A widely linear multichannel Wiener Filter for wind prediction

The desire to improve short-term predictions of wind speed\ud and direction has motivated the development of a spatial\ud covariance-based predictor in a complex valued multichannel\ud structure. Wind speed and direction are modeled as the\ud magnitude and phase of complex time series and measurements from multiple geographic locations are embedded in a complex vector which is then used as input to a multichannel Wiener prediction filter. Building on a C-linear\ud cyclo-stationary predictor, a new widely linear filter is developed\ud and tested on hourly mean wind speed and direction\ud measurements made at 13 locations in the UK over 6 years.\ud The new predictor shows a reduction in mean squared error\ud at all locations. Furthermore it is found that the scale of\ud that reduction strongly depends on conditions local to the\ud measurement site