CA1269423A - Demodulating an angle-modulated signal - Google Patents
Demodulating an angle-modulated signalInfo
- Publication number
- CA1269423A CA1269423A CA000543461A CA543461A CA1269423A CA 1269423 A CA1269423 A CA 1269423A CA 000543461 A CA000543461 A CA 000543461A CA 543461 A CA543461 A CA 543461A CA 1269423 A CA1269423 A CA 1269423A
- Authority
- CA
- Canada
- Prior art keywords
- signal
- pair
- derived
- signals
- frequency
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Lifetime
Links
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03D—DEMODULATION OR TRANSFERENCE OF MODULATION FROM ONE CARRIER TO ANOTHER
- H03D3/00—Demodulation of angle-, frequency- or phase- modulated oscillations
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03D—DEMODULATION OR TRANSFERENCE OF MODULATION FROM ONE CARRIER TO ANOTHER
- H03D3/00—Demodulation of angle-, frequency- or phase- modulated oscillations
- H03D3/007—Demodulation of angle-, frequency- or phase- modulated oscillations by converting the oscillations into two quadrature related signals
- H03D3/008—Compensating DC offsets
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03D—DEMODULATION OR TRANSFERENCE OF MODULATION FROM ONE CARRIER TO ANOTHER
- H03D7/00—Transference of modulation from one carrier to another, e.g. frequency-changing
- H03D7/16—Multiple-frequency-changing
- H03D7/165—Multiple-frequency-changing at least two frequency changers being located in different paths, e.g. in two paths with carriers in quadrature
Abstract
ABSTRACT
DEMODULATING AN ANGLE-MODULATED SIGNAL
A zero-IF receiver for an FM signal comprises quadrature-mixers (2,3) to which the input signal is applied and which mix the received carrier down to zero frequency. The signals appearing at the mixer outputs (11,12) are fed to a differentiate, multiply and subtract demodulator (22,23,25,27,31) via capacitors (17,19) which block the d.c. components of the signals, thereby preventing amplifier saturation effects, etc This d.c.-blocking would, unless steps were taken to prevent it, result in distortion of the output signal appearing at an output terminal (32). This distortion is prevented by providing an amplitude divider (33) in the output signal path. The signal fed to the divider control input (48) is derived from the output signals of the blocking capacitors by squaring them by means of multipliers (37,40), adding the results together in an adder (43) and applying the adder output signal to the control input (48) both directly and via a low-pass filter (46) the cut-off frequency of which is equal to that of the high-pass filters (15,16) of which the blocking capacitors effectively form part.
DEMODULATING AN ANGLE-MODULATED SIGNAL
A zero-IF receiver for an FM signal comprises quadrature-mixers (2,3) to which the input signal is applied and which mix the received carrier down to zero frequency. The signals appearing at the mixer outputs (11,12) are fed to a differentiate, multiply and subtract demodulator (22,23,25,27,31) via capacitors (17,19) which block the d.c. components of the signals, thereby preventing amplifier saturation effects, etc This d.c.-blocking would, unless steps were taken to prevent it, result in distortion of the output signal appearing at an output terminal (32). This distortion is prevented by providing an amplitude divider (33) in the output signal path. The signal fed to the divider control input (48) is derived from the output signals of the blocking capacitors by squaring them by means of multipliers (37,40), adding the results together in an adder (43) and applying the adder output signal to the control input (48) both directly and via a low-pass filter (46) the cut-off frequency of which is equal to that of the high-pass filters (15,16) of which the blocking capacitors effectively form part.
Description
DESCRIPTION
DEMODULATING AN ANGl.E-MODULATED SIGNAL
Th;s ;nvention relates to a method of demodulat;ng an angle-modulated s;gnal, compr;sing quardrature-m;xing said angle-modulated s;gnal w;th a local osc;llator s;gnal to produce a pa;r of s;gnals in each of ~h;ch a frequency within the bandw;dth of the angle-modulated signal has been translated to zero, d;fferentiating with respect to time a signal der;ved from one s;gnal of the pair and multiplying the result by a signal der;ved from the other signal of the pair to produce a first aux;l;ary s;gnal, d;fferent;ating ~ith respect to time the signal der;ved from the other s;gnal of the pair and multiplying the result by the s;gnal der;ved from the one.s;gnal of the pair to produce a second auxiliary s;gnal, form;ng an output signal proportional to the d;fference between sa;d f;rst and second aux;l;ary s;gnals, and vary;ng the proport;onal;ty factor ;n dependence upon the sum of the squares of the ;nstantaneous amplitudes of sa;d signals der;ved from the one and the other s;gnal of the pa;r. The ;nvent;on also relates to a demodulator for perform;ng such a method.
A method of the above k;nd ;s known from an art;cle by J.K. Goatcher, M.W~ Neale and I.A.WA Vance entitled "Noise considerations in an integrated c;rcuit VHF radio receiver" in the Proceedings of the IERE Clerk Maxwell Commemorative Conference on Radio Receivers and Associated Systems (IERE
Proceed;ngs No. 50), University of Leeds, 7th-9th July 1981, pages 49-51.
Demodulating an FM s;gnal by quadrature-m;x;ng it with a local oscillator signal to translate a frequency within ;ts bandwidth to zero, d;fferentiating each member of the resulting pair of signals with respect to t;me and multiply;ng the result by the other member of the pa;r, and subtracting the resuLts of the mul~iplications once from the other to produce an output signal is kno~n from Patent Specification G~-A-1530602.
