arXiv subjects

Yulei Wu

Publications and source records attributed to Yulei Wu.

17 recordsLinked to original sources

arXiv2026

Deadline-Driven Hierarchical Agentic Resource Sharing for AI Services and RAN Functions in AI-RAN

AI-RAN consolidates AI services and Radio Access Network (RAN) functions onto a unified, GPU-accelerated infrastructure at the network edge. However, compute sharing between real-time RAN functions and highly heterogeneous AI services requires coordination of scheduling decisions at mismatched timescales, and placement adaptation may require service migration across nodes with non-negligible interruptions. This paper proposes a hierarchical agentic framework (HAF) for compute sharing in AI-RAN that combines a large language model (LLM)-based agent for slow-timescale placement of AI services and RAN functions with a closed-form, deadline-aware convex algorithm for fast-timescale GPU/CPU allocation. The LLM agent is further equipped with a predictive critic that filters out migrations when the induced service interruption outweighs the expected service-level objective (SLO) benefit. Experimental results show that HAF reaches 90.0% overall SLO fulfillment, a 20.5% improvement over the strongest baseline, and raises AI service request fulfillment from 51% to 85.3%. Further evaluations show that HAF retains its advantage under diverse load conditions, while the critic consistently improves SLO fulfillment across multiple open-source LLM agents.

cs.DC↗

arXiv2026

Multi-Turn Distributed Inference with Mixture of Experts for 6G Edge--Cloud Networks

Mixture-of-Experts (MoE) architectures are increasingly deployed across 6G edge--cloud networks, where sparse activation reduces the computational footprint of each inference to only a fraction of the full expert set. However, MoE inference in edge-cloud networks creates a tension between KV state locality and elastic expert dispatch. KV state relocation incurs substantial transfer overhead, while expert computation benefits from spreading across the network to exploit available capacity. This tension is amplified in multi-turn inference, where each turn extends the KV state that must persist across the dialogue. To this end, we present StateFlow, a distributed inference policy that decouples persistent KV state from transient sparse computation. StateFlow pins KV state at a sticky serving site for cross-turn reuse and jointly optimizes expert dispatch and aggregation placement across the network. We further implement a real-world testbed with kernel-level network emulation and experimental results show that StateFlow sustains more than $2\times$ higher stable dialogue concurrency than the distributed baseline solutions and reduces turn-level p95 latency under multi-turn inference by 53.0\%.

cs.DC↗

arXiv2026

Multi-Agentic AI for Conflict-Aware rApp Policy Orchestration in Open RAN

Open Radio Access Network (RAN) enables flexible, AI-driven control of mobile networks through disaggregated, multi-vendor components. In this architecture, xApps handle real-time functions, whereas rApps in the non-real-time controller generate strategic policies. However, current rApp development remains largely manual, brittle, and poorly scalable as xApp diversity proliferates. In this work, we propose a Multi-Agentic AI framework to automate rApp policy generation and orchestration. The architecture integrates three specialized large language model (LLM)-based agents, Perception, Reasoning, and Refinement, supported by retrieval-augmented generation (RAG) and memory-based analogical reasoning. These agents collectively analyze potential conflicts, synthesize intent-aligned control pipelines, and incrementally refine deployment decisions. Experiments across diverse deployment scenarios demonstrate that the proposed system achieves over 70% improvement in deployment accuracy and 95% reduction in reasoning cost compared to baseline methods, while maintaining zero-shot generalization to unseen intents. These results establish a scalable and conflict-aware solution for fully autonomous, zero-touch rApp orchestration in Open RAN.

eess.SY↗

arXiv2026

Multi-Agentic AI for Fairness-Aware and Accelerated Multi-modal Large Model Inference in Real-world Mobile Edge Networks

