CA2162326A1 - Signal processor - Google Patents
Signal processorInfo
- Publication number
- CA2162326A1 CA2162326A1 CA002162326A CA2162326A CA2162326A1 CA 2162326 A1 CA2162326 A1 CA 2162326A1 CA 002162326 A CA002162326 A CA 002162326A CA 2162326 A CA2162326 A CA 2162326A CA 2162326 A1 CA2162326 A1 CA 2162326A1
- Authority
- CA
- Canada
- Prior art keywords
- signal
- signal processor
- output
- predetermined
- vectors
- 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.)
- Abandoned
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/26—Systems using multi-frequency codes
- H04L27/2601—Multicarrier modulation systems
- H04L27/2614—Peak power aspects
- H04L27/2621—Reduction thereof using phase offsets between subcarriers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/26—Systems using multi-frequency codes
- H04L27/2601—Multicarrier modulation systems
- H04L27/2626—Arrangements specific to the transmitter only
Abstract
The present inventions discloses a so called Discrete Multi Tone or DMT signal processor which is coupled to a transmitting end of a communication line and which reduces the need for the clipping of amplitude of the output signal OS'. The signal processor includes the cascade connection of a vector transformation circuit VTC, a modulator MOD and a peak detection circuit PDC. The VTC
circuit is provided with a plurality of input vectors IN
whereon a vector transformation is performed according to a signal at a control terminal (CT) of the peak detection circuit PDC which is indicative of the fact that the amplitude of the output signal OS provided by the modulator MOD exceeds a predetermined threshold value.
circuit is provided with a plurality of input vectors IN
whereon a vector transformation is performed according to a signal at a control terminal (CT) of the peak detection circuit PDC which is indicative of the fact that the amplitude of the output signal OS provided by the modulator MOD exceeds a predetermined threshold value.
Description
21~32B
, S I GNAL Pi~OCi~ S SOR
Tile present invention relates to a signal processor including first means adapted to provide an output signal which equals the sum of a plurality of modulated signals represented in a signal vector plane by a plurality of input vectors Such a signal processor is already known in tile art e.g. from the US paterlt 4, 679, 227 . Therein, the first means 10 is referred to as a modulator which generates a digitally encoded time series or an output signal representing a waveform comprising the sum of quadrature amplitude modulated (QAM) carrier frequencies or modulated signals. The output signal is forwarded by tile modulator to a digital to-analog converter DAC which provides an analog output signal to an I/O interface coupled to a communication line.
The output signal may have large amplitude peaks or so called spikes due to constructive interference of tile modulated signals which are summed. Consequently, the DAC
20 converter needs to use a large number of bits to convert these spikes with sufficient precision. ilowever, a large number of bits also means an expensive DAC. One may reduce the number of bits used by the DAC converter, by clipping the amplitude value of ~lle output signal to a maximum value. In this way, the DAC converter becomes less expensive, but on the other hand there is a chance tllat information modulated on the QAM carrier frequencies is lost. Indeed, clipping the amplitude of the output signal is equivalent with changing the amplitude and/or phase of these carrier frequencies and 30 thus it is equivalent witll changing the information which is modulated on these QAM carrier frequencies as phase and amplitude variations.
It is thus clear tilat clipping is to be avoided as much as possible.
An object of tile present invention ls to provide a signal processor of the above known type, but which reduces 21~23~
'-t ~
, S I GNAL Pi~OCi~ S SOR
Tile present invention relates to a signal processor including first means adapted to provide an output signal which equals the sum of a plurality of modulated signals represented in a signal vector plane by a plurality of input vectors Such a signal processor is already known in tile art e.g. from the US paterlt 4, 679, 227 . Therein, the first means 10 is referred to as a modulator which generates a digitally encoded time series or an output signal representing a waveform comprising the sum of quadrature amplitude modulated (QAM) carrier frequencies or modulated signals. The output signal is forwarded by tile modulator to a digital to-analog converter DAC which provides an analog output signal to an I/O interface coupled to a communication line.
The output signal may have large amplitude peaks or so called spikes due to constructive interference of tile modulated signals which are summed. Consequently, the DAC
20 converter needs to use a large number of bits to convert these spikes with sufficient precision. ilowever, a large number of bits also means an expensive DAC. One may reduce the number of bits used by the DAC converter, by clipping the amplitude value of ~lle output signal to a maximum value. In this way, the DAC converter becomes less expensive, but on the other hand there is a chance tllat information modulated on the QAM carrier frequencies is lost. Indeed, clipping the amplitude of the output signal is equivalent with changing the amplitude and/or phase of these carrier frequencies and 30 thus it is equivalent witll changing the information which is modulated on these QAM carrier frequencies as phase and amplitude variations.
It is thus clear tilat clipping is to be avoided as much as possible.
An object of tile present invention ls to provide a signal processor of the above known type, but which reduces 21~23~
'-t ~
the need for clipping of the output signal and which thus allows the use of a cheaper DAC converter without possible loss of information.
According to the present invention, this is acchieved due to the fact that said signal processor further includes a cascade connection of third means said first means and second means, said second means being adapted to activate a control signal at a control terminal t~lereof if tl-e amplitude of said output signal exceeds a predetermined 10 amplitude threshold, said third means being adapted to perform a predetermined transformation of said input vectors upon said control signal being activated, said second means further being adapted to provide said output signal at an output terminal of said signal processor.
Indeed, the amplitude value of the output signal is accurately modelled as a Gaussian F~andom Variable due to the fact that the output signal equals the sum of a plurality of modulated signals. qhe probability P that the amplitude of the output signal is bigger t~lan the amplitude threshold and 20 that the input vectors are transformed by the third means, is smaller than one i.e. P~l . As a result, the first means provides an output signal for which t~le probability that the amplitude is larger than t~le amplitude threshold is P.P which is smaller than P. The latter probability is unconditional due to the predetermined transformation.
A characteristic feature of the invention is that said plurality of input vectors are represented in said signal vector plane by a plurality of orthogonal coordinate pairs and that said first means is adapted to process said 30 output signal by means of an inverse fast fourier transformation of said plurality of orthogonal coordinate pai rs .
In this way, the first means provides the output signal in a very reliable way.
Another characteristic feature is that said predetermined transformation comprises a rotation in said ~ ~ 216~326 signal vector plane of said input vectors over respective predetermined angles.
In this way, a simple transformation is achieved.
Yet another characteristic feature is that said predetermined transformation further comprises a multiplication of the amplitudes of said input vectors wit~l respective predetermined factors, which realizes a more complex transformation.
