Performance evaluation of fixed and mobile on-street parking space detection systems Cristian Roman 1 , Ruizhi Liao 2 , Peter Ball 1 , Shumao Ou 1 , and Martin de Heaver 3 1 Department of Computing and Communication Technologies, Oxford Brookes University, UK 2 Chinese University of Hong Kong, China 3 Ethos Valuable Outcomes, UK {cristian.roman,pball,sou}@brookes.ac.uk,ray@uniaddress.com,martin.deheaver@ethosvo.org Abstract As the number of vehicles continues to grow, parking spaces are a premium in city streets. In addition, due to the lack of knowledge about street parking spaces, heuristic circling in the streets not only costs drivers’ time and fuel, but also increases city congestion. In the wake of the recent trend to build convenient, green and energy-efficient smart cities, common techniques adopted by high-profile smart parking systems are reviewed, and the performance of the various approaches are compared. A mobile sensing unit has been developed as an alternative to the fixed sensor approach. It is mounted on the passenger side of a car to measure the distance from the vehicle to the nearest roadside obstacle. By extracting parked vehicles’ features from the collected trace, a supervised learning algorithm has been developed to estimate roadside parking occupancy. Multiple road tests were conducted around Wheatley (Oxfordshire) and Guildford (Surrey) in the UK. In the case of accurate GPS readings, enhanced by a map matching technique, the accuracy of the system is above 90%. A quantity estimation model is derived to calculate the required number of sensing units to cover urban streets. The estimation is quantitatively compared to a fixed sensor solution. The results show that the mobile sensing approach can perform at the same level as fixed sensor solutions under certain conditions and substantially fewer sensors are needed compared to the fixed sensor system. Index Terms on-street/roadside parking, sonar/ultrasonic, supervised learning, map matching, space detection vehicle, fixed parking sensors, mesh network, crowdsourcing,