Novel techniques of array antenna design for satellite communication

Antennas with high directivity or narrow beam width are preferred to be used in satellite communications to offer better reception performance in terms of reducing interferences from dense geo orbital sources. Sources of interference mostly are nearby satellites, other nearby earth stations or nearby microwave towers. Consequently, array antennas with high number of elements are developed over the past decades to obtain high directivity, narrow beam width and low side lobe level to minimize interference from these unwanted sources. On the other hand, increasing number of the elements in the design of arrays causes to long microstrip lines in feed network leading to dielectric and ohmic losses. Also, the use of large number of power splitters leads to increasing insertion and isolation losses. The power-splitting network is also limiting the band width and enlarging the size. Therefore, optimization of feed network of large arrays has been investigated extensively. Moreover, miniaturizing the dimensions of antenna used in both satellite or ground station have been subject of developed studies. Consequently, main goal of this thesis is to determine methods for array antenna design with higher directivity, low side lobe level, effective feed network and smaller dimensions. Radiation characteristics of an array depends on the radiation pattern of the element antenna and the geometrical configuration along with relative displacement and excitation of the overall array described by the array factor. Most of the available design techniques are accomplished by optimizing the array factor, neglecting abilities of the element antennas. In Chapter 4 of this thesis, three novel approaches of array design are proposed by using different combinations of special element antennas. In comparison with existing design procedures, the novel proposed methods are based on effective uses of both array factor and element factor with respect to each other. Consequently, array antennas with small aperture, effective feeding or narrow beam width are designed by using three distinct novel methods. Accuracy of all the approaches are confirmed based on theoretical studies and calculations. However, validity of the techniques are also demonstrated with simulations or measurement of prototype antenna(only for section 4.2). All the designs and simulations are accomplished in the range between 10.7GHz and 12.7 GHz which is allocated as satellite downlink for KU band. In section 4.1 of this thesis, a novel method is proposed to design array antenna with desirable directivity but smaller aperture compared to the method used in available designs. The design approach is well-demonstrated in two-element array structure. In an array with two elements, although reducing the element spacing from λ to λ/2 causes to a wide beam, but the resulting beam has no side lobs. Consequently, a special element antenna having a beam with a more narrow main lobe and probable side lobs at any other directions can be used to achieve a beam with single narrow lobe. Then, this approach can be developed to design arrays with large number of elements with small element spacing. Accuracy of the method is demonstrated and studied by using Ansoft HFSS software simulations. The second novel approach (section 4.2) proposes a method to design a low loss array antenna. It is demonstrated that the same beam width can be achieved by using 2λ element spacing with half the number of elements compared to the design with conversional limit of λ element spacing. Consequently, reducing the number of elements in microstrip arrays leads to lowering the loss made by long feed lines and insertion or isolation loss made by the power splitters. Also, using larger element spacing lowers the coupling between adjacent elements. However, the method needs a directive element antenna to cancel additional lobes. An element antenna with a very directive beam is simulated and fabricated. Then, a prototype array antenna is also simulated and fabricated based on the method with 2λ spacing of that element antenna and the measured results are provided to confirm the method. Results of section 4.2 confirms that beam width of an array is determined by aperture length of the array; more clearly by the elements placed towards the ends of the array. Subsequent results confirms that those additional interior elements are mostly effective in cancelling minor lobes, while having negligible effect on the beam width. Therefore, placement of the interior elements towards the ends of the array is aimed to improve the beam width, while this method distorts the side lobes. This is considered as a favorite achievement given that in some cases such as satellite applications the beam width of main lobe is more required than the side lobe level. Consequently, a novel nonlinear array design is presented in section 4.3 to achieve a narrower beam width with acceptable side lobe level compared to linear array. Chapter 5 of this thesis presents the forth study on array antennas advantageous for the satellite communication applications. Although, the methods in Chapter 4 confirm the good results in terms of small dimensions, efficiency and beam width. However, quality of a directive antenna in suppressing the interferences is described by its side lobe level. Regarding that the most favorite approach for this aim is the Binomial array, study of Binomial arrays is presented in Chapter 5. Binomial arrays, favorably show low side lobe level due to smooth amplitude distribution from the center element toward those at the edges. As the other advantage, binomial arrays with the element spacing equal or less than λ/2 have no minor lobes. However, unfortunately, there are no compact expressions for the array factor and directivity for binomial arrays with no restriction in element spacing, except for λ/2 spacing. In chapter 5, a novel procedure is proposed to derive compact expressions for the array factor and directivity of binomial arrays. The properties of Pascal's triangle, upon which the excitation coefficients of the binomial array are based and distributed, are used to obtain the compact expressions. It is demonstrated that the excitation coefficients of an N-element binomial array can be expressed as an (N-2)th convolution of the two-element binomial array. Then, the array factor and directivity of the entire binomial array is obtained using the properties of Fourier transform.