iPHOEBE project investigates the technologies required for the development of a multi-wavelength (WDM) integrated photonic beamformer, with the novelty of operating through optical comb excitation coupled to multicore fiber (MCF), including an enhanced beam shaping by machine learning with operation in different sub-THz bands. These developments will enable the beamformer to work simultaneously at several frequencies above 95 GHz, responding to recent specifications for 6th generation (6G) cellular technology of experimental licenses by the FCC. In particular, correct operation at 35, 94, 140 and 220 GHz is of special interest since these frequencies present valleys in atmospheric attenuation.
The design and manufacturing of this integrated beamformer presents research challenges not resolved in the state-of-the-art, in particular: the generation of optical frequency combs and filtering each line of the comb, the phase modulation of different spectral lines to induce different delays to point the beam in different directions, the coupling of the light to multicore fiber (MCF) to feed each antenna element (AE), the machine-learning (ML) processing algorithms to adjust the delay of each AE and steer the beam, and the generation of radio frequency signals in the sub-THz band. The integrated beamformer will be designed and manufactured in Silicon Nitride (Si3N4) due to its tolerance to manufacturing imperfections, which is an important aspect given the complexity of the design.
The proposed photonic beamformer represents an advance on the state of the art since it can operate at much higher frequencies than those implemented in electronic technology, with greater bandwidth, allows feeding the array of antennas remotely and can be configured at different frequencies using machine learning.