Path planning is an important mean to alleviate traffic congestion and reduce travel cost in the internet of vehicles. Existing path planning strategies primarily rely on shortest path or single-agent algorithms. However, they encounter challenges related to global dynamic coordination safety and resource constraints. Therefore, we propose a Multi-Agent dynamic Path Planning strategy based on the Principal component Federation (MA3PF) to address the these challenges. In this strategy, we introduce the Heuristic Multi-Agent Deep Policy Gradient algorithm (H-MADPG), which incorporates future traffic states and leverages shared agent experiences through migration learning. This approach ensures fast convergence and addresses resource limitations through a combination of centralized evaluation and distributed decision-making. Next, we propose the Adaptive Client-based Principal component Federation learning algorithm (ACPFed) for real-time prediction of future traffic flow. This algorithm utilizes principal component analysis to select model parameters and incorporates Bayesian optimization to dynamically determine client weights. These enhancements improve global coordination security and reduce communication overhead. Experimental results on real and simulated datasets demonstrate that our proposed MA3PF method outperforms existing agent algorithms, such as DARP and MAPF. It achieves superior route planning in complex environments, resulting in a reduction of travel distance by 21.90% and time loss by 39.41%. Additionally, MA3PF improves the real-time prediction accuracy of future traffic states while achieving a significant 70% reduction in communication cost. Note to Practitioners—This paper was motivated by the problem of holding multi-vehicle path planning. Existing path planning approaches only consider path planning for a single vehicle, which 1) fail to consider the challenges posed by resource constraints and data security in multi-vehicle planning 2) neglect the influence of future traffic states on path planning. This paper suggests a new dynamic path planning strategy, which incorporates real-time information about future traffic states, ensures data security, and reduces communication overhead using a specially designed ACPFed algorithm, simultaneously leveraging transfer learning and distributed decision-making to address resource constraints. In this paper, we present a mathematical characterization of the multi-vehicle path planning problem, incorporating future traffic states, and analyze the problem to derive an effective heuristic scheme MA3PF. The experimental results on real and simulated datasets demonstrate that the proposed MA3PF achieves significant improvements in terms of route distance and time loss compared to the traditional agent algorithm. In future research, we will explore more complex road environments and incorporate the effects of safety attacks on dynamic agent path planning.

Recent research tends to address the biomedical entity linking problem in a unified framework solely based on surface form matching between mentions and entities. Specifically, these methods focus on addressing thevarietychallenge of the heterogeneous naming of biomedical concepts. Yet, theambiguitychallenge that the same word under different contexts can be used to refer to distinct concepts is usually ignored. To address this challenge, we propose BioPRO, a two-stage entity linking algorithm to enhance the biomedical entity representations based on context-infused prompt learning. The first stage includes a coarse-grained retrieval from a representation space defined by a bi-encoder that independently embeds the mention and entity's surface forms. Unlike previous one-model-fits-all systems, each candidate is then re-ranked with a fine-grained encoder based on prompt-tuning that sufficiently stimulates knowledge in contextual information of mentions and entities. Furthermore, the trained fine-grained encoder can be utilized to generate deep representations of bio-entities and boost candidate retrieval in the first stage. Extensive experiments show that our model achieves promising performance improvements compared with several state-of-the-art (SOTA) techniques on 4 biomedical corpora. We also observe by cases that the proposed context-infused prompt-tuning strategy is effective in solving both thevarietyandambiguitychallenges in the linking task.

