International Business Machines Corporation

Francis Paul Maddens

Dominique N Godard

Andre E Desblache

Francois Gabriel Caron

The method of adjusting the end of transmission in a data transmission system between a transmitting Data Terminal Equipment (DTE) and a receiving DTE connected respectively by means of a transmitting modem and a receiving modem linked therebetween by a telephone line, consisting in transmitting by the transmitting modem an End sequence of predetermined duration immediately after the data message and just preceding the signal energy drop, and composed of bits of the same polarity. The transmission of the End sequence by the transmitting modem is started as soon as the data message from the transmitting DTE includes an uninterrupted sequence of bits of the same polarity, the End sequence being completed by modem with bits of the same polarity until the predetermined duration, after a request signal for transmitting from the transmitting DTE has turned off.

Method of adjusting the end of transmission in a modem

The digital samples (An) representing the components in phase (X) and quadrature (Y) of data signals, are digitally integrated by the assembly comprised of multipliers (14 and 15) and an adder (16). The energy signal (Bn) which is obtained, is loaded into a register (17) so as to be provided with the energy signal obtained on the preceding sampling time. The value (Bs) of the energy signal from which the energy signal decrease rate exceeds a predetermiined value, is loaded into a register (21). Then a comparator (23) provides a binary 1 signal transmitted on the output line (25) as the energy drop signal when the value of the energy signal (Bn) has fallen below a predetermined fraction of the value (Bs) stored in register (21).

Device for detecting a data signal energy drop

A clock setting circuit is provided at a receiving modem for adjusting the phase of a timing signal defining the signal sampling instants. The received signal is filtered in two filters to derive a first signal having a phase .phi..sub.1 and a frequency f.sub.1 equal to f.sub.c - 1/2T, f.sub.c being the carrier frequency and 1/T being the transmission baud rate, and a second signal having a phase .phi..sub.2 and a frequency f.sub.2 equal to f.sub.c + 1/2T. The first and second derived signals are combined to derive an error signal indicative of the phase difference .phi..sub.2 - .phi..sub.1 which difference is used for adjusting the phase of a phase locked oscillator which provides the timing signal.

Growing community of inventors

Cagnes-sur-Mer, France

Francis Paul Maddens