The sense of realism in avatar animation is a widely pursued goal in social VR applications. A common approach to enhancing realism is improving the match between avatar motion and real-world human movement. However, experience with existing VR platforms may reshape users' expectations, suggesting that matching reality is not the only path to enhancing the sense of realism. This study examines how different levels of experience with a social VR platform influence users' criteria for evaluating the realism of avatar animation. Participants were shown a set of animations varying in the degree they reflected real-world motion and motion seen on the social VR platform VRChat. Results showed that users with no VRChat experience found animations recorded on VRChat unnatural and unrealistic, but experienced users in fact rated these animations as more likely to come from a real person than the motion-capture animations. Additionally, highly experienced users recognized the intent to imitate VRChat's style and noted the differences from genuine in-platform animations. All these results suggest users' expectations of and criteria for realistic animation were shaped by their experience level. The findings support the idea that realism in avatar animation does not solely depend on mimicking real-world movement. Experience with VR platforms can shape how users expect, perceive, and evaluate animation realism. This insight can inform the design of more immersive VR environments and virtual humans in the future.

Amid the increasingly severe security threat landscape of embedded systems, the security configuration information in chip technical documentation significantly influences the effectiveness of hardware security mechanisms. However, traditional approaches suffer from limitations including low efficiency in manual review, poor generalization capability of rule-based methods, and domain knowledge deficiencies in general-purpose Large Language Models. These short-comings may lead to systemic risks such as false positives in vulnerability identification and critical threat oversight. To address these issues, we proposes ChipGuard-BERT, a BERT-based technology that establishes a three-stage optimization mechanism comprising domain-specific pre-training, adversarial training enhancement, and knowledge-guided optimization, thereby achieving precise parsing of security configuration information. Experimental results demonstrate that ChipGuardBERT achieves an exact match rate of 71.5% on an independent test set, exhibiting robust recognition capabilities in chip manual terminology recognition. Comparative analysis with ChatGPT and the RAGflow local knowledge base reveals that the proposed method attains a 92.5% exact match rate in security configuration conflict detection tasks, outperforming other approaches.

The integration of time-sensitive networking (TSN) and fifth-generation (5G) offers a promising solution for real-time and reliable data transmission in the Industrial Internet of Things (IIoT). However, current research focuses on traffic scheduling in TSN-5G networks to support low latency. New challenges arise when TSN-5G networks leverage time-aware shaper (TAS) and frame replication and elimination for reliability (FRER) to achieve low latency and high reliability. Simply combining TAS and FRER (SCTF) requires scheduling all time-triggered (TT) flows and their replica flows, which substantially increases the computational complexity of gate control lists (GCLs) and severely weakens scheduling capabilities. Moreover, the packet elimination function (PEF) in FRER may induce packet misordering. In this paper, we propose an efficient and fault-tolerant time-aware shaper (EF-TAS) mechanism for TSN-5G networks. EF-TAS only allocates timeslots for TT flows, while replica TT (RT) flows are delivered using a best-effort strategy. Due to the potential violation of deadlines in RT flows, we design an adaptive cyclic GCL window (ACGW)-based hybrid scheduling (AHS) algorithm to schedule TT and RT flows differentially. The AHS algorithm utilizes network calculus to ensure the timely arrival of RT flows without affecting the deterministic transmission of TT flows. In particular, we provide upper bounds on the amount of reordering to quantify the disorder caused by PEF and analyze the impact of introducing the packet ordering function (POF) on EF-TAS performance. The evaluation results show that EF-TAS not only meets the reliability and deadline requirements but also significantly reduces the total number of GCL entries and the computation time of GCLs compared to state-of-the-art methods.

