Improved detection of distorted IR-UWB pulses using amplitude spectrum in indoor environments

In dense environments, such as residential, office or industrial sites, the predominant type of link encountered by a radio based real-time localization system (RTLS) is non-line-of-sight (NLOS). Often, the mobile nodes are separated from their related anchor nodes by the building's walls. In Impulse Radio Ultra Wideband (IR-UWB) based localization systems, the penetration of walls can cause significant distortions of the pulses. The grade of distortion depends on the walls' thickness, materials used and the pulses' angle of incidence. Pulse distortion has a negative impact on the performance of correlation in matched filter (MF) based algorithms in the time domain. This paper presents a detailed evaluation of an approach using the amplitude spectrum of the received signal for correlation. To evaluate this approach, simulations were performed for typical materials of walls, with varying wall thicknesses and angles of incidence and, in addition, different pulse types according to the regulations. For the parameter space using an UWB pulse with bandwidth compliant to FCC, improvements up to 15% were reached.