Blind recovery of smart meter wireless transmissions employing Independent Component Analysis in the presence of strong wideband interference
Raghuram Ranganathan, Robert C. Qiu, Shujie Hou, Husheng Li · 2011
The technique of Independent Component Analysis (ICA) has been applied in various statistical signal processing applications for Blind Source Separation (BSS), blind deconvolution, and so on. Recently, the Robust Principal Component Analysis (PCA) method was developed to separate a low rank matrix from its sparse component. In this paper, the novel approach of combining the concept of Independent Component Analysis (ICA) with the Robust PCA technique is presented to recover the simultaneous smart meter wireless transmissions in the presence of strong wideband interference. The low-rank property of the autocovariance matrix of the smart meter wireless signals, and the sparseness of the autocovariance matrix of the wideband interferer is exploited to effectively separate them using the Robust PCA method. Therein, a complex FastICA separation algorithm with a saddle point test called FicaCPLX is employed to separate the simultaneous transmissions of the smart meters. In this manner, by enabling simultaneous transmission, possible large delays incurred by using the Carrier Sense Multiple Access (CSMA) approach can be avoided.