Adaptive equalisation

Abstract

This invention is designed for use in transmission of data between downhole module in a wellbore and a controlling module at the surface. The invention provides an apparatus for receiving data signals from a telemetry module comprising first and second adaptive equalisers, and in which in a first modulation mode the coefficients of the first adaptive equaliser are updated until an error signal falls below a predetermined threshold and in a second modulation mode the coefficients of the first adaptive equaliser are locked and coefficients of the second adaptive equaliser are updated to continually minimise an error signal in which the number of bits encoded by the symbols of the first signal in an initial modulation mode is fewer than the number of bits encoded by the symbols of the second signal in a subsequent modulation mode.

Claims

1. A method of adaptive equalisation comprising the steps of: receiving a first signal via a channel in which the received first signal differs from a transmitted first signal due to channel distortion and/or noise, the transmitted first signal comprising a sequence of first symbols encoded by first inphase and quadrature components; repeating the steps of filtering a portion of the received first signal using a first filter having a plurality of first filter coefficients to obtain filtered first inphase and quadrature components of the received first signal; determining the first inphase and quadrature components of the transmitted first signal; determining a first error signal in dependence upon a first difference between the first inphase and quadrature components of the transmitted first signal and the filtered first inphase and quadrature components of the received first signal; updating the first filter coefficients to reduce said determined first error signal to create a reduced first error signal; and locking said first filter coefficients when the reduced first error signal is less than a predetermined threshold; and the method further comprises receiving a second signal via the channel in which the received second signal differs from a transmitted second signal due to the channel distortion and/or noise, the transmitted second signal comprising a sequence of second symbols encoded by second inphase and quadrature components; continually repeating the steps of filtering a portion of the received second signal using said first filter with said locked first filter coefficients to obtain a first filtered received second signal; filtering the first filtered received second signal using a second filter having a plurality of second filter coefficients to obtain filtered second inphase and quadrature components of the received second signal; determining the second inphase and quadrature components of the transmitted second signal; determining a second error signal in dependence upon a second difference between the second inphase and quadrature components of the transmitted second signal and the filtered second inphase and quadrature components of the received second signal; updating the second filter coefficients to reduce the determined second error signal to create a reduced second error signal; and wherein a number of bits encoded by the first symbols of the transmitted first signal in an initial modulation mode is fewer than a number of bits encoded by the second symbols of the transmitted second signal in a subsequent modulation mode.

2. The method according to claim 1, in which the transmitted first signal comprises a QPSK signal.

3. The method according to claim 1, in which the transmitted second signal is a QAM 16 signal.

4. The method according to claim 1, in which the first inphase and quadrature components of the transmitted first signal and the second inphase and quadrature components of the transmitted second signal are determined by using a least means squares algorithm.

5. The method according to claim 1, in which the first filter coefficients and the second filter coefficients are updated in dependence upon the determined first error signal and the determined second error signal respectively together with a factor which is greater than zero and less than 1/input signal power multiplied by a number of filter taps in the respective first filter or second filter.

6. An apparatus for receiving data signals from a telemetry module comprising: a first adaptive equaliser configured to receive a first signal transmitted via a channel in which the received first signal differs from a transmitted first signal due to channel distortion and/or noise, the transmitted first signal comprising a sequence of first symbols encoded by first inphase and quadrature components; the first adaptive equaliser comprising: a first filter configured to filter a portion of the received first signal using the first filter having a plurality of first filter coefficients to obtain received first inphase and quadrature components of the received first signal; a first slicer configured to determine the first inphase and quadrature components of the transmitted first signal; a first error calculator configured to determine a first error signal in dependence upon a first difference between the first inphase and quadrature components of the transmitted first signal and the received first inphase and quadrature components of the received first signal; and a first coefficient updater configured to: update the first filter coefficients to reduce said first error signal and to lock said first filter coefficients when the first error signal is less than a predetermined threshold; the first filter being further configured, once the first filter coefficients are locked, to receive and filter a second signal transmitted by the channel in which the received second signal differs from a transmitted second signal due to the channel distortion and/or noise, the transmitted second signal comprising a sequence of second symbols encoded by second inphase and quadrature components; and a second adaptive equaliser configured to receive a filtered second signal from the first filter, the second adaptive equaliser comprising: a second filter configured to filter a portion of the filtered second signal received from the first filter using a second filter having a plurality of second filter coefficients to obtain received second inphase and quadrature components of the received second signal; a second slicer configured to determine the second inphase and quadrature components of the second transmitted signal; a second error calculator configured to determine a second error signal in dependence upon a second difference between the second inphase and quadrature components of the second transmitted signal and the received second inphase and quadrature components of the received second signal; and a second coefficient updater configured to update the second filter coefficients to reduce said second error signal.

