Current trends in wireless communications require a Base Transceiver Station (BTS) to support an ever-increasing number of antennas, wider bandwidths, multiple frequency bands, and many simultaneous beams. This is a tall order when considering all-electronic implementations, especially when operating at high carrier frequencies. We describe here Fourier-optics based opto-electronic architectures that include analog and fully connected transmit (TX) and receive (RX) implementations that support simultaneous multi-band, multi-beam phased arrays where optical up- and down-conversion are used to generate IF/RF and RF/IF, respectively, signals for simultaneous multi-user and multi-beam transmission and reception over a wide range of frequencies. The proposed lens-based architectures have the ability to transmit and receive multiple distinct beams simultaneously even when only analog beamforming is performed, thus greatly simplifying the tasks of cell search and initial access without having to resort to complex matrix-based beamforming networks. Additionally, lens-based TX and RX beamforming is performed in the analog optical domain with zero power consumption and negligible latency bounded only by the time the light takes to travel through the lens. Photonic signal processing also reduces the number of RF components required at the remote radio unit (RRU) and the number of costly and power-hungry high-performance analog-to-digital/digital-to-analog converters (ADCs/DACs) required. The optical subsystems are ultra-wideband and frequency agnostic as only the front-end components (antennas, amplifiers) are RF frequency/band specific. Therefore, a single photonic system design may be operated at any band and, furthermore, existing installations may be modified/upgraded to operate in different bands with minimal component replacements needed. Finally, the presented architectures provide near unlimited beam-bandwidth product (BBP) with minimal power requirements and no external cooling. Theoretical analysis and experimental confirmation of the proposed architecture will both be reported, including a receiver with a nominal BBP of 36 GHz with a prototype system that consumed less than 300 W, thus yielding a power efficiency of beam formation of 8 W/GHz, which is more than a 6× improvement over the current state of the art.
Read full abstract