Generative AI (GenAI) has transformed applications in natural language processing and content creation, yet centralized inference remains hindered by high latency, limited customizability, and privacy concerns. Deploying large models (LMs) in mobile edge networks emerges as a promising solution. However, it also poses new challenges, including heterogeneous multi-modal LMs with diverse resource demands and inference speeds, varied prompt/output modalities that complicate orchestration, and resource-limited infrastructure ill-suited for concurrent LM execution. In response, we propose a Multi-Agentic AI framework for latency- and fairness-aware multi-modal LM inference in mobile edge networks. Our solution includes a long-term planning agent, a short-term prompt scheduling agent, and multiple on-node LM deployment agents, all powered by foundation language models. These agents cooperatively optimize prompt routing and LM deployment through natural language reasoning over runtime telemetry and historical experience. To evaluate its performance, we further develop a city-wide testbed that supports network monitoring, containerized LM deployment, intra-server resource management, and inter-server communications. Experiments demonstrate that our solution reduces average latency by over 80% and improves fairness (Normalized Jain index) to 0.90 compared to other baselines. Moreover, our solution adapts quickly without fine-tuning, offering a generalizable solution for optimizing GenAI services in edge environments.

eess.SY↗

arXiv2025

Lifecycle Management of Trustworthy AI Models in 6G Networks: The REASON Approach

Artificial Intelligence (AI) is expected to play a key role in 6G networks including optimising system management, operation, and evolution. This requires systematic lifecycle management of AI models, ensuring their impact on services and stakeholders is continuously monitored. While current 6G initiatives introduce AI, they often fall short in addressing end-to-end intelligence and crucial aspects like trust, transparency, privacy, and verifiability. Trustworthy AI is vital, especially for critical infrastructures like 6G. This paper introduces the REASON approach for holistically addressing AI's native integration and trustworthiness in future 6G networks. The approach comprises AI Orchestration (AIO) for model lifecycle management, Cognition (COG) for performance evaluation and explanation, and AI Monitoring (AIM) for tracking and feedback. Digital Twin (DT) technology is leveraged to facilitate real-time monitoring and scenario testing, which are essential for AIO, COG, and AIM. We demonstrate this approach through an AI-enabled xAPP use case, leveraging a DT platform to validate, explain, and deploy trustworthy AI models.

cs.NI↗

arXiv2024

Towards Practical Operation of Deep Reinforcement Learning Agents in Real-World Network Management at Open RAN Edges

Deep Reinforcement Learning (DRL) has emerged as a powerful solution for meeting the growing demands for connectivity, reliability, low latency and operational efficiency in advanced networks. However, most research has focused on theoretical analysis and simulations, with limited investigation into real-world deployment. To bridge the gap and support practical DRL deployment for network management, we first present an orchestration framework that integrates ETSI Multi-access Edge Computing (MEC) with Open RAN, enabling seamless adoption of DRL-based strategies across different time scales while enhancing agent lifecycle management. We then identify three critical challenges hindering DRL's real-world deployment, including (1) asynchronous requests from unpredictable or bursty traffic, (2) adaptability and generalization across heterogeneous topologies and evolving service demands, and (3) prolonged convergence and service interruptions due to exploration in live operational environments. To address these challenges, we propose a three-fold solution strategy: (a) advanced time-series integration for handling asynchronized traffic, (b) flexible architecture design such as multi-agent DRL and incremental learning to support heterogeneous scenarios, and (c) simulation-driven deployment with transfer learning to reduce convergence time and service disruptions. Lastly, the feasibility of the MEC-O-RAN architecture is validated on an urban-wide testing infrastructure, and two real-world use cases are presented, showcasing the three identified challenges and demonstrating the effectiveness of the proposed solutions.