Generative Artificial Intelligenge (GAI) is revolutionizing the world with its unprecedented content creation ability. Large Language Model (LLM) is one of its most embraced branches. However, due to LLM’s substantial size and resource-intensive nature, it is cloud-hosted, raising concerns about privacy, usage limitations, and latency. In this paper, we propose to utilize ubiquitous distributed wireless edge computing resources for real-time LLM inference. Specifically, we introduce a novel LLM edge inference framework, incorporating batching and model quantization to ensure high throughput inference on resource-limited edge devices. Then, based on the architecture of transformer decoder-based LLMs, we formulate an edge inference optimization problem which is NP-hard, considering batch scheduling and joint allocation of communication and computation resources. The solution is the optimal throughput under edge resource constraints and heterogeneous user requirements on latency and accuracy. To solve this NP-hard problem, we develop an OT-GAH (Optimal Tree-search with Generalized Assignment Heuristics) algorithm with reasonable complexity and$\frac {1}{2}$-approximation ratio. We first design the OT algorithm with online tree-pruning for single-edge-node multi-user case, which navigates the inference request selection within the tree structure to miximize throughput. We then consider the multi-edge-node case and propose the GAH algorithm, which recrusively invokes the OT in each node’s inference scheduling iteration. Simulation results demonstrate the superiority of OT-GAH batching over other benchmarks, revealing an over 45% time complexity reduction compared to brute-force searching.

Time-Sensitive networking (TSN) has put forward a series of standards, such as cyclic queuing and forwarding (CQF) and frame replication and elimination for reliability (FRER), to achieve deterministic latency and high reliability. However, most work studies these two mechanisms separately, while directly combining CQF and FRER (DCCF) will inevitably introduce distinct multiple-path delays, seriously impair scheduling capa-bilities and result in a large jitter. In this paper, we propose a Multi-path CQF (MCQF) mecha-nism. Firstly, MCQF enables flexible end-to-end delay calculation by extending the ping-pong queues of CQF to multi-queues. Then, we formulate a joint routing and scheduling mathematical model to maximize the number of schedulable flows and satisfy diverse latency and reliability requirements. Moreover, a hop-by-hop offset scheduling (HOS) algorithm is designed to achieve low jitter by aligning the packet delays on multiple disjoint paths. Evaluation results show that MCQF performs better than CQF on reliability. Compared to DCCF, MCQF greatly reduces the jitter and improves the schedulable flow number by about 31.9 %.

Generative Artificial Intelligence (GAI) is taking the world by storm with its unparalleled content creation ability. Large Language Models (LLMs) are at the forefront of this movement. However, the significant resource demands of LLMs often require cloud hosting, which raises issues regarding privacy, latency, and usage limitations. Although edge intelligence has long been utilized to solve these challenges by enabling real-time AI computation on ubiquitous edge resources close to data sources, most research has focused on traditional AI models and has left a gap in addressing the unique characteristics of LLM inference, such as considerable model size, auto-regressive processes, and self-attention mechanisms. In this paper, we present an edge intelligence optimization problem tailored for LLM inference. Specifically, with the deployment of the batching technique and model quantization on resource-limited edge devices, we formulate an inference model for transformer decoder-based LLMs. Furthermore, our approach aims to maximize the inference throughput via batch scheduling and joint allocation of communication and computation resources, while also considering edge resource constraints and varying user requirements of latency and accuracy. To address this NP-hard problem, we develop an optimal Depth-First Tree-Searching algorithm with online tree-Pruning (DFTSP) that operates within a feasible time complexity. Simulation results indicate that DFTSP surpasses other batching benchmarks in throughput across diverse user settings and quantization techniques, and it reduces time complexity by over 45% compared to the brute-force searching method.

With the emergence of time-sensitive IIoT applications, such as remote operation and industrial control, a long-distance deterministic forwarding service is highly desirable. However, most of the existing research is limited to local area networks, or requires costly replacement of core network devices. Enabling incremental deterministic networks based on off-the-shelf technologies is a significant challenge. This paper designs a core-agnostic and cost-effective solution named Hirail to achieve the smooth evolution of long-distance deterministic networks. Firstly, we investigate that a time-discrete shaper (TDS) can be deployed at the ingress node to enable millisecond-level bounded delay. TDS functions similarly to the concept of buying time-stamped tickets for each flow prior to getting on a high-speed rail, thus avoiding the expensive modification of core devices. Then, to alleviate the flow aggregation problem under long-distance links, we utilize the inband network telemetry to construct the delay-aware network map and conduct adaptive source routing based on the map. Finally, an adjustable buffer at the last hop is devised for jitter reduction. Evaluation results show that Hirail can meet the bounded delay and jitter demands, and outperforms other solutions in terms of performance and overhead.