The Industrial Internet of Things (IIoT) has found extensive applications in intelligent transportation. However, as the number of vehicles increases, the issue of traffic congestion becomes more prominent, emphasizing the need for accurate congestion propagation prediction to enhance traffic conditions. Current methods for predicting congestion propagation lack consideration for the influence of communication networks and do not incorporate the path characteristics of congestion propagation. Therefore, we propose a path-based congestion propagation model, UAU_SIS_Path, employing multigrain abstraction of traffic congestion and information propagation. Specifically, UAU_SIS_Path effectively captures the propagation dynamics of congestion in IIoT by leveraging the interaction of two-layer networks and the path propagation characteristics of traffic congestion. Subsequently, we validate the effectiveness of the UAU_SIS_Path model through theoretical analysis, establishing tight upper and lower bounds of the propagation threshold and elucidating the relationship between the propagation of congestion information in the information network and the diffusion of congestion in the road network. Finally, based on theoretical analysis, we examine the impact of our model on congestion control strategies, utilizing path replanning, and traffic restriction as congestion control strategies. Experimental results in simulated road network BA and real road networks of varying sizes (GC, TA, and As) demonstrate the stability and scalability of our model. In comparison to the traditional contact-based propagation model, our model achieves a reduction in congestion propagation rates of 58.2%, 66.9%, 32.6%, and 48.6%, respectively.

Multilingual biomedical entity linking (MBEL) aims to map language-specific mentions in the biomedical text to standardized concepts in a multilingual knowledge base (KB) such as Unified Medical Language System (UMLS).In this paper, we propose Con2GEN, a prompt-based controllable contrastive generation framework for MBEL, which summarizes multidimensional information of the UMLS concept mentioned in biomedical text into a natural sentence following a predefined template.Instead of tackling the MBEL problem with a discriminative classifier, we formulate it as a sequence-tosequence generation task, which better exploits the shared dependencies between source mentions and target entities.Moreover, Con2GEN matches against UMLS concepts in as many languages and types as possible, hence facilitating cross-information disambiguation.Extensive experiments show that our model achieves promising performance improvements compared with several state-of-the-art techniques on the XL-BEL and the Mantra GSC datasets spanning 12 typologically diverse languages.

Industrial Internet of Things (IIoT) technology is widely used in modern industrial fields like transportation, but data security remains a major challenge. The blockchain-based access control mechanism can address the data security issue by preventing unauthorized devices from accessing limited IIoT resources. However, most existing blockchain-based access control mechanism for IIoT still has scalability and privacy issues. To deal with the above-mentioned problems, we propose a new scalable and secure strategy for the blockchain-based access control framework for IIoT via sharding, which consists of two components. First, the network sharding scheme based on the access frequency set (N2SAF) is designed to 1) improve the scalability of our proposed strategy by transaction sharding to reduce the storage pressure on nodes, and 2) increase the transaction processing speed on a three-layer architecture based on cloud-edge-device in IIoT. Second, the privacy protection scheme based on a Bloom filter (P2BF) is designed to deal with the privacy leakage problem caused by anonymous address clustering. Simulation results show that our proposed strategy improves the scalability of the system compared to existing methods while ensuring the security of the shard network, especially in large-scale IIoT environments.

Biomedical entity linking aims to map mentions in biomedical text to standardized concepts or entities in a curated knowledge base (KB) such as Unified Medical Language System (UMLS). The latest research tends to solve this problem in a unified framework solely based on surface form matching between mentions and entities. Specifically, these methods focus on addressing the variety challenge of the heterogeneous naming of biomedical concepts. Yet, the ambiguity challenge that the same word under different contexts may refer to distinct entities is usually ignored. To address this challenge, we propose a two-stage linking algorithm to enhance the entity representations based on prompt learning. The first stage includes a coarser-grained retrieval from a representation space defined by a bi-encoder that independently embeds the mention and entity’s surface forms. Unlike previous one-model-fits-all systems, each candidate is then re-ranked with a finer-grained encoder based on prompt-tuning that utilizes the contextual information. Extensive experiments show that our model achieves promising performance improvements compared with several state-of-the-art techniques on the largest biomedical public dataset MedMentions and the NCBI disease corpus. We also observe by cases that the proposed prompt-tuning strategy is effective in solving both the variety and ambiguity challenges in the linking task.

