Loading...

Design and Implementation of Cyber-Phyical Network (Smart Building) using Machine Learning

Fahim, Mohammad | 2023

49 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 55862 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Hossein Khalaj, Babak; Farhadi, Alireza
  7. Abstract:
  8. Because of the increasing use of distributed computing and edge computing patterns, broadcast communications are becoming increasingly important. This communication should be such that each node can deliver its information to all or part of the network nodes. The quantity of delay should be small enough to not disturb the rest of the control processes.
    There are different topologies such as Mesh, Tree and Star networks, each of which has efficiencies according to the type of its media access control. In this thesis, a clustering architecture is proposed using the combination of Frequency Division Multiple Access with Carrier Sensing Multiple Access, and mathematical analysis is performed for the appropriate number of clusters to minimize the delay or energy in the all-to-all broadcast scenario. Also, a central algorithm for forming the network and selecting the cluster head is proposed.
    This algorithm has the self-configuration and self-organization and failure management property. In the design of this algorithm, we focused on the aspects of implementation and the real world challenges, and in order to evaluate the efficiency of the algorithms developed in this thesis, these algorithms have been applied to a smart building system based on edge computing (cyber-physical)
  9. Keywords:
  10. Machine Learning ; Clustering ; Intelligent Building ; Wireless Sensor Network ; Cyber-Physical Systems ; Edge Computing

 Digital Object List

 Bookmark

No TOC