Mobile urban sensing: Spatio-temporal observability analysis and optimization

Urban sensing systems must balance coverage, accuracy, and operational cost. To this aim, municipalities increasingly rely on opportunistic fleets of service vehicles as a cost-effective solution. However, constrained mobility and uneven spatial coverage make the design of effective monitoring systems a non-trivial problem. This paper addresses the challenge of dimensioning and deploying hybrid mobile-fixed urban sensing infrastructures based on the Edge-Fog-Cloud paradigm that leverage non-dedicated vehicle fleets while ensuring adequate spatio-temporal observability. To this end, we introduce a fleet-aware framework that combines spatio-temporal interpolation with novel observability and information-quality metrics and methodology to quantify how fleet configuration, route redundancy, and sampling density affect the accuracy of reconstructed environmental maps. The proposed methodology integrates real-world data analysis with “what-if” analysis to assess the impact of fleet sizing, data loss, and operational constraints, and to identify cost-effective hybrid mobile-fixed sensing configurations. The main contributions include the definition of spatial and temporal observability indexes, and value of fleet metrics, a greedy-based co-design procedure for optimal placement of fixed sensors, and an operational toolkit for fleet dimensioning and resource allocation. The experimental evaluation is conducted using a real-world dataset collected in the city of Benevento (Italy) by three sensing vehicles that cover an urban area of 11, 37 km 2 . Results demonstrate how informed decisions significantly improve interpolation accuracy and coverage efficiency. The methodology ultimately yields actionable recommendations for municipalities and operators about optimal configuration of hybrid sensing deployments, providing valuable insights into the fleet’s actual observability capabilities and balancing accuracy, cost, and operational effort.