Traffic flow prediction (TFP) is of great significance in the field of traffic congestion mitigation on the Internet of Vehicle(Iov). To be capable of a trade-off between data privacy protection and accurate prediction, we introduce a training paradigm based on Federated Learning (FL). However, the implementation of federal learning in practice is confronted with high communication and data heterogeneity. In this paper, Principal component analysis (PCA) is introduced to minimize the scale of data transmission on both the client and server. Due to the errors arising from the compression and reversion of the transmission model, we add an additional error term in the local objective function. To address the imbalanced data distribution and to accelerate the federal learning convergence, we then propose a mechanism that incorporates Bayesian optimization to dynamically determine the weights of clients during aggregation. With extensive experiments on real data, it can be demonstrated that communication costs can be minimized by 60-70% while ensuring fewer errors.

Social Internet of Things (SIoT) boosts the Internet of Things (IoT) by integrating the concept of social networking to advance the discovery of content and service. Another common and promising solution to IoT’s improvement is adopting the emerging multiaccess edge computing (MEC) paradigm by bringing computational and storage capacity at the edge. Concerning the MEC-enabled SIoT, this article aims to propose a network-embedding-based solution for scalable network navigation via leveraging the social similarity of SIoT. Different from the traditional SIoT, we characterize the social relationship from the contents point of view since MEC enables the edge with cache storage. Then, a novel heuristic embedding algorithm termed HeurEmb is proposed to embed the SIoT into the hyperbolic space to facilitate coordinate-based navigability. HeurEmb is a decentralized approach that leverages the social similarity to compute coordinates, bypassing the heavy computation of numerically maximizing the likelihood. Moreover, HeurEmb infers the virtual coordinates on multiple-level hyperbolic disks, which enabled the efficiently vertical and horizontal search for the destination. We analyze the complexity of HeurEmb and then evaluate its performance via the simulation study. The simulation results show that HeurEmb is efficient, achieving a decrease of up to two orders of magnitude of running time compared with the latest work, and effective, improving the success ratio and navigation path length. Finally, we apply HeurEmb in a content retrieval scenario of SIoT using a real-world trace. We make only a minor modification to HeurEmb-based forwarding, and the resulting strategy can achieve similar performance as the best benchmark.

Design an optimization model for task management among Mobile Terminal (MT), Macro cell Base Station (MBS), and multiple Small cell Base Stations (SBS) for the large-scale Mobile Edge Computing (MEC) system, is a challenging issue due to the large number of tasks and SBSs. Inspired by this, we propose a Parallel Optimal Task Allocation Mechanism (POTAM) framework for MEC, which includes Device to Device (D2D)-enabled computing, MBS computing and Edge Computation Resource Distribution (ECRD) computing. In POTAM, we exploit a parallel multi-block Alternating Direction Method of Multipliers (ADMM) based method to model both requirements of delay and energy consumptions, which formulates the task allocation under these requirements as a nonlinear 0–1 integer programming problem. To solve this problem, we develop an efficient combination of conjugate gradient, Newton and linear search techniques based algorithm with Logarithmic Smoothing and Cyclic Block coordinate Gradient Projection (CBGP) methods, which can guarantee convergence and reduce computational complexity with a good scalability. In order to allocate task cooperatively, an optimal approach is proposed, ECRD-A, which is used to find the shortest path among each node. Numerical results demonstrate the effectiveness of the POTAM and it can effectively reduce delay and energy consumption for a large-scale MEC system.

Long sequence time-series forecasting(LSTF) plays an important role in a variety of real-world application scenarios, such as electricity forecasting, weather forecasting, and traffic flow forecasting. It has previously been observed that transformer-based models have achieved outstanding results on LSTF tasks, which can reduce the complexity of the model and maintain stable prediction accuracy. Nevertheless, there are still some issues that limit the performance of transformer-based models for LSTF tasks: (i) the potential correlation between sequences is not considered; (ii) the inherent structure of encoder-decoder is difficult to expand after being optimized from the aspect of complexity. In order to solve these two problems, we propose a transformer-based model, named AGCNT, which is efficient and can capture the correlation between the sequences in the multivariate LSTF task without causing the memory bottleneck. Specifically, AGCNT has several characteristics: (i) a probsparse adaptive graph self-attention, which maps long sequences into a low-dimensional dense graph structure with an adaptive graph generation and captures the relationships between sequences with an adaptive graph convolution; (ii) the stacked encoder with distilling probsparse graph self-attention integrates the graph attention mechanism and retains the dominant attention of the cascade layer, which preserves the correlation between sparse queries from long sequences; (iii) the stacked decoder with generative inference generates all prediction values in one forward operation, which can improve the inference speed of long-term predictions. Experimental results on 4 large-scale datasets demonstrate the AGCNT outperforms state-of-the-art baselines.