The Internet of Things (IoT) ecosystem is undergoing a significant evolution through its integration with satellite networks, empowering remote and computation-intensive IoT tasks to leverage computing services via satellite links. Current research in this field predominantly focuses on minimizing latency and energy consumption in computation offloading, yet overlooks the substantial costs incurred by satellite resource utilization. To address this oversight, we introduce a cost-effective hybrid computation offloading (CE-HCO) paradigm in satellite-terrestrial integrated networks (STINs) in this article. First, we propose the 5G-based system framework facilitates gNB and user plane function functionalities on satellites and fosters collaboration between public cloud providers and satellite operators. The framework is in line with the latest 3GPP activities and business models in satellite computing. Then, we formulate the CE-HCO problem, aiming to minimize total computation offloading costs while satisfying diverse user latency requirements and adhering to satellite energy constraints. To tackle this NP-hard problem, we develop an algorithm employing the penalty method and successive convex approximation to simplify the complex mixed-integer nonlinear programming into tractable convex iterations. Simulation results show that our approach outperforms existing baselines in balancing performance and cost, and offer guidance on pricing policies for satellite computing services to promote future commercial growth.

The emerging space-terrestrial integrated network (STIN) assumes a pivotal role within the 6G vision, promising to deliver seamless global coverage and connectivity. Achieving advanced, high-reliability, and time-sensitive (TS) services in a resource-constrained and failure-prone space environment is critical, but also presents challenges. Existing space-terrestrial communication approaches either suffer from temporary link failures with unstable reliability, or intolerable service latency due to the extensive coverage and uneven traffic distribution. This paper presents FastTS, a heuristic resilient and performant scheduling strategy to achieve fault-tolerant and time-sensitive scheduling in futuristic STINs. First, we model the high-dynamic and failure-prone topology in space, and formulate the scheduling problem as a mixed non-linear problem with the objective of minimizing the average task completion time. To approach the optimal solution, joint time-variant routing and frame replication and elimination for reliability (FRER) redundancy under resource constraints are formally considered in our design. During the path-stable duration, FastTS prioritizes the multipath selection with higher redundancy scores, all while ensuring a bounded low latency for TS services based on time-sensitive networking (TSN) techniques. Specifically, our FastTS is divided into three phases: time-sensitive multipath generation (TMG), series-parallel redundancy scoring (SPRS), and SPRS-based time-variant routing (STR). Finally, simulation results show that FastTS exhibits outstanding performance improvements in terms of packet delay, scheduling success ratio, task completion time and packet loss rate, when compared to other state-of-the-art methods.

Recently, MEC has been integrated with satellite networks to process remote terrestrial computation tasks with superior coverage and delay. Since single satellite computation is hard to tackle spatially uneven computation workloads, computation peer offloading among multiple satellites is urgently needed to further improve service quality and resource utilization. However, considering limited resources, deficient energy, and costly overheads of communication and computation, how to enable efficient offloading cooperation in the time-varying satellite networks is a significant challenge. In this paper, we first design a satellite peer offloading scheme, where offloading is performed along multi-hop paths to explore collaborative computing capabilities. Second, we formulate the Multi-Hop Satellite Peer offloading (MHSPO) problem, aiming to jointly minimize the delay and energy consumption under system resources and backlog constraints. Then, to adapt to the network dynamics, the decision-making process with uncertain future workloads is optimized by leveraging the delayed online learning method under the Lyapunov framework. Finally, we develop a practical online distributed algorithm to solve the MHSPO problem, which is proven to achieve close-to-optimal performance. Extensive simulations show that multi-hop peer offloading among satellites improves edge computing performance efficiently.

