package PCI.ASQ.ica;
//作者,著作权人： 罗瑶光，浏阳
public class CorrelationICA{
  //比较简单的鸡尾酒频率盲分割
  public static double[] frequencyUpSplit(double[] originFrequency, double[] compareFrequency) {
    double[] output= new double[originFrequency.length];
    for(int i= 0; i< originFrequency.length; i++) {
      output[i]= originFrequency[i]- compareFrequency[i]> 0? originFrequency[i]: 0;
    }
    return output;  
  }
  
  public static double[] frequencyDownSplit(double[] originFrequency, double[] compareFrequency) {
    double[] output= new double[originFrequency.length];
    for(int i= 0; i< originFrequency.length; i++) {
      output[i]= originFrequency[i]- compareFrequency[i]< 0? compareFrequency[i]: 0;
    }
    return output;  
  }  
  
  public static double[] frequencyUpSplitWithScale(double[] originFrequency
      , double[] compareFrequency, double scale) {
    double[] output= new double[originFrequency.length];
    for(int i= 0; i< originFrequency.length; i++) {
      output[i]= originFrequency[i]- compareFrequency[i]> scale? originFrequency[i]: 0;
    }
    return output;  
  }
  
  public static double[] frequencyDownSplitWithScale(double[] originFrequency
      , double[] compareFrequency, double scale) {
    double[] output= new double[originFrequency.length];
    for(int i= 0; i< originFrequency.length; i++) {
      output[i]= originFrequency[i]- compareFrequency[i]< scale? compareFrequency[i]: 0;
    }
    return output;  
  }  
}