Asynchronous digital correlator and demodulators including a correlator of this type

• US 4,933,952 A
• Filed: 04/04/1989
• Issued: 06/12/1990
• Est. Priority Date: 04/08/1988
• Status: Expired due to Fees

• Current Assignee: LMT Radio Professionnelle
• Original Assignee: LMT Radio Professionnelle
• Inventors: Soufflet, Didier, Albrieux, Vincent
• Primary Examiner(s): CANGIALOSI, SALVATORE A
• Application Number:
  US07/332,968
• Time in Patent Office: 434 Days
• Field of Search: 375/1, 380/34
• US Class Current: 375/150
• CPC Class Codes: G06F 17/15 Correlation function computation including computation of convolution operations arithmetic circuits for sum of products per se, e.g. multiply-accumulators G06F7/5443; digital filters, e.g. FIR, IIR, adaptive filters H03H17/00
• CPC Class Codes: H04B 1/70718 with asynchronous demodulation, i.e. not requiring code synchronisation
• CPC Class Codes: H04B 1/709 Correlator structure
• CPC Class Codes: H04J 13/0025 M-sequences
• CPC Class Codes: H04L 23/02 adapted for orthogonal signalling
• CPC Class Codes: H04L 27/2278 using correlation techniques, e.g. for spread spectrum signals

Abstract

A demodulator to demodulate a complex signal formed by two carriers, in quadrature, the spectrum of which is spread by a pseudo-random signal, and which are modulated by an M-ary, Walsh, code comprises:

a chain of registers to store 256 bits of a pseudo-random signal;

a multiplexer to temporally multiplex the processing of the two carriers;

a chain of correlation macro-cells having a systolic structure enabling temporal multiplexing and having intermediate outputs providing for partial correlation values on 16 chips;

a device for the computation of linear combinations of the functions of partial correlation, to compute two functions of correlation corresponding to two carriers for each M-ary code;

a demultiplexing and computation device to compute the module of the function of correlation for each of the Walsh codes which may modulate the signal to be demodulated;

a device for the selection of the greatest value of the correlation functions and for the selection of the corresponding piece of data.

Claims

1. An asynchronous, digital correlator to correlate a binary reference signal (REF) with a complex signal formed by a first carrier and second carrier (I, Q), modulated by a pseudo-random function and by an M-ary orthogonal encoding, said two carriers being sampled and digitized, said correlator comprising:

a multiplexer to transmit, alternately, a value (I) of the first carrier, and a value (Q) of the second carrier, with a period equal to half their sampling period;

a first digital filter adapted to the pseudo-random function (PN), comprising a chain of registers to store L consecutive bits of the reference signal (REF) and a chain of M correlation macro-cells receiving the sequence of multiplexed values of the first carrier and the second carrier (I and Q), each macro-cell computing the value (CPI0,..., CPI15) of a first function, called a partial correlation function, between N bits of the reference signal and N values of the first carrier (I), and then computing the value (CPQ0,..., CPQ15) of a second function, called a partial correlation function, between N bits of the reference signal and N values of the second carrier (Q), N being equal to L/M;

a second digital filter adapted to at least one of the M-ary encoding codes and having computation means to determine the value of at least one linear combination of the values of the first functions of partial correlation (CPI0,... CPI15); and

then to determine the value of a linear combination of the values of the second functions of partial correlation (CPQ0,..., CPQ15), the coefficients of the linear combination being unchanged and being equal to ±

1; and

having means to demultiplex the values of the two linear combinations (FIi, FQi), and to determine the modulus (Fi) of a vector having, as its components, the values of these two linear combinations (FIi, FQi).

2. A correlator according to claim 1, wherein each correlation macro-cell has a systolic structure comprising:

a plurality of rows of N correlation cells, each cell receiving a bit (X0,..., X3) of multiplexed values (I, Q) and retransmitting this bit to a neighbouring cell in the same row, with a delay equal to a sampling half-period;

receiving a bit (R15, R15,..., R0, R0) of the reference signal and retransmitting this bit to a homologous cell in the previous line;

giving, with a delay equal to one sampling half-period, a carry-over bit (Cout) to a homologous cell in the following row;

receiving a partial correlation result bit (Sin), given by the following cell in the same row;

giving, with a delay equal to a sampling half-period, a partial correlation result bit (Sout), to the cell that precedes it in the same row;

and wherein each row of N correlation cells further comprises;

a chain of registers to delay the partial correlation result bit (SO,..., S7) that it gives, with a delay depending on the rank of the row in the macro-cell;

a chain of registers to delay the bit (X0,..., X3) of the multiplexed values (I, Q) that it receives, by a delay which is a function of the rank of the row in the macro-cell.

