Nonlinear self-interference cancellation in MIMO full-duplex transceivers under crosstalk

Abstract

This paper presents a novel digital self-interference canceller for an inband multiple-input-multiple-output (MIMO) full-duplex radio. The signal model utilized by the canceller is capable of modeling the in-phase quadrature (IQ) imbalance, the nonlinearity of the transmitter power amplifier, and the crosstalk between the transmitters, thereby being the most comprehensive signal model presented thus far within the full-duplex literature. Furthermore, it is also shown to be valid for various different radio frequency (RF) cancellation solutions. In addition to this, a novel complexity reduction scheme for the digital canceller is also presented. It is based on the widely known principal component analysis, which is used to generate a transformation matrix for controlling the number of parameters in the canceller. Extensive waveform simulations are then carried out, and the obtained results confirm the high performance of the proposed digital canceller under various circuit imperfections. The complexity reduction scheme is also shown to be capable of removing up to 65% of the parameters in the digital canceller, thereby significantly reducing its computational requirements.

Design and Implementation of a Neural Network Based Predistorter for Enhanced Mobile Broadband

“…The final output of the network after hidden layer K is given by (7) where the first element represents the real part of the signal, and the second element represents the imaginary part. In (7), W linear is a 2×2 matrix of the weights corresponding to the linear bypass. In practice, we fix it to be the identity matrix, I 2 , to reduce complexity though these weights could also be learned in systems with significant IQ imbalance.…”

Advanced Machine Learning Techniques for Self-Interference Cancellation in Full-Duplex Radios

“…One such technique is in-band full-duplex (FD), where information is transmitted and received simultaneously and on the same frequency band. While FD systems have long been considered impractical due to the strong self-interference (SI) caused by the transmitter to its own receiver, more recent work on the topic (e.g., [1]- [4]) has demonstrated that it is possible to achieve sufficient SI cancellation to make FD systems viable.…”

“…However, analog cancellation is generally expensive due to the additional analog circuitry and a residual SI signal typically still remains at the receiver, which is canceled in the digital domain. This requires modeling the non-linear effects of the different stages of the transceiver, such as digitalto-analog converter (DAC) and ADC non-linearities [5], IQ imbalance [5], [6], phase-noise [7], [8], and power amplifier (PA) non-linearities [4]- [6], [9]. Traditionally, this has been done using polynomial models, which have been shown to work well in practice.…”

Design and Implementation of a Neural Network Aided Self-Interference Cancellation Scheme for Full-Duplex Radios

“…In practice, however, the different stages of the transceiver introduce non-linearities to the signal, such as digital-to-analog converter (DAC) and analog-to-digital converter (ADC) non-linearities, IQ imbalance, and power amplifier (PA) non-linearities. Intricate memory polynomial models have to be used in order for the digital SI cancellation to be able to handle the aforementioned non-linearities (e.g., [4], [5], [6], [7], [8]). An alternative solution, which uses a neural network (NN) to reconstruct the non-linearities in order to generate the SI cancellation signal, was recently proposed in [9] and it was shown that it can achieve similar SI cancellation performance with the state-of-the-art polynomial model of [8], but with much lower computational complexity.…”

“…As such, communications applications require vastly different NN hardware accelerator architectures. Contribution: In this work, we present a hardware implementation of the SI cancellation method proposed in [9] in order to quantify and translate the computational complexity gains over the state-of-the-art polynomial based model of [8] into real-world hardware resource utilization gains. We provide FPGA and ASIC implementation results that clearly demonstrate the significant gains that can be achieved by our proposed NN-based canceller in terms of both the resource utilization and the achieved throughput.…”