QLoc: A Realistic Quantum Fingerprint-based Algorithm for Large Scale Localization

Fingerprinting is one of the most commonly used techniques for indoor localization due to their high accuracy in the presence of the wireless channel noise. Nonetheless, they require significant storage overhead and running time. This becomes even more important when scaling such systems to a worldwide scale.Quantum computing can speedup computation of intractable problems that are hard to solve on the traditional classical computers. In this paper, we propose a realistic quantum fingerprint based algorithm that builds on quantum computing concepts to scale. In particular, the proposed fingerprinting algorithm can provide exponential enhancement in both space and running time over the classical fingerprinting techniques. We explain the details of the quantum fingerprint construction, how to encode the fingerprint signal strength vectors in quantum states, and how to calculate the quantum similarity between the online signal strength measurement and the offline ones. In addition, we study the different practical implementation and deployment issues on IBM quantum computing cloud machines.Implementation and evaluation of our algorithm in a real testbed using quantum machines confirms its ability to correctly obtain the estimated location with an exponential enhancement in both time and space compared to the traditional classical fingerprinting techniques. This highlights the promise of our algorithm to scale fingerprinting localization to be used globally.