Access control is one of the important means to protect privacy in the Internet of Things (IoT), and access control model is mostly based on central trusted entity. The decentralized blockchain technology makes up for the shortcomings of the existing access control model for IoT. However, the transaction generation rate in IoT has brought huge challenges to the scalability of the integration framework of blockchain and IoT, including the Transaction Processing Speed (TPS) and the storage of each node.This paper proposes the Network Sharding Scheme (NsScheme) to improve the scalability of the blockchain-based access control framework in IoT. Based on the three-tier architecture of "cloud-edge-device" in IoT, we divide edge nodes into several shards. Each shard maintains a local blockchain, and cloud nodes maintain a global blockchain. Multiple blockchains process transactions in parallel to improve the Transaction Processing Speed (TPS) and reduce the storage of each node. In the meantime, we present the transactions allocation scheme that which shard processing the transaction depends on both parties in transaction, which can reduce the routing cost of querying transactions to O(1). Considering that the number of cross-shard transactions will affect the query cost of cross-shard transactions, we further introduce the network sharding algorithm based on Access Frequency Set (AFS) between nodes, which can effectively reduce the query cost of cross-shard transactions. The simulation results indicate that with the increase of the number of shards, NsScheme can linearly increase the TPS and reduce the storage of each node.

It has become a common practice to train large machine learning (ML) models across a cluster of computing nodes connected by RDMA-enabled networks. However, the communication overhead caused by parameter synchronization deteriorates the performance of such distributed ML (DML), especially in a large-scale setting. This paper tackles this issue by developing a traffic management scheme to support DML traffic, called TMDML (Traffic Management for DML), which needs only a minor modification to the existing RDMA congestion control scheme DCQCN. We assume that there is only one instance of DML workload running in a network. Existing literature has shown that Fat-Tree, a predominant topology in the data center, poorly supports DML compared with BCube. With our proposed TMDML, training DML in Fat-Tree can achieve better performance than that in BCube. We first study the impact of multi-bottlenecks on DML via NS-3-based simulations. The results show that DCQCN is inefficient for DML traffic in the multi-bottlenecks scenario. To mitigate the impact of multi-bottlenecks, we propose an optimization model to minimize the maximum flow completion time (FCT) while stabilizing the queues, and then apply the Lyapunov optimization technique to solve the problem. For all the practical purposes, we present two heuristic implementations of TMDML for different deployment requirements. We evaluate the performance of our proposals by simulation, comparing it with DCQCN. We use All-Reduce parameter synchronization in Fat-Tree and BCube with traffic trace of modern deep neural network models, including AlexNet, ResNet50, and VGG-16. Our proposals can achieve up to 59% of the time reduction.

With the development of blockchain technology, people always expect that blockchain can be applied to other fields. However, the low Transaction Processing Speed (TPS) and broadcast delay of blockchain still restrict the application of blockchain. To solve these problems, we propose a new scheme named GVScheme to improve the scalability of blockchain network. GVScheme introduces the role of guarantor based on trust value mechanism. The guarantor node will guarantee the block spread in the network. When the node receives the guarantee block from the guarantor node, the order of verification block and propagation block will be determined according to the trust value of the guarantor. By reducing the block verification time, the block propagation delay in the network will also be reduced. It is worth mentioning that our scheme keeps the minimum modification to the blockchain, and may even be directly applied to the blockchain network. Simulation results show that GVScheme can effectively reduce block propagation delay and the fork rate in blockchain network. When the block size and the number of nodes increase, GVScheme also shows great performance. Thus, under the same fork rate, the blockchain using GVScheme can allow less mining interval and larger block size limit.