A furt~er characteristic is t~at said signal 10 processor includes a counter register which is coupled between said second and said third means and which is incremented upon said control signal being activated, and that said output signal is provided at said output terminal by said second means if a value of said counter register exceeds a predetermined counter threshold.
As a result the need for clipping the output signal is reduced eYen more. Indeed, the probability that the amplitude of the output signal which is provided after K
input vector transformations, K being equal to the 20 predetermined counter threshold, is PAK (P to the power K) which is even smaller than P.P i.e. P^K P < 1.
Still another characteristic is tllat said predetermined transformation is performed according to said value of said counter register and that said third means is adapted to set a predetermined one of said input vectors in such a way as to being indicative of said vaLue of said counter register.
In this way, the output signal contains information, i.e. the value of the counter register, 30 concerning the predetermined transformation which is performed according to the latter value.
An advantage of the present invention is the fact that said second means further includes storagç means adapted to store said output signal during a time period determined by said value of said counter register.
- 216232~
> ~
According to the present invention, this is acchieved due to the fact that said signal processor further includes a cascade connection of third means said first means and second means, said second means being adapted to activate a control signal at a control terminal t~lereof if tl-e amplitude of said output signal exceeds a predetermined 10 amplitude threshold, said third means being adapted to perform a predetermined transformation of said input vectors upon said control signal being activated, said second means further being adapted to provide said output signal at an output terminal of said signal processor.
Indeed, the amplitude value of the output signal is accurately modelled as a Gaussian F~andom Variable due to the fact that the output signal equals the sum of a plurality of modulated signals. qhe probability P that the amplitude of the output signal is bigger t~lan the amplitude threshold and 20 that the input vectors are transformed by the third means, is smaller than one i.e. P~l . As a result, the first means provides an output signal for which t~le probability that the amplitude is larger than t~le amplitude threshold is P.P which is smaller than P. The latter probability is unconditional due to the predetermined transformation.
A characteristic feature of the invention is that said plurality of input vectors are represented in said signal vector plane by a plurality of orthogonal coordinate pairs and that said first means is adapted to process said 30 output signal by means of an inverse fast fourier transformation of said plurality of orthogonal coordinate pai rs .
In this way, the first means provides the output signal in a very reliable way.
Another characteristic feature is that said predetermined transformation comprises a rotation in said ~ ~ 216~326 signal vector plane of said input vectors over respective predetermined angles.
In this way, a simple transformation is achieved.
Yet another characteristic feature is that said predetermined transformation further comprises a multiplication of the amplitudes of said input vectors wit~l respective predetermined factors, which realizes a more complex transformation.
A furt~er characteristic is t~at said signal 10 processor includes a counter register which is coupled between said second and said third means and which is incremented upon said control signal being activated, and that said output signal is provided at said output terminal by said second means if a value of said counter register exceeds a predetermined counter threshold.
As a result the need for clipping the output signal is reduced eYen more. Indeed, the probability that the amplitude of the output signal which is provided after K
input vector transformations, K being equal to the 20 predetermined counter threshold, is PAK (P to the power K) which is even smaller than P.P i.e. P^K P < 1.
Still another characteristic is tllat said predetermined transformation is performed according to said value of said counter register and that said third means is adapted to set a predetermined one of said input vectors in such a way as to being indicative of said vaLue of said counter register.
In this way, the output signal contains information, i.e. the value of the counter register, 30 concerning the predetermined transformation which is performed according to the latter value.
An advantage of the present invention is the fact that said second means further includes storagç means adapted to store said output signal during a time period determined by said value of said counter register.
- 216232~
> ~
Indeed, the time delay between tlle input of the input vectors to the signal processor and the output of the output signal by the second means is constant lf the above time perlod is properly determined for each value of the counter register. More in particular, the time period T(M) for a value M of the counter register equals T(M~1 ) plus the time period needed by the third means to perform a transformation of tile input vectors Tthird plus the time perlod needed by the first means to provide the output signal 10 Tfirst i.e. T(M) = T(M~1 ) + Tthird ~ Tfirst.
A particular feature is that said storage means is adapted to store said output signal as a sequence of digital amplitude values, and that said second means is adapted to limit said digital amplitude values to said predetermined amplitude threshold.
In this way, a less expensive DAC converter may be used because the amplitude of the output signal is limited to the ampl i tude threshold value .
The above signal processor is coupled to the above 20 mentioned communication line at the tral1smittir~g side However, at the receiving side another signal processor complementary to the first one is coupled to the latter communication line. This complementary processor is adapted to process an input signal substantially equal to the above mentioned output signal provided by the signal processor coupled to the communication line at the transmitting side.
The present invention therefore further relates to a signal proc~ssor including first means adapted to provide a plurality of output vectors r~presenting a plurality of 30 modulated signals in a signal vector plane, an input signal of said signal processor being comprised of said modulated signals .
Characteristic features of this processor are that said signal processor includes the cascade connection of said first means and second means, said second means being adapted to perform a predetermined transformation of said output ~, 21~232~
A particular feature is that said storage means is adapted to store said output signal as a sequence of digital amplitude values, and that said second means is adapted to limit said digital amplitude values to said predetermined amplitude threshold.
In this way, a less expensive DAC converter may be used because the amplitude of the output signal is limited to the ampl i tude threshold value .
The above signal processor is coupled to the above 20 mentioned communication line at the tral1smittir~g side However, at the receiving side another signal processor complementary to the first one is coupled to the latter communication line. This complementary processor is adapted to process an input signal substantially equal to the above mentioned output signal provided by the signal processor coupled to the communication line at the transmitting side.
The present invention therefore further relates to a signal proc~ssor including first means adapted to provide a plurality of output vectors r~presenting a plurality of 30 modulated signals in a signal vector plane, an input signal of said signal processor being comprised of said modulated signals .
Characteristic features of this processor are that said signal processor includes the cascade connection of said first means and second means, said second means being adapted to perform a predetermined transformation of said output ~, 21~232~
vectors according to a value lndlcated by a predetermined one of said output vectors; and t~lat said plurality of output vectors arQ represented in said signal vector plane by a plurality of orthogonal coordinate pairs, and that said first means is adapted to process said plurality of orthognal coordinate pairs by means of a fast fourier transform of said input signal. Finally, said predetermined transformation comprises a rotation in said signal vector plane of said output vectors ovqr respective predetermined angles; and said 10 predetermined transformation further comprises a multiplication of the amplitudes of said output vectors wit}
respective predetermined factors.
