Cognitive Blind Interference Alignment for Macro-Femto Networks

Abstract

We propose a cognitive blind interference alignment (BIA) scheme for macro-femto two-tier networks where the users are equipped with a single reconfigurable antenna. The proposed scheme fully cancels the intracell interference in both the upper and lower tiers. Moreover, it provides the optimal Degrees of Freedom (DoF) in the macro-tier while the lower tier subject to inter-tier interference achieves non-zero DoF. To do so, the proposed scheme relies on a topology management scheme where the users belong to the macro-tier if and only if they can optimally treat the interference caused by the transmission at the femto-tier as noise. The proposed scheme does not require any channel state information at the transmitter or data sharing between the macro- and femto-tiers. An outer bound of the DoF is derived for the considered macro-femto network. It is demonstrated that the proposed scheme can attain the outer bound subject to optimal sum DoF for the macro-tier. Furthermore, compared to previous approaches, cognitive BIA can achieve better sum rates for macro-femto networks for femtocells subject to different levels of inter-tier interference.