Survey of Fog Computing: Fundamental, Network Applications, and Research Challenges

Mukherjee, Mithun; Shu, Lei; Wang, Di

doi:10.1109/comst.2018.2814571

Cited by 551 publications

(251 citation statements)

References 173 publications

Supporting

Mentioning

222

Contrasting

Unclassified

Order By: Relevance

“…Lee et al in [31] and Alrawais et al in [32] discuss the security and privacy issues and present solutions for mitigating the security threats. More details of Fog computing are in [33], [34]. Recently, several groups have also advocated the need for Fog Robotics.…”

Section: Fog Roboticsmentioning

confidence: 99%

“…Quality of service provisioning depends upon a number of factors such as communication latency, energy constraints, durability, size of the data, model placement over Cloud and/or Edge, computation times for learning and inference of the deep models, etc. This has motivated several models for appropriate resource allocation and service provisioning [34]. Chinchali et al use a deep reinforcement learning strategy to offload robot sensing tasks over the network [56].…”

Section: Flexibility Of Resource Placement and Allocationmentioning

confidence: 99%

See 1 more Smart Citation

A Fog Robotics Approach to Deep Robot Learning: Application to Object Recognition and Grasp Planning in Surface Decluttering

Tanwani

Mor

Kubiatowicz

et al. 2019

2019 International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

The growing demand of industrial, automotive and service robots presents a challenge to the centralized Cloud Robotics model in terms of privacy, security, latency, bandwidth, and reliability. In this paper, we present a 'Fog Robotics' approach to deep robot learning that distributes compute, storage and networking resources between the Cloud and the Edge in a federated manner. Deep models are trained on non-private (public) synthetic images in the Cloud; the models are adapted to the private real images of the environment at the Edge within a trusted network and subsequently, deployed as a service for lowlatency and secure inference/prediction for other robots in the network. We apply this approach to surface decluttering, where a mobile robot picks and sorts objects from a cluttered floor by learning a deep object recognition and a grasp planning model. Experiments suggest that Fog Robotics can improve performance by sim-to-real domain adaptation in comparison to exclusively using Cloud or Edge resources, while reducing the inference cycle time by 4× to successfully declutter 86% of objects over 213 attempts.

show abstract

Section: Fog Roboticsmentioning

confidence: 99%

Section: Flexibility Of Resource Placement and Allocationmentioning

confidence: 99%

A Fog Robotics Approach to Deep Robot Learning: Application to Object Recognition and Grasp Planning in Surface Decluttering

Tanwani

Mor

Kubiatowicz

et al. 2019

2019 International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

show abstract

“…The main purpose of dew computing is to fetch existing remote computational resources at a near vicinity to the dew‐user. Four trending paradigms are in this list, such as cloud computing, cloudlets, fog computing, and mobile edge computing (MEC) . Cloud computing provides the best solution for large enterprize related issues through virtualization techniques.…”

Section: Taxonomymentioning

confidence: 99%

Minimizing dependency on internetwork: Is dew computing a solution?

Ray

2018

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Since the inception of the internetwork facility, it has been continuously serving as the fundamental backbone of existing aspects, such as cloud, fog, and edge computing. Several application perspectives are in practice that mandate the involvement of application domains including smart retail, transportation, and data analytics. However, the deployed service platforms do not conform to facilitate the minimal and/or internetwork‐free orientation to current paradigms. Hence, to cater this approach, a dew computing paradigm that goals at bringing cloud, fog, or edge resources and services to the nearest of user's periphery without or minimal use of internetwork has been introduced. This article advocates the feasibility, applicability, and appropriation of dew computing while leveraging several real‐life case studies with special attention to internetwork‐free movement. From earlier published literary works, a taxonomy has been devised in the dominion of dew computing. Moreover, highlights are made on selective requirements that would be key to the advancement of dew computing. Finally, challenges are discussed to provide future research directions in the dew computing domain.

show abstract

“…It provides an easy, reliable, fast, secure, and close access for computational services. In Fog computing, service nodes are close to the end‐users, and they perform all mandatory computing services . The Fog computing concept is similar to natural “Fog” in the environment, as they are closed to ground/humans.…”

Section: Introductionmentioning

confidence: 99%

“…We can feel and access the Fog easily. This concept is applied to the computing domain, where the computing and processing units are deployed close to end‐users . Therefore, a layman user can feel real‐time computing and data processing.…”

Section: Introductionmentioning

confidence: 99%

Fog computing–based node‐to‐node communication and mobility management technique for 5G networks

Babu

Biswash

2019

Trans Emerging Tel Tech

View full text Add to dashboard Cite

In this paper, we present a Fog computing–based node‐to‐node communication architecture for 5G networks. It supports the minimum network infrastructure dependent communication. In order to establish the connection, associated next generation evolve node base station (gNB) imports the required information from nearby Fog servers. The retrieved pieces of information are processed in data analytic units (DAUs). The DAUs are in‐build processing units and closely tie‐up with Fog servers/gNBs. The Fog servers and data analytic units are frequently deployed in the coverage area. They always offer and receive the services from gNBs to establish node‐to‐node communication. A robust mobility management scheme is also proposed for dynamic mobile users to achieve node‐to‐node communication. The signaling cost, energy consumption, delay, and latency are the performance parameters for the proposed work. The numerical and system level simulation results are produced to validate the proposed contributions. We compare the obtained results with 4G/LTE cellular networks in order to check the effectiveness of work. We achieved 26% to 63% of performance improvement as compared to legacy networks w.r.t. listed parameters.

show abstract

Survey of Fog Computing: Fundamental, Network Applications, and Research Challenges

Cited by 551 publications

References 173 publications

A Fog Robotics Approach to Deep Robot Learning: Application to Object Recognition and Grasp Planning in Surface Decluttering

A Fog Robotics Approach to Deep Robot Learning: Application to Object Recognition and Grasp Planning in Surface Decluttering

Minimizing dependency on internetwork: Is dew computing a solution?

Fog computing–based node‐to‐node communication and mobility management technique for 5G networks

Contact Info

Product

Resources

About