Izegem , Belgium
Televic GSP (Rail)
With over 30 years of experience in designing, manufacturing and maintaining on-board communication and control systems, Televic Rail is a leading, trusted partner for railway operators and train builders worldwide.
Its Passenger Information Systems and Control Systems are high quality, tailor-made solutions that offer the flexibility, user-friendliness and stability that our clients ask for. Our various types of on-board control systems such as our bogie monitoring systems are innovative yet reliable products which are designed specifically for the railway business.
Trains and trams all around the world are equipped with Televic Rail solutions, from New Zealand to Canada, from China to the United States, from India to Belgium, England and France..
Network analysis, in its simplest definition, involves the analysis of network data and statistics, to identify trends and patterns. Operators can then use this analysis, and act on the data.
Televic GSP has many different projects, each providing full system solutions to our customers. These solutions consist of multiple devices communicating with one another, using various protocols. These "conversations" usually contain a lot of useful information about the status of our systems. However, currently not much is done with this information.
The objective of this topic is to do a thorough analysis of this data. After analysing the data, an anomaly detection model can be created, which alerts developers when something abnormal is happening. The internship will consist of 2 or 3 phases:
- Analysing the data: the data currently consists of unlabelled, raw network packages. The student will have to perform a deep analysis of the data, both on a packet-level (e.g., what does a packet contain, what can be learned from a single packet?) and on a global level (e.g., how are packets distributed over time? What does it mean when we get sudden bursts of data? Or periods of no packets at all?)
- Create useful visualizations using the data: this can be done in numerous ways and is completely up to the student. A student could for example use parse the packets and create a map of places where no GPS-data was able to reach the trains.
- In case the topic is chosen as a thesis, some form of automated anomaly detection could be added. The student will have to investigate how to create alerts when something irregular happens, but before the system completely breaks. This way, our engineers can act on those alerts and prevent mistakes, instead of repairing them.
- Level: Bachelor, Master
- Specialty: AI / Machine Learning, Software
- Type of work: Research: 5%, Implem.: 10%, Experim.: 75%
- Location: Televic
- Type of activities: Experimenting, Programming
- Number of students: 1