All signal processing systems need to be realized with analog circuits, digital hardware and/or software, or mixed-signal systems which constitute the combination af analog and digital subsystems. Such designs acknowledge that interaction with the physical environment is always analog while digital implementation becomes more and more advantageous in terms of hardware resources and powerconsumption. Going for “Green ICT”, therefore, often means to optimize a mixed-signalsystem design where digital methods assist the analog circuits to strike the best balance in system performance versus sustainable energy use.

On the digital side, the goal is best characterized as algorithm engineering, examples of which are found in our development of highly efficient multi-rate implementations of nonlinear filters and in the use of reconfigurable computing architecture where the challenge lies in the design of custom-precision number formats in both fixed point and floating point. Being able to choose an optimized computational precision according to the application
requirements and the internal properties of a signal processing algorithm allows substantial savings in resources without sacrificing resources. In particular, such choices can even be based on the online estimation of the relevant information in observed data and derived algorithm variables where probabilistic models from machine learning provide the necessary online analysis.