 Landweber spectral filter

ID: landweber-spectral-filter

Starting Landweber iteration at zero gives

x_{n} = \sum_{j} \frac{1 - ( 1 - τ σ _{j}^{2} ) ^{n}}{σ _{j}} ⟨ y, u_{j} ⟩ v_{j}

f_{n} (σ) = 1 - (1 - τ σ^{2})^{n}

. If the full inverse coefficient is called the filter instead, it is

f_{n} (σ) / σ

; these conventions describe the same operator.

 New to topics? Read the docs here!