IoT-Guard: Event-Driven Fog-Based Video Surveillance System for Real-Time Security Management

Tanin Sultana,Khan A Wahid

doi:10.1109/access.2019.2941978

Tanin Sultana, Khan A Wahid

Open Access

https://doi.org/10.1109/access.2019.2941978

Copy DOI

Abstract

In this paper, we design and implement a distributed Internet of Things (IoT) framework called IoT-guard, for an intelligent, resource-efficient, and real-time security management system. The system, consisting of edge-fog computational layers, will aid in crime prevention and predict crime events in a smart home environment (SHE). The IoT-guard will detect and confirm crime events in real-time, using Artificial Intelligence (AI) and an event-driven approach to send crime data to protective services and police units enabling immediate action while conserving resources, such as energy, bandwidth (BW), and memory and Central Processing Unit (CPU) usage. In this study, we implement an IoT-guard laboratory testbed prototype and perform evaluations on its efficiency for real-time security application. The outcomes show better performance by the proposed system in terms of resource efficiency, agility, and scalability over the traditional IoT surveillance systems and state-of-the-art (SoA) approaches.

Highlights

A smart home environment (SHE) consists of different applications of ubiquitous computing that integrates smartness into dwellings for comfort, healthcare, safety, security, and energy conservation
EVALUATION OF THE IOT-GUARD FRAMEWOK we describe the deployment of the testbed architecture in a laboratory environment
The pi camera connected with the Raspberry Pi 3 (RPi) will serve as a visual edge node for this experiment and the prototype

Summary

INTRODUCTION

A smart home environment (SHE) consists of different applications of ubiquitous computing that integrates smartness into dwellings for comfort, healthcare, safety, security, and energy conservation. Fog computing has the ability of responding quickly, and provides on-demand services by storing and processing data locally [10] This criterion encourages researchers to integrate fog computing in time-critical IoT applications (e.g., smart home security [SHS]) to improve real-time crime prevention. This approach can significantly reduce energy consumption and bandwidth due to the minimal amount of data transmission to the fog [9] Edge computing enables this event-driven approach in a target IoT surveillance application by delegating simple processing to camera-connected, constrained IoT-edge-node devices [15]. Based on all the discussions above, we propose IoT-guard, an event-driven edge-fog-integrated video surveillance framework, to perform real-time security management by aiding in crime prevention and predicting crime events at an SHE. The rest of the paper is organized as follows: Section II discusses the most relevant background works; Section III provides a detailed description of the proposed IoT-guard framework; Section IV and Section V present performance evaluation of the IoT-guard laboratory prototype and comparison with the state-of-the-art architectures, respectively and Section VI concludes the paper and outlines future research

RELATED BACKGROUND SURVEY

PERFORMANCE COMPARISON AND DISCUSSION

CONCLUSION