symmetric kernel (or similarity) matrix between observations - This is used as input to Kalman filtering and other clustering techniques.

position with detailed noise.

random variable with a specified distribution along an axis defined in XYZT.