Target detection in urban scenarios using netted radar multipath signals

In urban environments, multipath and obscuration by buildings decreases the ability of radar systems to detect targets and also increases the number of false alarms caused by image artefacts. To reduce these problems this paper proposes a network of low-cost easilydeployable radar sensors that are set up in a multiple-input multiple-output (MIMO) radar configuration. An analysis is given of the signal processing required to combine the data from several co-operating radar sensors while using building information and including the effects of multipath propagation. The key part of this idea is that the known building positions from a database are used to determine the origin of multipath signals. This process enables the removal of artefacts and allows the sensors to “see around the corner” of buildings.