Test-Time Adaptation (TTA) has recently emerged as a promising research direction, enabling vision-language models (VLMs) to adapt to unlabeled test data in zero-shot settings. Among TTA approaches, test-time prompt tuning has shown great potential for enhancing the practical applicability of VLMs. However, existing methods typically either focus on adapting a single modality or apply uniform optimization to both modalities, without explicitly defining modality-specific optimization objectives. Such a one-size-fits-all strategy often results in suboptimal performance under test-time conditions. To address this limitation, we propose Dual-modality Heterogeneous Prompt Tuning (DHPT), a novel framework designed to simultaneously capture fine-grained textual semantics and alleviate domain shift noise in the visual modality. Specifically, we leverage a large language model to provide textual cognition guidance for the text encoder, while on the vision side, we develop a lightweight calibration module that adaptively mitigates domain shift noise across different scales. Furthermore, we introduce a cluster-tight optimization objective that enhances the stability and generalizability of prompt tuning under distribution shifts. Extensive experiments conducted on 11 benchmark datasets demonstrate that DHPT consistently and significantly outperforms existing TTA methods for VLMs.

Transformers are increasingly becoming the dominant model in the field of computer vision, thereby catalyzing research efforts aimed at unraveling the interpretability of transformers. Existing explanation techniques, whether attention-based or gradient-based, furnish a dependable approach to quantifying the impact of input features on model predictions from the perspective of dissecting self-attention mechanisms. However, current research overlooks the block-to-block constraints, which result in misdirection in attribution. In this work, we propose a block-wise constraints-free interpretation method, Independent Block Level Attribution (IBA), which maintains the relative independence of each block in the model. The IBA reconfigures the model into mutually unaffected class-semantic blocks via class-semantic relevance, each of which performs the attribution computation independently, thus minimizing the influence of inter-block constraints on the model interpretation performance. Extensive perturbation and segmentation experiments unequivocally demonstrate the superiority of our method, showcasing its significant outperformance compared to current interpretation methods. Additionally, we also apply IBA to the text transformer to demonstrate the generalization of our method. Our code is available athttps://github.com/qinanin/TMM-IBA.git.

In standard reinforcement learning, since the uncertainty of task objectives is not adequately considered in the policy training, the policy achieves poor generalization for the out-of-distribution (OOD) tasks. Although considerable efforts have been made to enhance the generalization for OOD tasks, most of these methods overlook the structural information of task representations in latent space during the generation of extrapolative data, resulting in biased and blurred data embeddings, which then affect the policy generalization. To address this issue, we propose a context-based meta-reinforcement learning (meta-RL) method, namely latent variable distribution enhancement sampler (LVDES), which enhances the policy generalization on OOD tasks by providing efficient task representation space and accurate augmentation policy training data for OOD tasks. Specifically, the proposed LVDES consists of four modules: a task inference module, a task separation module, a latent enhancement module (LEM), and a policy module. The task inference module is used to identify the task. The task separation module (TSM) learns a representation space with highly structured separability. The LEM generates relevant additional task trajectories for augmenting policy training data. The policy module learns a policy to solve tasks. By using efficient task representation space and augmented trajectory data, the exploration efficiency and generalization of the policy for OOD tasks can be enhanced by our LVDES method. Extensive experiments are conducted to demonstrate the effectiveness of our method in comparison with existing methods on the MuJoCo and Meta-World benchmarks. The experimental results show that the task completion accuracy of our LVDES on OOD tasks is increased by 60.20%, with the average exploration time being reduced by 62.99% in comparison with the most effective current method, which demonstrates that our LVDES can achieve great policy generalization on OOD tasks.

