Amplitude-Modulated Characterization and Calibration of Phased Arrays Using Non-Coherent Detection

Abstract

Phased arrays have emerged as critical components in today’s communication, radar, and imaging systems, particularly in response to the pursuit of higher frequencies, specifi- cally the millimeter-wave (mmWave) band, attributed to shorter wavelengths. These arrays play an important role in directing beams towards desired directions, thereby helping the transmission of signals over longer distances. Utilizing electronic circuitry, arrays synthe- size the desired beam pattern by manipulating the amplitude and phase of each element’s radiation. However, the inherent variations in the electronic characteristics of each ar- ray element can impact the radiation response, leading to undesirable beam patterns. To mitigate these non-idealities, it is critical to calibrate the array during the manufacturing process. Additionally, factors such as the aging of the array and temperature variations can further affect its performance, in-situ calibration methods are necessary. Currently, most testing and calibration methods are conducted on an element-by-element basis (serially) using a vector network analyzer (VNA), which can be both expensive and time-consuming. Parallel testing is mostly done using coherent detection methods, introducing complexity. A promising solution is Code-Modulated Embedded Test (CoMET), which has demonstrated effectiveness as a non-coherent and parallel testing and calibration method, achieving high levels of accuracy and speed. CoMET employs a power detector to test and calibrate all elements within the array in parallel. This dissertation will concentrate on the investigation and utilization of Amplitude- Modulated CoMET (AM-CoMET) to address specific challenges encountered in conven- tional CoMET. The primary contributions of this study are listed as follows. Firstly, the research aims to establish a theoretical derivation and a framework for AM-CoMET tech- niques, the study seeks to validate the effectiveness of AM-CoMET through simulations and measurements using commercial phased arrays. AM-CoMET demonstrated a 0.1 dB RMS gain error and 0.85° RMS phase error for board-level test. For free-space testing, the RMS gain error is 0.4 dB and the RMS phase error is 1°. Secondly, the research investigates the impact of inherent non-idealities within the AM-CoMET codes on its performance. Additionally, the study has yielded a novel correction technique that demonstrates a significant improvement in gain and phase estimation by mitigating the impact of the inherent non-idealities. Multiple factors impact the effec- tiveness of the correction technique; however, the board-level measurements show an improvement of gain estimation to a near-ideal case. Additionally, the correction technique was able to reduce gain error by about 40 % in a free-space environment. Thirdly, this work aims to provide a generalized mathematical framework that captures both modulation schemes, amplitude modulation (AM) and phase modulation (PM). This framework enables the analysis of code unbalancing and other non-idealities, and directly correlates them to the extracted amplitude and phase responses. Additionally, methods for correcting and de-embedding errors caused by non-ideal code modulation are presented, without the need for complex equation solving. Lastly, the research utilizes CoMET for the characterization of the 1-dB compression point (P1dB), through the estimation of the third-order harmonic gain. Moreover, an alterna- tive approach to exploit CoMET for directly estimating the P1dB is introduced. A comparison between the two methods is provided, highlighting the advantages and drawbacks of each. CoMET demonstrated an accuracy of around 0.7 dB in characterizing the P1dB compared to a VNA. These results confirm the ability of CoMET to characterize linearity metrics for phased arrays, further strengthening its potential as a substitute for VNAs.