Ice-Mocha: Intelligent crowd engineering using mobile internet of things characterization and analytics
No Thumbnail Available
Authors
Meeting name
Sponsors
Date
Journal Title
Format
Subject
Abstract
Human casualties at entertaining, religious, or political crowd events often occur due to the lack of proper crowd safety management. Notably, for a moving crowd, a minor accident can create a panic for the people to start stampeding and trampling others. There are many smart video surveillance tools, inspired by the recent advanced Artificial Intelligence (AI) technology and Machine Learning (ML) algorithms, to detect and identify objects. However, it is still very challenging video surveillance alone to manage the crowd mobility in real-time for preventing any potential disaster — their scalability and capacity lack for providing the appropriate mobile crowd safety management in real-time. In this dissertation, we propose an Intelligent Crowd Engineering Platform using Mobility Internet of Things (IoT) Characterization and Analytics (ICE-MoCha). ICE-MoCha is an intelligent mobility monitoring system to enhance the reliability, scalability, energy efficiency, and practicality aspects of crowd safety management. Specifically, we tackle three essential components, including mobility characterization, mobility orchestration, and mobility energy efficiency. In mobility characterization, we build an IoT-based mobile crowd safety management system by predicting and preventing potential disasters through real-time Radio Frequency (RF) data characterization and analytics. The motivation is to improve the safety management method for the mobile crowd by filling up the scalability and capability gaps of the existing video surveillance via tightly integrating RF signal analytics. Among the many crowd mobility characteristics, we tackle object group identification, speed, direction detection, and density for the mobile group. We also apply them to group semantics to track the crowd status and predict any potential accidents and disasters. In mobility orchestration, we develop a new type of pervasive smart and mobile urban surveillance infrastructures to cope with the recent integration of wireless and mobile cyber-physical systems with emerging intelligent sensors. We design and test a prototype of an efficient and effective Edge system using light-weight agile software-defined control for mobile wireless nodes. It enhances the accuracy, efficiency, and productivity of dynamic big-data-driven real-time urban surveillance tasks such as continuous target tracking and situational awareness. Finally, in mobility energy efficiency, we investigate the facilities of networking the sensors. Specifically, when sensors form inline (i.e., linear topology in road and railway), it detects mobile data. It replies to the central sink via neighbor sensors. It causes energy inefficiency and imbalance issues, which provoke a complicated management hassle. We develop a framework of mobile data sensing and management that facilitates an efficient wake-up receiving method (WuRx) on sensors nodes to maximize the duty cycles without compromising network performance. We design and develop an IR-based WuRx approach using a pipelined relay algorithm. The results show that a constant IR WuRx delay is achievable regardless of the number of relaying nodes in the network.
Table of Contents
Introduction -- Literature review -- ICE-MoCha Solutions -- ICE-MoCha Implementations -- ICE-MoCha Experiments -- Conclusions and Future Directions
xv, 135 pages
xv, 135 pages
DOI
PubMed ID
Degree
Ph.D. (Doctor of Philosophy)