Unfortunately the output signal obtained by means o~ ~his basic ~;9~3
DEMODULATING AN ANGl.E-MODULATED SIGNAL
Th;s ;nvention relates to a method of demodulat;ng an angle-modulated s;gnal, compr;sing quardrature-m;xing said angle-modulated s;gnal w;th a local osc;llator s;gnal to produce a pa;r of s;gnals in each of ~h;ch a frequency within the bandw;dth of the angle-modulated signal has been translated to zero, d;fferentiating with respect to time a signal der;ved from one s;gnal of the pair and multiplying the result by a signal der;ved from the other signal of the pair to produce a first aux;l;ary s;gnal, d;fferent;ating ~ith respect to time the signal der;ved from the other s;gnal of the pair and multiplying the result by the s;gnal der;ved from the one.s;gnal of the pair to produce a second auxiliary s;gnal, form;ng an output signal proportional to the d;fference between sa;d f;rst and second aux;l;ary s;gnals, and vary;ng the proport;onal;ty factor ;n dependence upon the sum of the squares of the ;nstantaneous amplitudes of sa;d signals der;ved from the one and the other s;gnal of the pa;r. The ;nvent;on also relates to a demodulator for perform;ng such a method.
A method of the above k;nd ;s known from an art;cle by J.K. Goatcher, M.W~ Neale and I.A.WA Vance entitled "Noise considerations in an integrated c;rcuit VHF radio receiver" in the Proceedings of the IERE Clerk Maxwell Commemorative Conference on Radio Receivers and Associated Systems (IERE
Proceed;ngs No. 50), University of Leeds, 7th-9th July 1981, pages 49-51.
Demodulating an FM s;gnal by quadrature-m;x;ng it with a local oscillator signal to translate a frequency within ;ts bandwidth to zero, d;fferentiating each member of the resulting pair of signals with respect to t;me and multiply;ng the result by the other member of the pa;r, and subtracting the resuLts of the mul~iplications once from the other to produce an output signal is kno~n from Patent Specification G~-A-1530602.
Unfortunately the output signal obtained by means o~ ~his basic ~;9~3
2 20104-8348 method has a square-law dependence on the levels of ~he signals of the pair. Accordingly, in the method known from the above-mentioned article the two signals of the pair are additionally squared, the resul~s added together, and the signal resulting from thls addition is fed to the control input of an amplitude divider which operakes on the difference ~etween the auxiliary signals.
Receivers in which an FM signal is quadrature-mixed to translate a frequency within its handwidth to æero suffer from the disadvantage that the resulting d.c. component in each signal of the pair of signals obtained by means of the mixing operation cannot be distinguished from unwanted d.c. offsets which inevitably exist at the output of each mixer. During reception of the smallest detectable signal these offsets are typically 60dB
larger than the wanted d.c. components. Moreover, each signal of the pair has to be amplified by, typically, 90dB, without limiting, befoxe it is processed further. This is impracticable, using circuits operating from reasonable supply voltages, unless the d.c. components (both wanted and unwanted) are eliminated by suitable filtering, for example by the insertion of d.c.-blocking series capacitors or equivalent circuitry (see for example UK
Patent Applica~ion 8512491 which was published on November 26, 1986 under number GB-A-2175473 in ~he pa~h for each signal of the pair. The cut-off frequency of such a filter cannot be made arbitrarily small, since, at least in the case of portable or mobile receivers, the d.c. signals being eliminated cannot be regarded as time-invariant. Unfortunately the provision of such filters creates a hole in the effective input pass-band of the receiver, and this results in distorted recovery of the original -modulation by the method described in the aforesaid patent specification GB-A-1530602, and indeed by any other known method.
Employing, in addition, the simple square, add and divide mechanism described in the aforesaid article by Goatcher, Neale and Vance is found to make matters subjectively even worse, at least when the modulation is speech.
PH~ 33287 1 It is an object of the present invention to enable a cons;derable ;mprovement to be obtained ;n th;s respect.
Accord;ng to one aspect the ;nvention prov;des a method of demodulat;ng an angle-modulated s;gnal, comprising quardrature-mixing said angle-modulated signal with a local osc;llator s;gnal to produce a pa;r of s;gnals ;n each of wh;ch a frequency w;thin the bandwidth of the angle-modulated signal has been translated to zero, different;at;ng w;th respect to t;me a s;gnal der;ved from one s;gnal of the pa;r and mult;ply;ng the result by a signal der;ved from the other s;gnal of the pa;r to produce a first aux;liary s;gnal, d;fferent;at;ng with respect to time the signal der;ved from the other s;gnal of the pa;r and multiplying the result by the signal deriyed from the one signal of the pa;r to produce a second aux;l;ary s;gnal~ form;ng an output s;gnal proport;onal to the d;fference between sa;d f;rst and second aux;l;ary s;gnals, and vary;ng the proport;onal;ty factor ;n dependence upon the sum of the squares of the instantaneous amplitudes of sa;d s;gnals der;ved from the one and the other signal of the pa;r, characterised ;n that each of the s;gnals derived from the one and the other signal of the pair ;s a f;ltered vers;on of the correspond;ng s;gnal of the pa;r ;n ~h;ch any d.c~ component of the corresponding s;gnal of the pa;r has been removed and ;n that the proport;onal;ty factor ;s var;ed ;n ;nverse relat;onsh;p to an add;t;ve comb;nat;on of the squares of sa;d ;nstantaneous ampl;tudes and a low-pass filtered version of the squares of said ;nstantaneous ampl;tudes.
It has now been found that the d;stortion-produc;ng effect of the insert;on of d.c.-blocking means between the quadrature m;xers of the known demodulator and the d;fferent;ating and multiplying means can be very substantially reduced by varying the proportionality factor between the output signal and the difference bet~een the first and second auxiliary signal in inverse relationship to an additive combination of the sum of the squares of the said instantaneous amplitudes and a low-pass
Receivers in which an FM signal is quadrature-mixed to translate a frequency within its handwidth to æero suffer from the disadvantage that the resulting d.c. component in each signal of the pair of signals obtained by means of the mixing operation cannot be distinguished from unwanted d.c. offsets which inevitably exist at the output of each mixer. During reception of the smallest detectable signal these offsets are typically 60dB
larger than the wanted d.c. components. Moreover, each signal of the pair has to be amplified by, typically, 90dB, without limiting, befoxe it is processed further. This is impracticable, using circuits operating from reasonable supply voltages, unless the d.c. components (both wanted and unwanted) are eliminated by suitable filtering, for example by the insertion of d.c.-blocking series capacitors or equivalent circuitry (see for example UK
Patent Applica~ion 8512491 which was published on November 26, 1986 under number GB-A-2175473 in ~he pa~h for each signal of the pair. The cut-off frequency of such a filter cannot be made arbitrarily small, since, at least in the case of portable or mobile receivers, the d.c. signals being eliminated cannot be regarded as time-invariant. Unfortunately the provision of such filters creates a hole in the effective input pass-band of the receiver, and this results in distorted recovery of the original -modulation by the method described in the aforesaid patent specification GB-A-1530602, and indeed by any other known method.
