F. Ares

Optimal compromise among sum and difference patterns

Three methods to obtain compromised sets of excitations in monopulse antenna arrays are investigated. The techniques involve minimizing a scalar-valued performance measure, called cost function, via the use of simulated annealing method. The unknowns in the cost function are the weight factors of each sub-array or the relative amplitudes and the phases of the excitations of the array elements. Several examples are presented to illustrate the optimal results obtained.

Phase-only control of antenna sum and shaped patterns through null perturbation

Satisfactory sum or shaped antenna radiation patterns can often be synthesized by modifying just the phase distribution of an initial excitation by an arbitrary amplitude distribution. As examples, we calculate linear and circular apertures with uniform, Taylor, or cosine-section amplitude distributions, affording symmetric sum patterns with low sidelobe levels or symmetric shaped patterns with low ripple and sidelobe levels; linear aperture distributions, affording patterns with low sidelobe levels on one side of the beam; and planar arrays, with uniform or cosine-section amplitude distributions affording /spl phi/-symmetric or elongated-oval footprints.

Optimization of the Performance of Arrays With Failed Elements Using the Simulated Annealing Technique

A method is described which optimizes the performance of antenna arrays with defective elements, by numerically finding a new excitation for the remaining ones. Two approaches for recovering the pattems have been investigated. In the first, new excitations are obtained by changing their amplitude distribution, whereas in the second, only the phase distribution is modified while keeping the original amplitudes. These optimum values for new excitations are calculated by the application of the simulated annealing technique. Examples for both sum and shaped beam patterns are presented to illustrate the optimal results obtained.

Shaped power patterns produced by equispaced linear arrays : Optimized synthesis using orthogonal sin(Nx)/sin(x) beams

The major defects of the Woodward-Lawson method of shaped pattern synthesis (lack of control over side lobes and ripple, and high dynamic range ratios) can be overcome by perturbing the field samples weighting the component sin(Nx)/sin(x) beams. If a real field is not necessary, further improvement (especially as regards dynamic range ratio) can be achieved by optimizing the phases of the coefficients of the component beams. These procedures are illustrated by linear arrays producing a symmetric flat-topped beam and a cosecant squared pattern.

Element failure detection in linear antenna arrays using case-based reasoning

The present work proposes a novel case-based reasoning system for fault diagnosis in moderate or large linear antenna arrays. This system identifies the set of elements that are most likely to be defective, helping to significantly reduce the computational costs of their detection (e.g., using an optimization technique such as a genetic algorithm).

Sythesis of Satellite Footprints By Perturbation of Woodward-Lawson Solutions for Planar Array Antennas

By perturbing initial Woodward-Lawson solutions it is possible to design planar array antennas with moderate dynamic range ratios that produce power patterns with arbitrarily shaped beams and low ripple, such as those required of direct broadcast satellite antennas.

Recalculating linear array antennas to compensate for failed elements while maintaining fixed nulls

A new technique for the synthesis of linear array antenna patterns is presented. This technique allows to fix nulls in prescribed directions, to simulate the presence of failed elements and to obtain power patterns when both situations occur at the same time. These patterns are synthesized by finding the optimal configuration of the array factor roots using the simulated annealing technique. Examples of fixing nulls and/or failed elements in both sum and flat-topped beam patterns are presented.

Application of genetic algorithms and simulated annealing technique in optimizing the aperture distributions of antenna arrays

Simulated annealing and genetic algorithms are used to find optimum excitations for patterns with null-filling. These methods have the advantage that the optimum aperture distribution with a minimal variation between excitations of adjacent elements, is found without searching the entire solution space. A comparison between the performance of both methods shows that genetic algorithms are faster than simulated annealing, for this problem.

Planar arrays with square lattices and circular boundaries: sum patterns from distributions with uniform, amplitude or very low dynamic-range ratio

A combination of simulated-annealing methods with Kim and Elliotts (1988) generalization of the Tseng-Cheng (1968) distribution allows rapid synthesis of planar arrays with rectangular lattices, circular boundaries, and a very low dynamic-range ratio. These generate sum patterns with satisfactory characteristics.

Optimal compromise among sum and difference patterns through sub-arraying

A technique has been described which permits the design of linear and planar arrays for monopulse radar applications, using a small number of sub-arrays. The obtained results allows to build up an array antenna with a simple feed network. This technique let us keep other pattern parameters under control by simply including them in a cost function to to be minimized by a procedure such as simulated annealing.