Scalable Modular Beamformers for Energy-Efficient MIMO Applications Using Correlation Technologies

A scalable beamformer module built using unitary $1 \times 8$ sub-arrays integrating Antenna-in-Package (AiP) in a profile of 5mm pitch array is introduced. The $\mathbf{1}\times \mathbf{8}$ module can be stacked either horizontally or vertically to create dynamic phased arrays of 8, 16, 32 or 64 elements horizontally or $\mathbf{8}\times \mathbf{8},\mathbf{16}\times \mathbf{16}$ or $\mathbf{32}\times \mathbf{32}$. The unit is USB controlled and is bidirectional. Beams can be calibrated, and calibration stored in EEPROM for fast Look-Up-Table (LUT) access. The module offers 6-bit amplitude and 6-bit phase controls. The combined gain (when all elements sum up coherently) is 30dB for receive mode and 20dB for transmit mode. The module is capable of achieving $+\mathbf{53}\mathbf{dBm}$ EIRP when used in an $8 \times 8$ configuration. The unit can switch seamlessly between TX and RX modes and has internal DC-DC converters for maximum efficiency. Multiple-Input Multiple-Output (MIMO) correlators are introduced for building scalable multi-beamforming front-end-modules (FEM) with embedded cognition. Unified mmWave and baseband correlation technologies using energy density and Error Vector Magnitude (EVM) metrics are introduced for low complexity cost-effective multi-beamforming systems. The combination of convolutional accelerators with analog signal processing for linking beam-selection algorithms to stochastic-wave-shaping (SWS) will foster new system architectures with critical functionalities including direction of arrival (DOA) estimation and secure communications.

I. Introduction

Beamformer front-end-modules (FEM) enabling multi-user multiple-input multiple-output (MU-MIMO) operation for simultaneously connecting several user-equipment (UE) devices based on same time-frequency resources are known to achieve a higher capacity compared to single-user MIMO (SU-MIMO) [1]–[12]. However, in classical MIMO systems, fully digital beamforming requires that each antenna be driven by its dedicated RF chain, potentially leading to an impractical implementation of high system complexity, cost, and energy consumption [12]–[28].