A Privacy-Respecting Indoor Localization Approach for Identifying Shopper Paths by Using End-Users Mobile Devices

An established way to analyze shoppers' behavior at the point of sale consists of identifying their paths through the store as well as their approach behavior towards different shelves. Such proceeding allows e.g. for optimizing product placements or in-store advertising and guidance. Since there is a technological challenge in doing this inside the respective locations, there is a need for better localization methods than those using RFIDs or similar localization technologies (e.g. indoor GPS, CCTV, and different photo sensors) or by basing on human-based observations; at least due to privacy concerns. In this paper we introduce a multi-method approach for identifying shopper paths in the stores based on a combination of built-in sensors' capabilities of the end-users' mobile devices as well as a mobile product scanner application. Our approach allows for more privacy-preserving evaluation since the users could decide to provide accumulated paths data when paying at the point of sale. We also describe our prototypic implementation extending the Red pin system for iPhones, explain the architecture allowing also for anonymously sharing customers' paths in real-time, and address potential improvements for future work.