Digital Self-Interference Cancellation for In-Band Full-Duplex

Introduction : In-band full-duplex (FD) is a promising method to increase the spectral efficiency of current communications systems by transmitting and receiving data simultaneously in the same frequency band. In order for an FD node to operate correctly, the strong self-interference (SI) signal that is produced at the node’s receiver by its own transmitter needs to be effectively cancelled. A combination of SI cancellation in the analog and in the the digital domain is usually necessary in order to suppress the SI signal down to the level of the receiver noise floor. Analog cancellation can be either passive (i.e., through physical isolation between the transmitter and the receiver) or active (i.e., through the injection of a cancellation signal) and it is necessary in order to avoid saturating the analog front-end of the receiver. However, perfect cancellation in the analog domain is very challenging and costly to achieve, meaning that a residual SI signal is still present at the receiver after the analog cancellation stage. In principle, this residual SI signal should be easily cancellable in the digital domain, since it is caused by a signal that is fully known. Unfortunately, in practice this is not the case as several transceiver nonlinearities, such as various baseband non-linearities (e.g., digital-to-analog converter (DAC) and analog-to-digital converter (ADC)), IQ imbalance, phase-noise, and power amplifier (PA) non-linearities, distort the SI signal.
Difficulties and ML based solution : 1. Transmission and Receiving Data Mismatch: In digital domain, signals are sampled at specific rates. Digital cancellation requires that the transmitted digital signal passing through DAC to the emitting antenna has to be sampled at the same rate as the received analog signal will be sampled. If the transmission sample rate in the cancellation chain differs from the receiving ADC sample rate, the cancelling transmission data will mismatch the received data, which is a common problem in digital cancellation.
2. Real-time Iterative Cancellation [MFD]: In machine learning, training and loading are two separate processes, but the iterative self-interference cancellation with machine learning requires both processes have information exchanged. It is quite challenging for two chronologically separated processes to exchange information while meeting the real-time requirement demanded by wireless communications. Moreover, a machine learning full duplex node has to update the trained parameters each time and load the new trained parameters. Even the most advanced software defined radio platform USRP GNU- Radio available today is lack of the architecture to support the synchronized training and loading. Thus a new design of the architecture is required to support machine learning cancellation solutions.
The solution proposed in [MFD] is based on GNURadio SDR and USRP X310 platform.

Supervisor name: 
Dr. Mathini Sellathurai
Supervisor and Deputy email addresses:

Project Type:

Deputy name: 
Dr Khandaker