7. The apparatus according to claim 6, in which the transmitted first signal comprises a QPSK signal.

8. The apparatus according to claim 6, in which the transmitted second signal is a QAM 16 signal.

9. The apparatus according to claim 6, in which the first error calculator is configured to determine the first inphase and quadrature components of the transmitted first signal using a least means squares algorithm.

10. The apparatus according to claim 6, in which the first and second coefficient updaters are configured to update the first filter coefficients and the second filter coefficients respectively in dependence upon the first error signal and the second error signal respectively together with a factor which is greater than zero and less than 1/input signal power multiplied by a number of filter taps in the respective first filter or second filter.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a block diagram showing a QAM demodulator with an adaptive equaliser;

(2) FIG. 2 illustrates a conventional adaptive equaliser;

(3) FIGS. 3a, 3b and 3c illustrate sample error calculation using a slicer;

(4) FIGS. 4a and 4b illustrate an equalised QPSK constellation and QAM16 constellation from a signal received over a distorted channel or noisy environment;

(5) FIG. 5 illustrates a two stage adaptive equaliser;

(6) FIG. 6 is a flow chart illustrating steps in a method according to the present invention;

(7) FIG. 7 is a flow chart illustrating further steps in a method according to the present invention;

(8) FIG. 7 illustrates a two stage adaptive equaliser;

(9) and

(10) FIG. 8 illustrates constellation diagrams of the signal before and after each stage of adaptive equalisation.

DETAILED DESCRIPTION

(11) FIG. 1 is a block diagram showing a QAM demodulator with an adaptive equaliser.

(12) A mixer 100 receives an input signal 200 and generates inphase signal 201 and quadrature signal 202.

(13) Signals 201, 202 are down converted by a decimation and RRC filter 102 to generate down converted inphase signal 203 and down converted quadrature signal 204.

(14) A variable decimator 104, a symbol adjuster 105 and a loop filter 103 are used to detect the symbol timing error and sample the symbol at the ideal point to generate inphase symbol sample 205 and quadrature symbol sample 206.

(15) Symbol samples 205, 206 are processed by a phase error detector 109, a phase adjuster 102 and a loop filter 111 to determine a correction value 207 for use by a digital direct synthesiser (DDS) 110. The DDS 110 generates a carrier frequency to feed to the mixer 100.

(16) The symbol sample 205, 206 are processed by a symbol mapper 113 prior and equalised by adaptive equaliser 10 (which will be described in more detail later). Such symbol mappers are also referred to herein as slicers and mapping of a received symbol to the closest symbol is often referred to as slicing. After equalisation by the adaptive equaliser 10 symbol values are processed by a second symbol mapper 106 and a serial decoder 107 to a serial bit stream.

(17) FIG. 2 illustrates a conventional adaptive equaliser. It is to be noted that the terms adaptive equaliser and adaptive filter are used synonymously in the following description.

(18) An adaptive equaliser 10 adjusts filter coefficients of a finite impulse response (FIR) filter 12 based upon the difference between ideal sample values y(n) and the sample values y(n) output by the filter 12.

(19) A signal S(n) is distorted by distortions introduced by a channel 14 and/or by noise to produced received signal x(n). The received signal x(n) is processed by the FIR filter 12 having coefficients w(n) to produce signal y(n) according to the convolution equation:
y(n)=x(n)*w(n)

(20) The resulting signal y(n) is then processed by a slicer 16 as shown in FIGS. 3a, 3b and 3c.

(21) FIG. 3a is a constellation diagram illustrating a received symbol 21, together with a target symbol 20 representing the transmitted symbol. Vector 22 represents the difference between the two symbols otherwise know as the modulation error.

(22) In FIG. 3b a target symbol 23 is represented by inphase and quadrature components of 80. The received symbol 24 has and inphase component 25 I.sub.sample equal to 85 and a quadrature component 26 Q.sub.sample equal to 77. Slicer 16 receives the inphase and quadrature components 25, 26 and determines the target inphase component I and target quadrature component Q of the appropriate target symbol 23 from the received symbol 24 using a conventional least mean squares (LMS) algorithm.

(23) Error Calculator 17 then produces error signal e(n) based upon a calculation of the difference between the received symbol 24 and the target symbol 23:
e(n)_I=I.sub.sampleI
e(n)_Q=Q.sub.sampleQ

(24) Coefficient Updater 18 then updates the coefficients of the FIR filter 12 according to the equation:
w(n+1)=w(n)+u*e(n)*x(n)
where u is a step size chosen such that 0<u<1/input signal power multiplied by the no. of filter taps.

(25) FIGS. 4a and 4b are constellation diagrams illustrating equalized QPSK and QAM16 signals respectively using a conventional adaptive equaliser as illustrated in FIG. 1 for signals sent over the same channel as each other.

(26) As can be seen on FIG. 3a, QPSK constellation points do not overlap with adjacent constellation points. There is clear boundary between the 4 regions of constellation points. From QPSK modulation we could measure the sample error values on all points.

