University of Minnesota
Toward the Development of Scalable Phased Array Architectures for Communications and Healthcare
Abstract
dc:description.abstractAs the size of the multi-channel transceivers keeps scaling up to meet the growing signal-to-noise ratio (SNR) requirements, the scalability of such multi-channel systems encounters several practical challenges and poses a major bottleneck in the development of next-generation communications and healthcare systems. In communications, the expanding demand for higher data rates and increased bandwidths is projected to continue with 6G cellular communication utilizing phased arrays operating at mm wave and beyond. In healthcare, the advent of high-channel-count (e.g., 128 channels) ultra-high-field MRI receivers can significantly improve the SNR and reduce the image acquisition time. However, the design of such multi-channel frontends in a limited space poses several practical issues. The mm-wave phased arrays are expected to play a significant role in overcoming the high propagation loss at Ka-band frequencies and above. However, large-scale phased arrays suffer from significant element-to-element phase mismatches, necessitating novel solutions for scalable phase alignments. This work provides an innovative solution to the challenges of frequency distribution and phase alignment in phased arrays by presenting a theoretical framework of the novel 2-D-scalable method based on nearest-neighbor phase synchronization. The settling behavior and stability of the proposed scheme over long chains of elements, as well as in the presence of process, voltage, and temperature mismatches, are derived. A comprehensive study of the scaling of random phase error and its impact on the beam pattern is provided. In order to enable a scalable multi-chip architecture, a high-speed chip-to-chip interconnect is proposed in this work. The interconnect is based on chip-to-chip wire bonding and allows direct phase detection of adjacent elements at mm-wave range. To validate the scheme’s operation, we taped out a chip prototype for the scalable array at 30GHz. We also designed a 4-layer multi-chip PCB with an antenna array for this prototype. The proposed architecture was implemented on a 3 × 3 phased array, and various measurements were done to verify the operation. The measurements included phase noise characterization, over-the-air beamforming and beam steering, beam switching transients, and probing. In this prototype, we demonstrate a ±25° beam steering range across E- and H-planes with fast switching speed. We believe the suggested scalable architecture is a prominent solution for enabling large-scale mm-wave arrays without a dramatically increased startup calibration overhead and also provides improved robustness across various operation-time drifts. Moreover, this work addresses some of the major practical challenges of scalable high-channel-count MR receiver frontends by proposing an integrated 4-channel low noise amplifier (LNA) on chip. The LNA design considerations for optimum noise performance are provided. We have implemented prototype chips for 10.5T and 7T MRI and conducted benchtop measurements of the fabricated chips. The noise figure was measured to be 0.3dB. Moreover, the low noise figure is achieved at a lower power consumption and a smaller form factor, compared to competitor modules, making this work a desired solution for high-channel-count ultra-high-field MR receivers.
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Rouhafza, Alireza
Subjects
dc:subject × 6Rights
- Language dc:language.iso
- en
Identifiers
dc:identifier.*- Dc Identifier Citation
- Rouhafza, Alireza. (2026). Toward the Development of Scalable Phased Array Architectures for Communications and Healthcare. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/283449.