cs.NI↗

arXiv2024

Reasoning AI Performance Degradation in 6G Networks with Large Language Models

The integration of Artificial Intelligence (AI) within 6G networks is poised to revolutionize connectivity, reliability, and intelligent decision-making. However, the performance of AI models in these networks is crucial, as any decline can significantly impact network efficiency and the services it supports. Understanding the root causes of performance degradation is essential for maintaining optimal network functionality. In this paper, we propose a novel approach to reason about AI model performance degradation in 6G networks using the Large Language Models (LLMs) empowered Chain-of-Thought (CoT) method. Our approach employs an LLM as a ''teacher'' model through zero-shot prompting to generate teaching CoT rationales, followed by a CoT ''student'' model that is fine-tuned by the generated teaching data for learning to reason about performance declines. The efficacy of this model is evaluated in a real-world scenario involving a real-time 3D rendering task with multi-Access Technologies (mATs) including WiFi, 5G, and LiFi for data transmission. Experimental results show that our approach achieves over 97% reasoning accuracy on the built test questions, confirming the validity of our collected dataset and the effectiveness of the LLM-CoT method. Our findings highlight the potential of LLMs in enhancing the reliability and efficiency of 6G networks, representing a significant advancement in the evolution of AI-native network infrastructures.

cs.NI↗

arXiv2024

AI Model Placement for 6G Networks under Epistemic Uncertainty Estimation

The adoption of Artificial Intelligence (AI) based Virtual Network Functions (VNFs) has witnessed significant growth, posing a critical challenge in orchestrating AI models within next-generation 6G networks. Finding optimal AI model placement is significantly more challenging than placing traditional software-based VNFs, due to the introduction of numerous uncertain factors by AI models, such as varying computing resource consumption, dynamic storage requirements, and changing model performance. To address the AI model placement problem under uncertainties, this paper presents a novel approach employing a sequence-to-sequence (S2S) neural network which considers uncertainty estimations. The S2S model, characterized by its encoding-decoding architecture, is designed to take the service chain with a number of AI models as input and produce the corresponding placement of each AI model. To address the introduced uncertainties, our methodology incorporates the orthonormal certificate module for uncertainty estimation and utilizes fuzzy logic for uncertainty representation, thereby enhancing the capabilities of the S2S model. Experiments demonstrate that the proposed method achieves competitive results across diverse AI model profiles, network environments, and service chain requests.

cs.NI↗

arXiv2024

Federated Analytics for 6G Networks: Applications, Challenges, and Opportunities

Extensive research is underway to meet the hyper-connectivity demands of 6G networks, driven by applications like XR/VR and holographic communications, which generate substantial data requiring network-based processing, transmission, and analysis. However, adhering to diverse data privacy and security policies in the anticipated multi-domain, multi-tenancy scenarios of 6G presents a significant challenge. Federated Analytics (FA) emerges as a promising distributed computing paradigm, enabling collaborative data value generation while preserving privacy and reducing communication overhead. FA applies big data principles to manage and secure distributed heterogeneous networks, improving performance, reliability, visibility, and security without compromising data confidentiality. This paper provides a comprehensive overview of potential FA applications, domains, and types in 6G networks, elucidating analysis methods, techniques, and queries. It explores complementary approaches to enhance privacy and security in 6G networks alongside FA and discusses the challenges and prerequisites for successful FA implementation. Additionally, distinctions between FA and Federated Learning are drawn, highlighting their synergistic potential through a network orchestration scenario.

eess.SY↗

arXiv2023

General Phrase Debiaser: Debiasing Masked Language Models at a Multi-Token Level

The social biases and unwelcome stereotypes revealed by pretrained language models are becoming obstacles to their application. Compared to numerous debiasing methods targeting word level, there has been relatively less attention on biases present at phrase level, limiting the performance of debiasing in discipline domains. In this paper, we propose an automatic multi-token debiasing pipeline called \textbf{General Phrase Debiaser}, which is capable of mitigating phrase-level biases in masked language models. Specifically, our method consists of a \textit{phrase filter stage} that generates stereotypical phrases from Wikipedia pages as well as a \textit{model debias stage} that can debias models at the multi-token level to tackle bias challenges on phrases. The latter searches for prompts that trigger model's bias, and then uses them for debiasing. State-of-the-art results on standard datasets and metrics show that our approach can significantly reduce gender biases on both career and multiple disciplines, across models with varying parameter sizes.

