Nimbus: Towards Latency-Energy Efficient Task Offloading for AR Services

Cozzolino, Vittorio; Tonetto, Leonardo; Mohan, Nitinder; Ding, Aaron Yi; Ott, Jörg

doi:10.1109/tcc.2022.3146615

Cited by 24 publications

(9 citation statements)

References 78 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Works like [15] present task offloading as an enabler for different real applications. In this case, the real-time object detection for mobile Augmented Reality (AR) is presented, which would require too much energy if made fully in the device.…”

Section: Non Ml-based Solutionsmentioning

confidence: 99%

Deep Reinforcement Learning techniques for dynamic task offloading in the 5G edge-cloud continuum

Nieto,

de la Iglesia,

Lopez-Novoa

et al. 2024

J Cloud Comp

View full text Add to dashboard Cite

The integration of new Internet of Things (IoT) applications and services heavily relies on task offloading to external devices due to the constrained computing and battery resources of IoT devices. Up to now, Cloud Computing (CC) paradigm has been a good approach for tasks where latency is not critical, but it is not useful when latency matters, so Multi-access Edge Computing (MEC) can be of use. In this work, we propose a distributed Deep Reinforcement Learning (DRL) tool to optimize the binary task offloading decision, this is, the independent decision of where to execute each computing task, depending on many factors. The optimization goal in this work is to maximize the Quality-of-Experience (QoE) when performing tasks, which is defined as a metric related to the battery level of the UE, but subject to satisfying tasks’ latency requirements. This distributed DRL approach, specifically an Actor-Critic (AC) algorithm running on each User Equipment (UE), is evaluated through the simulation of two distinct scenarios and outperforms other analyzed baselines in terms of QoE values and/or energy consumption in dynamic environments, also demonstrating that decisions need to be adapted to the environment’s evolution.

show abstract

Section: Non Ml-based Solutionsmentioning

confidence: 99%

Deep Reinforcement Learning techniques for dynamic task offloading in the 5G edge-cloud continuum

Nieto,

de la Iglesia,

Lopez-Novoa

et al. 2024

J Cloud Comp

View full text Add to dashboard Cite

show abstract

“…To separate the original issue into two problems: a task offloading (TO) problem that improves the optimalvalue function corresponding to the RA problem, and a resource allocation (RA) [12] problem with fixed task offloading decision is proposed in [13]. Convex and quasiconvex optimization methods are used to tackle the RA problem, and an unique heuristic algorithm for the TO problem that yields a suboptimal result in polynomial time is provided [14].…”

Section: Related Workmentioning

confidence: 99%

Secure Communication Model for Dynamic Task Offloading in Multi-Cloud Environment

Usharani¹,

Krishna²,

Akki³

et al. 2022

IJRITCC

View full text Add to dashboard Cite

As the data is increasing day-by-day, the mobile device storage space is not sufficient to store the complete information and also the computation capacity also is a limited resource which is not sufficient for performing all the required computations. Hence, cloud computing technology is used to overcome these limitations of the mobile device. But security is the main concern in the cloud server. Hence, secure communication model for dynamic task offloading in multi-cloud environment is proposed in this paper. Cloudlet also is used in this model. Triple DES with 2 keys is used during the communication process between the mobile device and cloudlet. Triple DES with 3 keys is used by the cloudlet while offloading the data to cloud server. AES is used by the mobile device while offloading the data to the cloud server. Computation time, communication time, average running time, and energy consumed by the mobile device are the parameters which are used to evaluate the performance of the proposed system, SCM_DTO. The performance of the proposed system, SCM_DTO is compared with ECDH-SAHE and is proved to be performing better.

show abstract

“…The overall quality of the metaverse AR application depends on the communication delays and the capabilities of the above mentioned network entities that participate in the service creation. Hereafter, we assume a nominal frame rate as 15 frames/second and the rendering happens at every other frame ( 133.2ms interval) [Cozzolino et al(2022) Cozzolino, Tonetto, Mohan, Ding, and Ott] [Niu and Chen(2018)] [Naman et al(2013)Naman, Xu, and Taubman]. Thus, the service delay of the aforementioned work flow within the interval could be regarded as acceptable.…”

Section: Introductionmentioning

confidence: 99%

Optimal Mobility Aware Wireless Edge Cloud Support for the Metaverse

Huang¹,

Friderikos²

2023

Preprint

View full text Add to dashboard Cite

Mobile augmented reality (MAR) applications extended in the metaverse could provide mixed and immersive experiences by amalgamating the virtual and physical world. However, the joint consideration between MAR and metaverse seeks the reliable and high quality support for foreground interactions and background contents from these applications, which intensifies their consumption in energy, caching and computing resources. To tackle these challenges, a more flexible request assignment and resource allocation with more efficient processing are proposed in this paper through anchoring decomposed metaverse AR services at different edge nodes and proactively caching background metaverse region models embedded with target Augmented Reality Objects (AROs). Advanced terminals are also considered to further reduce service delay at an acceptable cost of energy consumption. We then propose and solve a joint optimization problem that explicitly considers the balance between service delay and energy consumption under the constraint of user perception quality in a mobility event. By also explicitly taking into account capabilities of user terminals, the proposed optimized scheme is compared to its terminal oblivious version in this paper. According to a wide set of numerical investigations, the proposed scheme owns advantages in service latency and energy efficiency over other nominal baseline schemes which neglect capacities of terminals, user physical mobility, service decomposition and the inherent multi modality of the metaverse MAR service.

show abstract

Nimbus: Towards Latency-Energy Efficient Task Offloading for AR Services

Cited by 24 publications

References 78 publications

Deep Reinforcement Learning techniques for dynamic task offloading in the 5G edge-cloud continuum

Deep Reinforcement Learning techniques for dynamic task offloading in the 5G edge-cloud continuum

Secure Communication Model for Dynamic Task Offloading in Multi-Cloud Environment

Optimal Mobility Aware Wireless Edge Cloud Support for the Metaverse

Contact Info

Product

Resources

About