Long-Short Term Memory (LSTM) is dead, long live Transformer. Recent advanced Transformers have been proven that it can outperform LSTM, especially in Natural Langue Processing (NLP). However, Transformers are still too large to put on an embedded device for indoor localization, which is essential for many applications.
Unlike outdoor localization, indoor localization is a daunting challenge since the Global Positioning System (GPS) signal is attenuated and scattered by roofs and walls. The WiFi radio frequency of smartphones is a popular mean for indoor localization because of its popularity, every smartphone has a WiFi interface. However, WiFi signals are also attenuated and scattered significantly in indoor environments due to walls, furniture, and human bodies. That makes the location estimation sensitive to the chance of the indoor environments.
This project aims at exploiting compressive Transformers for smartphone-based indoor localization.
The project will be divided into 3 main tasks:
- Study fingerprinting localization with Transformers
- Develop compressive Transformer for fingerprinting localisation
- Compare Transformer with and compressive Transformer based fingerprinting localisation
30% Theory, 50% Implementation, 20%Writing
Le Viet Duc, email@example.com, room ZI 5013