assessment (SWVA) framework based on Machine Learning, that can be frequently activated to assess the vulnerability of interconnected software components over edge systems. The performance of the SWVA framework is illustrated by assessing the vulnerability of 13 versions of a real-world 11 component software system, and comparing the ARNN results against the well-known ML models MLP, KNN, and Lasso. The results show the superior performance of SWVA, offering over 85% median accuracy and good scalability as the number of connected software components increases.

The complexity and heterogeneity of the Sixth Generation (6G) networks raise trust issues among network users. Implementing blockchain-based decentralized trust management in 6G networks ensures user trustworthiness while mitigating the risk of single-point failures. However, the throughput limitation of the blockchain restricts the efficiency of recording trust data, leading to the inability to promptly expose trustless users in the system. To address this issue, this article proposes a Multi-State Consensus (MSC), a blockchain-based trust management scheme that incorporates three novel consensus protocols. These three novel consensus protocols are utilized in different network states to efficiently record the trust data of trustless users, thereby exposing them more quickly. Additionally, in MSC, new users gain initial trust values through the multi-signatures of blockchain miners, thereby addressing the trust value cold-start problem. Experimental results demonstrate that MSC not only exposes trustless users more rapidly but also ensures the robustness of network operations.

Generative Artificial Intelligence (GAI) has recently emerged as a promising solution to address critical challenges of blockchain technology, including scalability, security, privacy, and interoperability. In this paper, we first introduce GAI techniques, outline their applications, and discuss existing solutions for integrating GAI into blockchains. Then, we discuss emerging solutions that demonstrate the effectiveness of GAI in addressing various challenges of blockchain, such as detecting unknown blockchain attacks and smart contract vulnerabilities, designing key secret sharing schemes, and enhancing privacy. Moreover, we present a case study to demonstrate that GAI, specifically the generative diffusion model, can be employed to optimize blockchain network performance metrics. Experimental results clearly show that, compared to a baseline traditional AI approach, the proposed generative diffusion model approach can converge faster, achieve higher rewards, and significantly improve the throughput and latency of the blockchain network. Additionally, we highlight future research directions for GAI in blockchain applications, including personalized GAI-enabled blockchains, GAI-blockchain synergy, and privacy and security considerations within blockchain ecosystems.

Deep learning-based models demonstrate a remarkable level of accuracy in network traffic identification. However, the black-box nature of neural networks often makes the identification results difficult to explain. Although some eXplainable Artificial Intelligence (XAI) methods have been applied to network traffic identification, most of them focus on model explainability and do not provide sufficient credibility. In emerging network systems that use proprietary protocols, low-credibility malicious traffic detection can result in severe consequences. Therefore, it is imperative to deeply understand network traffic features and trust the detection results. In this paper, we propose an explainable architecture for emerging network systems. This architecture enhances the explainability of malicious traffic detection from both input and output perspectives, aiming to understand network traffic data and improve the reliability of the results. The effectiveness of explaining inputs and outputs is verified through experimental analysis in case studies. Furthermore, we review the research on explainable models in the field of network traffic identification and summarize research opportunities.

The development of 6G mobile networks will produce many smart Internet of Things devices and data at the network’s edge. Advanced AI technology holds the potential to enable 6G mobile networks to collect and analyze this data for innovative applications and intelligent services. However, the inherent privacy constraints and limited communication resources within 6G mobile networks often make direct data transmission to servers undesirable. Federated learning (FL) is seen as a promising approach to address these problems. Yet, integrating FL into 6G mobile networks presents data heterogeneity issues. In this article, we design and propose a privacy-preserving FL scheme in 6 G mobile networks. The core idea of this scheme is to take both dataset size and the discrepancy between local and global category distributions into consideration to compute the weights of different clients and apply threshold Paillier cryptosystem to perform weighted aggregation on client-encrypted data. Security analysis and experimental results demonstrate the advantages of this scheme in guaranteeing privacy preservation and improving training accuracy. Finally, we present some future directions for the integration of FL and 6G mobile networks.