Employing, in addition, the simple square, add and divide mechanism described in the aforesaid article by Goatcher, Neale and Vance is found to make matters subjectively even worse, at least when the modulation is speech.
PH~ 33287 1 It is an object of the present invention to enable a cons;derable ;mprovement to be obtained ;n th;s respect.
Accord;ng to one aspect the ;nvention prov;des a method of demodulat;ng an angle-modulated s;gnal, comprising quardrature-mixing said angle-modulated signal with a local osc;llator s;gnal to produce a pa;r of s;gnals ;n each of wh;ch a frequency w;thin the bandwidth of the angle-modulated signal has been translated to zero, different;at;ng w;th respect to t;me a s;gnal der;ved from one s;gnal of the pa;r and mult;ply;ng the result by a signal der;ved from the other s;gnal of the pa;r to produce a first aux;liary s;gnal, d;fferent;at;ng with respect to time the signal der;ved from the other s;gnal of the pa;r and multiplying the result by the signal deriyed from the one signal of the pa;r to produce a second aux;l;ary s;gnal~ form;ng an output s;gnal proport;onal to the d;fference between sa;d f;rst and second aux;l;ary s;gnals, and vary;ng the proport;onal;ty factor ;n dependence upon the sum of the squares of the instantaneous amplitudes of sa;d s;gnals der;ved from the one and the other signal of the pa;r, characterised ;n that each of the s;gnals derived from the one and the other signal of the pair ;s a f;ltered vers;on of the correspond;ng s;gnal of the pa;r ;n ~h;ch any d.c~ component of the corresponding s;gnal of the pa;r has been removed and ;n that the proport;onal;ty factor ;s var;ed ;n ;nverse relat;onsh;p to an add;t;ve comb;nat;on of the squares of sa;d ;nstantaneous ampl;tudes and a low-pass filtered version of the squares of said ;nstantaneous ampl;tudes.
It has now been found that the d;stortion-produc;ng effect of the insert;on of d.c.-blocking means between the quadrature m;xers of the known demodulator and the d;fferent;ating and multiplying means can be very substantially reduced by varying the proportionality factor between the output signal and the difference bet~een the first and second auxiliary signal in inverse relationship to an additive combination of the sum of the squares of the said instantaneous amplitudes and a low-pass
3 f;ltered vers;on of the sum of the squares of the sa;d 342~
1 instantaneous ampl;tudes. It ;s thought that varying the proport;onal;ty factor ;n th;s way may result ;n effect;vely re;nsert;ng ;n the input s;gnals to the d;fferentiat;ng and mult;ply;ng means their average values which have been Lost by their transmission through the d~c. blocking means, the said low-pass filtered vers;on preferably having a cut-off frequency wh;ch ;s substant;ally equal to the low-frequency cut-off po;nt of each f;ltered vers;on of the corresponding signal of the pair.
AGcording to another aspect the ;nvention provides a demodulator for an angle-modulated signal, compris;ng an input for said signal, a local oscillator, quadrature-reLated mixing means to ;nputs of which said input and said local oscillator are coupled, for producing a pa;r of s;gnals in each of which a frequency w;th;n the bandw;dth of an angle-modulated signal applied in operation to said ;nput has been translated to zero, d;fferent;at;ng and mult;pl;er means for differentiating with respect to time a signal derived from one s;gnal of the pa;r and multiply;ng the result by a s;gnal der;ved from the other signal of the pa;r to produce a f;rst aux;liary signal, different;ating and mult;pl;er means for d;fferentiat;ng ~;th respect to time a s;gnal der;ved from the other s;gnal of the pair and multiplying the result by a s;gnal der;ved from the one s;gnal of the pa;r to produce a second auxiliary s;gnal, means for forming an output s;gnal proport;onal to the d;fference between said first and second aux;l;ary s;gnals, and further means for varying the proportionality factor in dependence upon the sum of the squares of the instantaneous amplitudes of sa;d s;gnals der;ved from the one and the other s;gnal of the pair~ characterised in that filter means are provided for form;ng each of the signals derived from the one and the other signal of the pa;r as a f;ltered version of the corresponding signal of the pa;r in ~hich any d.c. component of the corresponding s;gnal of the pa;r has been removed, and ;n that sa;d further means ;s arranged to vary the proport;onality factor ;n inverse relat;onship to an additive combination of the squares of said instantaneous ampl;~udes and a ~ 4~3 PHB 33287 1 low-pass f;ltered vers;on of the squares of sa;d instantaneous ampl;tudes.
An embodiment of the invent;on and some possible mod;f;cat;ons thereto w;ll now be descr;bed, by way of example, w;th reference to the accompany;ng d;agrammat;c draw;ng the s;ngle Figure of wh;ch ;s a d;agram of a demodulator for an FM
s;gnal.