Recent studies leverage time-aware shaper (TAS) and frame replication and elimination for reliability (FRER) techniques to achieve deterministic latency and high reliability for time-triggered (TT) flows. FRER requires transmitting TT flows on$k$disjoint paths to tolerate transient and permanent failures. However, directly allocating timeslot resources for all$k$TT flows will dramatically increase the computational complexity of gate control lists (GCLs), seriously impair scheduling capabilities, and result in a wastage of bandwidth. In this paper, we propose an ultra-reliable time-aware shaper (uTAS). uTAS only allocates timeslots for one TT flow to ensure deterministic transmission. The$k - 1$replica TT (RT) flows are delivered using a best-effort strategy. On this basis, we propose an adaptive window scheduling (AWS) algorithm based on network calculus, which aims to guarantee that RT flows reach their destinations within the deadline. Evaluation results show that uTAS can meet the flow reliability and deadline requirements. Compared to directly combining TAS and FRER (DCTF), uTAS reduces the total number of GCLs by approximately 72.9%.

The emerging time-critical applications pose intense demands for enabling large-scale deterministic networks. In this paper, we propose a new Programmable Cycle-Specified Queue (PCSQ) for wide-area deterministic packet scheduling. We implement the first end-to-end high-precision rotation dequeuing, which enables microsecond-level time slot resource reservation (noted as T) and especially jitter control of up to 2T. We prototype the PCSQ scheduler on an FPGA. The PCSQ-enabled switches can guarantee bounded delay and jitter transmission on a realistic testbed.

Time-Sensitive Networking (TSN) is emerging to support critical real-time applications in Industry 4.0. Recent proposals leverage Cyclic Queuing and Forwarding (CQF) to achieve bounded-delay transmission for cyclic flows in TSN. However, the CQF is not flexible enough in two aspects. First, it cannot achieve zero jitter. The Ping-Pong queue-based model in CQF will introduce the jitter of two cycles, which is inapplicable to industrial automation scenarios where isochronous flows require zero jitter. Second, it may require setting the maximum queue length to a fixed value in advance and scheduling the flows offline, which is challenging for dynamic traffic scheduling. In this paper, we firstly present a time-aware cyclic-queuing (TACQ) mechanism to enable zero jitter for CQF. TACQ consists of a novel no-wait shaper (NWS) and a cyclic-queuing shaper (CQS). The NWS handles isochronous flows by strictly limiting the transmission time of flows that do not overlap on each output port and each period. The CQS is extended from CQF to schedule cyclic flows. Then, we propose a variable time slot mechanism and a novel incremental routing and scheduling (IRAS) algorithm based on software-defined networking (SDN) to online schedule dynamic flows. Simulation results show that TACQ significantly reduces the delay of isochronous flows and achieves zero jitter compared with CQF. And the IRAS algorithm approaches 96.1% of the optimal solution in scheduling 2000 flows with a feasible per-flow computational time.

The virtual cloud network has been the first choice for most enterprises to expand local networks due to its convenience, flexibility, and elasticity. Cloud gateways are the key to steering traffic among these virtual networks, which require high throughput and low delay, usually in Tbps and microseconds (us), to improve the performance of cloud regions. However, as cloud traffic growth far exceeds Moore's law, previous software cloud gateways are facing the performance bottleneck that they are vulnerable to the attack of heavy-hitter flows. In this paper, we propose a high-performance and low-cost software cloud gateway for accelerating virtual cloud networks. By revisiting the Vector Packet Processing (VPP) framework, we design a custom control plane to enable various network functions of the cloud gateway, which enhances the routing and forwarding actions in the data plane. And we also provide a common interface for users to flexibly configure and manage the gateway. This pure software architecture has a low deployment cost. Compared with other software gateways with roughly the same price, the processing performance is close to the NIC's line speed, which is suitable for high concurrency cloud scenarios.