The above mentioned and other objects and features of tlle invention will become more apparent and the invention itself will be best understood by referring to the following description of an embodiment taken in conjunction with the accompanying drawings wherein:
Fig 1 shows a signal processor SPT according to the 20 invention which is coupled to the transmitting side of a communication line;
Fig 2 shows a signal procèssor SPR coupled to the receiving side of the above communlcation line.
Referring to Fig 1, the signal processor SPT has a plurality of input terminals IN coupled to an output terminal OS' via the cascade connection of a vector transformation circuit VTC, a modulator MOD and a peak detection circuit PDC. An input/output terminal T3 of t~le PDC circuit is coupled via a counter register CR to an input terminal Tl of 30 the VTC circuit. A control output terminal CT of the PDC
circuit is further connected to an input terminal T2 of the VTC circuit.
Referring to Fig 2, the signal processor SPR has an input terminal IS coupled to a plurality of output terminals OUT' via the cascade connection of a demodulator DEMOD and a ~ 2~23~
~3 ~
respective predetermined factors.
The above mentioned and other objects and features of tlle invention will become more apparent and the invention itself will be best understood by referring to the following description of an embodiment taken in conjunction with the accompanying drawings wherein:
Fig 1 shows a signal processor SPT according to the 20 invention which is coupled to the transmitting side of a communication line;
Fig 2 shows a signal procèssor SPR coupled to the receiving side of the above communlcation line.
Referring to Fig 1, the signal processor SPT has a plurality of input terminals IN coupled to an output terminal OS' via the cascade connection of a vector transformation circuit VTC, a modulator MOD and a peak detection circuit PDC. An input/output terminal T3 of t~le PDC circuit is coupled via a counter register CR to an input terminal Tl of 30 the VTC circuit. A control output terminal CT of the PDC
circuit is further connected to an input terminal T2 of the VTC circuit.
Referring to Fig 2, the signal processor SPR has an input terminal IS coupled to a plurality of output terminals OUT' via the cascade connection of a demodulator DEMOD and a ~ 2~23~
~3 ~
detection and vector transformation circuit DVTC. The operation of t~-e signal processor SPT is hereafter described.
The signal processor SPT provides a so called Discrete Multi Tone or DMT signal OS' at its like named output terminal. The signal OS' equals the sum of a plurallty of Quadrature Amplitude Modulated or QAM signals which are represented in a QAM signal vector plane by corresponding signal vectors or so called constellation points. The DMT
technique is well described in he above mentioned US patent.
These signal vectors are represented in the signal vector plane by orthogonal coordinate pairs IN which are input to the signal processor SPT via the lil~e named plurality of input terminals. The initial value of the counter register CR is zero and the control terminal CT of the PDC circuit is set low. The YTC circuit sets the orthogonal coordinates of a predetermined one of the above signal vectors to predetermined values so that the latter predetermined input vector is indicative of the value of the CR register. To be noted that the latter vector can not 20 represent information other tl~an the value of t~e counter register CR. When the value of the counter register is zero, the VTC circuit passes the orthogonal coordinated pairs IN' which are equal to the coordinate pairs IN except for the coordinate pair representing the above predetermined input vector, to the modulator MOD which performs an Inverse Fast Fourier Transform IFFT so as to provide an output signal OS
which equals the sum of the above mentioned QAM signals. The modulation technique of the IFFT is well known to a person skilled in the art and is therefore not described in detail.
30 The latter output signal OS comprises a series of digital amplitude values and is forwarded to tl~e PDC circuit wllich detects the peak amplitude value. The lattel peak amplitude value is then compared with a predetermined amplitude threshold value and the control signal provided at the control terminal CT is set low/high if the peak value is smaller/larger than the threshold value. Upon the control 2 ~ 6 , 1 signal being set high, the PDC circuit also increments the counter register CR via its terminal T3.
If the latter control signal is set low, the output signal OS is stored in a buffer register BUF of the PDC
circuit during a time interval T(M), M belng the value of the CR register ( in this case M=0 ) and tlle CR register is reset to zero via terminal T3. ~ow to determine T(M) is described further on.
If the control signal is set high, the signal at 10 the input terminal T2 of the VTC circuit is also set high upon wich the VTC circuit sets the orthogonal coordinates of the above mentioned predetermined one of the above signal vectors to predetermined values indicative of the incremented value of the CR register. The above predetermined input vector is then again indicative of the latter value.
Furthermore, the VTC circuit modifies the orthogonal coordinates of the other signal vectors so as to perform a vector transformation of the latter signal vectors. The vector transformation is a phase rotation of the signal 20 vectors in the signal vector plane over corresponding angles which are determined for each signal vector by the value M of the CR register. To be noted that the latter vector transformation may also comprlse multiplication of the amplitudes of each signal vector with factors determined for each signal vector by the value M of the CR register To be noted that for each value M, a different transformation is per f ormed .
The modified orthogonal coordinate pairs are then forwarded to the modulator which performs the IFFT
30 tranformation to provide another output signal OS which is passed to the PDC circuit. The PDC circuit then checl~s if the value of the CR register is smaller/equal to a counter threshold value. In case the value of the CR register is smaller than the latter threshold value, the PDC circuit operates as described above i.e. stores the output signal in the buffer register BUF during a time period T(M). In case 21~232 'l ~
The signal processor SPT provides a so called Discrete Multi Tone or DMT signal OS' at its like named output terminal. The signal OS' equals the sum of a plurallty of Quadrature Amplitude Modulated or QAM signals which are represented in a QAM signal vector plane by corresponding signal vectors or so called constellation points. The DMT
technique is well described in he above mentioned US patent.
These signal vectors are represented in the signal vector plane by orthogonal coordinate pairs IN which are input to the signal processor SPT via the lil~e named plurality of input terminals. The initial value of the counter register CR is zero and the control terminal CT of the PDC circuit is set low. The YTC circuit sets the orthogonal coordinates of a predetermined one of the above signal vectors to predetermined values so that the latter predetermined input vector is indicative of the value of the CR register. To be noted that the latter vector can not 20 represent information other tl~an the value of t~e counter register CR. When the value of the counter register is zero, the VTC circuit passes the orthogonal coordinated pairs IN' which are equal to the coordinate pairs IN except for the coordinate pair representing the above predetermined input vector, to the modulator MOD which performs an Inverse Fast Fourier Transform IFFT so as to provide an output signal OS
which equals the sum of the above mentioned QAM signals. The modulation technique of the IFFT is well known to a person skilled in the art and is therefore not described in detail.
