Artifact paths removal algorithm for ultra-wideband channels

• Main Author: Zahedi, Yasser K. Reza
• Format: Thesis
• Language: English
• Published: 2016
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utm.my/id/eprint/78632/1/YasserKRezaPFKE2016.pdf; http://eprints.utm.my/id/eprint/78632/; http://dms.library.utm.my:8080/vital/access/manager/Repository/vital:97120

Ultra-wideband (UWB) is a promising technology for achieving high data rate communications. When UWB channel measurements are conducted, channel impulse responses (CIRs) are extracted from measured UWB waveforms using CLEAN deconvolution algorithm. However, artifact paths that represent unreal received multipath components (MPCs) are generated during this process. These artifact paths are registered as part of the measured CIRs representing a reflected signal from a scatterer. In reality, these paths do not represent a real scattering environment and this affects accurate channel modeling. Therefore, removal of the artifact paths is important to conserve better and have a more real scattering environment. In this work, an algorithm was developed to remove artifact paths from measured CIRs. The algorithm development was achieved based on the concept of geometric elliptical modeling applied to wideband channels, where the effective path in each ellipse is utilized to represent the channel response of the ellipse. Several UWB channel measurements were conducted to obtain the measured UWB waveforms. In addition, the characteristics of the UWB channels were analyzed in terms of CIRs properties and their stationarity regions. The algorithm performance was evaluated by comparing the single-template CLEAN CIRs with the CIRs result from the application of the developed algorithm on single-template CLEAN CIRs. Results showed that the developed algorithm can successfully remove the artifact paths. Besides that, an enhancement in the received power was achieved. For a specific measured channel, the received power enhancement obtained was more than 5%. The algorithm is beneficial for enhancing accuracy of CIRs extracted from a single-template CLEAN algorithm. Consequently, more accurate channel characteristics are gained leading to improved channel modelling and different parameter extractions.