Pulse or digital communications – Repeaters – Testing
Patent
1985-01-09
1987-04-07
Griffin, Robert L.
Pulse or digital communications
Repeaters
Testing
375 83, 375 86, 375 90, 375 94, 371 43, H04L 2722
Patent
active
046566481
DESCRIPTION:
BRIEF SUMMARY
BACKGROUND OF THE INVENTION
The present invention relates to a process for the noncoherent demodulation of a linearly modulated signal where each symbol .alpha..sub.k has the same energy and a demodulator for performing this process. It is more particularly used in satellite links and in vehicle radio communications.
A digital signal x(t), in which t is a time variable, is generally a narrow band signal centered about a frequency f.sub.0 called the carrier frequency and of band width B. It is therefore a signal, whose spectral density is zero outside the frequency spacing, [f.sub.0 -B/2, f.sub.0 +B/2].
In signal theory, it is standard practice to represent this digital signal by its complex envelope .alpha.(t), the relation between x(t) and .alpha.(t) being given by the equation complex number, such as j.sup.2 =-1 and w.sub.0 is the ripple corresponding to the frequency f.sub.0. As the complex representation permits a clearer definition, it will be used throughout the remainder of the text.
Consideration will be given to a sequence of N symbols a.sub.0, . . . , a.sub.k, . . . , a.sub.N-1 where k is an integer and where each a.sub.k represents an information to be transmitted, where N equals approximately 64. These symbols are elements of an e.g. binary alphabet A. For minimizing the error rate in transmitting the sequence of symbols a.sub.0 . . . a.sub.N-1, it is standard practice to encode this sequence into another sequence of symbols .alpha..sub.0 . . . .alpha..sub.N-1 in which each symbol of said other sequence belongs to another alphabet .alpha., which is e.g. of the M-type, in which M is an integer.
This sequence of symbols .alpha..sub.0 . . . .alpha..sub.N-1 will effectively be transmitted by a modulated signal, the modulation being realized by the said signals. In the case of a linear modulation, the complex envelope .alpha.(t) of the modulated signal x(t) containing the information to be transmitted is then represented by the expression ##EQU1## in which T is the time interval between the transmission of two successive symbols and g(t) is a function, with real or complex values, describing the pulse response of all the emission filters so that .vertline..alpha..sub.k .vertline..sup.2 =1 and ##EQU2## dt, equal to E.sub.b is the energy per bit.
FIG. 1 is a diagrammatic representation illustrating the known chain of modulation, transmission and demodulation of symbols a.sub.0 . . . a.sub.N-1, which are sequentially received in a modulator 2. They follow one another spaced by a time interval T. The modulator 2 comprises a coding means 4 supplying at the output the sequence of symbols .alpha..sub.0 . . . .alpha..sub.N-1, which are spaced from one another by a time interval T. It also comprises a modulation means 6, which supplies the complex envelope .alpha.(t) in the form of its real part, Re(.alpha.(t)) and its imaginary part Im(.alpha.(t)). These two signals are frequency inverted by respectively modulating a signal cos (w.sub.0 t) supplied by an oscillator 8 and a signal -sin (w.sub.0 t) supplied by a phase shifter 10, which is connected by the input to oscillator 8. The two resulting modulated signals are summated and their sum constitutes the emitted signal x(t).
This emitted signal x(t) during transmission, is subject to disturbances, represented by the addition of a Gaussian white noise b(t) of bilateral spectral density N.sub.0 /2 in watt/hertz. Thus, demodulator 12 receives a signal y(t) equal to x(t)+b(t), which is frequency reinverted by modulating a first signal 2. cos (w.sub.0.t+.theta.(t)) from an oscillator 14 and a second signal -2. sin (w.sub.0.t+.theta.(t)) supplied by a phase shifter 16, which is connected to the same oscillator 14. The phase .theta.(t) of the signals is now known in the case of a non-coherent demodulation, but its variation is slow compared with the binary flow rate of transmission. The modulated signals are respectively designated Re(r(t)) and Im(r(t)). These are the real and imaginary components of the complex envelope r(t) of the signal y(t). These signals are
REFERENCES:
patent: 4087752 (1978-05-01), Melvin
patent: 4087787 (1978-05-01), Acampora
Griffin Robert L.
Huseman M.
LandOfFree
Process for the noncoherent demodulation of a digital signal does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Process for the noncoherent demodulation of a digital signal, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process for the noncoherent demodulation of a digital signal will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1090747