Disruptive Technologies
Are you interested?
If you are a professor or university student and you are interested in participating in the TUTORING program, register your information so that we can start the program.
Subject Area
Challenge aimed at talented students who want to develop their BsC/MsC in an industrial environment. Ideally, the candidate should have some experience in both Deep Learning (e.g. Transformers LLMs) and networking.
Introduction
Network monitoring systems generate vast amounts of data from heterogenous sources that are continuously collected into BigData platforms.
The efficient collection of this data involves a hard challenge for network operators, given the ever-increasing monitoring data and the complexity of the mobile network infrastructure as new generations are deployed.
In this vein, some recent works propose an unconventional use of Deep Learning models for data compression.
This can be useful to drastically reduce the amount of storage needed to keep historical network data.
This project is research-oriented, and is intended for candidates who dare to explore the boundaries of knowledge at the intersection of Deep Learning and Mobile Networks.
Challenge raised
In this project, the student will be tasked with exploring the use of Large Language Models and other Deep Learning techniques to compress large real-world network datasets.
The objective is to come up with an efficient solution that can achieve higher compression ratios than state-of-the-art online compression techniques (e.g., GZIP, lzop), thus offering potential CAPEX savings to the operator in terms of storage space, and enabling the possibility to set a longer retention period for network monitoring data.
