Projecte matriculat
Títol: Satellite-based NTN metric analysis and uncertainty modeling for operations management
Director/a: RINCÓN RIVERA, DAVID
Departament: ENTEL
Títol: Satellite-based NTN metric analysis and uncertainty modeling for operations management
Data inici oferta: 27-09-2023 Data finalització oferta: 27-04-2024
Estudis d'assignació del projecte:
MU AEROSPACE S&T 15- MU AEROSPACE S&T 21
- MU MASTEAM 2015
| Tipus: Individual | |
| Lloc de realització: Fora UPC | |
| Supervisor/a extern: Arnau Singla Manau | |
| Institució/Empresa: I2Cat Foundation | |
| Titulació del Director/a: MSc in Aeronautics, PhD in Network Eng. | |
| Paraules clau: | |
| Non-Terrestrial Network (NTN), constellation management, NB-IoT, scenario modeling | |
| Descripció del contingut i pla d'activitats: | |
| At i2CAT, a satellite operations planner is being developed for constellations acting as NTN. The idea is to apply task allocation methods (typically used for Earth observation missions) to telecommunications scenarios, specifically for providing IoT services. Some of the challenges (among others) include system heterogeneity, scenario uncertainties, limited resources, the integration of business policies into operations management, and the integration of the planner with a standard core network (3GPP compliant operations). We already have an initial version of this planner that has been systematically validated (unit tests) as well as in a realistic scenario. Among the tasks that continue in the development of this planner, there are two that may be of interest to undergraduate and master's students for their theses.
Firstly, an exhaustive study needs to be conducted on how the various implemented parameters of the planner affect the optimized plans obtained. An example of this would be how prioritizing different users affects the metrics of the optimized plans (latency per user, data transferred, etc.). Additionally, it's necessary to compare the results obtained with i2CAT's planner with other constellation management mechanisms (using the same relevant metrics). The activities to be performed could be as follows: 1. Research the relevant metrics to analyze in the selected scenarios. 2. Design different simulation scenarios to assess how optimization policies affect the results (metrics). 3. Simulate the scenarios with different policies and analyze the results using i2CAT tools and the student's own implementations. 4. Simulate the scenarios and compare the results with other constellation management mechanisms, using i2CAT tools and the student's own implementations. Secondly, it is necessary to model the inherent uncertainty in the selected scenarios (especially in terms of data generation). There are several methods to do this. The objective of this work package would be to identify these methods, implement a simple method, and validate it through simulations. The activities to be performed could be as follows: 1. Research methods for modeling uncertainty in IoT use cases, with a special focus on data traffic modeling. 2. Identify a simple method applicable to our areas of interest. 3. Implement the selected method in i2CAT software. 4. Validate the selected method using i2CAT tools. Depending on the motivation and the type of student (undergraduate or master's), it is possible to focus on one part, a combination, or both. There is also the possibility of splitting this proposal into two separate, independent theses proposals for two different students, if necessary. |
|
| Orientació a l'estudiant: | |
| The student will receive a scholarship. It could also be combined with an internship at the company.
Essential Requirements: ' Motivation for modeling and programming. ' Minimum background in terrestrial or satellite networks (one is sufficient; the other can be learned). ' Research in scientific publications. ' Handling and analysis of data resulting from simulations. ' Writing scientific articles and presenting results Useful Skills (not required, but having them will be helpful): ' The software is written in JAVA. ' Some data post-processing tools use Matlab. ' Modeling uncertainty involves topics such as statistics and machine learning. ' Minimum knowledge of key metrics for evaluating IoT scenarios. ' Minimum knowledge of satellite constellation design. ' Understanding of data generation models in IoT networks. If you are interested or have any question, please contact arnau.singla@i2cat.net and david.rincon@upc.edu |
|
| Requereix activitats hardware: No | |
| Requereix activitats software: Sí Sistema operatiu: Disc (Gb): | |
| Horari d'atenció a estudiants per a l'assignació de projecte: |
|