With its rapid development, AIGC applications have expanded to diverse generative content, including text, images, audio, and videos. To enhance the AIGC's output quality, prompt chains are proposed to structure the generation process. Owing to the prompt's data processing nature, one compromised prompt engineering-enabled server (PES) may propagate vulnerabilities across networks, leading to unintended content generation, system risks, and potential user trust issues. To mitigate these risks, this work adopts a zero-trust (ZT) security framework to protect inter-server communications. We define the problem of prompt chain deployment in ZT-enabled networks(PCD-ZT), with the objective of minimizing total service costs, encompassing both deployment and ZT verification costs. To address this, we introduce the verification-cost-balance (VCB) factor that helps reduce ZT verifications to save the overall service cost and accordingly propose an algorithm called sub-prompt-chain brand and bound (SCBB). Extensive simulations demonstrate that our SCBB achieves overall service cost reductions of 4.67 % and 13.89 %, and cuts verification costs by 11.86 % and 30.50 %, compared to the benchmarks extended from the state-of-the-art.

Recent advances in large language models (LLMs) have empowered robots and drones with autonomous decision-making capabilities. Due to the stringent real-time requirements of these applications, LLM inference must be performed at the network edge. However, hosting high-precision LLMs on a single edge server is often infeasible, creating challenges in efficiently distributing LLM deployments across edge networks. This work addresses these challenges by formulating and solving the profit maximization problem for distributed LLM inference deployment. We first formally define the Profit-Centric Inference Chain Deployment (PC-InCD) problem. To solve PC-InCD, we introduce a novel Local Maximal Profit (LMP) factor that enables effective edge server selection for hosting LLM sub-modules, and we propose the LMP-based Inference Chain Deployment (LMP-InCD) algorithm. Extensive simulations demonstrate that LMP-InCD significantly outperforms benchmark methods in maximizing profit across diverse network conditions.

Video traffic has become the principal part of data resources in the current cyberspace which brings many challenges such as security, stability and scalability of streaming transmission. Moreover, how to ensure high visual quality while obtaining a significant bit-rate reduction has always been the focus of the industry. By constructing a source distortion temporal propagation (SDTP) model, this paper proposes a temporal dependent RDO (TDRDO) algorithm to resolve the global RDO problem in the temporal domain. Besides, a fuzzy logic based rate control (FLRC) algorithm is proposed to robustly regulate encoding bit-rates. The two algorithms have previously been adopted by Audio Video Coding Standard Workgroup of China and integrated into the second generation (AVS2). Experimental results prove the excellence of the proposed algorithms, for significantly improving the AVS2 video coding performance and providing AVS2 with superb efficiency to compete with HEVC/H.265 in modern video compression.

Mixture of experts (MoE) has shown great potential in enhancing large language models, such as DeepSeek. In MoE, expert networks collaboratively process tokens routed by the gating network, enabling advanced capabilities such as semantic understanding, computational reasoning, and code generation. To address security threats such as DDoS in these dynamic environments, it is essential to implement tailored backup measures for these experts. However, traditional backup methods often lead to inefficiencies and excessive resource consumption. To overcome these challenges, we propose a novel approach for backing up experts in MoE. We formally define and mathematically model a new problem, termed the security-aware off-site pairwise protection (SOPP) problem, and prove its NP-hardness. To solve this problem, we develop three novel techniques of latency-aware MoE construction (LMC) to reduce backup latency, partitioned backup selection (PBS) to trade off security levels and resource consumption, as well as pairwise selective identifier (PSI) to determine the appropriate backup pairwise nodes. On the basis of these techniques, we propose an efficient heuristic algorithm called the off-site distributed pairwise nodes protection (OD-PNP), providing theoretical performance guarantees. Through extensive simulations and analyses, we demonstrate that our proposed algorithm outperforms state-of-the-art methods in terms of both protection efficiency and resource consumption.

