Source-linked AI summary
FOCAN: A Fog-supported Smart City Network Architecture for Management of Applications in the Internet of Everything Environments
Paola G. Vinueza Naranjo, Zahra Pooranian, Mohammad Shojafar, Mauro Conti, Rajkumar Buyya
TL;DR
Smart-city IoE applications generate distributed data, while remote Cloud processing can impose latency and traffic overhead. The paper proposes FOCAN, a multi-tier Fog architecture that selects nearby resources and organizes three communication types. The authors report an energy-efficient, scalable structure that minimizes average Fog-Node power consumption, with extensions needed for 5G, stream applications, and Mobile Edge Computing.
Problem
Remote Cloud processing can be unsuitable for latency-sensitive IoE applications, and prior work does not specify how to manage nearby Fog resources to reduce transfer delays and energy consumption.
Method
FOCAN is a multi-tier Fog architecture connecting heterogeneous IoE devices and Fog Nodes through interprimary, primary, and secondary communications.
Results
FOCAN is reported as a computation- and communication-efficient, scalable routing structure that minimizes average Fog-Node power consumption.
Takeaways & Limitations
FOCAN manages IoE applications through nearby Fog resources and differentiated communications intended to support low latency, energy awareness, and scalable smart-city services.
Takeaways & Limitations
Future extensions must address 5G management, stream applications, and real-time processing with Mobile Edge Computing techniques.
Abstract
from arXiv · showhide
Smart city vision brings emerging heterogeneous communication technologies such as Fog Computing (FC) together to substantially reduce the latency and energy consumption of Internet of Everything (IoE) devices running various applications. The key feature that distinguishes the FC paradigm for smart cities is that it spreads communication and computing resources over the wired/wireless access network (e.g., proximate access points and base stations) to provide resource augmentation (e.g., cyberforaging) for resource and energy-limited wired/wireless (possibly mobile) things. Moreover, smart city applications are developed with the goal of improving the management of urban flows and allowing real-time responses to challenges that can arise in users' transactional relationships. This article presents a Fog-supported smart city network architecture called Fog Computing Architecture Network (FOCAN), a multi-tier structure in which the applications running on things jointly compute, route, and communicate with one another through the smart city environment to decrease latency and improve energy provisioning and the efficiency of services among things with different capabilities. An important concern that arises with the introduction of FOCAN is the need to avoid transferring data to/from distant things and instead to cover the nearest region for an IoT application. We define three types of communications between FOCAN devices (e.g., interprimary, primary, and secondary communication) to manage applications in a way that meets the quality of service standards for the IoE. One of the main advantages of FOCAN is that the devices can provide the services with low energy usage and in an efficient manner. Simulation results for a selected case study demonstrate the tremendous impact of the FOCAN energy-efficient solution on the communication performance of various types of things in smart cities.
1. Introduction
The paper motivates Fog Computing for smart-city IoE applications that require lower latency and reduced traffic than remote Cloud processing can provide. It introduces FOCAN as a multi-tier framework for selecting nearby Fog resources and managing heterogeneous communications.
- Motivation: Cloud processing can be inefficient for latency-sensitive smart-city applications because data must travel to and from remote centers.Health and traffic monitoring are given as examples that cannot tolerate the resulting delay.
- Motivation: Fog Computing moves Cloud services toward network edges, reducing data-processing time and network-traffic overhead.
- Contributions: FOCAN is proposed as a generalized multi-tier smart-city architecture using Fog Computing for each device.
- Contributions: The framework develops resource allocation for thing-to-thing, thing-to-Fog-node, and Fog-node-to-Fog-node components.
- Evaluation: The paper evaluates the proposed solution through multiple communication types and numerical tests using real datasets and a simulated Fog platform.
2. Related Work
Prior work connects IoT, Cloud Computing, and Fog Computing to smart-city services, but does not explain how to select and manage nearby Fog resources to reduce transfer delays and energy consumption.
- IoT and smart cities: Earlier studies address Urban IoT characteristics, required smart-city services, and distributed coordination schemes.
- Cloud-based approaches: Cloud-based frameworks use IoT capabilities to create smart-city systems.
- Cloud limitations: Cloud Computing introduces latency, traffic congestion, lower throughput, and greater processing time for smart-city applications.
