The aim of this paper is to propose and evaluate joint interference alignment (IA) precoding with iterative non-linear frequency domain equalization for SC-FDMA based heterogeneous networks. We assume that a set of small cells coexist with a macro-cell within the same spectrum, and therefore interfere to each other. Thus, to enable transmissions from the small-cells, without any performance degradation at the macro-cell, we enforce that all interference generated by the small cells is aligned in an orthogonal subspace to the macro-cell received signal space. Then, we design an iterative nonlinear frequency domain equalizer at the macro-cell receiver that is able to recover the macro-cell spatial streams, in the presence of both small-cell and inter-carrier interferences. For the small cells we consider that the access points are connected through a limited capacity backhaul network to a central unit (CU) where the separation of the small-cell signals is also performed through an iterative nonlinear frequency domain equalizer. The matrices for these two nonlinear space-frequency equalizers are obtained by minimizing the overall mean square error of all data streams at each subcarrier. The results have shown that the proposed receiver structures are robust to the inter-system interferences and are able to efficient separate the macro and small cells spatial streams.