Network topology planning is an essential multi-phase process to build and jointly optimize the multi-layer network topologies in wide-area networks (WANs). Most existing practices target single-phase/layer planning, and are incapable of satisfying all rigorous topological structure constraints (e.g., dual-homing rings) defined by network standards and operators, especially in large-scale networks. These significantly limit their usability and performance in production networks. We consider a general topology planning problem with typical structure constraints over three essential phases (greenfield, reconfiguration, and site expansion) and topological layers (optical, IP, and routing topologies). We present, T3Planner, a novel practical solver to this problem in production. Specifically, we develop a structure-driven encoder based on graph neural network (GNN) for concise structure encoding, and design a new learning framework with optical-centric layer compression/reconstruction and rule-aided reinforcement learning (RL) for fast convergence and high performance. Extensive experiments on nine real topologies demonstrate that T3Planner scales to large optical networks with hundreds of sites, saves 46.6% cost, and supports$3.12\times $more demand when compared to related existing approaches.

Network slicing architecture is expected to fulfill network applications with heterogeneous requirements through efficient slice admission control (SAC) policies. Existing SAC approaches entirely rely on current limited observations to make admission decisions, ignoring the potential impact of future demands. The short-sighted behaviors lead to poor service performance and infrastructure providers’ (InPs’) revenue in practice. In this paper, we propose OACR$^{2}$, an online SAC approach based on deep reinforcement learning (DRL) that can exploit predictable future requests to make more precise admission control decisions for the long-term revenue, and reserve proper resources accordingly. Specifically, we design three novel schemes: (i) a requirement predictor based on long short-term memory (LSTM) and a novel input-output way to predict future unforeseen requests, (ii) a DRL admission controller based on the partially observable Markov decision process model to make precise admission decisions without accurate future request information, with the convergence strictly proved, and (iii) a decision defender to guarantee decision reliability. Extensive experiments on real-world traces demonstrate that compared to the No-wait, Wait-queue, and Wait-earliest time approaches, OACR$^{2}$improves InPs’ revenue and acceptance ratio by up to 40.9% and 16.7%, respectively, without sacrificing online inference time (within 0.9 milliseconds).

Knowledge distillation (KD), a technique widely employed in computer vision, has emerged as a de facto standard for improving the performance of small neural networks. However, prevailing KD-based approaches in video tasks primarily focus on designing loss functions and fusing cross-modal information. This overlooks the spatial-temporal feature semantics, resulting in limited advancements in model compression. Addressing this gap, our paper introduces an innovative knowledge distillation framework, with the generative model for training a lightweight student model. In particular, the framework is organized into two steps: the initial phase is Feature Representation, wherein a generative model-based attention module is trained to represent feature semantics; Subsequently, the Generative-based Feature Distillation phase encompasses both Generative Distillation and Attention Distillation, with the objective of transferring attention-based feature semantics with the generative model. The efficacy of our approach is demonstrated through comprehensive experiments on diverse popular datasets, proving considerable enhancements in video action recognition task. Moreover, the effectiveness of our proposed framework is validated in the context of more intricate video action detection task. Our code is available at https://github.com/aaai-24/Generative-based-KD.

Safety protective gear (SPG) detection based on the machine learning model plays an important role in improving outdoor personnel safety in the electric power industry. However, the detection method of transmitting video to the cloud faces a series of challenges, such as privacy disclosure and high latency. To solve this problem, we present FtlSPG, a federated transfer learning framework for SPG detection. In particular, under the three-layer pyramid architecture of “site-companyCloud-globalServer,” we propose a federated personalized model based on local batch normalization and dynamical weighting for the source domain with labeled video. Moreover, a federated domain adaptation model based on a federated deep adversarial network and model self-training is presented for the target domain with unlabeled video. Finally, we verify the effectiveness of FtlSPG in real-world power companies. Extensive experiments demonstrate that FtlSPG can significantly outperform existing schemes, in terms of privacy protection, detection precision, and response latency.

