A Cupola-Shaped Multimode Multiport Antenna for Aerial Direction Finding

Lukas Grundmann,Marc Gerlach,Wolfgang Schäfer,Dirk Manteuffel
2024-09-23
Abstract:A multimode multiport antenna (M$^3$PA), constisting of a single conducting structure, is proposed for use with the airborne collision avoidance system (ACAS). Its cupola-shaped surface is selected from a set of aerodynamically suitable and symmetric structures. The selection is based on a presented statistical evaluation procedure for direction-finding (DF) antennas and algorithms, which predicts the root mean square error (RMSE) of the estimation using characteristic modes (CMs). A demonstrator antenna and feed network is manufactured and its three ports are measured in an anechoic chamber. The results match the intended far-fields of an electric monopole and two magnetic dipoles. The presented flight test results show that the antenna is capable of unambiguous DF across the desired angular range in a real world environment.
Signal Processing
What problem does this paper attempt to address?
The main objective of this paper is to propose a Multi-Mode Multi-Port Antenna (M³PA) suitable for Airborne Collision Avoidance Systems (ACAS) for Direction Finding (DF). Specifically, the paper aims to address the following issues: 1. **Design Optimization**: A new design method for a multi-mode multi-port antenna is proposed, which features a low profile and can achieve unambiguous direction finding in actual flight environments. 2. **Statistical Evaluation Method**: The previously proposed evaluation method is extended to include the statistical behavior of direction finding algorithms, thereby better predicting the actual performance of the antenna. 3. **Practical Application Verification**: A sample antenna was manufactured and tested in an anechoic chamber, demonstrating its direction finding capability in real-world environments. Through these efforts, the paper showcases the potential application of the proposed antenna design on small aircraft such as Unmanned Aerial Vehicles (UAVs), particularly in the capability of automatically locating other airborne traffic participants.