... taken)¹

The noise means are denoted by $\bar{\mathbf{n}} = E[\mathbf{n}]$ and $\bar{\mathbf{v}} = E[\mathbf{v}]$, and are usually assumed to equal to zero.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... terms²

While ``second-order'' versions of the EKF exist, their increased implementation and computational complexity tend to prohibit their use.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...\space ³

$\mu$ is usually set to a small constant which can be related to the time-constant for RLS weight decay [3]. For a data length of 1000, $\mu \approx 1e-4$ was used.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... state-space⁴

The covariance of ${\bf u}$ is again adapted using the RLS-weight-decay method.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.