We consider the problem of the intelligent and efficient resource management framework in mobile-edge computing (MEC), which can reduce delay and energy consumption, and features distributed optimization and efficient congestion avoidance. In this article, we present a cooperative learning framework for resource management in MEC from an alternating direction method of multipliers (ADMMs) perspective, named the CL-ADMM framework. First, computing a task requires both the user personal data and corresponding program that processes it, to efficiently cache program in a group, a novel program popularity estimation scheme is proposed, which is based on a semi-Markov process model. Then, a greedy program cooperative caching mechanism is established, which can effectively reduce delay and energy consumption. Second, to address group congestion, a dynamic task migration scheme based on improved cooperative Q-learning is proposed, which can effectively reduce delay and alleviate congestion. Third, to minimize delay and energy consumption for resource allocation in a group, we formulate it as an optimization problem with a large number of variables, and then exploit a novel ADMM-based scheme to solve this problem, which can reduce the complexity of the problem with a new set of auxiliary variables, these subproblems are all convex problems that can be solved by using a primal-dual approach, which guarantees its convergence. Finally, we prove its convergence by using the Lyapunov theory. The numerical results demonstrate the effectiveness of the CL-ADMM framework in reducing delay and energy consumption in MEC.

OBJECTIVE: There have been various methods to deal with the erroneous training data in distantly supervised relation extraction (RE), however, their performance is still far from satisfaction. We aimed to deal with the insufficient modeling problem on instance-label correlations for predicting biomedical relations using deep learning and reinforcement learning. MATERIALS AND METHODS: In this study, a new computational model called piecewise attentive convolutional neural network and reinforcement learning (PACNN+RL) was proposed to perform RE on distantly supervised data generated from Unified Medical Language System with MEDLINE abstracts and benchmark datasets. In PACNN+RL, PACNN was introduced to encode semantic information of biomedical text, and the RL method with memory backtracking mechanism was leveraged to alleviate the erroneous data issue. Extensive experiments were conducted on 4 biomedical RE tasks. RESULTS: The proposed PACNN+RL model achieved competitive performance on 8 biomedical corpora, outperforming most baseline systems. Specifically, PACNN+RL outperformed all baseline methods with the F1-score of 0.5592 on the may-prevent dataset, 0.6666 on the may-treat dataset, and 0.3838 on the DDI corpus, 2011. For the protein-protein interaction RE task, we obtained new state-of-the-art performance on 4 out of 5 benchmark datasets. CONCLUSIONS: The performance on many distantly supervised biomedical RE tasks was substantially improved, primarily owing to the denoising effect of the proposed model. It is anticipated that PACNN+RL will become a useful tool for large-scale RE and other downstream tasks to facilitate biomedical knowledge acquisition. We also made the demonstration program and source code publicly available at http://112.74.48.115:9000/.

Blockchain is essentially a distributed ledger shared by all nodes in the system. All nodes in blockchain are equal, and each node holds all transactions and blocks in the network. As the network continues to expand, the data rises linearly. Participates are about to face the problem of storage limitation. Blockchain is hard to scale.This paper introduces ICIStrategy, a multi-node collaborative storage strategy based on intra-cluster integrity. In ICIStrategy, we divide all participates into several clusters. Each cluster requires holding all data of the network, whereas a node within the cluster does not need to maintain data integrity. It aims to solve the storage pressure by reducing the amount data that each participate need to store and reduce communication overhead by collaboratively storing and verifying blocks through in-cluster nodes. Moreover, the ICIStrategy could greatly save the overhead of bootstrapping. We show the mode of operation in our strategy. We further analysis the performance of ICIStrategy and conduct simulation experiments. The results of several comparative experiments show that our strategy just needs 25% of storage space needed by Rapidchain, which indeed solve the problem of storage limitation and improve the blockchain performance.