Knowledge and data dual-driven (KDDD) represents a novel paradigm that leverages the strengths of data-driven methods in feature representation and knowledge transfer, while also incorporating expertise accumulated by domain experts. This integration allows KDDD methods to enhance the interpretability, reliability, and robustness of fault diagnosis (FD) approaches, making them widely studied in the field of industrial equipment (IE) FD. Despite the existence of systematic and valuable reviews on IE FD, there remains a gap in the literature regarding the review of KDDD IE FD methods. Therefore, conducting a comprehensive investigation into KDDD IE FD methods is of utmost importance and necessity. Such an investigation will facilitate readers’ understanding of advanced technologies and enable the rapid design of effective solutions for real-world IE FD problems. In this survey, we first outline the limitations of data-driven and knowledge-based FD methods, highlighting the need for KDDD methods. Subsequently, we delve into the details of how domain knowledge can be effectively integrated with deep learning models. Additionally, we analyze challenges of KDDD methods in real-world IE FD applications, while also discussing novel solutions for prospective research directions. Finally, we conclude this survey, emphasizing the inspiration it offers to researchers interested in advancing IE FD, and its potential to stimulate practical IE FD research.

Weakly-supervised action localization aims to recognize and localize action instancese in untrimmed videos with only video-level labels. Most existing models rely on multiple instance learning(MIL), where the predictions of unlabeled instances are supervised by classifying labeled bags. The MIL-based methods are relatively well studied with cogent performance achieved on classification but not on localization. Generally, they locate temporal regions by the video-level classification but overlook the temporal variations of feature semantics. To address this problem, we propose a novel attention-based hierarchically-structured latent model to learn the temporal variations of feature semantics. Specifically, our model entails two components, the first is an unsupervised change-points detection module that detects change-points by learning the latent representations of video features in a temporal hierarchy based on their rates of change, and the second is an attention-based classification model that selects the change-points of the foreground as the boundaries. To evaluate the effectiveness of our model, we conduct extensive experiments on two benchmark datasets, THUMOS-14 and ActivityNet-v1.3. The experiments show that our method outperforms current state-of-the-art methods, and even achieves comparable performance with fully-supervised methods.

Vehicular edge networks involve edge servers that are close to mobile devices to provide extra computation resource to complete the computation tasks of mobile devices with low latency and high reliability. Considerable efforts on computation offloading in vehicular edge networks have been developed to reduce the energy consumption and computation latency, in which roadside units (RSUs) are usually considered as the fixed edge servers (FESs). Nonetheless, the computation offloading with considering mobile vehicles as mobile edge servers (MESs) in vehicular edge networks still needs to be further investigated. To this end, in this article, we propose a Deep-Reinforcement-Learning-based computation offloading with mobile vehicles in vehicular edge computing, namely, Deep-Reinforcement-Learning-based computation offloading scheme (DRL-COMV), in which some vehicles (such as autonomous vehicle) are deployed and considered as the MESs that move in vehicular edge networks and cooperate with FESs to provide extra computation resource for mobile devices, in order to assist in completing the computation tasks of these mobile devices with great Quality of Experience (QoE) (i.e., low latency) for mobile devices. Particularly, the computation offloading model with considering both mobile and FESs is conducted to achieve the computation tasks offloading through vehicle-to-vehicle (V2V) communications, and a collaborative route planning is considered for these MESs to move in vehicular edge networks with objective of improving efficiency of computation offloading. Then, a Deep-Reinforcement-Learning approach with designing rational reward function is proposed to determine the effective computation offloading strategies for multiple mobile devices and multiple edge servers with objective of maximizing both QoE (i.e., low latency) for mobile devices. Through performance evaluations, our results show that our proposed DRL-COMV scheme can achieve a great convergence and stability. Additionally, our results also demonstrate that our DRL-COMV scheme also can achieve better both QoE and task offloading requests hit ratio for mobile devices in comparison with existing approaches (i.e., DDPG, IMOPSOQ, and GABDOS).

Federated learning (FL) is capable of analyzing tremendous data from smart edge devices in Industrial Internet of Things (IIoTs), empowering numerous industrial applications. However, the increasing privacy concerns and deployment costs of IIoT environment have been posing new challenges for FL. This article proposes PCFed, a novel privacy-enhanced and communication-efficient FL framework to provide higher model accuracy with rigorous privacy guarantees and great communication efficiency. In particular, we develop a sampling-based intermittent communication strategy via a PID (proportional, integral, and derivative) controller on the cloud server to adaptively reduce the communication frequency. In addition, we design a budget allocation mechanism to balance the tradeoff between model accuracy and privacy loss. Then, we develop PCFed+, an enhanced variant for PCFed, with further consideration of infinite data streams on edge servers. Extensive experiments demonstrate that both PCFed and PCFed+ can significantly outperform existing schemes, in terms of communication efficiency, privacy protection, and model accuracy.

