Error analysis of a ramp-type p.c.m. source encoder for colour television signals

The principle of an inherently simple 8-bit p.c.m. source encoder for video signals is briefly discussed, and a mathematical model of the encoder is developed. Parameters in the model are each assigned a normal probability law in order to simulate the inevitable, small random variations arising from component spreads and alignment error. Monte Carlo sampling of these parameters is then used to relate the degree of instrument error to the degree of chrominance nonlinearity suffered by a PAL colour television signal. The analysis indicates that voltage errors in the encoder should invariably fall within a half-quantum if chrominance nonlinearity is to be acceptable for broadcast quality signals.