Distributed synchronous stochastic gradient descent (SGD) algorithms are widely used in large-scale deep learning applications, while it is known that the communication bottleneck limits the scalability of the distributed system. Gradient sparsification is a promising technique to significantly reduce the communication traffic, while pipelining can further overlap the communications with computations. However, gradient sparsification introduces extra computation time, and pipelining requires many layer-wise communications which introduce significant communication startup overheads. Merging gradients from neighbor layers could reduce the startup overheads, but on the other hand it would increase the computation time of sparsification and the waiting time for the gradient computation. In this paper, we formulate the trade-off between communications and computations (including backward computation and gradient sparsification) as an optimization problem, and derive an optimal solution to the problem. We further develop the optimal merged gradient sparsification algorithm with SGD (OMGS-SGD) for distributed training of deep learning. We conduct extensive experiments to verify the convergence properties and scaling performance of OMGS-SGD. Experimental results show that OMGS-SGD achieves up to 31% end-to-end time efficiency improvement over the state-of-the-art sparsified SGD while preserving nearly consistent convergence performance with original SGD without sparsification on a 16-GPU cluster connected with 1Gbps Ethernet.

Software Defined Networking (SDN) has attracted more and more attention due to its prominent features that are different from the traditional network. SDN is programmable through which controller can modify the rules in the switch. However, security was not considered in its initial design, and many manufacturers no longer support Transport Layer Security (TLS) due to the cost. Although many machine learning based approaches have been implemented in SDN, they all need features that experts extract from original data. However, the manual extraction increases the level of human interaction and decreases detection accurate. This paper presents a malicious network traffic classification method based on Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) to address these concerns. Our proposed method is implemented in Graphic Process Unit (GPU) enabled TensorFlow. We evaluated our proposal on three datasets. The results demonstrate that our proposal achieves improvements in term of detection accuracy and stability over existing approaches and strong potential for user in SDN security.

Delay Tolerant Network (DTN) is a multi-hop wireless network, which has an intermittent connection due to the mobility of nodes, short of wireless communication range, sleeping mode of nodes. As a result, delay tolerant network usually gets higher average delay and overhead. In this paper, we propose an Inter-session Network Coding based Clustering routing (INCC). Inter-session network coding codes messages from different flow together using the broadcast nature of wireless network to improve the transmission efficiency. Due to the dynamic topology structure, the inter-session network coding scheme cannot be applied to DTNs directly. Therefore, we cluster the node to gain more coding opportunities. We classify the nodes into different clusters according to their contact frequency. The packets could be performed network coding in the proposed scheme even if they have different destination nodes. Then, we use Ordinary Differential Equations (ODEs) to investigate the advantage of applying network coding in DTNs in terms of average delay. Finally, we evaluate the performance of INCC via simulations, and simulation results show that INCC can achieve lower average delivery delay and reduce the network load.

With the rapid development of cloud computing, many cloud service providers have been deploying more and more datacenters to provide better reliability and quality of service. The energy optimization problem has become an emerging concern. The current researches focus on how to either reduce the energy consumption of servers or reduce the consumption of inter-datacenters data transmission. However, energy optimization of joint inter-datacenter and servers has not been explored. In this paper, we first introduced the background to the energy consumption problem of geographically distributed datacenters. Then, based on the Service Level Agreement (SLA), this paper proposes an online control framework to minimize the energy consummation cost. The online control framework can dynamically make decisions by adjusting geographically load balancing, capacity right-sizing, server speed scaling, and flow programming. Finally, the simulation verifies the effectiveness of the proposed framework in cost saving.

Named Data Networking (NDN) is a promising instance of Information-Centric Networking (ICN). With the expansion of the network, unbounded namespace and query of the routing table in NDN can deteriorate routing performance. Since Hyperbolic Routing (HR) does not need to maintain a full routing table, it becomes a potential solution to this problem. Existing works assign coordinate based on betweenness centrality of nodes. The betweenness-based solution can fully embed the network into hyperbolic space; however, it brings a problem that packets are aggregated to high-betweenness nodes. In this paper, by jointly considering the betweenness centrality of nodes and popularity of contents while assigning hyperbolic coordinate, we first propose a content-based hyperbolic routing called Pop-Hyper. As result, packets are sent to nodes with high betweenness and high popularity. Then, a push mechanism based on Pop-Hyper called HyperPush is presented. Finally, we compare our proposals with the existing mechanisms in the 22-node and 100-node topology, respectively. The simulation results show that Pop-Hyper performs well in terms of hop count and packet loss; while HyperPush outperforms others in terms of network load, cache hit ratio and delays.

