Synthesis of Planar Circular Arrays with Quantized Amplitude Weights.

Zhiqiu He,Gang Chen

doi:10.3390/s21206939

Abstract

A new stepwise and radially processed method for synthesizing uniformly distributed circular planar arrays with quantized weights is proposed in this paper. This method is based on a generalized analytical equation describing that for high directivity focusing arrays, minimizing the weighted mean square error between the reference pattern and the synthesized pattern is equivalent to minimizing the mean square error between the radial cumulative distributions of the reference distribution and the synthesized distribution. This principle has been successfully performed for designing large concentric ring arrays, and in this paper, we extend its use for synthesizing uniformly distributed planar circular arrays with quantized weights. Various numerical examples and comparisons with several reported statistical methods in terms of the lowest Maximum SideLobe Level (MSLL) demonstrate the effectiveness of the proposed method.

Highlights

Radars with large uniformly distributed array antennas have been widely applied in atmosphere observation [1], military warning, and navigation [2] with an increasing speed
We extend the use of the analytical equation in [30] for synthesizing uniformly distributed circular arrays with quantized weights and low Maximum SideLobe Level (MSLL)
While the synthesis results show that the SLL of the reference pattern for the proposed method is 3–4 dB lower than the infimum of the MSLL of the array, the infimum of MSLL can refer to the results of the statistical methods

Summary

Introduction

Radars with large uniformly distributed array antennas have been widely applied in atmosphere observation [1], military warning, and navigation [2] with an increasing speed. Nature-inspired algorithms such as genetic search [5,6] and particle swarm optimization [7] have been exploited to design thinned arrays. The Iterative Fourier Technique (IFT) [8] as well as its derived IFT-based algorithms [9,10] and more recent dynamic programming [11] have been successfully applied to large array thinning. Another solution is the nonuniformly distributed arrays, e.g., [12,13,14,15,16]. Though they may outperform thinned arrays in many aspects theoretically, the interelement spacing is changed, which violates the nature of the uniform distribution

Objectives

Findings

Discussion

Conclusion