- Research gap: Existing work does not show how to choose and manage Fog resources to reduce delays and energy consumption from transferring data to far-away Fog nodes.
3. Model Overview
The model represents a heterogeneous smart city whose components and services communicate through the Internet of Everything and multiple access technologies to satisfy device requests.
- Model objective: The model describes how smart-city components and services communicate with one another and with Fog Computing.
- Connectivity: Heterogeneous components serve requests from diverse devices using technologies such as 3G/4G cellular, WiFi, and ZigBee.
- Connectivity: The Internet connects the devices and is labeled the Internet of Everything in the model.
4. Fog-supported Smart City Architecture
FOCAN connects heterogeneous IoE devices with Fog Nodes through two tiers, assigning local and distributed communication roles to support application processing, routing, and storage. Its communication design uses proximity, Fog-node routing, and differentiated communication types to address latency, QoS, mobility, and energy concerns.
- Architecture: FOCAN has an IoE tier of heterogeneous devices and an FN tier that processes and transfers incoming IoE traffic.
- IoE tier: IoE devices communicate directly through TCP/IP-based P2P links when nearby and use a Fog Node when outside local communication range.
- FN tier: Fog Nodes operate as virtualized networked data centers with computing, networking, storage, and configurable wired or wireless resources.
- Communication types: Primary and interprimary communications connect nearby things locally, while secondary communications connect geographically dispersed Fog Nodes.
- Communication characteristics: Primary communication improves QoS through high-speed, low-latency processing, while secondary Fog-node communication provides lower jitter for stream and pervasive applications.
- Supported capabilities: The architecture supports device and FN heterogeneity, mobility, dynamic application allocation, and multicast communication.
- Routing: FOCAN uses routing algorithms for FN-to-FN communication and supports application transfer through source and destination FN buffers.
5. Performance Evaluation and Validation
The evaluation uses iFogSim simulations of real-time smart-city traffic to assess FOCAN’s energy and communication performance, including comparisons with D2D.
- Simulation Setup: The experiments use iFogSim to simulate real-time Fog-based smart-city networks and web-based application demands for CPU and network resources.Incoming tasks represent I/O traffic requiring CPU and intra-/inter-network resources on city devices.
- Simulation Setup: Static Fog Nodes serve physically proximate spatial clusters using heterogeneous multicore servers.The deployment includes cluster radii such as 10 meters, 30 meters, and 3 feet, with multicore processing resources at each FN.
- Simulation Results: FOCAN is evaluated using average overall energy consumption per processing time across varied communication costs.Overall energy averages CPU and network costs, and the evaluation considers different communication rates.
- Simulation Results: The comparison evaluates FOCAN against the D2D approach in using simulated wired and wireless t2t, t2FN, and FN2FN links.The simulations report average energy and time per round for these communication activities.
- Simulation Results: FOCAN is more power efficient than D2D, whose per-connection power increases with fading, path loss, TCP timeouts, and packet retransmissions.The evaluation also reports average power for interprimary, primary, and secondary communications.
6. Open Issues and Challenges
The paper situates smart-city research within broader efforts to capture, curate, analyze, and visualize Big Data in educational settings.
- Open Issues and Challenges: Academic research develops techniques and technologies for capturing, curating, analyzing, and visualizing Big Data.An integrated educational system can combine network infrastructure, education information, and learning services.
7. Conclusion and Future Directions
FOCAN is a computation- and communication-efficient, scalable architecture for energy-aware management of applications across Fog Nodes. Future extensions target 5G, streaming applications, real-time processing, and Mobile Edge Computing.
- Conclusion: FOCAN minimizes power consumption across computing, intra-Fog communication, and wired/wireless transmission while managing IoE traffic through three communication categories.The categories are interprimary, primary, and secondary communication.
- Conclusion: A quantitative analysis shows that FOCAN effectively manages small urban areas and provides scalable, energy-aware Fog-supported application management.
- Future Directions: The framework can be extended to 5G management for the large number of things using stream applications across city regions.Examples include online video chatting and geographically distributed online gaming.
- Future Directions: Such streaming scenarios require real-time data processing together with Mobile Edge Computing techniques for robust frameworks.