cs.CL↗

arXiv2023

STATGRAPH: Effective In-vehicle Intrusion Detection via Multi-view Statistical Graph Learning

In-vehicle network (IVN) is facing complex external cyber-attacks, especially the emerging masquerade attacks with extremely high difficulty of detection while serious damaging effects. In this paper, we propose the STATGRAPH, which is an effective and fine-grained intrusion detection methodology for IVN security services via multi-view statistical graph learning on in-vehicle controller area network (CAN) messages with insight into their variations in periodicity, payload and signal combinations. Specifically, STATGRAPH generates two statistical graphs, timing correlation graph (TCG) and coupling relationship graph (CRG), in every CAN message detection window, where edge attributes in TCGs represent temporal correlation between different message IDs while edge attributes in CRGs denote the neighbour relationship and contextual similarity. Besides, a lightweight shallow layered graph convolution network is trained based on graph property of TCGs and CRGs, which learns the universal laws of various patterns more effectively and further enhance the performance of detection. To address the problem of insufficient attack types in previous intrusion detection, we select two real in-vehicle CAN datasets covering five new instances of sophisticated and stealthy masquerade attacks that are never investigated before. Experimental result shows STATGRAPH improves both detection granularity and detection performance over state-of-the-art intrusion detection methods. Code is available at https://github.com/wangkai-tech23/StatGraph.

cs.NI↗

arXiv2023

Incremental DRL-Based Resource Management for Dynamic Network Slicing in an Urban-Wide Testbed

Multi-access edge computing provides localized resources within mobile networks to address the requirements of emerging latency-sensitive and computing-intensive applications. At the edge, dynamic requests necessitate sophisticated resource management for adaptive network slicing. This involves optimizing resource allocations, scaling functions, and load balancing to utilize only essential resources under constrained network scenarios. However, existing solutions largely assume static slice counts, ignoring the re-optimization overhead associated with management algorithms when slices fluctuate. Moreover, many approaches rely on simplified energy models that overlook intertemporal resource scheduling and are predominantly evaluated through simulations, neglecting critical practical considerations. This paper presents an incremental cooperative Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm for resource management in dynamic edge slicing. The proposed approach optimizes long-term slicing benefits by reducing delay and energy consumption while minimizing retraining overhead in response to slice variations. Furthermore, we implement an urban-wide edge computing testbed based on OpenStack and Kubernetes to validate the algorithm's performance. Experimental results demonstrate that our incremental MADDPG method outperforms benchmark strategies in aggregated slicing utility and reduces training energy consumption by up to 50% compared to the re-optimization approach.

eess.SY↗

arXiv2022

Towards Zero Touch Networks: From the Perspective of Hierarchical Language Systems

With ever-increasing complexity and dynamicity of communication networks, intelligent network operation and maintenance has become more important to network operators. With the fast development of artificial intelligence, concepts such as "Zero Touch", "Intent-based", "Knowledge-defined" and "Self-driving" networks have become well-known in networking community for a great vision of making networks automatically manageable and responsive to user demands. This article discusses how to achieve Zero Touch Networks from the perspective of language-like systems. We propose a novel hierarchical `language' framework dedicated for networks to enable the Zero Touch Network, which covers from symbolizing network components, a unified framework for understanding network systems, to the logical description with network semantics. A case study based on the proposed language framework is provided. Finally, we discuss the challenges and open issues of intelligence models for zero touch networks.

cs.NI↗

arXiv2022

Knowledge-powered Explainable Artificial Intelligence (XAI) for Network Automation Towards 6G

