How do you incorporate the unique patterns of each antenna in and array into the overall steering vector calculation (using Antenna and Phased Array toolboxes)?

I have a conformal array of 8 antennas, with each having a unique pattern, and some having strong mutual coupling. I want to use all of the antenna patterns in the overall array analysis, not just a representative pattern of an embedded element. I don't see how this cam be performed with the Antenna and Phased Array toolboxes. Specifically I want to characterize the steering vectors to use as calibration points for beamforming and direction finding analysis.

Answer 1

[Honglei Chen]

You can use HeterogeneousConformalArray in Phased Array System Toolbox to model arrays with unique patterns for each element. You can find the reference page at

https://www.mathworks.com/help/phased/ref/phased.heterogeneousconformalarray-class.html

For steering vectors, if you set IncludeElementResponse to true, it will include the element response in the resulting steering vector.

HTH

[Kirk]

I found this object after sending the question. It seems that I need to perform the modeling of an array in a tool like EzNEC to model the array and then import the models to each of the antenna models for each element in the array. It does not seem like I can model a simple dipole antenna, model N times in an array, and then have the complex response of that antenna be used within the phased array object. I have been performing modeling of antennas in arrays for some time outside of Matlab, and then import the frequency selective response of the antenna to truly model the phase/amplitude response of the antenna. It seems that the combination of the Antenna and Phased Array System toolboxes don't simplify that process. While the tools are interesting, they fall woefully short of performing full EM modeling of array elements (beyond the pattern function, which only allows the magnitude response of each array antenna). If I am missing something, I would like to know, but as yet, these tools are too geared towards theoretical arrays and not real arrays of varying antenna configurations and responses.

[Honglei Chen]

Hi Kirk, Antenna Toolbox does have a method of moments solver in it so it could do EM modeling. Could you elaborate a bit more how you see the two should work together? I'll be very interested in your comment and if you don't want to share it here, please feel free to shoot me an email through my author page. Thanks.