Congestion control is one of the most critical issues in Named Data Networking (NDN). Compared to traditional Internet, NDN has some new features: receiver-driven, in-network caching, hop-by-hop forwarding, etc. These new features pose new challenges for designing congestion control mechanism. Congestion in NDN is mainly caused by Data packets, thus, shaping the transmission rate of Interest packet can regulate the returning rate of Data packet. In this paper, we propose an Interest shaping mechanism to tackle congestion in NDN by controlling Interest packet transmission rate in an optimized way. We formulate the rate allocation problem as a global optimization problem via jointly considering congestion control and traffic management. In order to achieve traffic management objective, we add an extra term in utility function to penalize over-loaded links. By applying partial dual decomposition, we solve the optimization problem with a gradient-based algorithm and we prove that gradient-based algorithm converges to optimality of optimization. Then, we present a practical implementation of this algorithm. Finally, we conduct simulation in ndnSIM to evaluate performance of the proposed mechanism by comparing with other existing methods. Simulation results show that our proposed mechanism can achieve high ratio of satisfied Interest, low delay and packet drop rate. The proposed mechanism can also achieve fairness among flows though the link utilization may not be high.

Answer selection in community question answering (cQA) is a challenging task in natural language processing. The difficulty lies in that it not only needs the consideration of semantic matching between question answer pairs but also requires a serious modeling of contextual factors. In this letter, we propose an attentive deep neural network architecture so as to learn the deterministic information for answer selection. The architecture can support various input formats through the organization of convolutional neural networks, attention-based long short-term memory, and conditional random fields. Experiments are carried out on the SemEval-2015 cQA dataset. We attain 58.35% on macroaveraged F1, which outperforms the Top-1 system in the shared task by 1.16% and improves the state-of-the-art deep-neural-network-based method by 2.21%.

In community question answering (cQA), the quality of answers are determined by the matching degree between question-answer pairs and the correlation among the answers. In this paper, we show that the dependency between the answer quality labels also plays a pivotal role. To validate the effectiveness of label dependency, we propose two neural network-based models, with different combination modes of Convolutional Neural Net-works, Long Short Term Memory and Conditional Random Fields. Extensive experi-ments are taken on the dataset released by the SemEval-2015 cQA shared task. The first model is a stacked ensemble of the networks. It achieves 58.96% on macro averaged F1, which improves the state-of-the-art neural network-based method by 2.82% and outper-forms the Top-1 system in the shared task by 1.77%. The second is a simple attention-based model whose input is the connection of the question and its corresponding answers. It produces promising results with 58.29% on overall F1 and gains the best performance on the Good and Bad categories.

In this paper, we consider the joint opportunistic routing and channel assignment problem in multi-channel multi-radio (MCMR) cognitive radio networks (CRNs) for improving aggregate throughput of the secondary users. We first present the linear programming optimization model for this joint problem, taking into account the feature of CRNs-channel uncertainty. Then considering the queue state of a node, we propose a new scheme to select proper forwarding candidates for opportunistic routing. Furthermore, a new algorithm for calculating the forwarding probability of any packet at a node is proposed, which is used to calculate how many packets a forwarder should send, so that the duplicate transmission can be reduced compared with MAC-independent opportunistic routing & encoding (MORE) [11]. Our numerical results show that the proposed scheme performs significantly better that traditional routing and opportunistic routing in which channel assignment strategy is employed.