In the drawing a demodulator for an FM s;gnal of nom;nal carr;er frequency fc rece;ved on an ;nput 1, for exampLe an aer;al, compr;ses quadrature related mix;ng means ;n the form of a pa;r of m;xers 2 and 3 and a 90 clegree phase-sh;fter 4O The ;nput 1 is coupled v;a a band-pass anti-harmon;c f;lter 5 ~hose pass-band is centred on the frequency fc to a f;rst ;nput 6 of m;xer 2 directly and to a first input 7 of mixer 3 via the phase-shifter 4~ Second ;nputs 8 and 9 of the m;xers 2 and 3 are fed w;th the output s;gnal of a local osc;llator 10, wh;ch generates an output frequency fc~ ~h;ch output s;gnal is non-coherent w;th the input signal applied to input 1. The resulting frequency-difference signals appearing at the mixer outputs 11 and 12 (in which signals the nominal carrier frequency fc hag been translated to substantially zero) are fed via low-pass adjacent-channel selectivity filters 13 and 14 respectively to d.c~-blocking filters 15 and 16 respectively, ~hich are shown diagrammatically as compris;ng a ser;es capacitor 17 and a parallel resistor 18, and a series capacitor 19 and a parallel resistor 20, respect;vely. In pract;ce the f;lters 15 and 16 (wh;ch could alternatively be ;ncluded ;n the ;nput s;gnal paths to the f;lters 13 and 14 respect;vely) may take the form of series capacitors or their equivalent (see for example the a~oresaid ~ r~in~ UK Patent Appl;cat;on 8512491) bet~een respect;ve pa;rs of stages of a cascade combinat;on of ampl;f;er stages (not shown).
The output s;gnal of f;lter 15 ;s fed to a first ;nput 21 of a mult;pl;er stage 22 via a d;fferent;ating circuit 23, and to a
1 instantaneous ampl;tudes. It ;s thought that varying the proport;onal;ty factor ;n th;s way may result ;n effect;vely re;nsert;ng ;n the input s;gnals to the d;fferentiat;ng and mult;ply;ng means their average values which have been Lost by their transmission through the d~c. blocking means, the said low-pass filtered vers;on preferably having a cut-off frequency wh;ch ;s substant;ally equal to the low-frequency cut-off po;nt of each f;ltered vers;on of the corresponding signal of the pair.
AGcording to another aspect the ;nvention provides a demodulator for an angle-modulated signal, compris;ng an input for said signal, a local oscillator, quadrature-reLated mixing means to ;nputs of which said input and said local oscillator are coupled, for producing a pa;r of s;gnals in each of which a frequency w;th;n the bandw;dth of an angle-modulated signal applied in operation to said ;nput has been translated to zero, d;fferent;at;ng and mult;pl;er means for differentiating with respect to time a signal derived from one s;gnal of the pa;r and multiply;ng the result by a s;gnal der;ved from the other signal of the pa;r to produce a f;rst aux;liary signal, different;ating and mult;pl;er means for d;fferentiat;ng ~;th respect to time a s;gnal der;ved from the other s;gnal of the pair and multiplying the result by a s;gnal der;ved from the one s;gnal of the pa;r to produce a second auxiliary s;gnal, means for forming an output s;gnal proport;onal to the d;fference between said first and second aux;l;ary s;gnals, and further means for varying the proportionality factor in dependence upon the sum of the squares of the instantaneous amplitudes of sa;d s;gnals der;ved from the one and the other s;gnal of the pair~ characterised in that filter means are provided for form;ng each of the signals derived from the one and the other signal of the pa;r as a f;ltered version of the corresponding signal of the pa;r in ~hich any d.c. component of the corresponding s;gnal of the pa;r has been removed, and ;n that sa;d further means ;s arranged to vary the proport;onality factor ;n inverse relat;onship to an additive combination of the squares of said instantaneous ampl;~udes and a ~ 4~3 PHB 33287 1 low-pass f;ltered vers;on of the squares of sa;d instantaneous ampl;tudes.
An embodiment of the invent;on and some possible mod;f;cat;ons thereto w;ll now be descr;bed, by way of example, w;th reference to the accompany;ng d;agrammat;c draw;ng the s;ngle Figure of wh;ch ;s a d;agram of a demodulator for an FM
s;gnal.
In the drawing a demodulator for an FM s;gnal of nom;nal carr;er frequency fc rece;ved on an ;nput 1, for exampLe an aer;al, compr;ses quadrature related mix;ng means ;n the form of a pa;r of m;xers 2 and 3 and a 90 clegree phase-sh;fter 4O The ;nput 1 is coupled v;a a band-pass anti-harmon;c f;lter 5 ~hose pass-band is centred on the frequency fc to a f;rst ;nput 6 of m;xer 2 directly and to a first input 7 of mixer 3 via the phase-shifter 4~ Second ;nputs 8 and 9 of the m;xers 2 and 3 are fed w;th the output s;gnal of a local osc;llator 10, wh;ch generates an output frequency fc~ ~h;ch output s;gnal is non-coherent w;th the input signal applied to input 1. The resulting frequency-difference signals appearing at the mixer outputs 11 and 12 (in which signals the nominal carrier frequency fc hag been translated to substantially zero) are fed via low-pass adjacent-channel selectivity filters 13 and 14 respectively to d.c~-blocking filters 15 and 16 respectively, ~hich are shown diagrammatically as compris;ng a ser;es capacitor 17 and a parallel resistor 18, and a series capacitor 19 and a parallel resistor 20, respect;vely. In pract;ce the f;lters 15 and 16 (wh;ch could alternatively be ;ncluded ;n the ;nput s;gnal paths to the f;lters 13 and 14 respect;vely) may take the form of series capacitors or their equivalent (see for example the a~oresaid ~ r~in~ UK Patent Appl;cat;on 8512491) bet~een respect;ve pa;rs of stages of a cascade combinat;on of ampl;f;er stages (not shown).
