Machine Learning for Accelerated IBFD Tuning in 5G Flexible Duplex Networks

The fifth-generation wireless system framework provides the option to evaluate the performance of in-band full-duplex (IBFD) operation through flexible duplexing. The resulting self-interference, however, must be mitigated within a fraction of a symbol duration for successful communication. This paper introduces the use of machine learning to accelerate the tuning of multi-tap adaptive RF cancellers. The tuning performance of a prototype system using a two-tap canceller was measured over a 20 MHz bandwidth centered at 2.5 GHz, and demonstrated more than 38 dB of cancellation with less than 10 tuning iterations. This tuning speed is significantly faster than previous approaches, and illustrates that this novel application of machine learning to RF canceller tuning can enable IBFD operation in dynamic interference environments.