过滤流数据以减少噪声，卡尔曼滤波器c#

我正在将数据从惯性传感器流式传输到 C# 应用程序中。数据有点嘈杂，所以我需要添加一个过滤器来平滑它。我有一个卡尔曼滤波器实现，在给定数组时效果很好，但我无法理解如何在恒定数据流上使用它。

I have:

double sensorData; //the noisy value, constantly updating from another class.

过滤器：

public static double[] noisySine = new double[20] { 40, 41, 38, 40, 45, 42, 43, 44, 40, 38, 44, 45, 40, 39, 37, 41, 42, 70, 44, 42 };
    public static double[] clean = new double[20];

      public static void KalmanFilter(double[] noisy)  
            {                  
                double A = double.Parse("1"); //factor of real value to previous real value
                // double B = 0; //factor of real value to real control signal
                double H = double.Parse("1"); 
                double P = double.Parse("0.1");
                double Q = double.Parse("0.125");  //Process noise. 
                double R = double.Parse("1"); //assumed environment noise.
                double K;
                double z;
                double x;

                //assign to first measured value
                x = noisy[0];
                for (int i = 0; i < noisy.Length; i++)  
                {
                    //get current measured value
                    z = noisy[i];

                    //time update - prediction
                    x = A * x;
                    P = A * P * A + Q;

                    //measurement update - correction
                    K = P * H / (H * P * H + R);
                    x = x + K * (z - H * x);
                    P = (1 - K * H) * P;
                    //estimated value
                    clean[i] = x;
                    Console.WriteLine(noisy[i] + " " + clean[i]);
                }
            }

我如何流式传输双精度值而不是数组，并返回（过滤后的）双精度值？

谢谢。

创建这个类：

public class KalmanFilter
{
    private double A, H, Q, R, P, x;

    public KalmanFilter(double A, double H, double Q, double R, double initial_P, double initial_x)
    {
        this.A = A;
        this.H = H;
        this.Q = Q;
        this.R = R;
        this.P = initial_P;
        this.x = initial_x;
    }

    public double Output(double input)
    {
        // time update - prediction
        x = A * x;
        P = A * P * A + Q;

        // measurement update - correction
        double K = P * H / (H * P * H + R);
        x = x + K * (input - H * x);
        P = (1 - K * H) * P;

        return x;
    }
}

并使用该类：

KalmanFilter filter = new KalmanFilter(1, 1, 0.125, 1, 0.1, noisySine[0]);
for (int i = 0; i < noisy.Length; i++) clean[i] = filter.Output(noisySine[i]);