basically the postprocess material is based on this core part:
It’s just repeated 4 times for the horizontal and vertical directions (this is the computing for the left pixel, as you can see on the ConstantFloat2 node that is [1,0] ). Then everything is added-up and saturated.
The ScreenPositionToWorldPosition function is the custom one I created, mentioned above.
PingPosition is the float3 location of the actor that activates the scan, set in BP through a MaterialParameterCollection.
Hope this helps