An Asynchronous Distributed Data Collection Approach for Mobile Group Consumption

Mobile group consumption refers to that a group of people such as couples, colleagues, etc. use mobile devices to query commodities and perform consumption. To support the marketing of it, a proper data collection system is needed. Existing such systems are centralized, which suffers from performance bottleneck, one-point failure, overhead of server, etc. Moreover, existing systems are based on synchronous clock, which is often impossible due to hardware constraint or privacy concerns, or at least incur large synchronization overhead. In this paper, we propose an asynchronous distributed approach to collect data of mobile consumption for the first time. We formally build the system model of mobile group consumption in a distributed way and without synchronized clock. Then we propose a three-layer solution for distributed procession, including the procession for local sub-regions, consecutive sub-regions, and multiple sub-regions. After that, a detailed approach is proposed to analyze the causality of two marketing events happened in the system based on pure message exchange. Extensive simulation results show that the proposed approach is effective for data collection of mobile group consumption and is complementary to the classic method based on synchronized clock.