A Master Thesis Has Been Discussed in Computer Eng. Dept. at U.O.T.
The research entitled :- " Decreasing Latency for Smart Manfacturing using dual Edge-Server"
For the student (Rawa Ammar Redzuki) on five chapters, 1st chapter is the introduction and 2nd chapter is the theoretical background, 3rd chapter is the proposed system, 4th chapter results and chapter five conclusions and future work.
This research aims to decrease the delay to improve the performance of the smart factory
By studying the effect of the number of sensors on the waiting time in each local server built using the Matlab program, in addition to studying the current cloud computing and peripheral computing, knowing the problems faced by previous researchers and working to solve them, and giving a high performance in monitoring the system by placing a set of lamps that give Alert the status of each sensor in the smart factory.
This thesis discusses the time factor which is considered as a basic method for improving the performance of smart factories. This proposed system focuses on reducing time and achieving real time requirements. Using local servers, which are represented in a simulation using Matlab to study the effect of the number of sensors on the time factor of each server. Then the terminal servers are practically represented using two computers installed with the (XAMP) program.
Where the terminal servers are used to reduce the time because they are closer to the factory than the cloud server, and the data processing process can be completed without the need to stop the factory in the event of any failure In the factory, it is represented by five types of sensors connected to the Arduino board, which sends the sensor readings to the Raspberry pi. value tidy server
The local servers make sure that the Raspberry pi sends new readings and a decision is made, or that the key switch value itself has not changed five times and no decision has been made. It waits until it receives new sensor readings from the Raspberry pi to process it. Also, this system monitors the performance of the smart factory by sending cases after making a decision to the (Arduino) to translate them by turning off or turning on a specific lamp for each case.
The results of this study were to significantly reduce the delay by using the terminal servers. When comparing the total time of the terminal servers with the total time of the cloud server, we notice a significant difference in the time of sending and receiving sensor values, as the time was reduced from 128ms to 2ms using terminal computing.
The discussion committee consisted of:
1- Asst. Prof. Dr. Muhammad N. Abdullah/ Head.
2- Asst. Prof. Dr. Nadia Adnan /from the University of Baghdad- College of Eng. Dept. of Computer Eng./ Member.
3- Lect. Dr. Ahmed Saad Karim / Member.
4-Assr. Prof. Dr. Hassan Jalil Hassan /Member and Supervisor.
5- Lect. Dr. Ghaydaa M. Abdul Sahib / Member and Supervisor.