The output s;gnal of f;lter 15 ;s fed to a first ;nput 21 of a mult;pl;er stage 22 via a d;fferent;ating circuit 23, and to a
4~:~
PHs 33287 second ;nput 24 of a multiplier stage 25 directly. Sim;larly, the output s;gnal of f;lter 16 ;s fed to a first input 26 of the mult;plier stage 25 via a differentiating circuit 27 and to a second input 28 of the multiplier stage 22 directly. The differentiat;ng circuits may be, for example, s;mple ser;es-capac;tor parallel-resistor circuits hav;ng time-constants shorter than the reciprocal of the highest frequency expected at the outputs of the filters 15 and 16, or more complicated circu;ts~ for e~ample act;ve d;fferentiating circu;ts, hav;ng the same property. The signals appear;ng at the outputs 29 and 30 of the multiplier stages 22 and 25 are subtracted one from the other in a subtractor 31, for example an amplifier having inverting and non inverting ;nputs, and the result;ng di.fference signal ;s appl;ed to an output term;nal 32 v;a an ampl;tude d;v;der or controllable ga;n stage 33, for example of-the k;nd obta;nable from Analog Dev;ces Inr. under the type number AD535, and a low-pass post-filter 34. The signal appear;ng at output 32 ;s a demodulated vers;on of the ;nput s;gnal appl;ed to ;nput 1.
The output s;gnals of the filters 15 and 16 are also applied to both ;nputs 35 and 36 of a multiplier 37 and both inputs 38 and 39 of a mult;plier 40, respectively~ Mult;pliers 37 and 40 therefore produce at the;r outputs 41 and 42 signals representat;ve of the squares of the ;nstantaneous ampl;tudes of the output s;gnals of f;lters 15 and 16 respect;vely. These squared s;gnals are added together ;n a s;gnal adder 43 and the result ;s fed to a further signal adder 44, both directly to an ;nput 45 thereof and also v;a a lo~-pass f;lter 46 to another ;nput 47 thereof~ The output of adder 44 ;s fed to the d;v;s;on factor control input 48 of the ampl;tude d;v;der 33. The proport;onal;ty factor between the s;gnal fed to the output term;nal 32 and the output s;gnal of signal subtractor or d;f~erencer 31 ;s thus made to vary inversely ~;th the sum of the squares o, the ;nstantaneous amplitudes of the s;gnals appearing at the outputs of the filters 15 and 16 and a lo~-pass filtered ~ 3 PHB 33287 or averaged vers;on of the sum of the squares of said instantaneous ampLitudes, the low-pass filtering being carried out by means of filter 46 the cut-off frequency of which is preferably substantially equal to the low-frequency cut-off points of the f;lters 15 and 16.
If filters 15 and 16 were replaced by direct connections and filter 46 were replaced by an open circuit~ the circuit shown would correspond with that described in the article by Goatcher, Neale and Vance, with all the problems of coping w;th d~cn components in the s;gnal paths from filters 13 and 14 to the components 22~23,25 and 27 that that ~ould entail. However, as pointed out above, mere ;nclus;on of the filters 15 and 16 to overcome these problems would result in c~nsiderable distortion of the signal appearing at output 32. It has surprisingly been found that inclusion, in addition, of fil~er 4O results in this distortion being reduced to a very low level, indeed a level cons;derably below that wh;ch would be obtained if the feed-forward loop comprising the components 33,37,40,43,44 and 46 were omitted completelyn It will be appreciated that many modifications are possible to the circuit shown. For example the amplitude divider 33 in the output signal path from differencer 31 may be replaced by a respective similarly-controlled amplitude d;v;der ;n the output s;gnal path from multipl;er 22 and the output signal path from multipler 25. As another example it is not essential to provide two parallel signal paths from the output of adder 43 to the control input 48 of divider 33. All that is required is that the transmission of the path from adder 43 to control input 48 should correspond to that g;ven by the parallel path arrangement shown, i.e. at relatively low frequencies be double ~hat it is at relatively high frequencies; such a transmiss;on character;st;c could be obta;ned by means of a s;ngle filter w;thout a path circumvent;ng ;t tor by a respective such f;lter hav;ng the same characterist;cs at the output of each multiplier 37 and 40, the components 44-47 being replaced by a direct connection from the ~2~ 3 PHB 33287 1 output of adder 43 to control ;nput 48 ;n th;s case). The actual ga;n or lo5s in this filter is, in principle, immaterial. As yet another example the take-off points to the multipliers 37 and 40 from the s;gnal channels to the components 22,23,25,27 could be moved further back along these channels towards the s;gnal input 1, provided that dupl;cates of those components ;n the two channels ~hich are no longer present in the ;nput paths to the mult;pl;ers are then themselves prov;ded in the input paths to the multipliers. As is kno~n, the phase-shifter 4 may be provided instead in the signal path from local oscillator 10 to either the ;nput 8 of m;xer 2 or the ;nput 9 of m;xer 3.
Although a hard-~;red analog circu;t has been descr;bed it w;ll be ev;dent that some or all of the f~nct;ons performed thereby may be carr;ed out by means of d;g;tal components such as dig;tal filters and/or a su;tably programmed microcomputer, provided that the relevant s;gnals are first converted to d;gital form. Thus, for example, the funct;ons performed by the arrangement compr;s;ng the components 22,23,25,27,31,33,37,40 43,46 and 44 may be performed by a su;tably programmed m;crocomputer.
In a practical example a circuit of the kind described had an effective input bandw;dth of 15kHz and an ;nput s;gnal had 3kHz peak dev;at;on. The cut-off frequency of the f;lters 15 and 16 ~as 500Hz, as ~as the cut-off frequency of f;lter 46. F;lter 46 ~as a simple f;rst-order filter. In spite of the high cut-off frequency of the filters 15 and 16 the output signal at output 32 ~as found to conta;n very l;ttle distortion ;ndeed. Obviously the cut-off frequenc;es of the f;lters 15 and 16 must not be so h;gh that the s;gnal frequenc;es appl;ed to these f;lters never exceed them.
Although ;n the embod;ment descr;bed the output frequency of osc;llator 10 ;s equal to the nominal carrier frequency fc of the FM ;nput s;gnal, th;s is not essent;al. The demodulator may alternat;vely be of the so-called "offset" type ;n ~h;ch the osc;llator output frequency ;s different from fc but still equal ~ 4~ PHB 33287 to a frequency within the bandwidth of the input signal, so that this latter frequency in the input signal is translated to zero by means of the mixers 2 and 3.
