Optimised quantisation system for PCM signals - uses signed quantisation of actual values from range giving reduced redundancy

摘要

The method is particularly suited for the optimised encoding of DPCM signals. Information is preserved by the use of the signed quantisation of actual valves rather than of the increments which could be ambiguous i.e. positive or negative. A theoretical sub division is made initially into an expected quantum (Me) and a remaining quantum (Mr). The centre of the expected quanta (Me) is determined by one or several functions (E) and optimisation of stepping is established.