Fingerprint-based location sensing technologies play an increasingly important role in pervasive computing applications due to their accuracy and minimal hardware requirements. However, typical fingerprint-based schemes implicitly assume that communication occurs over the same channel (frequency) during both training and runtime. When this assumption is violated, the mismatches between training and runtime fingerprints can significantly deteriorate the localization performance.
In this work, we propose a novel, scalable, multi-channel fingerprint-based indoor localization system that employs modern mathematical concepts based on the Sparse Representations and Matrix Completion theories. The contribution of our work is threefold. First, we investigate the impact of channel changes on the fingerprint characteristics and the effects of channel mismatch on state-of-the-art localization schemes. Second, we propose a novel fingerprint collection technique that significantly reduces the calibration time, by formulating the map construction as an instance of the Matrix Completion problem. Third, we propose the use of sparse Bayesian learning to achieve accurate location estimation. Experimental evaluation on real data highlights the superior performance of the proposed framework in terms of reconstruction error and localization accuracy.
The key contributions of this work include:
- Investigation of the effect of multiple wireless frequency channels on RSS based fingerprint localization. This marks a significant departure from typical localization schemes that do not consider channel mismatch between the training and the runtime phases.
- Design of a novel training scheme, based on Matrix Completion, for efficiently generating a complete (across space and frequency) training signature map by randomly sub-sampling the available channels at various locations. The proposed training scheme substantially reduces the effort and time spent during training.
- Development of a novel location sensing technique based on Sparse Bayesian Learning (SBL), motivated by the inherent sparsity of indoor localization. SBL is able to robustly identify the MS location by solving an under-determined system of equations arising from the recovered measurements.
- Experimentally validate our method on real data collected from a representative active indoor office area that includes walls, cubicles, windows, and moving people. These experimental data allow the quantification of the effects of multiple channels on both state-of-art and the proposed technique.