Indoor positioning for wearables relies on combining complementary technologies to overcome the absence of GPS signals and the complexity of built environments. Radio-based positioning, inertial sensing, visual methods, and sensor fusion each contribute strengths and weaknesses. Proven research and standards guide practical deployments and determine trade-offs among accuracy, power consumption, and privacy.
Radio-based methods
Wi-Fi fingerprinting compares received signal patterns to a mapped database and was popularized by Niraj Bahl and Venkata Padmanabhan at Microsoft Research who demonstrated site-survey approaches that work in many buildings. Bluetooth Low Energy beacons provide lower-power proximity cues useful for roomscale tracking, while Ultra-Wideband UWB offers centimeter-level accuracy in line-of-sight and is increasingly used in wearable scenarios where fine localization matters. Time-of-flight and angle-of-arrival processing reduce multipath ambiguity but require specialized hardware. Fadel Adib at MIT Media Lab showed how RF techniques can even track motion through occlusions, highlighting radio approaches’ ability to operate in visually cluttered spaces.
Inertial and vision-based methods
Wearables commonly embed accelerometers and gyroscopes to run Pedestrian Dead Reckoning PDR, estimating steps and heading when external signals are unavailable. The Kalman filter developed by Rudolf E. Kálmán at Research Institute for Advanced Studies remains a foundational algorithm for smoothing inertial estimates. Visual simultaneous localization and mapping SLAM provides high accuracy when cameras are available; Andrew J. Davison at Imperial College London pioneered real-time monocular SLAM that underpins many vision systems. Vision works well in feature-rich settings but raises privacy and lighting concerns.
Fusion, challenges, and consequences
Combining modalities mitigates individual weaknesses: inertial drift corrected by radio anchors, visual odometry refined by UWB anchors. Niki Trigoni at University of Oxford and others have emphasized probabilistic sensor fusion as essential for robust wearable tracking. Practical consequences include infrastructure cost for anchors, ongoing calibration such as Wi-Fi fingerprint updates, and variable performance across building materials where concrete or metal increases multipath. Cultural and human factors matter: continuous localization can enable safer eldercare and efficient industrial workflows but also creates surveillance risks that require clear consent and data governance. Environmental considerations include battery constraints on small wearables and the energy cost of dense anchor networks. In deployment, choosing methods requires balancing accuracy, latency, power, privacy, and the social context in which tracking will be used. In practice, hybrid systems that exploit UWB or BLE anchors with inertial smoothing offer the most reliable, real-world solution for wearable indoor localization.