In network function virtualization (NFV), service providers can accommodate services from clients by deploying software-based functions, called service functions (SFs), on commodity servers. The deployed SFs are connected to SF forwarders for efficient management in these commodity servers. Many efforts have been put into protecting SFs from failures. However, little attention has been paid to protecting SF forwarders from failures. In this paper, we investigate how to efficiently allocate backup computing resources when facing SF and SF forwarder failures. To protect the deployed SFs with the minimum backup computing resources, we define a new SF forwarder off-site shared protection problem and approve its NP-hardness. An efficient heuristic algorithm is proposed and proved to be logarithmic approximate. Extensive simulation results show that the proposed algorithm significantly outperforms the approaches directly extended from the existing work.

Emerging machine learning techniques enable Internet-connected devices to generate service function chain (SFC) requests for reliability-sensitive applications. To facilitate reliable SFC provisioning, prior works have proposed the SFC dedicated protection approach for fog and cloud networks, which generally have abundant resources and connectivity. However, with limited resources and connectivity, it might be inefficient to directly employ these approaches at the network edge. In this work, we study how to efficiently embed and provide dedicated protection when accommodating SFCs at the network edge. We formally define the problem of SFC embedding with dedicated protection (SFCE-DP) to minimize the bandwidth resources usage at the network edge. Based on the proposed technique of augmenting path in a layered graph, we construct an efficient algorithm, called augmenting-path-based SFC embedding with dedicated protection (AP-SDP), to optimize SFCE-DP. When the network resources are limited, our results show that AP-SDP significantly outperforms the benchmark that is directly extended from the state-of-the-art.

The ultra-reliable and low latency communication (URLLC) service in 5G/6G will be delivered through a sequence of software-based network functions, also known as a service function chain (SFC). To satisfy the ultra-reliable requirement of the URLLC service, fault-tolerance in URLLC SFC processes is required. However, achieving deterministic fault-tolerance in URLLC SFC delivery is challenging as physical/virtual network failures and hardware/software failures have to be jointly consid-ered. In this work, we first introduce an augmented SF protection graph, called k-connected service function slicing (KC-SFS), which can facilitate the SF protection against multiple concurrent physical/virtual node and physical link failures. Based on the KC-SFS, we define a new problem called deterministic fault-tolerant service function slicing (DFT-SFC) and formulate it with a mathematical model. To solve DFT-SFC, we propose an efficient heuristic algorithm, called service function slice embedding (SFSE), which employs the k-connected network slicing technique (k-NST). Via thorough mathematical analysis, we prove that k-NST achieves 2-approximation. Meanwhile, our extensive experimental results show that the proposed SFSE guarantees deterministic fault-tolerance and outperforms the schemes directly extended from the stste -of - the art.

In network function virtualization (NFV), the customer may request a set of virtual network functions (VNFs) that the customer traffic will go through. To accommodate such requests, the service providers have to embed the requested VNFs onto the substrate network (SN) to form an actual traffic forwarding path called service function path (SFP). In the elastic optical network (EON), how to protect the running network services against VNF failures becomes an attractive research focus. Most existing work concentrates on the protection or restoration of the physical node or fiber link failures in the SN. Few research attention has been paid to the failure of the virtual machines running a VNF. In this paper, we study how to protect VNFs when a VNF failure occurs in an EON. We define a new VNF failure protection cover (VFPC) problem and mathematically formulate VFPC. We propose a protection cover list based VNF protection (PCL-VP) algorithm against any single VNF failure while satisfying the latency requirement. Extensive simulations and analysis show the effectiveness of the proposed algorithm.

The ultra-fast speed and massive capacity in 5G networks push huge amounts of data to networks. With network function virtualization, these data will go through multiple service functions (SFs) and big data processing/analysis. As a result, the processing delay from such SFs and data processing/analysis can significantly impact the delivery of latency-sensitive services. To reduce the processing delay, network function parallelism techniques are introduced to allow multiple SFs running parallelly for the same request. In this work, we study how to apply network function parallelism into SF chaining and embedding to optimize the latency. When physical nodes have unlimited computing resource, we propose the mixed integer programming based parallelism-aware SFC optimization (MIP-PS) algorithm. Our analysis proves the proposed MIP-PS is integer-approximation. When physical nodes have limited computing resource, we propose the latency factor based parallelism-aware SFC optimization (LF-PS) algorithm. Our extensive simulations demonstrate that our proposed schemes outperform the approaches extended directly from the existing work.