Recent proposals leverage Cyclic-Queuing and Forwarding (CQF) to achieve bounded-delay transmission for cyclic flows in Time-Sensitive Networking (TSN). However, the Ping-Pong queue-based model in CQF will introduce the jitter of two cycles, which is inapplicable to industrial automation scenarios where isochronous flows such as synchronized frames and motor control loops require zero jitter.We present TACQ, a first time-aware cyclic-queuing mechanism to support the co-transmission of cyclic flows and isochronous flows. Our key insight is that isochronous flows should enable zero-jitter while minimally impacting the bounded-delay of cyclic flows. To achieve this goal, we design a novel no-wait shaper (NWS) that handles isochronous flows with as little time-slot as possible. For cyclic flows, we extend the CQF to schedule flows with a double-closed state on the Tx-gate and compute the open time on the Rx-gate. Simulation results show that TACQ significantly reduces the delay of isochronous flows by 81.3% and achieves zero jitter compared with CQF. And the NWS effectively schedules 87.2% flows at 500 flows level in feasible execution time.

Recent proposals leverage Time-Aware Shaper (TAS) to achieve precise transmission in Time-Sensitive Networking (TSN). However, most of the proposals require the information of all time-triggered flows to be known in advance and synthesize the gate control list of each switch offline, making the mechanisms they designed inapplicable to industrial automation scenarios where the devices are changed dynamically and the flows should be scheduled online. In this paper, we propose an online routing and scheduling mechanism of TAS for time-sensitive networks. In order to maximize the number of schedulable flows and reduce bandwidth waste, we devise the variable time slot mechanism and minimize the sending start time of each flow. Based on these mechanisms, a novel incremental routing and scheduling (IRAS) algorithm is designed to achieve per-flow deployment, with a pre-routing algorithm to reduce synthesis time. The evaluations show that the IRAS algorithm approaches 96.5 % of the optimal solution in scheduling 2000 flows, and has a feasible per-flow computational time from sub-seconds to less than ten seconds.

Mobile edge computing (MEC) is proposed to boost high-efficient and time-sensitive 5G applications. However, the "microburst" may occur even in lightly-loaded scenarios, which leads to the indeterministic service latency, hence hindering the deployment of MEC. Deterministic IP networking (DIP) has been proposed to provide bounds on latency, and high reliability in the large-scale networks. Nevertheless, the direct migration of DIP into the MEC network is non-trivial owing to its original design for the Ethernet with homogeneous devices. Meanwhile, DIP also faces the challenges on the network throughput and scheduling flexibility. In this paper, we delve into the adoption of DIP for the MEC networks and some of the relevant aspects. A deterministic MEC (D-MEC) network is proposed to deliver the deterministic MEC service. In the D-MEC network, the cycle mapping and cycle shifting are designed to enable: (i) seamless and deterministic transmission with heterogeneous underlaid resources; and (ii) traffic shaping on the edges to improve the resource utilization. We also formulate a joint configuration to maximize the network throughput with deterministic QoS guarantees. Extensive simulations verify that the proposed D-MEC network can achieve a deterministic MEC service, even in the highly-loaded scenarios.

A measure of software dependability called trustability is described. A program p has trustability T if we are at least T confident that p is free of faults. Trustability measurement depends on detectability. The detectability of a method is the probability that it will detect faults, when there are faults present. Detectability research can be used to characterize conditions under which one testing and analysis method is more effective than another. Several detectability results that were only previously described informally, and illustrated by example, are proved. Several new detectability results are also proved. The trustability model characterizes the kind of information that is needed to justify a given level of trustability. When the required information is available, the trustability approach can be used to determine strategies in which methods are combined for maximum effectiveness. It can be used to determine the minimum amount of resources needed to guarantee a required degree of trustability, and the maximum trustability that is achievable with a given amount of resources. Theorems proving several optimization results are given. Applications of the trustability model are discussed. Methods for the derivation of detectability factors, the relationship between trustability and operational reliability, and the relationship between the software development process and trustability are described.

A measure of software dependability called trustability is described. A program p has trustability T if we are at least T confident that p is free of faults. Trustability measurement depends on detectability. The detectability of a method is the conditional probability that it will detect faults. The trustability model characterizes the kind of information needed to justify a given level of trustability. When the required information is available, the trustability approach can be used to determine strategies in which methods are combined for maximum effectiveness. It can be used to determine the minimum amount of resources needed to guarantee a required degree of trustability, and the maximum trustability that is achievable with a given amount of resources.>

No abstract available.