A SYSTEM FOR PROCESSING AUDIO FREQUENCY INFORMATION FOR FREQUENCY MODULATION

摘要

Generally in land mobile FM transmitters a speech signal is pre-emphasised between 300 Hz and 3 kHz whereafter it is filtered to provide a sharp cut-off and in order to maintain channel integrity the maximum peak frequency deviation is limited. In consequence as the level of the pre-emphasised speech signal increases it is limited to a maximum peak output level which corresponds to the maximum peak frequency deviation. Heretofore the filters have had such a step response that overshoot and ringing is produced when filtering the amplitude limiter signals. As the peak of the overshoot is treated by the frequency modulator as the maximum peak output level then the level of undistorted speech for frequency modulation is typically 60% to 70% of the maximum peak output level.
The level of undistorted speech can be raised to within 90% of the transmitter peak deviation by directly connecting the amplitude limiter (30) to a filter amplifier (32) which behaves as a level sensitive filter when the speech signal has been limited and prevents ringing and overshoot being produced. The output from the filter amplifier (32) is coupled to a low pass filter (38), typically a fourth-order Butterworth filter, via an attenuator (34) which reduces the amplitude of the signal to prevent the onset of further clipping.