Abstract: In cellular systems, inter-cell interference will severely degrade the system performance. In uplink transmission systems, through base-station (BS) cooperative receiving, the inter-cell interference in a cooperated cluster can be converted into useful signals. To implement BS cooperation, a large amount of information needs to be exchanged through backhaul links. When the backhaul capacity is limited, the received signals at each BS must be first locally processed before it is transferred to the central processor. In this paper, four local processing schemes are considered in a unified framework including direct quantization of the received signals at each antennas, single-source compression with source coding, distributive multiplesource compression exploiting the correlation among signals received at multiple BSs, and local decoding at some BSs. The connections between the achievable sum rate and the backhaul constraint are studied, and the impact of user positions and antenna configurations on system performance is analyzed. Finally, the implementation complexities and proper operation scenarios of each scheme are discussed.