Computational Complexity Closed-Form Upper Bounds Derivation for Fingerprint-Based Point-of-Interest Recognition Algorithms

Place recognition and user positioning methods are becoming trending topics among the research community thanks to the need of providing cutting-edge solutions to enable innovative and efficient Location-Based Services (LBS) and applications. However, the mobile framework suffers from typical constraints on energy and computational capabilities due to the inherent characteristics of wireless mobile devices. For this reason, in order to be suitable for deploying in such framework, a place recognition algorithm should be thrifty in terms of resource consumption. In this paper we consider six different WiFi-based outdoor positioning approaches and we analytically derive an upper-bound for their computational burden in terms of number of FLoating-point OPerations (FLOPs).