It ~ill be evident that by simply prov;d;ng an integrator at the output 32 the demodulator descr;bed may be made su;table for demodulat;ng a phase rather than frequency modulated ;nput s;gnal ;f desired.
PHs 33287 second ;nput 24 of a multiplier stage 25 directly. Sim;larly, the output s;gnal of f;lter 16 ;s fed to a first input 26 of the mult;plier stage 25 via a differentiating circuit 27 and to a second input 28 of the multiplier stage 22 directly. The differentiat;ng circuits may be, for example, s;mple ser;es-capac;tor parallel-resistor circuits hav;ng time-constants shorter than the reciprocal of the highest frequency expected at the outputs of the filters 15 and 16, or more complicated circu;ts~ for e~ample act;ve d;fferentiating circu;ts, hav;ng the same property. The signals appear;ng at the outputs 29 and 30 of the multiplier stages 22 and 25 are subtracted one from the other in a subtractor 31, for example an amplifier having inverting and non inverting ;nputs, and the result;ng di.fference signal ;s appl;ed to an output term;nal 32 v;a an ampl;tude d;v;der or controllable ga;n stage 33, for example of-the k;nd obta;nable from Analog Dev;ces Inr. under the type number AD535, and a low-pass post-filter 34. The signal appear;ng at output 32 ;s a demodulated vers;on of the ;nput s;gnal appl;ed to ;nput 1.
The output s;gnals of the filters 15 and 16 are also applied to both ;nputs 35 and 36 of a multiplier 37 and both inputs 38 and 39 of a mult;plier 40, respectively~ Mult;pliers 37 and 40 therefore produce at the;r outputs 41 and 42 signals representat;ve of the squares of the ;nstantaneous ampl;tudes of the output s;gnals of f;lters 15 and 16 respect;vely. These squared s;gnals are added together ;n a s;gnal adder 43 and the result ;s fed to a further signal adder 44, both directly to an ;nput 45 thereof and also v;a a lo~-pass f;lter 46 to another ;nput 47 thereof~ The output of adder 44 ;s fed to the d;v;s;on factor control input 48 of the ampl;tude d;v;der 33. The proport;onal;ty factor between the s;gnal fed to the output term;nal 32 and the output s;gnal of signal subtractor or d;f~erencer 31 ;s thus made to vary inversely ~;th the sum of the squares o, the ;nstantaneous amplitudes of the s;gnals appearing at the outputs of the filters 15 and 16 and a lo~-pass filtered ~ 3 PHB 33287 or averaged vers;on of the sum of the squares of said instantaneous ampLitudes, the low-pass filtering being carried out by means of filter 46 the cut-off frequency of which is preferably substantially equal to the low-frequency cut-off points of the f;lters 15 and 16.
If filters 15 and 16 were replaced by direct connections and filter 46 were replaced by an open circuit~ the circuit shown would correspond with that described in the article by Goatcher, Neale and Vance, with all the problems of coping w;th d~cn components in the s;gnal paths from filters 13 and 14 to the components 22~23,25 and 27 that that ~ould entail. However, as pointed out above, mere ;nclus;on of the filters 15 and 16 to overcome these problems would result in c~nsiderable distortion of the signal appearing at output 32. It has surprisingly been found that inclusion, in addition, of fil~er 4O results in this distortion being reduced to a very low level, indeed a level cons;derably below that wh;ch would be obtained if the feed-forward loop comprising the components 33,37,40,43,44 and 46 were omitted completelyn It will be appreciated that many modifications are possible to the circuit shown. For example the amplitude divider 33 in the output signal path from differencer 31 may be replaced by a respective similarly-controlled amplitude d;v;der ;n the output s;gnal path from multipl;er 22 and the output signal path from multipler 25. As another example it is not essential to provide two parallel signal paths from the output of adder 43 to the control input 48 of divider 33. All that is required is that the transmission of the path from adder 43 to control input 48 should correspond to that g;ven by the parallel path arrangement shown, i.e. at relatively low frequencies be double ~hat it is at relatively high frequencies; such a transmiss;on character;st;c could be obta;ned by means of a s;ngle filter w;thout a path circumvent;ng ;t tor by a respective such f;lter hav;ng the same characterist;cs at the output of each multiplier 37 and 40, the components 44-47 being replaced by a direct connection from the ~2~ 3 PHB 33287 1 output of adder 43 to control ;nput 48 ;n th;s case). The actual ga;n or lo5s in this filter is, in principle, immaterial. As yet another example the take-off points to the multipliers 37 and 40 from the s;gnal channels to the components 22,23,25,27 could be moved further back along these channels towards the s;gnal input 1, provided that dupl;cates of those components ;n the two channels ~hich are no longer present in the ;nput paths to the mult;pl;ers are then themselves prov;ded in the input paths to the multipliers. As is kno~n, the phase-shifter 4 may be provided instead in the signal path from local oscillator 10 to either the ;nput 8 of m;xer 2 or the ;nput 9 of m;xer 3.
Although a hard-~;red analog circu;t has been descr;bed it w;ll be ev;dent that some or all of the f~nct;ons performed thereby may be carr;ed out by means of d;g;tal components such as dig;tal filters and/or a su;tably programmed microcomputer, provided that the relevant s;gnals are first converted to d;gital form. Thus, for example, the funct;ons performed by the arrangement compr;s;ng the components 22,23,25,27,31,33,37,40 43,46 and 44 may be performed by a su;tably programmed m;crocomputer.