Cognitive radio (CR) is a promising technology to improve spectrum utilization. However, spectrum availability is uncertain which mainly depends on primary user's (PU's) behaviors. This makes it more difficult for most existing CR routing protocols to achieve high throughput in multi-channel cognitive radio networks (CRNs). Inter-session network coding and opportunistic routing can leverage the broadcast nature of the wireless channel to improve the performance for CRNs. In this paper we present a coding aware opportunistic routing protocol for multi-channel CRNs, cognitive radio opportunistic routing (CROR) protocol, which jointly considers the probability of successful spectrum utilization, packet loss rate, and coding opportunities. We evaluate and compare the proposed scheme against three other opportunistic routing protocols with multichannel. It is shown that the CROR, by integrating opportunistic routing with network coding, can obtain much better results, with respect to throughput, the probability of PU-SU packet collision and spectrum utilization efficiency.

It is challenging to provide QoS in wireless ad hoc networks. One of the main problems that affects QoS in wireless ad hoc networks is the multi-hop nature of the network which generates high volume of control overhead. Such overhead is necessary to create/maintain network routes. Congested routes further aggravate the existing problem by generating more control packets for performing route maintenance. In addition, established routes are not well-protected and are subject to the changes in traffic load, which further reduces the QoS provided and wastes network resources. Therefore, in this paper, we propose a flow admission control scheme at the routing layer to mitigate the above problem by taking pre-emptive measures through bandwidth and congestion estimation. The scheme can reduce unnecessary control overhead for route maintenance and select a proper route for a flow, thus improve the overall network performance. Moreover, established routes can be protected from any changes in traffic load for the duration of the route lifetime. Our extensive simulation results demonstrate that the proposed scheme achieves good network performance.

This paper investigates the performance of network coding (NC) in unreliable wireless networks and the integration with TCP protocol. As the wireless nodes have limited processing capacity and energy, it will be difficult for them to deal with complex problems. Coding and decoding with traditional NC will cause a large overhead for wireless nodes. It is necessary to improve the NC scheme for applying to wireless networks. Furthermore, unreliable wireless channels will result in a lot of unnecessary retransmissions in wireless networks. In this paper, we propose a joint generation network coding scheme to improve the performance of NC in wireless networks. We first analyze the impact of the probability of decoding under lossy wireless channels in the traditional NC and joint generation NC. Then, we design a scheme that integrates the proposed network coding scheme with TCP. By adopting the joint generation NC, we could avoid unnecessary retransmissions in wireless networks due to the loss of acknowledgment in TCP protocol. Our simulation results demonstrate that joint generation NC could greatly reduce retransmissions.

We explore the problem of designing on-line virtual private networks (VPN) over wavelength division multiplexing networks (WDM) to facilitate the guarantee of diverse quality of service (QoS) requirements of different traffic streams especially that with very stringent delay jitter requirement and that with bursty characteristic. Each VPN consists of different traffic streams with various quality sensitivity. The proposed QoS model provides three types QoS services for various VPNs, static, optical burst switching (OBS) and alternative VPNs. All the three types provide transmission in the optical core networks without any O/E/O conversion. The static type has a set of dedicated or shared lightpaths which have been pre-allocated and the traffic from this VPN has constant bit rate (CBR). The OBS type need to setup the lightpath using a so called optical burst switching (OBS) while a traffic stream with variable bit rate (VBR) request is occurring. The alternative one has to setup a lightpath based on two-way reservation protocols such as tell-and-wait (TAW) for the traffic with unspecified bit rate (UBR) or available bit rate (ABR). The proposed model is suitable for carrying the multimedia applications which have various sensitivity of delay and jitter. By employing dynamic VPNs directly over WDM layer, we can eliminate the often complex QoS related functions in the upper protocol layers, thus increase the overall communication efficiency. Furthermore, this could be the first step toward providing fine grained service over optical wide-area networks. In addition, we present preliminary results based on simulation and provide a network designer with a quantitative assessment of proportion for different traffic types with different QoS service requirements.

We introduce an analytical model to obtain the performance bounds for a single hop optical network based on WDM. We model the network under saturation traffic loading while ignoring the specific channel access scheme as a multi-class closed queueing network (BCMP network), for which a product-form solution for the steady-state probabilities exists. Therefore the exact solutions can be derived. In addition, we study the effect on the system performance of the number of stations in the network, the number of transmitters/receivers at each station, and the traffic distribution pattern.