Inferring indoor floorplans from wireless signals

Neural Radiance Fields (NeRFs) have been remarkably successful at synthesizing novel views of 3D scenes by optimizing a volumetric scene function. This scene function models how an optical ray accumulates colors on its path and eventually delivers this color to the camera pixel it impinges upon. Radio frequency (RF) or audio signals can also be viewed as a vehicle for delivering information about the environment to a sensor. However, unlike camera pixels, an RF/audio sensor receives a mixture of signals that contains many environmental reflections. Is it still possible to infer the environment using such mixed signals? We show that with redesign, the core NeRF framework has the potential to solve this inverse problem. We focus on a specific application of inferring the indoor floorplan of a home from WiFi measurements made at multiple locations inside the home. Our inferred floorplans look promising, and benefit downstream signal prediction applications. Our work also uncovers a number of problems for continued research.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Electrical & Computer Engr

Grantor

University of Illinois Urbana-Champaign

Year dc:date

2025

Author dc:creator

• Amballa, Chaitanya

Contributors dc:contributor

• Roy Choudhury, Romit

• NeRF
• Indoor floorplan

Statement dc:rights

• Copyright 2025 Chaitanya Amballa

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/129507

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F129507