الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 165 |  |  |
| | | from | 165 | | |
Abstract
With the rapid increase in wireless applications such as mobile base stations, radars, etc.., the antenna arraybecomes the key issue in solving these problems. Since the synthesis of the radiation pattern from beamwidth and SLL points of view depends mainly on the feeding network. Most goals are achieved via the excitation coefficients optimizations. This may make the feeding network more complicated since the unequal power dividers utilized in beamforming almost has large dynamic range ratio. In this thesis, the optimization is performed on the spacing between elements using the known neural network algorithm. The optimization goals are the HPBW to be nearly 7°in the elevation plane, around 65° in the azimuth plane and a SLL less than -20 dB. The feeding network is implemented using the Wilkinson power dividers and the phase compensation is done using additional sections of meandered transmission lines. Thethesis is divided into two parts: the synthesis part and the new multiband design with two Simultaneous Sectors Mobile Base Station Antenna to cover two areas simultaneously with feeding network of custom Wilkinson power dividers with minimum side lobe level, maximum main lobe gain and minimum number of elements. The synthesis algorithm is based on the use of the neural network algorithm to estimate the distances between antenna elements in order to synthesis the desired radiation pattern. The proposed theories are verified experimentally by designing, fabricating and testing a complete base station antenna. Good results are obtained comparing the simulated and experimental results.