Fast SQUID Pseudo Random Generators

Digital and counting techniques provide very comprehensive measuring methods. Even in the field of noise measurements which seems to be the domain of analogous signals digital methods become more interesting which try to replace the conventional methods both in generating and detecting noise signals. The generation of random signals by fast digital methods is the subject which will be considered here. In the field of noise measurement, the generation of well defined noise power within the rf and hf range has become an important task. A number of applications in engineering, such as long-distance communication, navigation, localization, remote sensing and others need random electromagnetic wave generators in this frequency range. Even in meteorological and medical investigations and not forgetting radioastronomical observations, the measurement of noise spectra provides new ways of gathering information about systems. Almost all of these applications rely on the fact that electromagnetic noise waves have a power spectrum which is widely distributed in frequency. Since noise measurement is constituted by the intercomparison of two noise powers, noise sources of high accuracy are needed, the power of which can be exactly calculated. The convenient method adapted in the RF frequency range is to generate thermal noise in a dissipative element. Here, the noise power density is given via Nyquist’s theorem by the temperature. However, these thermal systems suffer from the fact that on account of thermal effects, particulary the heating, they are not reliable over long time periods. As will be shown in this chapter, pseudo random generators based on fast superconducting switches present a new solution to this problem of generating noise power spectra. They allow the output noise power to be defined in terms of the measurements of a voltage, a resistance and a frequency. The work described was sponsored by the “Stiftung Volkswagenwerk”.