Quantization in Uplink Multi-Cell Processing with Fixed-Order Successive Interference Cancellation Scheme under Backhaul Constraint

We study the uplink Multi-Cell Processing (MCP) model where each base station (BS) is connected to a centralized processor (CP) via a noiseless backhaul link with fixed and finite capacity. User data is decoded in the CP. In this paper, a fixed-order Successive Interference Cancellation (SIC) decoding scheme is adopted in the CP and the rate region is also derived. Further, we extend the MCP model with fixed-order SIC scheme to include the constraint that each user has the requirement of signal-to- interference-and-noise ratio (SINR). We analyze the option of quantized levels and give the theoretical lower bound and upper bound of the quantized levels. The analysis reveals that the quantized levels can affect whether each user uses its own maximized power to achieve the SINR requirement or not. Finally, numerical simulation shows the lower bound and the upper bound are very closed to the theoretical results.