Network function virtualization (NFV) is introduced to effectively deliver end-to-end network services for the emerging Internet of Things (IoT), multiaccess edge computing, and 5G communication techniques. In NFV, the network service request can be accommodated in the form of a service function chain (SFC). The SFC will have to reserve abundant resources, such as link bandwidth, service functions, and computation in the physical network to meet the demands of customers. Minimizing the cost from the resource reservation in NFV remains challenging, even though a few works in the literature proposed cost-optimization methodologies with assumptions to guarantee their correctness. In this article, we comprehensively investigate how to minimize the cost when delivering network services as SFCs with provable bounds and fewer assumptions. We formally define the problem of minimum cost service function chaining and embedding (MC-SFCE) and propose an algorithm, namely, cost factor-based SFCE optimization with shortcut (COFO-SC), for MC-SFCE. Novel mathematical analysis is provided to demonstrate the correctness of our approaches and related bounds. Our extensive simulations and analysis also show that the proposed COFO-SC outperforms the schemes directly extended from the existing work.

In network function virtualization (NFV), the client’s service requests will go through multiple service functions (SFs). The instances of the required SFs will be hosted on the geographically-distributed physical nodes in physical networks (PNs). The failure of any physical nodes or virtual nodes (i.e., SF instances) will impact the delivery of services and the client’s experience. It is essential for service providers to take the node failure and network reliability into account. Different from physical node failures that will affect the PN’s topology and connectivity, virtual node failures impact the services delivery of certain clients. As a result, applying traditional backup schemes that are designed for physical node failures may not efficiently provide protection for virtual node failures. In this work, we define and mathematically formulate a new latency-bounded off-site virtual node protection (LOVNP) problem in NFV. After proving the NP-hardness of the LOVNP problem, we introduce a novel shared protection technique called pairwise node protection to effectively facilitate the protection of node failures in NFV. Then, we propose an efficient heuristic algorithm called protection centrality based pairwise node protection (PC-PNP) to optimize the LOVNP problem and prove that PC-PNP has a logarithm-approximation boundary. Our extensive simulations and analysis show that the proposed algorithm significantly outperforms the algorithms that are extended from the existing work.

In Network Function Virtualization (NFV), to satisfy the Service Functions (SFs) requested by a customer, service providers will composite a Service Function Chain (SFC) and embed it onto the shared Substrate Network (SN). For many latency-sensitive and computing-intensive applications, the customer forwards data to the cloud/server and the cloud/server sends the results/models back, which may require different SFs to handle the forward and backward traffic. The SFC that requires different SFs in the forward and backward directions is referred to as hybrid SFC (h-SFC). In this paper, we, for the first time, comprehensively study how to optimize the latency in Hybrid SFC composition and Embedding (HSFCE). When each substrate node provides only one unique SF, we prove the NP-hardness of HSFCE and propose the first 2-approximation algorithm to jointly optimize the processes of h-SFC construction and embedding, which is called Eulerian Circuit based Hybrid SFP optimization (EC-HSFP). When a substrate node provides various SFs, we extend EC-HSFP and propose the efficient Betweenness Centrality based Hybrid SFP optimization (BC-HSFP) algorithm. Our extensive simulations and analysis show that EC-HSFP can hold the 2-approximation, while BC-HSFP outperforms the algorithms directly extended from the state-of-art techniques by an average of 20%.

The emerging multiaccess edge computing (MEC) architecture brings the needed computing resource to the network edge. Many 5G and Internet of Things (IoT) applications are latency sensitive and computation intensive in MEC systems. To flexibly provide and manage the network service requests in MEC systems, network function virtualization (NFV) can be employed to create a chain of service functions (SFs), namely, SF chain (SFC). Through SFC, the customer forwards user data to the edge server/cloud, and the edge server/cloud may return the processed results/models to the customer. When the forward and backward traffic is carrying different content, different SFs may be required for the forward and backward traffic, which requires a hybrid SFC (h-SFC). In this article, we study how to minimize the latency cost when embedding an h-SFC in MEC systems. We define a new problem called minimum latency hybrid SFC embedding (ML-HSFCE) and propose an algorithm, namely, optimal hybrid SFC embedding (Opt-HSFCE) to optimally embed a given h-SFC in MEC systems. Our extensive simulations and analysis show that the proposed Opt-HSFCE needs much less runtime compared with the brutal force algorithm and significantly outperforms the schemes that are directly extended from the existing techniques.

