Modelling and design theory of multi-faceted antenna arrays for hemispherical scan coverage

Hemispherical scan coverage, where the antenna beam may be positioned anywhere above horizon, is required in many applications such as satellite communications, and surveillance and tracking radar systems. This work involves the modelling and design of multi-faceted antenna arrays for providing general hemispherical coverage. Two classes of multi-faceted antenna arrays are used, namely, pyramids and pyramidal frusta, and four stages of their design process are considered. The first stage concerns the geometric design of the antenna and presents simple and novel methods to determine the optimal antenna geometry. Investigations of how the maximum scan angle is affected by far-field approximation and by the number of faces are also presented. The second stage focuses on the design of the circularly polarized circular patch antenna element. Its polarization purity is investigated over angular sectors about broadside. The results show that there is a well-defined optimal patch size which produces extremely pure polarization. This size corresponds to specific low values of the relative substrate permittivity. The third stage deals with the design of the planar array supported by each of the antenna faces. It presents a comparison between the two methods typically used to compute the element spacing over all coverage ranges and using different antenna configurations. The final stage presents a performance evaluation framework which compares planar, pyramidal, and pyramidal frustum arrays. Four increasingly detailed levels of evaluation are performed: using the geometric structure only; then incorporating the array factor; and finally including both the modelled; and the simulated element patterns. The results demonstrate that the performance advantages of multi-faceted arrays over planar arrays increase as the coverage range becomes wider and/or closer to horizon, and when the directional radiation characteristics of the antenna element are considered. They also show that no clear improvement is gained by using more than about 10 faces, given the added area, cost, and complexity involved. Moreover, under most circumstances, there is no significant difference in performance among pyramids with 4 to 7 faces or among pyramidal frusta with 5 to 9 faces.