An edge-cloud collaborative intelligent (ECCI) platform is of great significance for the agile development and rapid deployment of ECCI applications, which are essential for realizing smart industry in the era of Industry 4.0. However, the existing platforms lack considering the high real-time latency demand of industrial operations, which severely hinders the development of smart industry and may even lead to severe industrial accidents. To effectively reduce the response latency of industrial applications, in this article, we propose an ECCI platform IndustEdge. It takes time-sensitive networking as the deterministic transport for the link layer, and provides an extensible ECCI orchestration component to reduce the system level latency. Furthermore, IndustEdge has an ECCI algorithm library for different collaborative modes and provides the complete life cycle management for ECCI applications. We implement platforms for both the real-world prototype and emulated-world emulation, and conduct two case studies to evaluate the effectiveness of IndustEdge.

The densified deployment of heterogeneous networks coexisting with a variety of overlapping cells has emerged as a viable solution for next generation wireless networks. Despite numerous advantages, the heterogeneity and denseness also raise complicated handover management issue. Nonetheless, most existing handover methods generally depend on one or more objective attributes, and rarely consider the subjective demands of personalized users and specific applications that demand differentiated services. Through decomposing the control plane and data plane, software defined network(SDN) offers a flexible architectural paradigm to overcome these challenges. In this article, we first develop an SDN-driven handover architecture that is capable of perceiving global network status and requirements from various perspectives, including the physical layer, users, and applications. Then, a context-aware multi-criteria handover mechanism is developed in the SDN edge to provide differentiated services. Considering the numerous complicated factors, the handover decision is made based on a hierarchical fuzzy inference system to process diverse attributes and vague requirements described in natural language. Finally, we evaluate the performance of our proposed scheme through a combination of extensive simulations and real-world experiments. The results demonstrate that our solution outperforms the baseline handover schemes, more efficiently providing differentiated services with respect to throughput, bandwidth cost, and application satisfaction, and is efficient and feasible in practice.

Latent Dirichlet Allocation (LDA) is a popular topic modeling technique for hidden semantic discovery of text data and serves as a fundamental tool for text analysis in various applications. However, the LDA model as well as the training process of LDA may expose the text information in the training data, thus bringing significant privacy concerns. To address the privacy issue in LDA, we systematically investigate the privacy protection of the main-stream LDA training algorithm based on Collapsed Gibbs Sampling (CGS) and propose several differentially private LDA algorithms for typical training scenarios. In particular, we present the first theoretical analysis on the inherent differential privacy guarantee of CGS based LDA training and further propose a centralized privacy-preserving algorithm (HDP-LDA) that can prevent data inference from the intermediate statistics in the CGS training. Also, we propose a locally private LDA training algorithm (LP-LDA) on crowdsourced data to provide local differential privacy for individual data contributors. Furthermore, we extend LP-LDA to an online version as OLP-LDA to achieve LDA training on locally private mini-batches in a streaming setting. Extensive analysis and experiment results validate both the effectiveness and efficiency of our proposed privacy-preserving LDA training algorithms.

Latency-aware service placement is promising in reducing the overall service response latency of proliferating edge-cloud collaborative smart manufacturing systems. However, intuitive latency estimators used by existing service placement approaches cannot accurately depict the nonlinear end-to-end (E2E) latency of multihop microservices with complex dependencies, which is severely hindering the effectiveness of latency-aware service placement. To address this issue, in this article, we present a microservice placement mechanism for edge-cloud collaborative smart manufacturing (MPCSM), where a microservice placement algorithm latency-aware edge-cloud collaborative placement supported by an accurate data-driven E2E latency estimation method is proposed. We build a real-world collaborative prototype, and conduct a case study on semiconductor manufacturing to elaborate the construction of our latency estimator. Results of extensive experiments demonstrate that the error of our E2E latency estimator is up to 10× less than that of existing ones, and the overall service latency with MPCSM is up to 10× less than that with existing service placement approaches.