Network Function Virtualization (NFV) takes advantages of the emerging technologies in virtualization and automation to offer new ways in design, deployment, and management of networking services. In NFV, the proprietary hardware-based network functions are replaced by the software-based modules named as Virtual Network Functions (VNFs) or Service Functions (SFs). A network service request from the customer can be formed by multiple SFs. To satisfy a network service request, the service provider has to chain the SFs in the request into a Service Function Chain (SFC) and embed the constructed SFC onto the shared substrate network. In this paper, we comprehensively study how to composite and embed an SFC onto a shared substrate network with unique service function. We formulate this problem with the Integer Linear Programming (ILP) technique. We also propose an efficient heuristic algorithm with 2-approximation boundary, namely, Spanning Closed Walk based SFC Embedding (SCW-SFCE). Our extensive simulations and analysis show that the proposed approach can achieve near-optimal performance in a small network and outperform the Nearest Neighbour (NN) algorithm.

Network Function Virtualization (NFV) technology decouples network functions from proprietary hardware equipments. As a result, Internet Service Providers (ISPs) implement software-based network functions on generic highvolume substrate network devices. In NFV, a Service Function Chain (SFC) is defined as an ordered set of abstract network functions running on specific substrate nodes (e.g., servers). A challenging issue in NFV management and orchestration is how to optimize the Dependence-aware SFC Embedding in substrate Optical networks (D_SFCE_O). In this paper, we propose a novel algorithm, namely, Dependence-aware SFC embedding with Least-Used consecutive subcarriers (D_SFC_LU), which jointly optimizes SFC design, SFC mapping and spectrum allocation in optical networks. To minimize resource consumption, D_SFC_LU takes advantages of the proposed techniques: Impact Factor based Node Selection (IFNS), Chain Node Mapping (CNM) and Chain-Fit (CF) spectrum allocation. Our simulation and analysis demonstrate that D_SFC_LU can efficiently embed a network requests while minimizing the required substrate resource in optical networks.

In Network Function Virtualization (NFV), Service Function Chain (SFC) is composed of Virtual Network Function (VNF) nodes that are chained via VNF links. SFCs can be specified as unidirectional or bidirectional. A unidirectional SFC (u-SFC) demands the traffic being forwarded via the VNFs in one direction, while a bidirectional SFC (b-SFC) requires bidirectional traffic flows. In this paper, for the first time, we investigate the problem of how to efficiently deploy a hybrid SFC (h-SFC), whereas some VNF nodes are required to process bidirectional traffic while others only handle unidirectional traffic. We define a new problem called hybrid SFC Deployment (h-SFCD). When each substrate node provides one unique VNF, we prove the NP-hardness of the h-SFCD problem and propose an approximate algorithm, namely, 2-approximation Hybrid Service function chain Deployment in Unique function networks (2-HSD-U). Our experimental results show that the proposed 2-HSD-U algorithm significantly outperforms the heuristic algorithm based on the traditional Nearest-Neighbor technique.

Transforming infrastructures, buildings and services with the sensed data from the Internet of Things (IoT) technique has drawn wide attention. Enormous video data from city surveillance cameras poses huge challenges of transmission, storage and analysis, which necessitates new video compression technologies. The fusion of video data generated from smart city could be used to support city management and urban policy. Based on the specific characteristics of surveillance video, which are successive pictures have very strong correlations and each picture can be divided into background and foreground, this work proposes a block-level background modeling (BBM) algorithm to support long-term reference structure for efficient surveillance video coding. A rate–distortion optimization for surveillance source (SRDO) algorithm is also developed to improve the coding performance. Experimental results show that the proposed BBM and SRDO can significantly improve the compression performance, which can effectively support diverse video applications in smart city.