> reflective layers behind the sensor Reminds me of this surprising little tidbit from E. T. Jaynes' 1991 paper "Theory of Radar Target Discrimination" (https:

> reflective layers behind the sensor Reminds me of this surprising little tidbit from E. T. Jaynes' 1991 paper "Theory of Radar Target Discrimination" (https://bayes.wustl.edu/etj/articles/craddis.pdf https://bayes.wustl.edu/etj/articles/craddis.pdf) One might then think naively that we should take equal care to match the receiver to its transmission line so that all the energy captured by the receiving antenna is actually delivered to the receiver. However, this is not the case for a good receiver. [...] Quite generally, in order to deliver the maximum signal/noise ratio at its output, an ideal receiver must reflect all the energy incident up on it. It is only in the limit of an infinitely bad receiver, in which all the output noise is generated internally, that matched input impedance becomes the condition for maximum signal/noise ratio at the output. Actual receivers are somewhere between ideal and infinitely bad, and so they perform best when partially matched, so that a part of the incident energy is reflected and radiated back out the receiver antenna. In fact, an ideal receiver does not run on energy at all, but reflects back all the energy incident on it! [...] This fact surprises many people on first hearing; but we note that it is so general that it remains true in quantum theory, at optical frequencies where hf >> kT and the Nyquist noise formula no longer holds. For initiation of a photo chemical reaction it is not necessary that the light energy be absorbed. For example, it might be thought that the eyes of animals adapted to seeing in the dark would have pupils that act as perfect black bodies, absorbing all the incident light energy. On the contrary, it is a familiar fact that the animals with best night vision have eyes that reflect the incident light strongly, looking like search-lights in the dark.