The emerging edge-cloud collaborative Deep Learning (DL) paradigm aims at improving the performance of practical DL implementations in terms of cloud bandwidth consumption, response latency, and data privacy preservation. Focusing on bandwidth efficient edge-cloud collaborative training of DNN-based classifiers, we present CDC, a Classification Driven Compression framework that reduces bandwidth consumption while preserving classification accuracy of edge-cloud collaborative DL. Specifically, to reduce bandwidth consumption, for resource-limited edge servers, we develop a lightweight autoencoder with a classification guidance for compression with classification driven feature preservation, which allows edges to only upload the latent code of raw data for accurate global training on the Cloud. Additionally, we design an adjustable quantization scheme adaptively pursuing the tradeoff between bandwidth consumption and classification accuracy under different network conditions, where only fine-tuning is required for rapid compression ratio adjustment. Results of extensive experiments demonstrate that, compared with DNN training with raw data, CDC consumes 14.9× less bandwidth with an accuracy loss no more than 1.06%, and compared with DNN training with data compressed by AE without guidance, CDC introduces at least 100% lower accuracy loss.

The electrocardiogram (ECG) signal, as one of the most important vital signs, can provide indications of many heart-related diseases. Nonetheless, in the case of telehealth context, the automated analysis and accurate detection of ECG signals remain unsolved issues, because the poor data quality collected by the wearable devices and unprofessional users further increases the complexity of hand-crafted feature extraction, ultimately affecting the efficiency of feature extraction and the detection accuracy. To address this issue and improve the detection accuracy, in this paper we present a novel detection scheme with the raw ECG signal in wearable telehealth system. Our system benefits from the concept of big data, sensing and pervasive computing and the emerging deep learning technology. In particular, a Deep Heartbeat Classification (DHC) scheme is proposed to analyze the ECG signal for arrhythmia detection. Distinct from existing solutions, the detection model in DHC can be trained directly on the raw ECG signal without hand-crafted feature extraction. A cloud-based prototypical system is also designed and implemented with the functions of data acquisition, wireless transmission, back-end data management, and ECG detection. The experimental results demonstrate that our prototypical system is feasible and effective in real-world practice, and extensive experimentation based on the MIT-BIH database demonstrates that the proposed DHC scheme outperforms baseline schemes.

With the explosive growth of mobile devices and subsequent traffic volume, densified deployment of Heterogeneous Network (HetNet) coexisting with a variety of cells with overlay coverage has emerged as a viable solution for future 5G networks. Despite many advantages, this new architecture also introduces numerous new network management issues, such as frequent handovers. Although a number of handover mechanisms have been proposed, these methods generally depend on one or more objective attributes from the perspective of the users and network, and do not consider the subjective demands of personalized users and specific applications that demand differentiated network services. In this paper, we first develop an software defined networking (SDN)-driven handover architecture that is capable of perceiving global network statements and requirements from all perspectives, including the physical layer, users, and applications. Then, a context-aware multi-criteria handover mechanism is developed in the SDN controller to provide differentiated services. Considering the many complicated factors, the handover decision is made based on a hierarchical fuzzy inference system to process diverse attributes and fuzzy information described in natural language. The evaluation results demonstrate that our scheme outperforms the baseline Received Signal Strength Indicator (RSSI)-based handover scheme, more efficiently providing differentiated services with respect to throughput, bandwidth cost, and application satisfaction.

