Searcharxiv⌕ Search

arXiv subjects

Sheng Zhou

Publications and source records attributed to Sheng Zhou.

At least 163 records · Page 9Linked to original sources

Distributed Task Replication for Vehicular Edge Computing: Performance Analysis and Learning-based Algorithm

In a vehicular edge computing (VEC) system, vehicles can share their surplus computation resources to provide cloud computing services. The highly dynamic environment of the vehicular network makes it challenging to guarantee the task offloading delay. To this end, we introduce task replication to the VEC system, where the replicas of a task are offloaded to multiple vehicles at the same time, and the task is completed upon the first response among replicas. First, the impact of the number of task replicas on the offloading delay is characterized, and the optimal number of task replicas is approximated in closed-form. Based on the analytical result, we design a learning-based task replication algorithm (LTRA) with combinatorial multi-armed bandit theory, which works in a distributed manner and can automatically adapt itself to the dynamics of the VEC system. A realistic traffic scenario is used to evaluate the delay performance of the proposed algorithm. Results show that, under our simulation settings, LTRA with an optimized number of task replicas can reduce the average offloading delay by over 30% compared to the benchmark without task replication, and at the same time can improve the task completion ratio from 97% to 99.6%.

cs.NI↗

Urgency of Information for Context-Aware Timely Status Updates in Remote Control Systems

As 5G and Internet-of-Things (IoT) are deeply integrated into vertical industries such as autonomous driving and industrial robotics, timely status update is crucial for remote monitoring and control. In this regard, Age of Information (AoI) has been proposed to measure the freshness of status updates. However, it is just a metric changing linearly with time and irrelevant of context-awareness. We propose a context-based metric, named as Urgency of Information (UoI), to measure the nonlinear time-varying importance and the non-uniform context-dependence of the status information. This paper first establishes a theoretical framework for UoI characterization and then provides UoI-optimal status updating and user scheduling schemes in both single-terminal and multi-terminal cases. Specifically, an update-index-based scheme is proposed for a single-terminal system, where the terminal always updates and transmits when its update index is larger than a threshold. For the multi-terminal case, the UoI of the proposed scheduling scheme is proven to be upper-bounded and its decentralized implementation by Carrier Sensing Multiple Access with Collision Avoidance (CSMA/CA) is also provided. In the simulations, the proposed updating and scheduling schemes notably outperform the existing ones such as round robin and AoI-optimal schemes in terms of UoI, error-bound violation and control system stability.

cs.IT↗

AoI-Delay Tradeoff in Mobile Edge Caching with Freshness-Aware Content Refreshing

Mobile edge caching can effectively reduce service delay but may introduce information staleness, calling for timely content refreshing. However, content refreshing consumes additional transmission resources and may degrade the delay performance of mobile systems. In this work, we propose a freshness-aware refreshing scheme to balance the service delay and content freshness measured by Age of Information (AoI). Specifically, the cached content items will be refreshed to the up-to-date version upon user requests if the AoI exceeds a certain threshold (named as refreshing window). The average AoI and service delay are derived in closed forms approximately, which reveals an AoI-delay tradeoff relationship with respect to the refreshing window. In addition, the refreshing window is optimized to minimize the average delay while meeting the AoI requirements, and the results indicate to set a smaller refreshing window for the popular content items. Extensive simulations are conducted on the OMNeT++ platform to validate the analytical results. The results indicate that the proposed scheme can restrain frequent refreshing as the request arrival rate increases, whereby the average delay can be reduced by around 80% while maintaining the AoI below one second in heavily-loaded scenarios.

cs.NI↗

How Does BN Increase Collapsed Neural Network Filters?

Improving sparsity of deep neural networks (DNNs) is essential for network compression and has drawn much attention. In this work, we disclose a harmful sparsifying process called filter collapse, which is common in DNNs with batch normalization (BN) and rectified linear activation functions (e.g. ReLU, Leaky ReLU). It occurs even without explicit sparsity-inducing regularizations such as $L_1$. This phenomenon is caused by the normalization effect of BN, which induces a non-trainable region in the parameter space and reduces the network capacity as a result. This phenomenon becomes more prominent when the network is trained with large learning rates (LR) or adaptive LR schedulers, and when the network is finetuned. We analytically prove that the parameters of BN tend to become sparser during SGD updates with high gradient noise and that the sparsifying probability is proportional to the square of learning rate and inversely proportional to the square of the scale parameter of BN. To prevent the undesirable collapsed filters, we propose a simple yet effective approach named post-shifted BN (psBN), which has the same representation ability as BN while being able to automatically make BN parameters trainable again as they saturate during training. With psBN, we can recover collapsed filters and increase the model performance in various tasks such as classification on CIFAR-10 and object detection on MS-COCO2017.