30 The latter output signal OS comprises a series of digital amplitude values and is forwarded to tl~e PDC circuit wllich detects the peak amplitude value. The lattel peak amplitude value is then compared with a predetermined amplitude threshold value and the control signal provided at the control terminal CT is set low/high if the peak value is smaller/larger than the threshold value. Upon the control 2 ~ 6 , 1 signal being set high, the PDC circuit also increments the counter register CR via its terminal T3.
If the latter control signal is set low, the output signal OS is stored in a buffer register BUF of the PDC
circuit during a time interval T(M), M belng the value of the CR register ( in this case M=0 ) and tlle CR register is reset to zero via terminal T3. ~ow to determine T(M) is described further on.
If the control signal is set high, the signal at 10 the input terminal T2 of the VTC circuit is also set high upon wich the VTC circuit sets the orthogonal coordinates of the above mentioned predetermined one of the above signal vectors to predetermined values indicative of the incremented value of the CR register. The above predetermined input vector is then again indicative of the latter value.
Furthermore, the VTC circuit modifies the orthogonal coordinates of the other signal vectors so as to perform a vector transformation of the latter signal vectors. The vector transformation is a phase rotation of the signal 20 vectors in the signal vector plane over corresponding angles which are determined for each signal vector by the value M of the CR register. To be noted that the latter vector transformation may also comprlse multiplication of the amplitudes of each signal vector with factors determined for each signal vector by the value M of the CR register To be noted that for each value M, a different transformation is per f ormed .
The modified orthogonal coordinate pairs are then forwarded to the modulator which performs the IFFT
30 tranformation to provide another output signal OS which is passed to the PDC circuit. The PDC circuit then checl~s if the value of the CR register is smaller/equal to a counter threshold value. In case the value of the CR register is smaller than the latter threshold value, the PDC circuit operates as described above i.e. stores the output signal in the buffer register BUF during a time period T(M). In case 21~232 'l ~
the value of the CR reglster is equal to the counter treshold value, the PDC circuit clips the amplitude of the output signal to a predetermined amplitude threshold Athres and forwards the output signal to the output terminal OS ' of the SPT processor, the CR register is reset to zero and the control signal at terminal CT is set low. IQ this way, only a limited number of vector transformations are performed on the same plurality of input vectors and the amplitude of the output signal is limited to the amplitude threshold value.
With respect to the time interval T(M) that the output signal OS is stQred in the buffer register BUF of the PDC circuit for a value M of tl-e CR register the following is to be noted: T(M~1 ) = T(M1 - Tvtc - Tmod, i.e. the time interval T(M+1 ) for a value M+1 of t~e CR register equals T(M) minus the time needed to perform a vector transformation in the VTC circuit and minus the time to perform an IFFT
transformation in the modulator MOD. In this way, the signal processor SPT provides the output signal at its terminal OS' after a constant time interval since the input of the 20 plurality of orthogonal coordinate pairs to the signal processor SPT.
The advantage of the operation of such a signal processor SPT is now briefly explained with reference to the article 'Effect of amplitude clipping in DMT-ADSL
transceivers ' in Electronic Letters 22nd July 1993 Vol . 29 No. 15 page 1354 and 1355 by D. Mestdagh, J. Spruyt and B.
siran. The output signal OS equals the sum of N QAM signals.
For large values of N, say N>= 10, the amplitude of the output signal is accurately modelled as a Gaussian Random 30 Value with a zero mean value. After the first pass through the signal processor SPT, the modulator MOD provides an output signal OS for which the probability that itsamplitude is larger than the threshold amplitude Athres, is P. For the amplitude which is a Gaussian Random Value, this probability is readily determined by a person skilled in the art. If Mmax vector transformations are performed on the same plurality of ` 2~ ~232~
g input vectors, the probability that the amplitude of the output signal OS provided by the modulator MOD after Mmax passes, is larger than the amplitude threshold Athres, is P^Mmax (P to the power Mmax). Indeed, this probability is unconditional due to Mmax consecutive vector transformations which are independent of each other. As P<1 it is obvious that P^Mmax P.
The signal processor SPT is coupled to a digital-to-analog DAC converter which uses B1 bits to represent the 10 amplitude values of ti-e output signal provided by tile SPT
processor. The use of the above signal processor SPT reduces the probability that the amplitude of the output slgnal OS is larger than the amplitude threshold Atllres. Consequently, the amplitude value may be clipped to the latter amplitude threshold so that the DAC converter can use B2 bits instead of B1 bits, B2 being smaller than B1, without signal quality loss. The signal ~uality is measured as noise to signal ratio (N/S) and (N/S) B1 (_B2) is readlly determined by a person skilled in the art for the DAC using B1 (B2) bits for a 20 maximimum amplitude Amax (Atllres). i~owever, tile clipping of the amplitude introduces extra noise and the noise to signal ratio (N/S)_ clip due to the clipping is also easlly deducted by someone skilled in the art, from the probability distribution of ti~e amplitude of the output signal OS. In order to maintain the same signal quality the following condition (1 ) must be satisfied:
(N/S)_ B1 =) (N/S)_B2 + (N/S)_clip_Mmax.
(N/S)_clip_Mmax being the noise-to-signal ratio caused by clipping of the anplitude of the output signal OS after Mmax 30 vector transformations. Developping this equation results in a relation between Athres,Amax,B1 and s2 whereby tile choice of Athres is related to the probability distribution of the amplitude of ti~e output signal OS and related to the number of vector transformations performed by the VTC circuit i.e.
Mmax. Moreover, the values of Athres and of Mmax directly affect the value of (N/S)_clip_Mmax.
2~62~
As a result of the above, the use of tl~e signal processor SPT according to the invention reduces the probabllity that the amplitude of the output. signal OS is larger than the amplitude threshold, therefore the amplitude value may be clipped to the amplitude threshold value and t~le DAC coupled after the signal Frocessor SPT only needs B2 bits to cover the amplitude range. A11 this is possible without signal quality loss as explained above. To be noted that Athres may be chosen in function of the number of vector 10 transformations Mmax.
The output signal provided at the output terminal OS' of the SPT processor is converted to an analog signal in the DAC converter (not shown) and is transmitted over a communication line (not shown) coupled to the DAC converter via an interface circuit (not shown). At the other end of the communication line, a simllar interface circuit (not shown) couples the communication line to an analog-to-digital ADC
converter (not shown) which supplies an input signal, substantially like the above output signal, to the input 20 terminal IS of the signal processor SPR as shown in Fig 2.
The operation of tile signal processor SPR shown in Fig 2 is hereaf ter described .
