Ultra-Wideband Channel Modeling using Singularity Expansion Method

Ultra-wideband (UWB) communications is expected to revolutionize high data-rate, short-distance wireless communications, providing data-rates in excess of 100 Mbps. However, the wireless channel distorts the transmitted signal by dispersing the signal energy over time. This degrades the output signal-to-noise ratio (SNR) of a correlation based matched-filter receiver, limiting the achievable data-rate and user capacity. Most wideband channel models do not account for all the identified dispersion mechanisms namely the frequency dispersion, the resonant dispersion and the multipath dispersion.

The objective of this research is to model resonant dispersion based on the Singularity

Expansion Method (SEM) and provide guidelines for UWB receiver design to meet the data capacity. The original contribution of this research is a novel pole dispersion channel model that includes resonant dispersion characterization. An empirical investigation supports our claim that a correlation type matched-filter receiver using a template signal based on the pole dispersion channel model overcomes distortion related losses. Various physical mechanisms responsible for dispersion in UWB communication systems are described in detail. The applicability of the proposed dispersive channel model is evaluated using the optimal matched filter (OMF) receiver.

The SEM approach, which was originally proposed for target identification using short pulse radars, offers limited benefits of due to its susceptibility to noise. A combined fuzzy-statistical approach is proposed to improve the robustness of resonant dispersion channel modeling in presence of noise. A natural extension of this doctoral research is to improve buried landmine detection as well as breast tumor detection by applying statistical and fuzzy analysis to the backscatter response. Moreover, radar target identification using UWB short pulses stands to gain tremendously from this research.