In this paper, we propose a privacy-aware user tracking system which traces users by their footprints. Tracking by footprints could reduce discomfort for privacy-sensitive users compared with the use of face recognition by cameras in existing tracking systems, since a footprint is weakly identifiable personal information, i.e., it cannot directly identify a specific individual. In this context, the proposed system is said to be privacy-aware. In the proposed tracking system, the footprint data are collected from spatially disjoint pressure sensor sheets put on significant locations in a target area. The trajectory of a specific user is estimated by extracting a sequence of sensor sheets that can observe the footprint of the user. The main task to be handled in the system is to recognize correspondence between footprints measured across multiple pressure sensor sheets. Moreover, we have conducted a basic evaluation to confirm the feasibility of the proposed system. The result shows that it can distinguish users by 94%.