Calibration of 5G Millimeter-Wave Active Antenna Array Platform With Search-Based Optimization Method

Array calibration becomes challenging when not only the main lobe gain but also the side lobes become important for interference mitigation from different directions. This article presents an antenna array calibration method that uses multiple variable optimization techniques to calibrate phase errors. A starting point of the calibration method is the radiation pattern without phase calibration. This radiation pattern is set as the target pattern for the phase error search, where phase errors resulting in the best matching radiation pattern are found using simulations to model the antenna array used. The search space increases as a function of the array size, so the multistart search method is used to avoid getting stuck in local optimums. In this article, a prototype millimeter-wave (mmW) antenna array platform is used to verify the functionality of the calibration method. A simulation model of the antenna array is created that can be used for the search. The radiation pattern measurement results show that the proposed method can achieve a similar calibration performance compared with a comprehensive scattering-parameter phase calibration. Comprehensive calibration requires a phase-coherent measurement setup and the control of individual antenna paths. However, the proposed method allows all elements of the antenna array to be active during the calibration measurement, as in the normal operation mode of the array. The method presented uses only amplitude measurement information, simplifying the calibration procedure and measurement setup. Additionally, phase calibration calculations can be performed offline with normal computing hardware.