Estimating the Number of Persons in an Unknown Indoor Environment by Applying Wireless Acoustic Sensors and Blind Signal Separation

Determining the number of persons and their locations in an unknown indoor environment is one of the first tasks that must be done in military, rescue or intelligence operations. If there is no functioning security system in the building, one must use a monitoring system, which is rapidly deployable but simultaneously as invisible as possible. Wireless sensor network is an attractive candidate to fill such requirements. Many sensor types are suitable for indoor situation modeling, but there are also several technical challenges rising from the limited energy and communication resources of the sensor nodes. This paper focuses on estimating the number of persons in an unknown indoor environment by using low-power acoustic sensors. The number of persons is estimated by applying blind signal separation to the acoustic signal collected by the sensor nodes.