The demodulator DEMOD performs a so called Fast Fourier Transformation on the input signal and provides a plurality of orthogonal coordinate pairs OUT at a like named plurality of output terminals. These coordinate pairs represent a corresponding plurality of signal vectors in the above mentioned signal vector plane. The signal vectors represent QAM modulated signals the sum of which equals t~e 30 input signal. The coordinate pairs OUT are passed to t}le detection and vector transformation circuit DVTC.
The DVTC circuit first of all detects which value is indicated by the signal vector represented by a predetermined one of the coordlnate pairs. This is done by means of comparing the coordinate values of tile latter predetermlned coordinate pair, with predetermined coordinate 21~232~
.
values whlch are each indicative of a value like the above value and which are substantially equal to the above mentloned predetermined values indicative of the value of tlle CR register. The DVTC circuit then modifies all the other orthogonal coordinate pairs according to the above value so as to perform the inverse transformation of the above mentioned vector transformation performed by the VTC ci}cuit of the SPT processor. Finally the DVTC circuit provides the modified orthogonal coordinate pairs at the output terminals 10 OUT' of the SPR processor.
It is shown that at the receiving side of the communication line there is no signal quality loss caused by the use of a B2 bits ADC converter coupled to the SPR signal processor .
The analog output signal which is supplied to the communication line via the interface circuit by the B2 bits DAC converter, is submitted to phase distortion while propagating over the latter communication line. This process is equivalent to performing a vector transformation on the 20 signal vectors representing t~e above mentioned QAM signals.
Consequently, the probability that the amplitude of the signal which is provided to the ADC converter by the interface circuit at the receiving side of the communication line, is larger than the amplitude threshold ~t~res which is the maximum range of t~e B2 bits ADC converter, is P. Indeed, although the amplitude of t~-e output signal is limited to the amplitude threshold at the transmitting side of t~le communication line, it is to be considered that tlle amplitude of the output signal received at the receiving side of the 30 line may be larger then the amplitude threshold because t~le propagation over the latter line Ls equivalent wit~ a vector transformation. In order to avoid signal quality loss at t~le receiving side of the communication line we thus have the condition (2):
(N/S)_B2 + (N/S)_clip_1 <= (N/S)_Bl, which is like the condition at the transmitting side of the line. Indeed, ~ 21~232 `i ~
the use of a B2 bits converter involves clipplng of the amplitude and as explalned above thls introduces a noise to si~nal ratio (N/S)_ clip_1 caused by the cllpping of the amplitude after the propagation over the communication line whlch ls equlvalent of one slngle vector transformation. To be noted that:
(N/S)_ clip_n2 = (N/S)_clip_n1 . P^(n1-n2), P being the above mentioned probability that the amplltude ls larger than Athres. The comblnatlon of (1 ) and (2) results in:
2(N/S)_B1 >= 2(N/S)_B2 + (N/S)_clip_Mmax +
(N/S )_clip_1 .
or 2(N/S)_B1 >= 2(N/S)_B2 + (N/S)_clip_1 . (1 +
P^ (Mmax-1 ) ) which may be simplified to 2(N/S)_B1 >= 2(N/S)_B2 + (N/S)_clip_1 because P^ (Mmax-1 ) 1 .
So, the overal operatlon of the SPR and SPT
processors allows the use of less expensive DAC and ADC
20 converters without slgnal quality being lost if conditlons (1 ) and (2) are fullfilled.
To be noted that t~at t~e received signal may be attenuated due to the propagation over the communication line. Therefore, a so called automatic gain control or AGC
circuit is coupled before the ADC converter so as to compensate this attenuation with a corresponding amplification of t~le received signal.
If an echo canceller tnot shown) is used in the digital domain i.e. coupled to the output terminal of the SPR
30 processor, a like condition may be derived for the echoed signal. Let us suppose that a large analog echo signal is added to the received DMT signal. The latter analog echo slgnal is e.g. an attenuated part of another DMT signal which is transmitted on the same communication line but in the opposite direction of the recelved DMT signal. This other DMT
signal is provlded by another SPT processor (not shown) which ~1~2~2~
,~. ~
is also coupled to this echo canceller. Consequently, the signal which is supplied by the above AGC circuit to the ADC
converter contains a large echo component and a much smaller received DMT signal component. Suppose that the range and bit precision of the DAC and ADC converters is identical, then it is clear that the full range of the DAC converter can not be used because the ADC converter needs enough precision for the sum of the echo signal and the received signal together. As the echo signal is not subjected to any phase distortion, and 10 thus no extra vector transformation, the echo signal is not clipped at the ADC converter. For the path of the echo signal the following condition has to be fulfilled in order to maintain the same signal quality for the echoed signal:
2(N/S)_B1 >= 2(N/S)_82 + (N/S)_clip_Mmax.
It is important to notice that the (N/S)_B2 (_B1 ) is calculated for the maximum amplitude which may be converted by the DAC or ADC converters and not for the amplitude threshold Athres !
Consequently, in case of an echo cancelation scheme the 20 number of bits used by t!~e ADC and DAC converters may be reduced even more. Indeed, the gain in number of blts ls readily obtained as 2_log(Atl~res_~1/Athres_n2), Athres_x being the threshold amplitude determined for x vector transformations and n1 < n2.
To be noted that in the above mentioned echo cancelling scheme tlle DAC is over dimensioned. It is clear that the latter overdimensioning of the bit precision may be optimised .
While t!le principles of the invention have been 30 described above in connection with specific apparatus, it is to be clearly understood that t~lis description is made only by way of example and not as a limitation on the scope of the invention .
With respect to the time interval T(M) that the output signal OS is stQred in the buffer register BUF of the PDC circuit for a value M of tl-e CR register the following is to be noted: T(M~1 ) = T(M1 - Tvtc - Tmod, i.e. the time interval T(M+1 ) for a value M+1 of t~e CR register equals T(M) minus the time needed to perform a vector transformation in the VTC circuit and minus the time to perform an IFFT
transformation in the modulator MOD. In this way, the signal processor SPT provides the output signal at its terminal OS' after a constant time interval since the input of the 20 plurality of orthogonal coordinate pairs to the signal processor SPT.
The advantage of the operation of such a signal processor SPT is now briefly explained with reference to the article 'Effect of amplitude clipping in DMT-ADSL
transceivers ' in Electronic Letters 22nd July 1993 Vol . 29 No. 15 page 1354 and 1355 by D. Mestdagh, J. Spruyt and B.
siran. The output signal OS equals the sum of N QAM signals.
