Deep learning improves indoor location accuracy using device signals and orientation
Orientation-Aware Deep Learning Framework for RSSI-Based Indoor Localization in RIS-Aided mmWave Systems
Networking and Internet Architecture
Summary
Indoor devices can be located using signals they receive, but changes in how the device is held make this tricky. The authors created a deep learning system that figures out both where a device is and how it is oriented by looking only at signal strength data. Their method combines multiple ways of analyzing the signals and pays attention to how signal patterns change over time. Tests in a lab showed their method was more accurate than previous ones and worked well even when the device was in new locations or when signals from multiple devices overlapped.
What this means in practice
- •For wireless network engineers: Enhance indoor positioning systems in mmWave environments by accounting for device orientation using only received signal strengths.
- •For indoor navigation system developers: Improve location accuracy for users in complex indoor settings where signals overlap, by applying orientation-aware signal processing.
Authors
Varun S. Advani, Ahmed Nasser, Mohamed Y. Selim, Ahmed M. Eltawil
Abstract
Reconfigurable intelligent surface (RIS)-aided millimeter wave (mmWave) systems provide directional measurements of received power for indoor localization, but device orientation changes the fingerprints. This work proposes an orientation-aware dual-stream gated (O-DSG) deep learning framework for received signal strength indicator (RSSI)-based indoor localization. An orientation classifier estimates device orientation from RSSI, so inference needs only received power. An embedding of the predicted angle and posterior entropy controls feature modulation and soft expert gating. Parallel convolutional and statistical streams are fused through bidirectional long short-term memory (BiLSTM) processing and temporal attention. Single-user measurements in an indoor laboratory at 28 GHz under non-line-of-sight (NLoS) conditions yield a median localization error of 0.141 m, a 25.8% improvement over the closest evaluated baseline (supplied with the measured orientation), and 92.3% orientation classification accuracy across four discrete classes. In spatial regions withheld from training, O-DSG has an average median error of 0.247 m, against 0.392 m for a validation-selected sequence model. A separate offline evaluation constructs non-orthogonal multiple access (NOMA)-style power mixtures from the single-user measurements and measures localization accuracy and computational scaling with up to eight emulated users whose power contributions overlap, under the specified mixture model.