cs.LG↗

Beyond Age: Urgency of Information for Timeliness Guarantee in Status Update Systems

Timely status updating is crucial for future applications that involve remote monitoring and control, such as autonomous driving and Industrial Internet of Things (IIoT). Age of Information (AoI) has been proposed to measure the freshness of status updates. However, it is incapable of capturing critical systematic context information that indicates the time-varying importance of status information, and the dynamic evolution of status. In this paper, we propose a context-based metric, namely the Urgency of Information (UoI), to evaluate the timeliness of status updates. Compared to AoI, the new metric incorporates both time-varying context information and dynamic status evolution, which enables the analysis on context-based adaptive status update schemes, as well as more effective remote monitoring and control. The minimization of average UoI for a status update terminal with an updating frequency constraint is investigated, and an update-index-based adaptive scheme is proposed. Simulation results show that the proposed scheme achieves a near-optimal performance with a low computational complexity.

cs.IT↗

On the Coverage and Capacity of Ultra-Dense Networks with Directional Transmissions

We investigate the performance of a downlink ultra-dense network (UDN) with directional transmissions via stochastic geometry. Considering the dual-slope path loss model and sectored beamforming pattern, we derive the expressions and asymptotic characteristics of the coverage probability and constrained area spectrum efficiency (ASE). Several special scenarios, namely the physically feasible path loss model and adjustable beam pattern, are also analyzed. Although signal-to-interference-plus-noise ratio collapsing still exists when the path loss exponent in the near-field is no larger than 2, using strategies like beam pattern adaption, can avoid the decrease of the coverage probability and constrained ASE even when the base station density approaches infinity.

cs.IT↗

Channel Fingerprint Based Beam Tracking for Millimeter Wave Communications

Beamforming structures with fixed beam codebooks provide economical solutions for millimeter wave (mmWave) communications due to the low hardware cost. However, the training overhead to search for the optimal beamforming configuration is proportional to the codebook size. To improve the efficiency of beam tracking, we propose a beam tracking scheme based on the channel fingerprint database, which comprises mappings between statistical beamforming gains and user locations. The scheme tracks user movement by utilizing the trained beam configurations and estimating the gains of beam configurations that are not trained. Simulations show that the proposed scheme achieves significant beamforming performance gains over existing beam tracking schemes.

cs.IT↗

AI-Assisted Low Information Latency Wireless Networking

The 5G Phase-2 and beyond wireless systems will focus more on vertical applications such as autonomous driving and industrial Internet-of-things, many of which are categorized as ultra-Reliable Low-Latency Communications (uRLLC). In this article, an alternative view on uRLLC is presented, that information latency, which measures the distortion of information resulted from time lag of its acquisition process, is more relevant than conventional communication latency of uRLLC in wireless networked control systems. An AI-assisted Situationally-aware Multi-Agent Reinforcement learning framework for wireless neTworks (SMART) is presented to address the information latency optimization challenge. Case studies of typical applications in Autonomous Driving (AD) are demonstrated, i.e., dense platooning and intersection management, which show that SMART can effectively optimize information latency, and more importantly, information latency-optimized systems outperform conventional uRLLC-oriented systems significantly in terms of AD performance such as traffic efficiency, thus pointing out a new research and system design paradigm.

cs.IT↗

Flexible Functional Split and Power Control for Energy Harvesting Cloud Radio Access Networks

Functional split is a promising technique to flexibly balance the processing cost at remote ends and the fronthaul rate in cloud radio access networks (C-RAN). By harvesting renewable energy, remote radio units (RRUs) can save grid power and be flexibly deployed. However, the randomness of energy arrival poses a major design challenge. To maximize the throughput under the average fronthaul rate constraint in C-RAN with renewable powered RRUs, we first study the offline problem of selecting the optimal functional split modes and the corresponding durations, jointly with the transmission power. We find that between successive energy arrivals, at most two functional split modes should be selected. Then the optimal online problem is formulated as an Markov decision process (MDP). To deal with the curse of dimensionality of solving MDP, we further analyze the special case with one instance of energy arrival and two candidate functional split modes as inspired by the offline solution, and then a heuristic online policy is proposed. Numerical results show that with flexible functional split, the throughput can be significantly improved compared with fixed functional split. Also, the proposed heuristic online policy has similar performance with the optimal online one, as validated by simulations.

