Quantization Noise Reduction by Digital Signal Processing-Assisted Analog-to-Digital Converter for In-Band Full-Duplex Systems

Abstract

In-band full-duplex (IBFD) systems are attracting attention as a key technology for ideally doubling channel capacity for beyond 5G and 6G. First, we propose an analytical model of the effect of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$N$ </tex-math></inline-formula> -bit analog-to-digital converter (A/D) quantization noise on channel capacity quasi-upper limit of the desired signal at the base station of the IBFD systems. We derive averaged and peak powers of input signal to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$N$ </tex-math></inline-formula> -bit A/D and quantization noise power of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$N$ </tex-math></inline-formula> -bit A/D to formulate uplink (UL) channel capacity quasi-upper limit. Numerical analysis results by the proposed analytical model show that the UL channel capacity quasi-upper limit degrades by 16% and 80% in the 12- and 8-bit A/D, respectively. Second, we propose digital signal-processing (DSP)-assisted dual- and quad-A/Ds to reduce the quantization noise effect in IBFD systems. When A/D resolution bit is 8-bit, the proposed DSP-assisted dual- and quad-A/Ds improve the UL channel capacity quasi-upper limit by up to 92.3 and 192%, respectively, compared to conventional A/D for peak-to-average power ratio condition of 15 dB and analog self-interference (SI) cancellation capability of 30 dB. The proposed dual- and quad-A/Ds can reduce the burden on the analog SI cancellation circuitry by 6 and 12 dB, respectively.