The development of the Internet of Things (IoT), cyber physical systems, and ubiquitous mobile terminals enable video content to be shared and consumed in new and innovative ways. Nonetheless, the stochastic and unpredictable nature of heterogeneous wireless networks with mobile clients presents a significant challenge to the increasing demand of quality of experience (QoE). In this paper, we study the problem of maximizing the user's QoE of viewing streaming video via automatic bitrate adaptation in heterogeneous wireless networks. To be specific, a stochastic optimization problem is first formulated by considering fundamental uncertainties of heterogeneous wireless networks (i.e., the stochastic throughput). A dynamic bitrate adaptation scheme is then designed based on the Lyapunov optimization framework. Our proposed scheme conducts video bitrate selection based on the current queue buffer state and real-time throughput, and is capable of balancing the tradeoff between a user's QoE and buffer occupation (i.e., memory utilization). The performance of our proposed scheme is investigated through a combination of extensive analysis, simulations, and experiments in a real-world testbed. Simulation results demonstrate that our scheme outperforms the baseline with regard to QoE and bandwidth utilization. In addition, the experimental results in real-world testbed validate the scheme's efficiency and practicality in real-world system.

In recent years, the smart grid has been recognized as an important form of the Internet of Things application. In the smart grid, as an energy-based cyber-physical system, the advanced metering infrastructure (AMI) will be developed to monitor and control the power grid by integrating computing and networking components to ensure stable and efficient operation. The AMI is vulnerable to cyber attacks, especially data integrity attacks. There have been a number of research efforts on detecting such attacks. Nonetheless, most of existing schemes either rely on predefined thresholds or require external knowledge. This may lead to low detection accuracy when the thresholds are improperly defined, and where there is a lack of the external knowledge. To address these issues, in this paper, we propose a Gaussian-mixture model-based detection scheme to mitigate data integrity attacks. Not relying upon the predefined thresholds or external knowledge, our developed scheme operates through narrowing the range of normal data, which can be obtained through clustering the historical data and learning minimum and maximum values or distance values to each center of individual clusters. To evaluate the effectiveness of our proposed scheme, we conduct performance simulation based on the ElectricityLoadDiagrams20112014 data set, and then analyze the effectiveness of the proposed scheme with respect to detection accuracy and overhead. The results of our investigation show that our scheme could achieve a higher detection rate, and a lower error rate, in comparison to existing schemes based on the Min-Max model.

In the smart grid, the Advanced Metering Infrastructure (AMI) will be deployed to monitor and control the power grid by integrating both computing and networking components to achieve stable and efficient operation. The AMI is vulnerable to cyber attacks, especially in the form of data integrity attacks. A number of research efforts have been devoted to detecting such attacks. Nonetheless, the majority of existing schemes either rely on a pre-defined threshold, or require external knowledge. This leaves open the possibility for low detection accuracy when the threshold is improperly defined, and where there is a lack of the requisite external knowledge. To address this issue, in this paper we propose a Gaussian-Mixture Model-based Detection (GMMD) scheme to combat data integrity attacks. Not relying upon the pre-defined threshold or external knowledge, our scheme operates by narrowing the range of normal data that can be obtained by clustering the historical data and learning the minimum and maximum values of individual clusters. To validate the effectiveness of our scheme, we conduct performance evaluation based on the ElectricityLoadDiagrams20112014 data set, and analyze the effectiveness of the proposed scheme with respect to detection accuracy.The results of our investigation demonstrate that our scheme can achieve a higher detection rate, and lower error rate, in comparison with existing schemes based on the Min-Max model.

Wireless Mesh Networks (WMNs) have potential to provide convenient broadband wireless Internet access to mobile users. With the emergence of Software-Defined Networking (SDN) paradigm that separates control plane and data plane, WMNs can be easily deployed and managed. In addition, by exploiting the broadcast nature of the wireless medium and the spatial diversity of multi-hop wireless networks, intra-flow network coding has shown a greater benefit in comparison with traditional routing paradigms in data transmission for WMNs. In this paper, we develop a novel OpenCoding protocol, which combines the SDN technique with intra-flow network coding for WMNs. Our developed protocol can simplify the deployment and management of the network and improve network performance. In OpenCoding, a controller working on the control plane makes routing decisions for mesh routers and the hop-by-hop forwarding function is replaced by network coding functions in data plane. Through a simulation study, we show the effectiveness of the OpenCoding protocol in comparison with existing schemes. Our data shows that OpenCoding outperforms both traditional routing and intra-flow network coding schemes.