For large values of N, say N>= 10, the amplitude of the output signal is accurately modelled as a Gaussian Random 30 Value with a zero mean value. After the first pass through the signal processor SPT, the modulator MOD provides an output signal OS for which the probability that itsamplitude is larger than the threshold amplitude Athres, is P. For the amplitude which is a Gaussian Random Value, this probability is readily determined by a person skilled in the art. If Mmax vector transformations are performed on the same plurality of ` 2~ ~232~
g input vectors, the probability that the amplitude of the output signal OS provided by the modulator MOD after Mmax passes, is larger than the amplitude threshold Athres, is P^Mmax (P to the power Mmax). Indeed, this probability is unconditional due to Mmax consecutive vector transformations which are independent of each other. As P<1 it is obvious that P^Mmax P.
The signal processor SPT is coupled to a digital-to-analog DAC converter which uses B1 bits to represent the 10 amplitude values of ti-e output signal provided by tile SPT
processor. The use of the above signal processor SPT reduces the probability that the amplitude of the output slgnal OS is larger than the amplitude threshold Atllres. Consequently, the amplitude value may be clipped to the latter amplitude threshold so that the DAC converter can use B2 bits instead of B1 bits, B2 being smaller than B1, without signal quality loss. The signal ~uality is measured as noise to signal ratio (N/S) and (N/S) B1 (_B2) is readlly determined by a person skilled in the art for the DAC using B1 (B2) bits for a 20 maximimum amplitude Amax (Atllres). i~owever, tile clipping of the amplitude introduces extra noise and the noise to signal ratio (N/S)_ clip due to the clipping is also easlly deducted by someone skilled in the art, from the probability distribution of ti~e amplitude of the output signal OS. In order to maintain the same signal quality the following condition (1 ) must be satisfied:
(N/S)_ B1 =) (N/S)_B2 + (N/S)_clip_Mmax.
(N/S)_clip_Mmax being the noise-to-signal ratio caused by clipping of the anplitude of the output signal OS after Mmax 30 vector transformations. Developping this equation results in a relation between Athres,Amax,B1 and s2 whereby tile choice of Athres is related to the probability distribution of the amplitude of ti~e output signal OS and related to the number of vector transformations performed by the VTC circuit i.e.
Mmax. Moreover, the values of Athres and of Mmax directly affect the value of (N/S)_clip_Mmax.
2~62~
As a result of the above, the use of tl~e signal processor SPT according to the invention reduces the probabllity that the amplitude of the output. signal OS is larger than the amplitude threshold, therefore the amplitude value may be clipped to the amplitude threshold value and t~le DAC coupled after the signal Frocessor SPT only needs B2 bits to cover the amplitude range. A11 this is possible without signal quality loss as explained above. To be noted that Athres may be chosen in function of the number of vector 10 transformations Mmax.
The output signal provided at the output terminal OS' of the SPT processor is converted to an analog signal in the DAC converter (not shown) and is transmitted over a communication line (not shown) coupled to the DAC converter via an interface circuit (not shown). At the other end of the communication line, a simllar interface circuit (not shown) couples the communication line to an analog-to-digital ADC
converter (not shown) which supplies an input signal, substantially like the above output signal, to the input 20 terminal IS of the signal processor SPR as shown in Fig 2.
The operation of tile signal processor SPR shown in Fig 2 is hereaf ter described .
The demodulator DEMOD performs a so called Fast Fourier Transformation on the input signal and provides a plurality of orthogonal coordinate pairs OUT at a like named plurality of output terminals. These coordinate pairs represent a corresponding plurality of signal vectors in the above mentioned signal vector plane. The signal vectors represent QAM modulated signals the sum of which equals t~e 30 input signal. The coordinate pairs OUT are passed to t}le detection and vector transformation circuit DVTC.
The DVTC circuit first of all detects which value is indicated by the signal vector represented by a predetermined one of the coordlnate pairs. This is done by means of comparing the coordinate values of tile latter predetermlned coordinate pair, with predetermined coordinate 21~232~
.
values whlch are each indicative of a value like the above value and which are substantially equal to the above mentloned predetermined values indicative of the value of tlle CR register. The DVTC circuit then modifies all the other orthogonal coordinate pairs according to the above value so as to perform the inverse transformation of the above mentioned vector transformation performed by the VTC ci}cuit of the SPT processor. Finally the DVTC circuit provides the modified orthogonal coordinate pairs at the output terminals 10 OUT' of the SPR processor.
It is shown that at the receiving side of the communication line there is no signal quality loss caused by the use of a B2 bits ADC converter coupled to the SPR signal processor .
The analog output signal which is supplied to the communication line via the interface circuit by the B2 bits DAC converter, is submitted to phase distortion while propagating over the latter communication line. This process is equivalent to performing a vector transformation on the 20 signal vectors representing t~e above mentioned QAM signals.
Consequently, the probability that the amplitude of the signal which is provided to the ADC converter by the interface circuit at the receiving side of the communication line, is larger than the amplitude threshold ~t~res which is the maximum range of t~e B2 bits ADC converter, is P. Indeed, although the amplitude of t~-e output signal is limited to the amplitude threshold at the transmitting side of t~le communication line, it is to be considered that tlle amplitude of the output signal received at the receiving side of the 30 line may be larger then the amplitude threshold because t~le propagation over the latter line Ls equivalent wit~ a vector transformation. In order to avoid signal quality loss at t~le receiving side of the communication line we thus have the condition (2):
(N/S)_B2 + (N/S)_clip_1 <= (N/S)_Bl, which is like the condition at the transmitting side of the line. Indeed, ~ 21~232 `i ~
the use of a B2 bits converter involves clipplng of the amplitude and as explalned above thls introduces a noise to si~nal ratio (N/S)_ clip_1 caused by the cllpping of the amplitude after the propagation over the communication line whlch ls equlvalent of one slngle vector transformation. To be noted that:
(N/S)_ clip_n2 = (N/S)_clip_n1 . P^(n1-n2), P being the above mentioned probability that the amplltude ls larger than Athres. The comblnatlon of (1 ) and (2) results in:
2(N/S)_B1 >= 2(N/S)_B2 + (N/S)_clip_Mmax +
(N/S )_clip_1 .
or 2(N/S)_B1 >= 2(N/S)_B2 + (N/S)_clip_1 . (1 +
P^ (Mmax-1 ) ) which may be simplified to 2(N/S)_B1 >= 2(N/S)_B2 + (N/S)_clip_1 because P^ (Mmax-1 ) 1 .
So, the overal operatlon of the SPR and SPT
processors allows the use of less expensive DAC and ADC
20 converters without slgnal quality being lost if conditlons (1 ) and (2) are fullfilled.