cs.NI↗

Device Scheduling with Fast Convergence for Wireless Federated Learning

Owing to the increasing need for massive data analysis and model training at the network edge, as well as the rising concerns about the data privacy, a new distributed training framework called federated learning (FL) has emerged. In each iteration of FL (called round), the edge devices update local models based on their own data and contribute to the global training by uploading the model updates via wireless channels. Due to the limited spectrum resources, only a portion of the devices can be scheduled in each round. While most of the existing work on scheduling focuses on the convergence of FL w.r.t. rounds, the convergence performance under a total training time budget is not yet explored. In this paper, a joint bandwidth allocation and scheduling problem is formulated to capture the long-term convergence performance of FL, and is solved by being decoupled into two sub-problems. For the bandwidth allocation sub-problem, the derived optimal solution suggests to allocate more bandwidth to the devices with worse channel conditions or weaker computation capabilities. For the device scheduling sub-problem, by revealing the trade-off between the number of rounds required to attain a certain model accuracy and the latency per round, a greedy policy is inspired, that continuously selects the device that consumes the least time in model updating until achieving a good trade-off between the learning efficiency and latency per round. The experiments show that the proposed policy outperforms other state-of-the-art scheduling policies, with the best achievable model accuracy under training time budgets.

cs.NI↗

Energy-Aware Analog Aggregation for Federated Learning with Redundant Data

Federated learning (FL) enables workers to learn a model collaboratively by using their local data, with the help of a parameter server (PS) for global model aggregation. The high communication cost for periodic model updates and the non-independent and identically distributed (i.i.d.) data become major bottlenecks for FL. In this work, we consider analog aggregation to scale down the communication cost with respect to the number of workers, and introduce data redundancy to the system to deal with non-i.i.d. data. We propose an online energy-aware dynamic worker scheduling policy, which maximizes the average number of workers scheduled for gradient update at each iteration under a long-term energy constraint, and analyze its performance based on Lyapunov optimization. Experiments using MNIST dataset show that, for non-i.i.d. data, doubling data storage can improve the accuracy by 9.8% under a stringent energy budget, while the proposed policy can achieve close-to-optimal accuracy without violating the energy constraint.

cs.IT↗

Closed-Form Whittle's Index-Enabled Random Access for Timely Status Update

We consider a star-topology wireless network for status update where a central node collects status data from a large number of distributed machine-type terminals that share a wireless medium. The Age of Information (AoI) minimization scheduling problem is formulated by the restless multi-armed bandit. A widely-proven near-optimal solution, i.e., the Whittle's index, is derived in closed-form and the corresponding indexability is established. The index is then generalized to incorporate stochastic, periodic packet arrivals and unreliable channels. Inspired by the index scheduling policies which achieve near-optimal AoI but require heavy signaling overhead, a contention-based random access scheme, namely Index-Prioritized Random Access (IPRA), is further proposed. Based on IPRA, terminals that are not urgent to update, indicated by their indices, are barred access to the wireless medium, thus improving the access timeliness. A computer-based simulation shows that IPRA's performance is close to the optimal AoI in this setting and outperforms standard random access schemes. Also, for applications with hard AoI deadlines, we provide reliable deadline guarantee analysis. Closed-form achievable AoI stationary distributions under Bernoulli packet arrivals are derived such that AoI deadline with high reliability can be ensured by calculating the maximum number of supportable terminals and allocating system resources proportionally.

cs.NI↗

Context-Aware Information Lapse for Timely Status Updates in Remote Control Systems

Emerging applications in Internet of Things (IoT), such as remote monitoring and control, extensively rely on timely status updates. Age of Information (AoI) has been proposed to characterize the freshness of information in status update systems. However, it only considers the time elapsed since the generation of the latest packet, and is incapable of capturing other critical information in remote control systems, such as the stochastic evolution and the importance of the source status. In order to evaluate the timeliness of status updates in remote control systems, we propose a context-aware metric, namely the context-aware information lapse. The context-aware information lapse characterizes both the stochastic evolution of the source status and the context-aware importance of the status. In this paper, the minimization of average context-aware lapse in a multi-user system is considered, and a corresponding user scheduling policy is proposed based on Lyapunov optimization. Numerical results show that compared to AoI-based policy, the context-aware-lapse-based policy can achieve a substantial improvement in terms of error-threshold violation probability and control performance.

cs.IT↗

Closed-Form Analysis of Non-Linear Age-of-Information in Status Updates with an Energy Harvesting Transmitter

Timely status updates are crucial to enabling applications in massive Internet of Things (IoT). This paper measures the data-freshness performance of a status update system with an energy harvesting transmitter, considering the randomness in information generation, transmission and energy harvesting. The performance is evaluated by a non-linear function of age of information (AoI) that is defined as the time elapsed since the generation of the most up-to-date status information at the receiver. The system is formulated as two queues with status packet generation and energy arrivals both assumed to be Poisson processes. With negligible service time, both First-Come-First-Served (FCFS) and Last-Come-First-Served (LCFS) disciplines for arbitrary buffer and battery capacities are considered, and a method for calculating the average penalty with non-linear penalty functions is proposed. The average AoI, the average penalty under exponential penalty function, and AoI's threshold violation probability are obtained in closed form. When the service time is assumed to follow exponential distribution, matrix geometric method is used to obtain the average peak AoI. The results illustrate that under the FCFS discipline, the status update frequency needs to be carefully chosen according to the service rate and energy arrival rate in order to minimize the average penalty.