Communication networks are becoming increasingly complex towards 6G. Manual management is no longer an option for network operators. Network automation has been widely discussed in the networking community, and it is a sensible means to manage the complex communication network. Deep learning models developed to enable network automation for given operation practices have the limitations of 1) lack of explainability and 2) inapplicable across different networks and/or network settings. To tackle the above issues, in this article we propose a new knowledge-powered framework that provides a human-understandable explainable artificial intelligence (XAI) agent for network automation. A case study of path selection is developed to demonstrate the feasibility of the proposed framework. Research on network automation is still in its infancy. Therefore, at the end of this article, we provide a list of challenges and open issues that can guide further research in this important area.

cs.NI↗

arXiv2021

Blockchain-empowered Edge Intelligence for Internet of Medical Things Against COVID-19

We have witnessed an unprecedented public health crisis caused by the new coronavirus disease (COVID-19), which has severely affected medical institutions, our common lives, and social-economic activities. This crisis also reveals the brittleness of existing medical services, such as over-centralization of medical resources, the hysteresis of medical services digitalization, and weak security and privacy protection of medical data. The integration of the Internet of Medical Things (IoMT) and blockchain is expected to be a panacea to COVID-19 attributed to the ubiquitous presence and the perception of IoMT as well as the enhanced security and immutability of the blockchain. However, the synergy of IoMT and blockchain is also faced with challenges in privacy, latency, and context-absence. The emerging edge intelligence technologies bring opportunities to tackle these issues. In this article, we present a blockchain-empowered edge intelligence for IoMT in addressing the COVID-19 crisis. We first review IoMT, edge intelligence, and blockchain in addressing the COVID-19 pandemic. We then present an architecture of blockchain-empowered edge intelligence for IoMT after discussing the opportunities of integrating blockchain and edge intelligence. We next offer solutions to COVID-19 brought by blockchain-empowered edge intelligence from 1) monitoring and tracing COVID-19 pandemic origin, 2) traceable supply chain of injectable medicines and COVID-19 vaccines, and 3) telemedicine and remote healthcare services. Moreover, we also discuss the challenges and open issues in blockchain-empowered edge intelligence.

cs.CR↗

arXiv2020

Blockchain-based Privacy Preservation for 5G-enabled Drone Communications

5G-enabled drones have potential applications in a variety of both military and civilian settings (e.g., monitoring and tracking of individuals in demonstrations and/or enforcing of social / physical distancing during pandemics such as COVID-19). Such applications generally involve the collection and dissemination of (massive) data from the drones to remote data centres for storage and analysis, for example via 5G networks. Consequently, there are security and privacy considerations underpinning 5G-enabled drone communications. We posit the potential of leveraging blockchain to facilitate privacy preservation, and therefore in this article we will review existing blockchain-based solutions after introducing the architecture for 5G-enabled drone communications and blockchain. We will also review existing legislation and data privacy regulations that need to be considered in the design of blockchain-based solutions, as well as identifying potential challenges and open issues which will hopefully inform future research agenda.

cs.NI↗

arXiv2019

Two Dimensional Router: Design and Implementation

Higher dimensional classification has attracted more attentions with increasing demands for more flexible services in the Internet. In this paper, we present the design and implementation of a two dimensional router (TwoD router), that makes forwarding decisions based on both destination and source addresses. This TwoD router is also a key element in our current effort towards two dimensional IP routing. With one more dimension, the forwarding table will grow explosively given a straightforward implementation. As a result, it is impossible to fit the forwarding table to the current TCAM, which is the de facto standard despite its limited capacity. To solve the explosion problem, we propose a forwarding table structure with a novel separation of TCAM and SRAM. As such, we move the redundancies in expensive TCAM to cheaper SRAM, while the lookup speed is comparable with conventional routers. We also design the incremental update algorithms that minimize the number of accesses to memory. We evaluate our design with a real implementation on a commercial router, Bit-Engine 12004, with real data sets. Our design does not need new devices, which is favorable for adoption. The results also show that the performance of our TwoD router is promising.

cs.NI↗