To be noted that t~at t~e received signal may be attenuated due to the propagation over the communication line. Therefore, a so called automatic gain control or AGC
circuit is coupled before the ADC converter so as to compensate this attenuation with a corresponding amplification of t~le received signal.
If an echo canceller tnot shown) is used in the digital domain i.e. coupled to the output terminal of the SPR
30 processor, a like condition may be derived for the echoed signal. Let us suppose that a large analog echo signal is added to the received DMT signal. The latter analog echo slgnal is e.g. an attenuated part of another DMT signal which is transmitted on the same communication line but in the opposite direction of the recelved DMT signal. This other DMT
signal is provlded by another SPT processor (not shown) which ~1~2~2~
,~. ~
is also coupled to this echo canceller. Consequently, the signal which is supplied by the above AGC circuit to the ADC
converter contains a large echo component and a much smaller received DMT signal component. Suppose that the range and bit precision of the DAC and ADC converters is identical, then it is clear that the full range of the DAC converter can not be used because the ADC converter needs enough precision for the sum of the echo signal and the received signal together. As the echo signal is not subjected to any phase distortion, and 10 thus no extra vector transformation, the echo signal is not clipped at the ADC converter. For the path of the echo signal the following condition has to be fulfilled in order to maintain the same signal quality for the echoed signal:
2(N/S)_B1 >= 2(N/S)_82 + (N/S)_clip_Mmax.
It is important to notice that the (N/S)_B2 (_B1 ) is calculated for the maximum amplitude which may be converted by the DAC or ADC converters and not for the amplitude threshold Athres !
Consequently, in case of an echo cancelation scheme the 20 number of bits used by t!~e ADC and DAC converters may be reduced even more. Indeed, the gain in number of blts ls readily obtained as 2_log(Atl~res_~1/Athres_n2), Athres_x being the threshold amplitude determined for x vector transformations and n1 < n2.
To be noted that in the above mentioned echo cancelling scheme tlle DAC is over dimensioned. It is clear that the latter overdimensioning of the bit precision may be optimised .
While t!le principles of the invention have been 30 described above in connection with specific apparatus, it is to be clearly understood that t~lis description is made only by way of example and not as a limitation on the scope of the invention .
Claims (12)
1. Signal processor (SPT) including first means (MOD) adapted to provide an output signal (OS') which equals the sum of a plurality of modulated signals represented in a signal vector plane by a plurality of input vectors ( IN), characterized in that said signal processor further includes a cascade connection of third means (VTC) said first means and second means (PDC), said second means being adapted to activate a control signal at a control terminal thereof (CT) if the amplitude of said output signal exceeds a predetermined amplitude threshold, said third means being adapted to perform a predetermined transformation of said input vectors upon said control signal being activated, said second means further being adapted to provide said output signal at an output terminal of said signal processor.
2. Signal processor according to claim 1, characterized in that said plurality of input vectors are represented in said signal vector plane by a plurality of orthogonal coordinate pairs and that said first means is adapted to process said output signal by means of an inverse fast fourier transformation of said plurality of orthogonal coordinate pairs .
3. Signal processor according to claim 1, characterized in that said predetermined transformation comprises a rotation in said signal vector plane of said input vectors over respective predetermined angles.
4. Signal processor according to claim 3, characterized in that said predetermined transformation further comprises a multiplication of the amplitudes of said input vectors with respective predetermined factors.
5. Signal processor according to claim 1, characterized in that said signal processor includes a counter register (CR) which is coupled between said second and said third means and which is incremented upon said control signal being activated, and that said output signal is provided at said output terminal by said second means if a value of said counter register exceeds a predetermined counter threshold.
6. Signal processor according to claim 5, characterized in that said predetermined transformation is performed according to said value of said counter register and that said third means is adapted to set a predetermined one of said input vectors in such a way as to being indicative of said value of said counter register.
7. Signal processor according to claim 5, characterized in that said second means further includes storage means (BUF) adapted to store said output signal during a time period determined by said value of said counter register.
8. Signal processor according to claim 6, characterized in that said storage means is adapted to store said output signal as a sequence of digital amplitude values, and that said second means is adapted to limit said digital amplitude values to said predetermined amplitude threshold.
9. Signal processor (SPR) including first means (DEMOD) adapted to provide a plurality of output vectors (OUT) representing a plurality of modulated signals in a signal vector plane, an input signal (IS) of said signal processor being comprised of said modulated signals, characterized in that said signal processor includes the cascade connection of said first means and second means (DVTC), said second means being adapted to perform a predetermined transformation of said output vectors according to a value indicated by a predetermined one of said output vectors.
10. Signal processor according to claim 9, characterized in that said plurality of output vectors are represented in said signal vector plane by a plurality of orthogonal coordinate pairs, and that said first means is adapted to process said plurality of orthognal coordinate pairs by means of a fast fourier transform of said input signal.
11. Signal processor according to claim 9, characterized in that said predetermined transformation comprises a rotation in said signal vector plane of said output vectors over respective predetermined angles.