In a practical example a circuit of the kind described had an effective input bandw;dth of 15kHz and an ;nput s;gnal had 3kHz peak dev;at;on. The cut-off frequency of the f;lters 15 and 16 ~as 500Hz, as ~as the cut-off frequency of f;lter 46. F;lter 46 ~as a simple f;rst-order filter. In spite of the high cut-off frequency of the filters 15 and 16 the output signal at output 32 ~as found to conta;n very l;ttle distortion ;ndeed. Obviously the cut-off frequenc;es of the f;lters 15 and 16 must not be so h;gh that the s;gnal frequenc;es appl;ed to these f;lters never exceed them.
Although ;n the embod;ment descr;bed the output frequency of osc;llator 10 ;s equal to the nominal carrier frequency fc of the FM ;nput s;gnal, th;s is not essent;al. The demodulator may alternat;vely be of the so-called "offset" type ;n ~h;ch the osc;llator output frequency ;s different from fc but still equal ~ 4~ PHB 33287 to a frequency within the bandwidth of the input signal, so that this latter frequency in the input signal is translated to zero by means of the mixers 2 and 3.
It ~ill be evident that by simply prov;d;ng an integrator at the output 32 the demodulator descr;bed may be made su;table for demodulat;ng a phase rather than frequency modulated ;nput s;gnal ;f desired.
Claims (5)
OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:
1. A method of demodulating an angle-modulated signal, comprising quardrature-mixing said angle-modulated signal with a local oscillator signal to produce a pair of signals in each of which a frequency within the bandwidth of the angle-modulated signal has been translated to zero, differentiating with respect to time a signal derived from one signal of the pair and multiplying the result by a signal derived from the other signal of the pair to produce a first auxiliary signal, differentiating with respect to time the signal derived from the other signal of the pair and multiplying the result by the signal derived from the one signal of the pair to produce a second auxiliary signal, forming an output signal proportional to the difference between said first and second auxiliary signals, and varying the proportionality factor in dependence upon the sum of the squares of the instantaneous amplitudes of said signals derived from the one and the other signal of the pair, characterised in that each of the signals derived from the one and the other signal of the pair is a filtered version of the corresponding signal of the pair in which any d.c. component of the corresponding signal of the pair has been removed and in that the proportionality factor is varied in inverse relationship to an additive combination of the squares of said instantaneous amplitudes and a low-pass filtered version of the squares of said instantaneous amplitudes.
2. A method as claimed in Claim 1, wherein said low-pass filtered version has a cut-off frequency which is substantially equal to the low-frequency cut-off point of each said filtered version of the corresponding signal of the pair.
3. A demodulator for an angle-modulated signal, comprising an input for said signal, a local oscillator, quadrature-related mixing means to inputs of which said input and said local oscillator are coupled, for producing a pair of signals in each of which a frequency within the bandwidth of an angle-modulated signal applied in operation to said input has been translated to zero, differentiating and multiplier means for differentiating with respect to time a signal derived from one signal of the pair and multiplying the result by a signal derived from the other signal of the pair to produce a first auxiliary signal, differ-entiating and multiplier means for differentiating with respect to time a signal derived from the other signal of the pair and multi-plying the result by a signal derived from the one signal of the pair to produce a second auxiliary signal, means for forming an output signal proportional to the difference between said first and second auxiliary signals, and further means for varying the proportionality factor in dependence upon the sum of the squares of the instantaneous amplitudes of said signals derived from the one and the other signal of the pair, characterised in that filter means are provided for forming each of the signals derived from the one and the other signal of the pair as a filtered version of the corresponding signal of the pair in which any d.c. component of the corresponding signal of the pair has been removed, and in that said further means is arranged to vary the proportionality factor in inverse relationship to an additive combination of the squares of said instantaneous amplitudes and a low-pass filtered version of the squares of said instantaneous amplitudes.
4. A demodulator as claimed in Claim 3, wherein said further means comprises low-pass filter means, adder means to a first input of which the output of the low-pass filter means is coupled, means for applying a signal proportional to said sum of the squares to both the input of the low-pass filter means and to a second input of the adder means, and amplitude divider means for varying the amplitude of said output signal in inverse proportion to the output signal of said adder means.
5. A demodulator as claimed in Claim 4, wherein the cut-off frequency of the low-pass filter means is substantially equal to the low-frequency cut-off point of the filter means.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| GB08618855A GB2193405A (en) | 1986-08-01 | 1986-08-01 | Demodulating an angle-modulated signal |
| GB8618855 | 1986-08-01 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA1269423A true CA1269423A (en) | 1990-05-22 |
Family
ID=10602104
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA000543461A Expired - Lifetime CA1269423A (en) | 1986-08-01 | 1987-07-30 | Demodulating an angle-modulated signal |
Country Status (9)
| Country | Link |
|---|---|
| US (1) | US4766392A (en) |
| EP (1) | EP0255175A3 (en) |
| JP (1) | JPS6386905A (en) |
| KR (1) | KR880003479A (en) |
| AU (1) | AU586394B2 (en) |
