Spatial DCT-based least square estimation in multi-antenna multi-cell interference channels

Interference management techniques in multicell multiple input multiple output (MIMO) networks require accurate channel state information (CSI). A popular technique for acquiring this CSI in time division duplex (TDD) systems is uplink training by exploiting the reciprocity of the wireless medium. Recently, the pilot contamination problem has been identified as one of the limiting factors for such kind of CSI acquisition. In an effort to tackle the problem of contamination, we utilize the compression capability of discrete cosine transform (DCT) and covariance matrices to spatially separate the estimate for the scenario of correlated single input multiple input (SIMO) channels. A least square estimation framework is proposed by utilizing the DCT to separate the overlapping spatial paths that create the interference. The spatial domain is thus exploited to mitigate the contamination which is able to discriminate across the interfering users. We validate our algorithms by simulation and compare them to the state of the art techniques.