ABOUT ME:
I am Matteo Ciprian and I am a Machine Learning Engineer, currently working in Cambridge (UK) in the field of Machine Learning, Sensor Fusion, and applied AI.

- Youtube channel:
https://www.youtube.com/channel/UCF--...
- Personal Website:
https://www.matteociprian.com/
- Github:
https://github.com/MatCip
- Instagram:
https://www.instagram.com/cip_mat/?hl=it
- Facebook:
https://www.facebook.com/matteo.cipri...

============================================================
ABOUT THE VIDEO:

=== Contents of Lectures 004 ====
Welcome to Lecture 004, an insightful exploration of "Machine Learning Techniques for GNSS." Joined by Ross van der Merwe, we venture into the nuanced landscape of diverse machine learning methodologies relevant across various stages of the GNSS pipeline. Ross begins by providing a comprehensive overview of the GNSS pipeline's logical blocks. Subsequently, we delve into the practical applications of machine learning within each block, encompassing critical use cases such as enhancing resilience against Spoofing, addressing Multipath complexities, and facilitating seamless Adaption. As we conclude the lecture, Ross sheds light on lesser-known applications and offers perceptive insights into future trajectories within this fascinating domain.


- LECTURER: Rossouw van der Merwe is a GNSS Engineer working in research and development on signal processing for robust GNSS receivers. He holds B.Eng, B.Eng (Hons), and M.Eng. degrees in Electronic Engineering from the University of Pretoria, South Africa, awarded in 2014, and 2016, respectively. In June 2023, he successfully defended his doctorate in GNSS spoofing detection at Friedrich-Alexander Universität in Erlangen, Germany. From 2016 to 2022, he was employed as a GNSS Researcher at Fraunhofer IIS in Germany, contributing to areas such as signal processing, interference management, and array processing. Starting in 2022, Rossouw has been engaged as a GNSS Engineer at Focal Point Positioning in the UK.


- TIMESTAMPS
00:00 Introduction
03:59 GNSS Pipeline
12:44 Appendix 1: Model Input, I-Q data
15:43 Three main applications of ML in GNSS: Resilience, Multipath, Adaption
16:44 1.a Resilience: Interference and Jammers
32:15 1.b Resilience: Spoofing
40:30 Support the channel
40:38 2. Multipath
49:55 3. Adaption
56:56 Other applications
1:01:15 Conclusions

BIBLIOGRAPHY / LINKS:

1) https://www.c4reports.org/aboveusonlystars

2) Toward a Unified PNT — Part 2:
https://www.gpsworld.com/toward-a-unified-pnt-part-2/

3) Evaluation of Low-Complexity Adaptive Full Direct-State Kalman Filter for Robust GNSS Tracking
https://www.mdpi.com/1424-8220/23/7/3658



============================================================

- ABOUT "STATES OF AI"
Welcome to "States of the AI: Lectures & Applications", a playlist of lectures about Artificial Intelligence and Data Science. In each lecture, I (Matteo Ciprian) interview a domain expert about the "Start of the Arts" of a specific topic related to AI and Machine Learning.

1) FORMAT:
The format is simple. For each topic X, I ask my guests to clarify:
- What is X.
- Why is X interesting from the point of view of the technology.
- What are the main practical applications of X ?
- A review of the research on X done so far.
- What are the future perspectives and developments of X in the short/medium/long term?

2) TARGET:
The goal of these lectures is to provide an easily accessible/digestible review of a specific topic related to AI and Data Science. Ideal if you want to keep up with the State of the Arts in a specific field/discipline.

3) AUDIENCE: Tech experts and enthusiasts, university students in STEM areas, novel researchers in the STEM field, and AI experts.

4) REQUIREMENTS: The requirements vary from lesson to lesson. Suggested a good mathematical background and basic knowledge of Artificial Intelligence and Machine Learning.



============================================================
- Music:
Intro Future Chill Logo
https://pixabay.com/music/id-143866/