The vehicular edge computing system integrates the resources of vehicles, and provides services for other vehicles pedestrians with task offloading. However, offloading environment is dynamic uncertain, fast varying network topologies, wireless channel states, workloads. These uncertainties bring extra challenges to In this paper, we consider among propose a solution that enables learn delay performance their neighboring while computation tasks. We design an adaptive learning based (ALTO)...

Travel-time prediction constitutes a task of high importance in transportation networks, with web mapping services like Google Maps regularly serving vast quantities travel time queries from users and enterprises alike. Further, such requires accounting for complex spatiotemporal interactions (modelling both the topological properties road network anticipating events -- as rush hours that may occur future). Hence, it is an ideal target graph representation learning at scale. Here we present...

With the proliferation of mobile applications, Mobile Cloud Computing (MCC) has been proposed to help devices save energy and improve computation performance. To further quality service (QoS) MCC, cloud servers can be deployed locally so that latency is decreased. However, computational resource local generally limited. In this paper, we design a threshold-based policy QoS MCC by cooperation Internet resources, which takes advantages low abundant resources simultaneously. This also applies...

Mobile edge computing is a novel technique in which mobile devices offload computation-intensive tasks with stringent delay requirements to the cloud. However, limited computational resource cloud may result Quality of Service degradation. In this paper, we address issue by coordinating heterogeneous includes and remote Considering offloading delay-bounded tasks, study into scheduling order maximize probability that can have met. The problem formulation proved be concave, an optimal...

Vehicular cloud computing (VCC) is proposed to effectively utilize and share the storage resources on vehicles. However, due mobility of vehicles, network topology, wireless channel states available vary rapidly are difficult predict. In this work, we develop a learning-based task offloading framework using multi-armed bandit (MAB) theory, which enables vehicles learn potential performance its neighboring with excessive resources, namely service (SeVs), minimizes average delay. We propose an...

In vehicular cloud computing (VCC) systems, the computational resources of moving vehicles are exploited and managed by infrastructures, e.g., roadside units, to provide services. The offloading tasks collection results rely on successful transmissions between infrastructures during encounters. this paper, we investigate how timely services in VCC systems. particular, seek minimize deadline violation probability given a set be executed clouds. Due uncertainty vehicle movements, task...

A cloud computing multi-objective task scheduling optimization based on fuzzy self-defense algorithm is proposed. Select the shortest time, degree of resource load balance and cost completion as goal scheduling, establish a mathematical model to measure effect construct objective function scheduling. The experimental results show that method can improve performance maximum deadline violation rate virtual machine utilization.

Base station (BS) sleeping is an effective way to improve the energy-efficiency of cellular networks. However, it may bring extra user-perceived delay. We conduct a theoretical study into impact BS on both and consider hysteresis sleep three typical wake-up schemes, namely single sleep, multiple N-limited schemes. model system as M/G/1 vacation queue, which captures setup time, mode-changing cost, well counting or detection cost during mode. Closed-form expressions for average power...

Base station (BS) sleeping operation is one of the effective ways to save energy consumption cellular networks, but it may lead longer delay customers. The fundamental question then arises: How much can be traded off by a tolerable delay? In this paper, we characterize tradeoffs between total and overall in BS with sleep mode operations queueing models. Here, includes not only transmitting power also basic (for baseband processing, amplifier, etc.) switch-over working mode, transmission...

Edge cloud is a promising architecture in order to address the latency problem mobile computing. However, as compared with remote clouds, edge clouds have limited computational resources, and higher operating costs. In this paper, we design policies which carry out assignment of tasks that are generated at subscribers an online fashion. The proposed achieve optimal power-delay trade-off system. Here, delay experienced by computing task includes time spent waiting for transmission cloud,...

In a vehicular edge computing (VEC) system, some vehicles with surplus resources can provide computation task offloading opportunities for other or pedestrians. However, the network is highly dynamic, fast varying channel states and loads. These dynamics are difficult to model predict, but they have major impact on quality of service (QoS) offloading, including delay performance reliability. Meanwhile, in VEC often redundant due high density vehicles. To improve QoS exploit abundant...

By Mobile Edge Computing (MEC), computation-intensive tasks are offloaded from mobile devices to cloud servers, and thus the energy consumption of can be notably reduced. In this paper, we study task offloading in multi-user MEC systems with heterogeneous clouds, including edge clouds remote clouds. Tasks forwarded via wireless channels, they further Internet. Our objective is minimize total multiple devices, subject bounded-delay requirements tasks. Based on dynamic programming, propose an...

