GB716172A - Improvements in or relating to the correction of coded electric signals - Google Patents

Abstract

716,172. Digital electric calculating apparatus. SOC. D'ELECTRONIQUE ET D'AUTOMATISME. Jan. 18, 1950 [Jan. 19, 1949 ; Feb. 17, 1949], No. 1348/50. Class 106 (1) An electric elementary carry-over operator for correcting a train of coded pulses which is the result of combining two or more coded pulse trains, comprises an input tube stage (16) to which the incorrectly coded train is applied, an output tube stage (20), a transmission channel (P) between the input and output stages, in a channel (Q) branched off the transmission channel a peak-limiting amplifier stage (17) amplituderesponsive to produce pulses of uniform amplitude for any pulse in the transmission channel equal to, or greater than, radix value B (or, in an alternative, than a predetermined multiple of the radix), a carry channel (R) from the output of the amplifier stage (17) to the input stage (16) including resistance means (22) for limiting the amplitude of the amplifier output pulse appearing at the end of the carry channel to unit amplitude and thereby form a carry train, means (delay networks 23) for causing the carry train to be applied to the input stage with a phase-lag relative to the incorrectly-coded train of one pulse period, a cancellation channel (S) from the output of the amplifier stage (17) to the output stage (20), resistance means (21) in the cancellation channel for limiting the amplitude of the amplifier output pulses appearing at the end of the cancellation channel to radix amplitude and thereby forming a cancellation train, and means for applying the cancellation train pulses to the output stage (20) in phase with the corresponding pulses of the incorrectly coded train. The tube stage 15 merely changes the polarity of the incoming pulses. In the case of an uncorrected binary pulse train all pulses in the transmission channel P of two or more units amplitude cause the amplifier 17 to produce pulses of uniform amplitude which are applied to the cancellation channel S and the carry channel R, respectively. The resistances 21 and 22 are chosen so that, on the one hand. two-unit cancellation pulses of opposite polarity to the pulses-of the uncorrected binary pulse train are applied to the channel P while, on the other hand, single-unit carry pulses, after passing through a one-pulse-period delay line 23, are applied with the same polarity as the uncorrected pulse train to the input tube stage 16. This valve, and/or the others, may be a pentode, and the carry input may be connected to a separate grid. If the incoming train is of irregular periodicity it will generally be associated with pilot pulses, in which case a trigger circuit, actuated by pilot impulses, may replace the delay line 23. For pulse trains of regular periodicity the delay element may be interposed in the direct transmission channel (Fig. 3, not shown). Circuits similar to that shown but having threshold devices 17 adjusted for different amplitudes, may be combined to provide a correcting circuit for pulses with higher radices. Moreover, incorrected trains resulting from the combination of both positive and negative pulse trains may be handled by a suitable arrangement (Fig. 8, not shown). Reference is made to Specification 716,226 and U.S.A. Specification 2,429,227.