3. A correlator according to claim 2, wherein each correlation cell has:

an exclusive-NOR gate, of which one input receives a bit (Xin) of a value (I, Q) of the multiplexed signals, and one input receives a bit (R) of the reference signal;

a flip-flop retransmitting said bit (Xin) of a value of the multiplexed signals, under the control of a clock signal (CKS) at a frequency which is twice the sampling frequency;

an adder, of which one input is connected to the output of the exclusive-NOR gate;

one output receives a partial correlation result bit (Sin) given by the following cell on the same line; and

one output receives a carry-over bit (Cin) given by the homologous cell in the previous line;

a flip-flop retransmitting a partial correlation result bit (Sout) given by a first output of the adder, under the control of the clock signal (CKS);

a flip-flop retransmitting a carry-over bit (Cout) given by a second output of the adder, under the control of the clock signal (CKS).

4. A digital demodulator to demodulate a complex signal consisting of a first carrier and a second carrier (I, Q), in quadrature, which has its spectrum spread by a pseudo-random function and which is modulated by an M-ary orthogonal encoding, according to a PSK or MSK or QPSK modulation;

wherein the correlator according to any one of the claims 1 to 3, receives a reference signal (REF) formed by a pseudo-random signal (PN) identical to the one that has been used to spread the spectrum and given by a pseudo-random signal generator;

and wherein the second filter has means;

to compute the values (FIn, FQn) of a first linear combination and a second linear combination of functions of partial correlation, to demultiplex these two values and to compute the modulus (Fn) of a vector having, as its components, these two values, for each of the M M-ary codes;

and further comprising means to determine the greatest modulus (Fj) from among these M moduli and to deduce, therefrom, that code (j) which modulates the two carriers (I, Q).

5. A digital demodulator to demodulate a complex signal formed by a first carrier and a second carrier, in quadrature, the spectrum of which is spread by a pseudo-random function, and which are modulated by an M-ary orthogonal encoding, according to a PSK or MSK or QPSK modulation;

comprising;

a correlator according to any one of the claims 1 to 3;

a generator giving a pseudo-random, binary signal identical to the one that has been used to spread the spectrum;

a generator of M-ary orthogonal codes, giving a binary signal according to one of these codes;

an exclusive-OR gate to give the correlator a reference signal (REF'"'"') formed by the product of the binary signals (PN, Wi) respectively given by the two generators;

and wherein the second filter of the correlator has means to compute the values (FIi, FQi) of a first linear combination and a second linear combination of functions of partial correlation, to demultiplex these two values and to compute the modulus (Fn) of a vector having, as its components, these two values, for at least one of the M M-ary codes (Wi, Wh).

6. A demodulator according to claim 5, to demodulate a complex signal formed by a first carrier and a second carrier (I, Q), in quadrature, the spectrum of which is spread by a pseudo-random function, and which are modulated by an M-ary Walsh encoding, wherein the computing means to sucessively compute the value (FIi, FQi) of a first linear combination and a second linear combination of the functions of partial correlation (CPI0'"'"',..., CPI15'"'"', CPQ0'"'"',..., CPQ15'"'"') comprise:

a plurality of adders to add all the values (CPI0'"'"', CPI2'"'"',..., CPI14'"'"';

CPQ0'"'"', CPQ2'"'"', CPQ14'"'"') of the functions of partial correlation having even rank;

a plurality of adders to add all the values (CPI1'"'"', CPI3'"'"',..., CPI15'"'"';

CPQ1'"'"', CPQ3'"'"', CPQ15'"'"') of the functions of partial correlation having odd rank;

an adder to add the values given by these two pluralities of adders and thus successively give the value (FIi, FQi) of a first component and a second component of the function of correlation for a first Walsh code (Wi) which is the one given by the generator of codes;

a subtractor to subtract, from one another, the values given by the two pluralities of adders and thus successively give the value (FIh, FQh) of a first component and a second component of the function of correlation for a second Walsh code (Wh).