Voice scrambler using syllabic masking

Abstract

Voice signals are combined with a masking signal that changes frequencies between discrete frequencies at a selected rate and in a random manner. The combined voice spectrum and masking signal is frequency translated by heterodyning the combined signals and a carrier signal, with the carrier signal varying between discrete frequencies at a selected rate that is faster than the selected rate of the masking signal. The scrambled output is coupled to a utilization means through a communication link such as telephone lines or radio frequency waves. The utilization means includes an unscrambler in which the scrambled input is heterodyned with a carrier signal that is synchronized with the carrier signal in the scrambler and has the same discrete frequencies and at the same time as the carrier signal in the scrambler. The combined voice spectrum and masking signal is thus unscrambled and is applied in series to a mixer, a notch filter and another mixer for removing the mask signal. The local oscillator signal for each mixer is the mask signal translated to a higher frequency and synchronized with the mask signal in the scrambler unit.