12. Signal processor according to claim 11, characterized in that said predetermined transformation further comprises a multiplication of the amplitudes of said output vectors with respective predetermined factors.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP94203724.3 | 1994-12-22 | ||
| EP94203724A EP0719001A1 (en) | 1994-12-22 | 1994-12-22 | DMT modulator |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA2162326A1 true CA2162326A1 (en) | 1996-06-23 |
Family
ID=8217479
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA002162326A Abandoned CA2162326A1 (en) | 1994-12-22 | 1995-11-07 | Signal processor |
Country Status (5)
| Country | Link |
|---|---|
| US (1) | US5768318A (en) |
| EP (1) | EP0719001A1 (en) |
| AU (1) | AU703482B2 (en) |
| CA (1) | CA2162326A1 (en) |
| IL (1) | IL115799A (en) |
Families Citing this family (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0735731B1 (en) * | 1995-03-31 | 2004-05-12 | Victor Company Of Japan, Limited | Multicarrier modulator demodulator, with arrangements for reducing peak power |
| EP0836303B1 (en) * | 1996-10-14 | 2003-02-26 | Ntt Mobile Communications Network Inc. | Method and apparatus for reduction of peak to average power ratio |
| FR2758030B1 (en) | 1996-12-31 | 1999-03-26 | Sgs Thomson Microelectronics | METHOD AND DEVICE FOR SHAPING A CLIPPING NOISE OF A MULTI-PORTABLE MODULATION |
| US6466923B1 (en) | 1997-05-12 | 2002-10-15 | Chroma Graphics, Inc. | Method and apparatus for biomathematical pattern recognition |
| US6314102B1 (en) | 1997-07-10 | 2001-11-06 | Alcatel | Telecommunications system for providing both narrowband and broadband services to subscribers |
| US6081502A (en) * | 1997-09-18 | 2000-06-27 | Orckit Communications Ltd. | Method and apparatus for reducing probability of clipping |
| JPH11298436A (en) * | 1998-02-13 | 1999-10-29 | Sony Corp | Modulation method, modulating device, demodulation method and demodulating device |
| US6911186B2 (en) * | 1998-11-05 | 2005-06-28 | Sharper Image Corporation | Electro-kinetic air transporter and conditioner device with enhanced housing configuration and enhanced anti-microorganism capability |
| WO2000072546A1 (en) * | 1999-05-21 | 2000-11-30 | Fujitsu Limited | Signal transmission apparatus |
| US6674810B1 (en) | 1999-05-27 | 2004-01-06 | 3Com Corporation | Method and apparatus for reducing peak-to-average power ratio in a discrete multi-tone signal |
| AU6039800A (en) * | 1999-07-08 | 2001-01-30 | Telefonaktiebolaget Lm Ericsson (Publ) | Transmit power control for mcpa-equipped base stations |
| US6570929B1 (en) | 1999-07-08 | 2003-05-27 | Telefonaktiebolaget Lm Ericsson (Publ) | Power control scheme for maximizing carrier signal-to-noise ratio in multicarrier transmitters |
| US6694148B1 (en) | 1999-07-26 | 2004-02-17 | Telefonaktiebolaget Lm Ericsson (Publ) | Transmit power control for MCPA-equipped based stations |
| CA2291493A1 (en) * | 1999-12-03 | 2001-06-03 | Catena Technologies Canada, Inc. | Peak to average power ratio reduction in communication systems |
| GB2372917A (en) * | 2001-03-02 | 2002-09-04 | Mitel Corp | Discrete multi-tone peak reduction |
| US7075978B2 (en) * | 2001-08-06 | 2006-07-11 | Broadcom Corporation | Multi-tone transmission |
| US7133443B2 (en) * | 2001-08-06 | 2006-11-07 | Broadcom Corporation | Multi-tone transmission |
| US7020188B2 (en) | 2001-08-06 | 2006-03-28 | Broadcom Corporation | Multi-tone transmission |
| JP2006314081A (en) * | 2005-04-08 | 2006-11-16 | Matsushita Electric Ind Co Ltd | Communications apparatus and communication method thereof |
| JP4635947B2 (en) * | 2005-10-12 | 2011-02-23 | パナソニック株式会社 | Power line communication device, integrated circuit, and power line communication method |
| JP5152967B2 (en) | 2007-10-12 | 2013-02-27 | パナソニック株式会社 | COMMUNICATION METHOD, COMMUNICATION DEVICE, AND COMMUNICATION SYSTEM, |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4064464A (en) * | 1976-04-13 | 1977-12-20 | Westinghouse Electric Corporation | Amplitude stabilized power amplifier |
| US4438511A (en) * | 1980-11-10 | 1984-03-20 | Telebit Corporation | Packetized ensemble modem |
| US4679227A (en) * | 1985-05-20 | 1987-07-07 | Telebit Corporation | Ensemble modem structure for imperfect transmission media |
| GB9218874D0 (en) * | 1992-09-07 | 1992-10-21 | British Broadcasting Corp | Improvements relating to the transmission of frequency division multiplex signals |
| US5302914A (en) * | 1992-10-20 | 1994-04-12 | At&T Bell Laboratories | Method and apparatus for reducing the peak-to-average power in multi-carrier RF communication systems |
| JPH07283743A (en) * | 1994-04-12 | 1995-10-27 | Fujitsu Ltd | Multicarrier radio transmission device |
-
1994
- 1994-12-22 EP EP94203724A patent/EP0719001A1/en not_active Withdrawn
-
1995
- 1995-10-27 IL IL11579995A patent/IL115799A/en not_active IP Right Cessation
- 1995-11-07 CA CA002162326A patent/CA2162326A1/en not_active Abandoned
- 1995-11-20 US US08/560,938 patent/US5768318A/en not_active Expired - Fee Related
- 1995-12-14 AU AU40416/95A patent/AU703482B2/en not_active Ceased
Also Published As
| Publication number | Publication date |
|---|---|
| IL115799A (en) | 1998-12-27 |
| EP0719001A1 (en) | 1996-06-26 |
| US5768318A (en) | 1998-06-16 |
| IL115799A0 (en) | 1996-01-19 |
| AU4041695A (en) | 1996-06-27 |
| AU703482B2 (en) | 1999-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2162326A1 (en) | Signal processor | |
| US5018088A (en) | Adaptive locally-optimum detection signal processor and processing methods | |
| EP0396662B1 (en) | Adaptive waveform radar | |
| US5727026A (en) | Method and apparatus for peak suppression using complex scaling values | |
| US6526105B1 (en) | Time domain equalization for discrete multi-tone systems | |
| US7787557B2 (en) | Apparatus for equalizing clipping noise signals of receiver systems and method thereof | |
| US5383224A (en) | Cross polarization interference canceller | |
| US5386495A (en) | Method and apparatus for determining the signal quality of a digital signal | |
| US7187722B2 (en) | Power reduction | |
| US7916801B2 (en) | Time-domain equalization for discrete multi-tone systems | |
| US7809078B2 (en) | OFDM modulator | |
| US8363704B1 (en) | Method of transmitting information using a bandwidth limited communications channel | |
| US7636389B2 (en) | Systems and methods that provide frequency domain supplemental training of the time domain equalizer for DMT | |
| CA2248717A1 (en) | Multiband detector | |
| JP3388938B2 (en) | Diversity receiver | |
| US7327793B2 (en) | DMT peak reduction without affecting transmission signal | |
| US6947530B1 (en) | Adaptive pilot tone location in a transmission system | |
| JPS648937B2 (en) | ||
| US8175198B2 (en) | Matched filter and receiver | |
| Van Saders et al. | Study of nonlinear variants of the least mean square (LMS) algorithm | |
| KR0152273B1 (en) | Transmission reception apparatus of image signal using adaptive factor | |
| Bozzoni et al. | A general analysis of the performance of the oscillating limiter with noiseless signals | |
| BE1007464A3 (en) | Transmission system with quality determination of the received signal | |
| Ghobrial et al. | Discrete wavelet transform domain adaptive decision feedback equalization | |
| JPH05292138A (en) | Receiver estimating maximum likelihood series |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FZDE | Discontinued |