Improving TCP performance over Wireless network is critical to enhancing the Quality of Service of mobile users. However, most conventional approaches can neither discover the hidden drawbacks, nor provide both fast and accurate mechanisms to adapt to unique Wireless TCP specifications, thus cannot guarantee desirable performance. In this article, we propose WiTracer, a novel solution for wireless TCP performance enhancement. WiTracer uses several criterion to extract packet losses and Round-Trip-Time (RTT) information, thus to infer potential channel variations. Hereafter, WiTracer builds a simple and robust design to decouple packet loss recovery from TCP congestion control, and mitigates spurious performance degradation. Therefore, satisfactory throughput is maintained, with no modification to existing Reno compatible TCP suite. Extensive simulation and implementation results show that WiTracer outperforms other benchmark approaches and performs well in varying wireless environments.

Admission control (AC) is an effective mechanism for providing bandwidth assurance in wireless mesh networks. Early AC schemes over multirate WMNs typically use a pre-chosen rate or a MAC-layer adapted rate for each link, denying data sessions that could have been admitted should a better multirate AC be available. Taking full advantage of multirate WMNs, we present a rate-adaptive admission control protocol (RaAC) for IEEE 802.11-based WMNs. RaAC consists of three major components: (1) a rate adaption algorithm to meet the bandwidth requirement of the data session and satisfy the channel condition of the PHY layer; (2) a new path-selection metric to balance between hop counts, bandwidth, rates, and other network parameters; and (3) a routing-coupled, distributed, rate-adaptive admission control algorithm to admit data sessions with bandwidth assurance. Through simulations, we show that RaAC is efficient and effective in meeting bandwidth requirements.

Opportunistic routing (OR) is a viable approach for improving performance of wireless communications. Previous studies on OR have focused on cost minimization, performance of multiple rates, congestion control, and other issues. Bandwidth assurance over OR, however, has not been adequately investigated. To bridge this gap, we present a bandwidth-aware opportunistic routing (BOR) with admission control (AC) protocol named BOR/AC. In particular, by analyzing the expected available bandwidth (EAB) and the expected transmission cost (ETC) in OR, we first devise a new metric called BCR (bandwidth-cost ratio) to determine the priority of relays in the forwarding candidates set. Admission control is then applied to admit or reject traffic flows based on estimated expected available bandwidth. Extensive simulation results show that BOR/AC consistently achieves much better performance than existing opportunistic routing protocols.

Accurately measuring the available bandwidth information is critical for providing QoS assurance, especially for the bandwidth-limited 802.11-based wireless networks. However, the shared nature of wireless medium and IEEE 802.11 MAC pose great challenges for estimating the bandwidth accurately. This paper tends to tackle this issue. In particular, we first formally define the available bandwidth in IEEE 802.11 network by considering its unique characteristics. We then present our solution, Passive Available Bandwidth Estimation (PABE). In PABE, the effective link capacity is analyzed by considering the random factors in transmission, and the available channel idle time is estimated by passively monitoring the medium based on a new, lower threshold bandwidth obtained during the normal operations of IEEE 802.11. Our approach incurs very low cost to the network without any explicit message overhead. Through extensive simulation, our data validate that our approach consistently achieves much better performance than other existing algorithms in term of estimation accuracy.

Admission control plays an important role in providing Quality of Service (QoS) guarantees for wireless mesh networks (WMNs). Multipath routing can improve network performance in reliability and load balancing. However, when the multipath routing are adopted in 802.11-based WMNs, the transmission with bandwidth assurance is facing rigorous challenges. In this paper, a novel joint design of multipath routing and admission control protocol is presented, named MRAC. In MRAC, the multipath routing with bandwidth assurance is formulated as an optimization problem based on the analysis of available bandwidth and the bandwidth consumption. Based on the formulation, a heuristic solution is proposed to admit the data session through two parallel paths with bandwidth assurance. Through extensive simulations, the effectiveness of the MRAC is demonstrated in term of satisfying bandwidth requirement.