cs.IT↗

Heterogeneous Coded Computation across Heterogeneous Workers

Coded distributed computing framework enables large-scale machine learning (ML) models to be trained efficiently in a distributed manner, while mitigating the straggler effect. In this work, we consider a multi-task assignment problem in a coded distributed computing system, where multiple masters, each with a different matrix multiplication task, assign computation tasks to workers with heterogeneous computing capabilities. Both dedicated and probabilistic worker assignment models are considered, with the objective of minimizing the average completion time of all computations. For dedicated worker assignment, greedy algorithms are proposed and the corresponding optimal load allocation is derived based on the Lagrange multiplier method. For probabilistic assignment, successive convex approximation method is used to solve the non-convex optimization problem. Simulation results show that the proposed algorithms reduce the completion time by 80% over uncoded scheme, and 49% over an unbalanced coded scheme.

cs.IT↗

HAHE: Hierarchical Attentive Heterogeneous Information Network Embedding

Heterogeneous information network (HIN) embedding has recently attracted much attention due to its effectiveness in dealing with the complex heterogeneous data. Meta path, which connects different object types with various semantic meanings, is widely used by existing HIN embedding works. However, several challenges have not been addressed so far. First, different meta paths convey different semantic meanings, while existing works assume that all nodes share same weights for meta paths and ignore the personalized preferences of different nodes on different meta paths. Second, given a meta path, nodes in HIN are connected by path instances while existing works fail to fully explore the differences between path instances that reflect nodes' preferences in the semantic space. rTo tackle the above challenges, we propose aHierarchical Attentive Heterogeneous information network Embedding (HAHE) model to capture the personalized preferences on meta paths and path instances in each semantic space. As path instances are based on a particular meta path, a hierarchical attention mechanism is naturally utilized to model the personalized preference on meta paths and path instances. Extensive experiments on several real-world datasets show that our proposed \model model significantly outperforms the state-of-the-art methods in terms of various data mining tasks.

cs.SI↗

Optimizing Information Freshness in Broadcast Network with Unreliable Links and Random Arrivals: An Approximate Index Policy

With the rapid growth of real-time Internet of Things (IoT) applications, the need for fresh information has surged. Age of Information (AoI) is a tailor-made metric to characterize the information freshness perceived by the devices. In this paper, we investigate the problem of scheduling updates to minimize AoI in broadcast network. In this case, a central controller, e.g. a base station, collects status updates from different sources and schedules them to corresponding clients. Different from previous work, we consider both stochastic status updates and unreliable links. The problem is first modeled as an infinite horizon average constrained cost Markov decision problem (CMDP). With Lagrangian relaxation, an approximation of Whittle's index is derived and a scheduling policy is designed based on the approximate index. The results in previous work can be view as degenerate cases of the approximation index policy either with reliable links or periodic arrival constraints. Simulation results demonstrate the near-optimal performance of the proposed policy.

cs.IT↗

Improving Device-Edge Cooperative Inference of Deep Learning via 2-Step Pruning

Deep neural networks (DNNs) are state-of-the-art solutions for many machine learning applications, and have been widely used on mobile devices. Running DNNs on resource-constrained mobile devices often requires the help from edge servers via computation offloading. However, offloading through a bandwidth-limited wireless link is non-trivial due to the tight interplay between the computation resources on mobile devices and wireless resources. Existing studies have focused on cooperative inference where DNN models are partitioned at different neural network layers, and the two parts are executed at the mobile device and the edge server, respectively. Since the output data size of a DNN layer can be larger than that of the raw data, offloading intermediate data between layers can suffer from high transmission latency under limited wireless bandwidth. In this paper, we propose an efficient and flexible 2-step pruning framework for DNN partition between mobile devices and edge servers. In our framework, the DNN model only needs to be pruned once in the training phase where unimportant convolutional filters are removed iteratively. By limiting the pruning region, our framework can greatly reduce either the wireless transmission workload of the device or the total computation workload. A series of pruned models are generated in the training phase, from which the framework can automatically select to satisfy varying latency and accuracy requirements. Furthermore, coding for the intermediate data is added to provide extra transmission workload reduction. Our experiments show that the proposed framework can achieve up to 25.6$\times$ reduction on transmission workload, 6.01$\times$ acceleration on total computation and 4.81$\times$ reduction on end-to-end latency as compared to partitioning the original DNN model without pruning.

cs.NI↗