| CA (1) | CA1269423A (en) |
| DK (1) | DK395387A (en) |
| FI (1) | FI873314A (en) |
| GB (1) | GB2193405A (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2210742A (en) * | 1987-10-05 | 1989-06-14 | Philips Nv | Frequency difference detector (fdd) and a carrier modulated receiver including such a fdd |
| GB2215545A (en) * | 1988-03-16 | 1989-09-20 | Philips Electronic Associated | A direct-conversion receiver |
| CH676404A5 (en) * | 1988-11-22 | 1991-01-15 | Ascom Radiocom Ag | |
| US5241702A (en) * | 1990-09-06 | 1993-08-31 | Telefonaktiebolaget L M Ericsson | D.c. offset compensation in a radio receiver |
| US5249203A (en) * | 1991-02-25 | 1993-09-28 | Rockwell International Corporation | Phase and gain error control system for use in an i/q direct conversion receiver |
| JPH04310037A (en) * | 1991-04-09 | 1992-11-02 | Nec Corp | Fsk receiver |
| JPH05175733A (en) * | 1991-12-20 | 1993-07-13 | Kenwood Corp | Fm demodulator |
| DE4236546C1 (en) * | 1992-10-29 | 1994-05-05 | Hagenuk Telecom Gmbh | Homodyne receiver and direct conversion method |
| US6127884A (en) * | 1999-05-24 | 2000-10-03 | Philips Electronics North America Corp. | Differentiate and multiply based timing recovery in a quadrature demodulator |
| DE50109162D1 (en) | 2000-05-16 | 2006-05-04 | Infineon Technologies Ag | CIRCUIT ARRANGEMENT FOR TOLERANCE CORRECTION IN A FREQUENCY DEMODULATOR |
| US7519333B2 (en) * | 2000-07-03 | 2009-04-14 | Texas Instruments Incorporated | Radio architecture for use with frequency division duplexed systems |
| SE521838C2 (en) * | 2001-02-16 | 2003-12-09 | Nat Semiconductor Corp | Method and apparatus for automatic gain control |
| US6847255B2 (en) * | 2001-06-01 | 2005-01-25 | Broadband Innovations, Inc. | Zero IF complex quadrature frequency discriminator and FM demodulator |
| DE10131676A1 (en) | 2001-06-29 | 2003-01-16 | Infineon Technologies Ag | Receiver arrangement with AC coupling |
| US6904538B2 (en) * | 2001-11-20 | 2005-06-07 | Agere Systems Inc. | System and method for differential data detection |
| JP2006148627A (en) * | 2004-11-22 | 2006-06-08 | Fujitsu Ltd | Demodulator of frequency modulation signal and demodulation method of frequency modulation signal |
| US8369820B2 (en) * | 2007-09-05 | 2013-02-05 | General Instrument Corporation | Frequency multiplier device |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| HU175236B (en) * | 1977-01-10 | 1980-06-28 | Hiradastech Ipari Kutato | Method and apparaus to receive and generate frquency modulated signals |
| GB2106734B (en) * | 1981-09-15 | 1986-01-15 | Standard Telephones Cables Ltd | Radio receiver |
| US4651107A (en) * | 1984-01-11 | 1987-03-17 | Nec Corporation | Demodulator for digital FM signals |
| GB2166324A (en) * | 1984-10-25 | 1986-04-30 | Stc Plc | A multi-mode radio transceiver |
| GB2175473A (en) * | 1985-05-17 | 1986-11-26 | Philips Electronic Associated | D.c. block capacitor circuit |
-
1986
- 1986-08-01 GB GB08618855A patent/GB2193405A/en not_active Withdrawn
-
1987
- 1987-07-21 EP EP87201394A patent/EP0255175A3/en not_active Withdrawn
- 1987-07-29 FI FI873314A patent/FI873314A/en not_active Application Discontinuation
- 1987-07-29 US US07/079,548 patent/US4766392A/en not_active Expired - Fee Related
- 1987-07-29 DK DK395387A patent/DK395387A/en not_active Application Discontinuation
- 1987-07-30 KR KR1019870008333A patent/KR880003479A/en not_active Application Discontinuation
- 1987-07-30 CA CA000543461A patent/CA1269423A/en not_active Expired - Lifetime
- 1987-07-31 AU AU76353/87A patent/AU586394B2/en not_active Ceased
- 1987-08-01 JP JP62193615A patent/JPS6386905A/en active Pending
Also Published As
| Publication number | Publication date |
|---|---|
| EP0255175A3 (en) | 1989-03-15 |
| GB2193405A (en) | 1988-02-03 |
| EP0255175A2 (en) | 1988-02-03 |
| JPS6386905A (en) | 1988-04-18 |
| DK395387D0 (en) | 1987-07-29 |
| DK395387A (en) | 1988-02-02 |
| US4766392A (en) | 1988-08-23 |
| FI873314A (en) | 1988-02-02 |
| AU7635387A (en) | 1988-02-04 |
| GB8618855D0 (en) | 1986-09-10 |
| FI873314A0 (en) | 1987-07-29 |
| KR880003479A (en) | 1988-05-17 |
| AU586394B2 (en) | 1989-07-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA1269423A (en) | Demodulating an angle-modulated signal | |
| CA1199377A (en) | Baseband demodulator for fm signals | |
| US5507036A (en) | Apparatus with distortion cancelling feed forward signal | |
| US4789897A (en) | Frequency converting apparatus for converting an RF television signal to a video signal employing low IF techniques | |
| US4736390A (en) | Zero IF radio receiver apparatus | |
| DE3516492C2 (en) | ||
| US6192225B1 (en) | Direct conversion receiver | |
| US4633315A (en) | Receiver for RF-signals comprising a pair of parallel signal paths | |
| US4776039A (en) | Receiver for mobile communication systems | |
| KR100279031B1 (en) | Frequency Modulation Receiver | |
| US4811425A (en) | Apparatus for reducing the effects of local oscillator leakage in mixers employed in zero IF receivers | |
| US4878251A (en) | Interference signal suppressor for a radio receiver | |
| DE2645950A1 (en) | RADIO RECEIVER FOR FREQUENCY MODULATED SIGNALS | |
| GB2324919A (en) | Modulation or frequency conversion by time sharing | |
| CA1299249C (en) | Receiver comprising parallel signal paths | |
| US4599743A (en) | Baseband demodulator for FM and/or AM signals | |
| CA1299250C (en) | Frequency difference detector (fdd) and a carrier modulated receiver including such a fdd | |
| US4677690A (en) | Baseband demodulator for FM and/or AM signals | |
| KR100792236B1 (en) | Receiver of eliminating image signal | |
| JP2005536081A (en) | Second-order Nyquist slope filter | |
| US3873931A (en) | FM demodulator circuits | |
| JPH02157667A (en) | Phase detector and frequency demodulator | |
| US6995595B2 (en) | Direct conversion receiver having a DC offset eliminating function | |
| US4974086A (en) | Frequency conversion circuit for a color television channel | |
| US3706946A (en) | Deviation modifier |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MKLA | Lapsed |