Since commercial and residential buildings have become the largest energy consumers across all sectors, building energy efficiency has attracted increasing attention in recent years. Many studies suggest occupancy detection is critical in promoting building energy efficiency because it is premised on the idea of avoiding unnecessary waste while providing sufficient service. In this paper, we propose the integration of an iBeacon-enabled indoor positioning system (IPS) and a Variable Air Volume (VAV) heating, ventilation, and air-conditioning (HVAC) system to optimize system control and save energy based on high-resolution occupancy detection. The proposed system aims to match thermal service with the spatial distribution of occupants and redefine occupancy as a dynamic spatial occupancy distribution (DSOD) occupancy matrix. For this reason, this paper proposes measuring spatial occupancy by meshing large indoor spaces into zones and patches, and uses a feature-scaled artificial neural network algorithm to map the spatial IPS signal patterns. After acquiring the detailed spatial distribution, we also developed a ventilation control mechanism based on occupancy distribution. To validate the proposed control mechanism, we compared it with other traditional controllers in an on-site experiment and through a computational fluid dynamics (CFD) simulation. The results suggest that a 20% energy saving potential can be realized when the proposed approach is properly implemented. (C) 2017 Elsevier Ltd. All rights reserved.