RSSI fingerprinting is a radio-based indoor localization method in which the received signal strength indicator (RSSI) values measured from multiple wireless access points at known reference locations are collected to build a spatial "radio map," which is then used to estimate an unknown device's position by matching its observed RSSI readings against stored fingerprints using classification or regression algorithms such as K-Nearest Neighbor or Naive Bayes. It matters to the field because it enables infrastructure-light indoor positioning without requiring line-of-sight or precise propagation models, making it practical for IoT deployments across diverse wireless technologies including ZigBee, Z-Wave, and Wi-Fi. Key variants differ primarily in the matching algorithm employed — deterministic approaches like KNN identify the closest fingerprint by signal distance, while probabilistic approaches like Naive Bayes model the likelihood of observed signals given a location — as well as in the underlying wireless technology, which affects range, resolution, and localization accuracy.
Source Papers
- An Overview on IEEE 802.11bf: WLAN Sensing ↗ — An Overview on IEEE 802.11bf: WLAN Sensing
- Memoryless Techniques and Wireless Technologies for Indoor Localization With the Internet of Things ↗ — Memoryless Techniques and Wireless Technologies for Indoor L
- NeRF2: Neural Radio-Frequency Radiance Fields ↗ — NeRF2: Neural Radio-Frequency Radiance Fields