Kalman filter


Kalman filter can be applied to aggregate multiple sources of data with Gaussian errors, while allowing addition of two noisy data. This is due to two closed properties of normal distribution.

Let a, b be two independent random variables. s.t.
a ~ Normal( u1, sd1 )
b ~ Normal( u2, sd2 )

1.  a+b ~ Normal( u1 + u2,  sqrt( sd1^2 + sd2^2) )

2. P( a == x && b == x) = P( a == x ) * P( b == x ) = normal(x, u3, sd3)
normal(x,u,sd) = exp( (x-u)^2/sd^2) / sqrt(2*pi) / sd
u3 = (u1*sd2^2 + u2*sd1^2)/(sd1^2+sd2^2)
sd3 = 1/ sqrt( 1/sd1^2 + 1/sd2^2 )


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s