Channel charting (CC) is an emerging application of self-supervised machine learning to wireless communication which leverages the fact that wireless communications systems continuously collect data about the electromagnetic propagation channel. This high-dimensional data, known as channel state information (CSI), is acquired at fast rates but typically discarded immediately after use. CC recycles acquired CSI data by means of dimensionality reduction to learn a so-called channel chart, which reveals information about relative positions of signal sources. If two signals come from nearby positions, they are mapped to nearby locations on the chart. The objective of this project is to develop the concept of Channel Charting as a Service, which assists a number of network functions and user-level applications with a CC-based pseudo-location management architecture accessible through a dedicated application programming interface.

2023

2026