AN IMPROVED METHOD OF SCALING FOR REAL-TIME SIGNAL-PROCESSING APPLICATIONS

Implementing digital signal processing algorithms using fixed-point ar ithmetic is a difficult task, involving trade-offs to balance the effi ciency and noise performance of a given realization. One of the most i mportant components of such design is to minimize the noise generated by quantization and overflow effects. This is generally accomplished b y the scaling of signals and coefficients in the fixed-point realizati on based upon knowledge of signal features and statistics. This paper presents a new method for structuring this design task. It uses a mode l based on scaled-fractional numbers to simplify both the concepts req uired to realize fixed-point arithmetic versions of algorithms, as wel l as the real effort required for such implementations. This model has been implemented as a data type in a high level computer language to allow direct implementation of fixed-point arithmetic versions of algo rithms.