The thesis is concerned with the current use of ray-optics based propagation prediction tools for the determination of wideband characteristics of the wireless urban propagation channel. Reliable prediction of propagation channel parameters such as delay spread, coherence bandwidth etc. is important for the deployment of wideband systems. Deterministic propagation models, based on ray-optical techniques make use of site-specific topographic databases, and aimed to yield reliable area coverage and wideband propagation channel parameters. Such models determine wideband wireless channel characteristics by Fourier transforming the radio channel impulse response at a center frequency. This can lead to errors in the derived wideband parameters. A three-dimensional ray-optics based propagation prediction tool is developed. The model uses image theory, Geometrical Optics (GO) and the Uniform Theory of Diffraction (UTD) to determine wideband characteristics of the urban wireless communication channel. Higher order diffractions which can contribute significantly in shadow region are incorporated in addition to first order diffractions. The model is thoroughly validated with single frequency and swept frequency measurements in scaled down controlled environment. Systematic errors are identified in deriving wideband response. Errors (up to 27 dB and 4.1 rad) in the lit regions corresponds to the individual tiny deep nulls which are quite insignificant. In the shadow region average errors gradually increases away from the center frequency and rises up to 1 dB and 0.05 rad. Error sign reversal is observed in the transition regions. Errors corresponding to realistic bandwidths are less than 0.5 dB and 0.03 rad. Errors in frequency correlation function corresponds to shadow regions and frequency separation of 0.25 GHz to 7 GHz. It is concluded that the coherence bandwidth provides more information of the dispersive nature of the channel compared to delay spread.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:633234 |
| Date | January 1997 |
| Creators | Sheikh, Muhammad Imran |
| Publisher | University of Birmingham |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0016 seconds