Introduction In high-stakes environments such as aviation, monitoring cognitive, and mental health is crucial, with electroencephalogram (EEG) data emerging a keytool for this purpose. However traditional methods like linear models Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU) architectures often struggle to capture the complex, non-linear temporal dependencies in EEG signals. These approaches typically fail integrate multi-scale features effectively, resulting suboptimal...

A problem of much current practical interest is the replacement wiring infrastructure connecting approximately 200 sensor and actuator nodes in automobiles by an access point. This motivated considerable savings automobile weight, simplification manufacturability, future upgradability. key issue how to schedule on shared point so as provide regular packet delivery. In this other similar applications, mean inter-delivery times packets, i.e., throughput, not sufficient guarantee...

Mobile edge computing is a novel technique to offer cloud-based computation offloading services mobile users with short delay. However, the Cloud Service Providers (CSPs) of cloud are generally different from CSPs remote Internet cloud. Considering competition between heterogeneous clouds, we study optimal provisioning computational resource in We firstly analyze Nash equilibrium prices market when amount given. Based on pricing policy, further design an algorithm optimize capacity so that...

Base station (BS) sleeping is an effecting way to improve the energy-efficiency of cellular networks. Considering BS sleep mode operation under different scenarios, we focus on three wake-up policies: single vacation (SV) policy, multiple (MV) policy and N policy. A hysteresis time also considered avoid frequent mode-changing operation. By modeling systems as M/G/1 queues, derive two performance measures interest, expected system response energy consumption per bit. The impacts parameters...

In networked cyber-physical systems, the inter-delivery time of data packets becomes an important quantity interest. However, providing a guarantee that times are "small enough" difficult task in such systems due to unreliable communication medium and limited network resources. We design scheduling policies meet requirements multiple clients connected over wireless channels. formulate problem as infinite-state risk-sensitive Markov decision process, where large exceedances for different...

Cellular network performance depends heavily on the configuration of its parameters. Current practice parameter relies largely expert experience, which is often suboptimal, time-consuming, and error-prone. Therefore, it desirable to automate this process improve accuracy efficiency via learning-based approaches. However, such approaches need address several challenges in real operational networks: lack diverse historical data, a limited amount experiment budget set by operators, highly...

With the proliferation of mobile applications, Mobile Cloud Computing (MCC) has been proposed to help devices save energy and improve computation performance. To further quality service (QoS) MCC, cloud servers can be deployed locally so that latency is decreased. However, computational resource local generally limited. In this paper, we design a threshold-based policy QoS MCC by cooperation Internet resources, which takes advantages low abundant resources simultaneously. This also applies...

The membrane of Gram-negative bacteria (GNB) is especially robust due to the additional, unique highly asymmetric outer membrane, with lipopolysaccharides (LPS) as main component. This LPS layer serves a protective barrier against antibiotics, host immune responses, and other environmental stresses. However, constructing model membranes containing that capture structural asymmetry for fundamental studies GNB cell wall remains an open challenge. In this context, we discuss how recent...

Base station (BS) sleep mode operation is one of the effective ways to save energy, but it may lead longer delay customers. In order evaluate tradeoffs between energy consumption and customer delay, we model BS as an N-policy M/G/1 vacation queue with setup close-down times, where enters if no customers arrive during time after becomes empty starts when sees N arrivals its period. Several closed-form formulas are derived demonstrate mean by changing N. It shown that relationship linear time....

In this paper, we propose and study opportunistic bandits - a new variant of where the regret pulling suboptimal arm varies under different environmental conditions, such as network load or produce price. When load/price is low, so cost/regret (e.g., trying configuration). Therefore, intuitively, could explore more when low exploit high. Inspired by intuition, an Adaptive Upper-Confidence-Bound (AdaUCB) algorithm to adaptively balance exploration-exploitation tradeoff for bandits. We prove...

In cyber-physical systems such as automobiles, measurement data from sensor nodes should be delivered to other consumer actuators in a regular fashion. But, practical over unreliable media wireless, it is significant challenge guarantee small enough inter-delivery times for different clients with heterogeneous channel conditions and requirements. this paper, we design scheduling policies aiming at satisfying the requirements of clients. We formulate problem risk-sensitive Markov Decision...

Making idle servers sleep is considered to be a key approach reducing energy consumption of various information and communication systems. Optimal sleeping policies for single server have been derived only non-bursty traffic in prior work. In this paper, the first time, we study optimal operation with bursty answer question whether can bring extra benefit or not. Key factors including switchover as well delay performance are considered. We formulate problem partially observable Markov...