Simulation Environment for Investigating Crowd-sensing Based Urban Parking

This paper introduces a simulation environment developed for analyzing crowd-sensing based applications in the Smart City application domain. As a case study, an urban parking application scenario is investigated and presented. In this scenario, smart citizens collect and share parking related events, such as leaving or occupying a free parking spot. These events can be presented on a real-time city map and used in navigation software, thereby helping others alleviate parking related issues, such as the time spent while cruising for parking. The simulation environment, implemented in Java, allows to investigate and assess the critical user base of crowd-sensing based smart city applications, and the requisites (benefits, challenges) of introducing such applications. Our simulation results show that considerable gain (approx. 15%) can be achieved in the cruising time even with relatively low (30%) user base in a medium size city (Novi Sad, Serbia). Moreover, the proposed simulation environment can be used also in real field measurements by replacing/extending input data from real users.