Backscatter-Assisted Computation Offloading for Energy Harvesting IoT Devices via Policy-based Deep Reinforcement Learning

Xie, Ying; Xu, Zhengzhuo; Zhong, Yuxing; Xu, Jing; Gong, Shimin; Wang, Yi

doi:10.1109/iccchinaw.2019.8849964

Cited by 30 publications

(15 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, MEC has attracted widespread attention, the DRL [16], [17], DDPG [18], convex optimization [19], [20] algorithms are introduced to determine when/where/how to perform task offloading. DRL can be used to find the online offloading schemes.…”

Section: Related Workmentioning

confidence: 99%

“…Ref. [18] proposed a DDPG-based hybrid MEC offloading algorithm to balance each users power consumption in data transmission and task computing. In [20], the authors designed an offloading mechanism of MEC system based on content prediction, including a content prediction model based on Long Short Term Memory (LSTM) and a task offloading strategy based on cross-entropy (CE) method, to maximize the total system throughput.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Task Offloading and Serving Handover of Vehicular Edge Computing Networks Based on Trajectory Prediction

Yang

Chen

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Vehicular edge computing (VEC) has emerged as a promising paradigm to ensure the realtime task processing caused by the emerging 5G or high level intelligent assisted driving applications. The computing tasks can be processed via the edge services deployed at the roadside units (RSUs) or moving vehicles. However, the high dynamic topology of the vehicular communication system and the time-varying available computing resources in RSUs make a challenge of the efficient task offloading of vehicles. In this paper, we consider an efficient task offloading scheme for VEC networks based on trajectory prediction, we focus on the serving handover between the adjacent RSUs. The moving vehicles can cooperate with RSUs or the surrounding vehicles for task processing. To reduce the latency of task transmission between vehicles, we present a cooperative vehicle selection method based on trajectory prediction. Then, we propose an efficient task offloading scheme based on deep reinforcement learning (DRL), while the dynamically available computing and communication resources are considered jointly. The simulation results show that the proposed task offloading scheme has great advantages in improving the utility of vehicles.INDEX TERMS Internet of vehicles, edge computing, trajectory prediction, task offloading, deep reinforcement learning

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Task Offloading and Serving Handover of Vehicular Edge Computing Networks Based on Trajectory Prediction

Yang

Chen

et al. 2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Under the common sense that conventional RF communications is not a possible solution to IoT devices, Xie, Yutong, et al raised deep reinforcement learning scheme concentrating on the optimized power allocation strategies despite uncertainties and locations [85]. FPGA can implement high-performance embedding requirements for applications.…”

Section: Dynamic Resource Provisioning For Cps In Edge Computingmentioning

confidence: 99%

Dynamic resource provisioning for cyber-physical systems in cloud-fog-edge computing

Zhang

et al. 2020

J Cloud Comp

View full text Add to dashboard Cite

Cyber-Physical Systems(CPS) serves as an interdisciplinary effort that incorporates cyber vector as well as physical vector. The latter can generate exponentially growing amounts of data. How to process CPS big data systematically and efficiently is the key to the breakthrough of offering prospective and personalized services for each individual user and entity involved. In recent years, much research has been proactively conducted on advancing specific scenario or algorithm. However, few surveys value their integration. For good measure, the synthesis remains a fundamental challenge. Subsequently, we fill the gap in the literature by constructing cloud computing, fog computing and edge computing as a whole to inspire on new architectures and cross utilizations. Moreover, bringing the enthusiasm of traditionally solitude entities into play is crucial. In this exploratory study, we examine definitions of CPS as well as the three aforementioned computing paradigms and then shed new light on comprehensively established frameworks. We also survey on the application level of Cloud-Fog-Edge Computing in CPS respectively and dive into diversified algorithms and strategies to embed big data applications into a more intelligent and convenient society with current deficiencies and future research directions followed.

show abstract

“…The multi-user scenario model is more complicated, for the resource allocation must consider the interaction between users. The time scheduling strategy of the multi-user AmBC system was studied in [10,19,24,26]. RL and DRL have certain advantages in dealing with resource allocation problems in time-varying communication networks because of their characteristics of learning in environment interaction.…”

Section: Performance Evaluationmentioning

confidence: 99%

“…Because of energy constraints, BDs sometimes require offloading computing tasks to nearby computing servers through active transmission or low-power backscatter communications. The authors of [24] proposed a deep reinforcement learning algorithm (DRL) is to implement the optimal unloading strategy in a hybrid unloading AmBC network.…”

Section: Introductionmentioning

confidence: 99%

Prioritized Uplink Resource Allocation in Smart Grid Backscatter Communication Networks via Deep Reinforcement Learning

et al. 2020

View full text Add to dashboard Cite

With the rapid increase in the number of wireless sensor terminals in smart grids, backscattering has become a very promising green technology. By means of backscattering, wireless sensors can either reflect energy signals in the environment to exchange information with each other or capture the energy signals to recharge their batteries. However, the changing environment around wireless sensors, limited radio frequency and various service priorities in uplink communications bring great challenges in allocation resources. In this paper, we put forward a backscatter communication model based on business priority and cognitive network. In order to achieve optimal throughput of system, an asynchronous advantage actor-critic (A3C) algorithm is designed to tackle the problem of uplink resource allocation. The experimental results indicate that the presented scheme can significantly enhance overall system performance and ensure the business requirements of high-priority users.Electronics 2020, 9, 622 2 of 16 compared to the active WPCN. The authors of [4] maximized the throughput of backscatter and radio frequency (RF) WSNs by formulating the optimization objectives.Harvest-the-transmit (HTT) protocol is employed in a traditional RF-powered cognitive network (CRN) where the secondary transmitter (ST) harvests the energy in primary signals and then transmits information to the receiver using the energy stored in the battery actively. Recently, ambient energy signals have been used to transmit ST information by backscattering. The overhead of transmitting data is further reduced. Therefore, it is advisable to combine ambient backscatter communication with RF-powered CRN. The authors of [5] analyzed and compared the performance and throughput of RF-powered CRN in overlay and underlay scenarios. Internet of Things (IoT) is regarded as an ultimate solution to connect everything and has aroused extensive interest in academia and industry. However, energy limitation and implementation cost are the challenges in the construction of IoT. In [6], ambient backscatter communication (AmBC) was introduced as a green communication paradigm that could exploit the ambient wireless signals to power sensors and backscatter the information.There have been many optimization problems for RF-powered backscatter networks. The authors of [7] studied a full-duplex communication mode in the AmBC network where a full-duplex communication device transmitted the energy signal while the receiving signal was reflected back from the backscatter devices (BDs). An iterative algorithm was proposed to optimize the minimum throughput for all BDs by jointly scheduling time, power resources and adjustment reflection coefficients. The authors of [8] introduced backscatter communication into a multi-user device-to-device (D2D) network where the device could harvest energy from RF signals. A game-theoretic approach was proposed to balance relaying performance and energy harvesting.The above algorithms need to know the complete environment information. However, ...

show abstract

Backscatter-Assisted Computation Offloading for Energy Harvesting IoT Devices via Policy-based Deep Reinforcement Learning

Cited by 30 publications

References 10 publications

Task Offloading and Serving Handover of Vehicular Edge Computing Networks Based on Trajectory Prediction

Task Offloading and Serving Handover of Vehicular Edge Computing Networks Based on Trajectory Prediction

Dynamic resource provisioning for cyber-physical systems in cloud-fog-edge computing

Prioritized Uplink Resource Allocation in Smart Grid Backscatter Communication Networks via Deep Reinforcement Learning

Contact Info

Product

Resources

About