A method of processing time-discrete measured values, which can be described in their time characteristic by a first exponential function which has a first time constant, the method comprising the steps of: detecting a first measured value and storing the first measured value, detecting a second measured value and storing the second measured value according to a defined time interval with respect to the detection of the first measured value, filtering the first measured value and second measured value by calculating a sum of the first measured value and a weighted difference between the second measured value and the first measured value, thereby generating time-discrete output values which can be described in their time characteristic by a second exponential function having a second time constant different from the first time constant, and outputting the output values is disclosed.