Aerial Multiport Direction-Finding Antenna Design Using Characteristic Mode Analysis

authored by: Lukas Grundmann
supervised by: Dirk Michael Manteuffel
Abstract: Direction-finding antenna systems are of particular importance in the aeronautical context, with the airborne collision avoidance system (ACAS) being among the most prominent applications. The ACAS is a long established, fully automated system. By detecting other aircraft, communicating with them and issuing resolution advisories if necessary, it contributes to the prevention of collisions. With ever more challenging integration scenarios for such systems, multimode multiport antennas (M³PAs) become an interesting alternative to the commonly employed antenna arrays. As M³PAs implement multiple ports on a single physical antenna structure, many existing analysis methods are not applicable. Therefore, alternative evaluation methods and design guidelines are developed. In these, characteristic modes (CMs), which constitute an eigenmode decomposition of the scattering behavior of arbitrary antenna structures, are employed. Thereby, the radio frequency behavior of a possible antenna structure is evaluated without the need to design a set of matched ports first. Particular care is taken of the CMs' resonance behavior, which dictates the bandwidth of the antenna. Here, the crossing of specific eigenvalue traces over frequency is allowed for some symmetry groups, but prohibited for others, leading to microscopic crossing avoidances (MICAs) in the latter case. Through additional geometry modifications, these transform into macroscopic crossing avoidances (MACAs). Designing the antenna such that MACAs are avoided in the frequency range of the application is found to stabilize the direction-finding behavior of the antenna system over frequency. With the far-fields as a common computational interface, both CMs and ports can be investigated using the proposed procedures. Firstly, to this end, a purely deterministic evaluation procedure is introduced. It is based on the correlation between the values measured at the antenna ports for different directions of arrival (DoAs). Therefore, it provides detailed physical insight into the behavior of the antenna and is particularly well suited in early design stages for establishing a fundamental understanding of the investigated antenna structure. This procedure is employed as a design guidance in the development of a cuboid demonstrator antenna with three matched ports, which is found to provide good direction-finding performance in an ACAS-inspired flight test. Secondly, a direction-finding algorithm is incorporated into the antenna evaluation method, with multiple signal classification (MUSIC) being used as the example. The correlation-based, deterministic metric is replaced by the statistical results obtained from the algorithm under the influence of noise. A reference for the otherwise arbitrary signal-to-noise ratio is introduced through a set of idealized reference antenna patterns. This approach is best suited for later design stages, where a basic understanding of the antenna has already been established and the utilized algorithm is known. To show the feasibility of the statistical evaluation procedure, a second, cupola-shaped demonstrator antenna is designed. Additional requirements, such as improved aerodynamic behavior and angular coverage are also introduced and fulfilled, showcasing the flexibility of the proposed procedure. In a second flight test, this demonstrator antenna also reveals the successful improvement of the estimation accuracy for the azimuth angle, compared to the cuboid antenna. As both evaluation procedures are designed to be inter-operable, they form a combined, generally applicable design guidance for direction-finding antenna systems, in particular for, but not limited to, aerial applications.
Organisation(s): Institute of Microwave and Wireless Systems
Type: Doctoral thesis
No. of pages: 162
Publication date: 21.11.2025
Publication status: Published
Electronic version(s): https://doi.org/10.15488/20074 (Access: Open)
: Details in the research portal "Research@Leibniz University"

BibTeX

@phdthesis{3c9643d5e4af42c9aecd2cb38cb3a2f6,
title = "Aerial Multiport Direction-Finding Antenna Design Using Characteristic Mode Analysis",
abstract = "Direction-finding antenna systems are of particular importance in the aeronautical context, with the airborne collision avoidance system (ACAS) being among the most prominent applications. The ACAS is a long established, fully automated system. By detecting other aircraft, communicating with them and issuing resolution advisories if necessary, it contributes to the prevention of collisions. With ever more challenging integration scenarios for such systems, multimode multiport antennas (M³PAs) become an interesting alternative to the commonly employed antenna arrays. As M³PAs implement multiple ports on a single physical antenna structure, many existing analysis methods are not applicable. Therefore, alternative evaluation methods and design guidelines are developed. In these, characteristic modes (CMs), which constitute an eigenmode decomposition of the scattering behavior of arbitrary antenna structures, are employed. Thereby, the radio frequency behavior of a possible antenna structure is evaluated without the need to design a set of matched ports first. Particular care is taken of the CMs' resonance behavior, which dictates the bandwidth of the antenna. Here, the crossing of specific eigenvalue traces over frequency is allowed for some symmetry groups, but prohibited for others, leading to microscopic crossing avoidances (MICAs) in the latter case. Through additional geometry modifications, these transform into macroscopic crossing avoidances (MACAs). Designing the antenna such that MACAs are avoided in the frequency range of the application is found to stabilize the direction-finding behavior of the antenna system over frequency. With the far-fields as a common computational interface, both CMs and ports can be investigated using the proposed procedures. Firstly, to this end, a purely deterministic evaluation procedure is introduced. It is based on the correlation between the values measured at the antenna ports for different directions of arrival (DoAs). Therefore, it provides detailed physical insight into the behavior of the antenna and is particularly well suited in early design stages for establishing a fundamental understanding of the investigated antenna structure. This procedure is employed as a design guidance in the development of a cuboid demonstrator antenna with three matched ports, which is found to provide good direction-finding performance in an ACAS-inspired flight test. Secondly, a direction-finding algorithm is incorporated into the antenna evaluation method, with multiple signal classification (MUSIC) being used as the example. The correlation-based, deterministic metric is replaced by the statistical results obtained from the algorithm under the influence of noise. A reference for the otherwise arbitrary signal-to-noise ratio is introduced through a set of idealized reference antenna patterns. This approach is best suited for later design stages, where a basic understanding of the antenna has already been established and the utilized algorithm is known. To show the feasibility of the statistical evaluation procedure, a second, cupola-shaped demonstrator antenna is designed. Additional requirements, such as improved aerodynamic behavior and angular coverage are also introduced and fulfilled, showcasing the flexibility of the proposed procedure. In a second flight test, this demonstrator antenna also reveals the successful improvement of the estimation accuracy for the azimuth angle, compared to the cuboid antenna. As both evaluation procedures are designed to be inter-operable, they form a combined, generally applicable design guidance for direction-finding antenna systems, in particular for, but not limited to, aerial applications.",
author = "Lukas Grundmann",
year = "2025",
month = nov,
day = "21",
doi = "10.15488/20074",
language = "English",
publisher = "Leibniz Universit{\"a}t Hannover",
school = "Leibniz University Hannover",
}