Identification Using Ground Reaction Force Patterns.

This paper investigates the feasibility of identification using ground reaction force patterns. A prototype sensing array is used to measure the distribution of forces between the foot and the floor during normal walking. Data is collected from ten subjects and is analyzed to determine whether patterns of ground reaction forces can be used to reliably differentiate one subject from another within a trial population. In the limited trials described, recognition rates of 86% are achieved using data from a single step. By averaging sequential steps into composite samples, recognition rates approaching 100% are demonstrated. Success with this form of identification suggests applications for conference rooms, secure areas and other intelligent room facilities where pointing of cameras and microphones is required or where records of the present location, and identity of individuals must be kept.