Due to discrepancies in the SN-Ia-cosmology-project, at the time an increasing expansion was postulated instead of the previously assumed decreasing expansion. At the beginning of this work it is stated that this is a fallacy resulting from mutually contradictory premises, mainly geometric damping with and EM wave propagation without expansion. It is shown that the prevalent propagation function applies locally only, since Maxwell's equations neither take into account, imply nor condition the expansion of the universe.In succession, an alternative propagation function with expansion is developed, which behaves like the classic Maxwell solution in the first approximation for z ≤ 0.1. This repeats the positive comparison I made earlier with the observational data of the SN-Ia-cosmology-project supplementing it by the latest high-z data z ≥ 0.9, at which point the MLE model is confirmed for this area too.Applying this model consistently, an additional evaluation mb(r) is carried out. A new, unexpected deviation emerges at r ≥ 0.1R. The observational data there is darker than calculated. The reason is the Hubble parameter, which depends on time and distance. Route sections that are further away expand faster than those that are closer. The greater the distance, the greater the value of H and the expansion speed v = Hr. With the help of a correction factor m from [75], a function mb(r) is set up that correctly traces the deviating distribution. This is also proof that the expansion rate decreases over time and does not increase.