Top Percentile FilterΒΆ

Top Percentile Filter algorithm replaces the original data values with a chosen percentile of a certain number of neighbor points (e.g. their left \(M\) and right \(M\) data points). Note that the choice of filter width \(L = 2M + 1\) has a great impact on the processed results. It is equivalent to the Centered Moving Median algorithm if the 50th percentile is chosen.

Input Parameters

Parameter Type Constraint Description Remarks
\(Y[n]\) \(Y[n] \in \mathbb R^N\) \(N \in \mathbb{N}\) Input data sequence of length \(N\) None
\(L\) \(L \in \mathbb N\) \(L = 2M + 1, \quad M \in \mathbb{N}\) None None
\(p\) \(p \in \mathbb{R}\) \(0 \leq p \leq 1\) A specified percentile of a certain number of data points considered None

Output Parameters

Parameter Type Constraint Description Remarks
\(\hat{Y}[n]\) \(\hat{Y}[n] \in \mathbb R^N\) \(N \in \mathbb{N}\) Output data sequence of length \(N\) None

Single Steps using the Algorithm