A novel empirical model of the k-factor for radiowave propagation in Southern Africa for communication planning applications
JavaScript is disabled for your browser. Some features of this site may not work without it.
Please be advised that the site will be down for maintenance on Sunday, September 1, 2024, from 08:00 to 18:00, and again on Monday, September 2, 2024, from 08:00 to 09:00. We apologize for any inconvenience this may cause.
| dc.contributor.advisor | Baker, D.C. | en |
| dc.contributor.postgraduate | Palmer, Andrew J | en |
| dc.date.accessioned | 2013-09-07T12:52:14Z | |
| dc.date.available | 2004-09-22 | en |
| dc.date.available | 2013-09-07T12:52:14Z | |
| dc.date.created | 2003-10-09 | en |
| dc.date.issued | 2005-09-22 | en |
| dc.date.submitted | 2004-09-22 | en |
| dc.description | Thesis (PhD)--University of Pretoria, 2005. | en |
| dc.description.abstract | The objective of this study was to provide an adequate model of the k-factor for scientific radio planning in South Africa for terrestrial propagation. An extensive literature survey played an essential role in the research and provided verification and confirmation for the novelty of the research on historical grounds. The approach of the research was initially structured around theoretical analysis of existing data, which resulted from the work of J. W. Nel. The search for analytical models was extended further to empirical studies of primary data obtained from the South African Weather Service. The methodology of the research was based on software technology, which provided new tools and opportunities to process data effectively and to visualise the results in an innovative manner by a means of digital terrain maps (DTMs) and spreadsheet graphics. MINITAB | en |
| dc.description.availability | unrestricted | en |
| dc.description.department | Electrical, Electronic and Computer Engineering | en |
| dc.identifier.citation | regression analysis of the data, in addition to the neural network algorithm. The comparative studies and evaluation of the research were done by comparing predictions with existing data and primary data of groundbased meteorological observations obtained from the South African Weather Service. A key finding of the study were that the expression for the cumulative distribution of the k-factor as a percentage of time for events could be represented by a truncated normal or gaussian function. The existing data were recreated according to the basic model. This model was extended further in terms of a dual cumulative distribution taking into account what appeared to be seasonal effects by using “dry” and “wet” terms. The results were validated successfully by comparison between observed and predicted data. The extended cumulative distribution model was in turn used with a DTM to generate height dependent contours of k-factor values for South Africa for different values of time availability. Use was also made of a point form of the refractivity gradient and ground based meteorological data obtained from the South African Weather Service, to predict the long-term average of the k-factor for South Africa. A contour map to illustrate the results was generated using the DTM. It has thus been possible to extrapolate values of the k-factor observed at a limited number of sites to cover the whole country using the algorithms developed and the DTM for the country. A number of new analytical models were proposed and evaluated. Recommendations include that current research should continue in order to explain an apparent anomaly in the observed data around Alexander Bay. The vertical gradient of temperature will require further work to redefine its characteristics under different climatic and geographic conditions. Kriging, as a technique similar to the Neural Network approach, could be investigated further for applications using great circle distances between observation sites. The proposed model should be investigated further to determine whether it has potential global applications. A search for a generally applicable global representation of the k-factor should continue using the results of the work reported | en |
| dc.identifier.other | here. | en |
| dc.identifier.upetdurl | http://upetd.up.ac.za/thesis/available/etd-09222004-071145/ | en |
| dc.identifier.uri | http://hdl.handle.net/2263/28102 | |
| dc.language.iso | en | |
| dc.publisher | University of Pretoria | en_ZA |
| dc.rights | © software was used for regression analysis of the data, in addition to the neural network algorithm. The comparative studies and evaluation of the research were done by comparing predictions with existing data and primary data of groundbased meteorological observations obtained from the South African Weather Service. A key finding of the study were that the expression for the cumulative distribution of the k-factor as a percentage of time for events could be represented by a truncated normal or gaussian function. The existing data were recreated according to the basic model. This model was extended further in terms of a dual cumulative distribution taking into account what appeared to be seasonal effects by using “dry” and “wet” terms. The results were validated successfully by comparison between observed and predicted data. The extended cumulative distribution model was in turn used with a DTM to generate height dependent contours of k-factor values for South Africa for different values of time availability. Use was also made of a point form of the refractivity gradient and ground based meteorological data obtained from the South African Weather Service, to predict the long-term average of the k-factor for South Africa. A contour map to illustrate the results was generated using the DTM. It has thus been possible to extrapolate values of the k-factor observed at a limited number of sites to cover the whole country using the algorithms developed and the DTM for the country. A number of new analytical models were proposed and evaluated. Recommendations include that current research should continue in order to explain an apparent anomaly in the observed data around Alexander Bay. The vertical gradient of temperature will require further work to redefine its characteristics under different climatic and geographic conditions. Kriging, as a technique similar to the Neural Network approach, could be investigated further for applications using great circle distances between observation sites. The proposed model should be investigated further to determine whether it has potential global applications. A search for a generally applicable global representation of the k-factor should continue using the results of the work reported her | en |
| dc.subject | Effective earth radius | en |
| dc.subject | Radio-wave propagation | en |
| dc.subject | Refractivity gradient | en |
| dc.subject | Vertical model of the troposphere | en |
| dc.subject | Neural network algorithm | en |
| dc.subject | Digital terrain model (dtm) | en |
| dc.subject | The k-factor | en |
| dc.subject | Multiple regression – minitab | en |
| dc.subject | UCTD | en_US |
| dc.title | A novel empirical model of the k-factor for radiowave propagation in Southern Africa for communication planning applications | en |
| dc.type | Thesis | en |
