Abstract: The users’ performances are affected by the interferences in cellular systems. Especially for the users on the boundary whose communication qualities are poor. The interference alignment (IA) is a promising technology to effectively mitigate interferences and improve system throughput in wireless communications which has received lots of attentions. In this paper, we propose a linear IA algorithm that can achieve high degrees of freedom with less antennas for the users on the boundary of two-cell multiple-input multiple-output (MIMO) cellular systems. The algorithm uses the precoding vectors to eliminate the inter-user interference (IUI) and uses the interference suppression vectors to remove the inter-cell interference (ICI). Assuming each Base station equipped with M antennas communicate with K users each equipped with N antennas. Just let M=K and N=K+1, the proposed algorithm can complete downlink and uplink IA and can achieve a total DOF of 2K with low overhead. The algorithm doesn’t require inter-cell feedback in the downlink and it requires only a small amount of inter-cell feedback in the uplink. Simulation results show that the proposed algorithm can achieve better performance than the conventional schemes.