(27) However, for the same channel with a QAM 16 signal, the constellations overlap with adjacent constellation points as shown in FIG. 4b. Therefore when these points are processed by the slicer 16 they will be misidentified and the sample errors will be incorrect because ideal sample is from adjacent constellation point. If the percentage of symbols which are misidentified is greater than 10% adaptive equalization will not converge and becomes unstable.

(28) An improved adaptive equalisation scheme will now be described with reference to FIGS. 5 to 8.

(29) Referring to FIGS. 5 and 6 at step 60 a first signal is received via a channel by a first adaptive equaliser 50. The received signal differs from the transmitted signal due to channel distortion and/or noise. The signal comprises a sequence of symbols encoded by inphase and quadrature components, for example the signal may be a QPSK or QAM16 encoded signal as is well known in the art. In the preferred embodiment the signal is a QPSK signal.

(30) In the preferred embodiment the module starts in 100 kbps bandwidth mode by default. This corresponds to QPSK modulation. When the system is started up the channel conditions are unknown and using QPSK modulation the chance of adaptive equalisation convergence to correct filter coefficients is much better than when using a higher order modulation schemes such as QAM16, QAM64. However if it is believed the channel conditions are good enough it is possible to start with a higher order modulation. The first modulation/bandwidth mode will be referred to below as the initial modulation. The subsequent modulation mode is one in which the number of bits encoded by the symbols of the first signal is fewer than the number of bits encoded by the symbols of the second signal.

(31) At step 61 a portion of the received first signal using a first filter, such as a conventional FIR filter having a plurality of first filter coefficients. Such filtering is conventionally represented by the convolution equation
y(n)=x(n)*w(n)
as described previously where x(n) represent the first signal, w(n) represents the filter coefficients and y(n) represents the filtered signal.

(32) At step 62 the filtered signal is sliced within a first coefficient updater 54 to determine the likely inphase and quadrature components of the originally transmitted signal as described above with reference to FIGS. 2a, 2b and 2c.

(33) At step 63 the difference between the likely (sometimes called the ideal) inphase and quadrature components of the originally transmitted signal and the filtered signal is calculated by the first coefficient updater 54 to produce an error signal e(n).

(34) If the error is less than a predetermined threshold at step 64 then if the and the QAM mode is determined by a decision module 53 to be equal to the initial modulation the first filter coefficients are locked at step 65. Otherwise the first filter coefficients are updated at step 66 as described previously with reference to coefficient updater 18.

(35) Steps 61 to 64 are repeated until the coefficients are locked at step 65.

(36) FIG. 7 illustrates the method steps that are carried out after the first filter coefficients in the first adaptive equaliser have been locked.

(37) At step 70 a second signal is received via a channel. The received signal differs from the transmitted signal due to channel distortion and/or noise. The signal comprises a sequence of symbols encoded by inphase and quadrature components, for example the signal may be QAM16 or higher order QAM encoded signal as is well known in the art. In the preferred embodiment the signal is a QAM16 signal. The number of bits encoded by the symbols of the first signal is fewer than the number of bits encoded by the symbols of the second signal ie the second signal is a higher order QAM signal than the first signal which was used to lock the filter of the first adaptive equaliser.

(38) At Step 71 the signal is filtered by the first adaptive filter 50 which now has the coefficients locked.

(39) FIG. 8 shows QAM16 modulation using a first adaptive equaliser 50 having a FIR filter which has been locked. Before entering QAM 16 mode the QPSK modulation has been able to obtain good convergence of the filter's coefficients for the filter.

(40) It can be seen that the constellation 80 of the second received signal QAM16 is severely affected by noise and Channel distortions. The signal degradation is worse than that of shown in FIG. 8a.

(41) The signal is filter by the filter of the first adaptive equaliser. A constellation diagram 81 of the output signal y(n) is shown. This filtered signal is then processed by a second adaptive equaliser 10 in the conventional way as follows:

(42) At step 72 the filtered second signal is filtered using a second adaptive filter 51 having a plurality of second filter coefficients.

(43) At step 73 the filtered signal is sliced in a second coefficient updater 55 to determine the likely inphase and quadrature components of the originally transmitted signal as described above with reference to FIGS. 2a, 2b and 2c.

(44) At step 74 the difference between the likely (or ideal) inphase and quadrature components of the originally transmitted signal and the filtered signal is calculated by the second coefficient updater 55 to produce a (second) error signal e(n).

(45) At step 75 the (second) error signal e(n) is used by the second coefficient updater to update the second filter coefficients:
w(n+1)=w(n)+u*e(n)*y(n)

(46) A constellation diagram 82 of the output signal y(n) from the second adaptive filer 51 is shown in FIG. 8.

(47) By applying this scheme 200 kbps and 300 kbps bandwidths may be achieved on channels with severe channel distortions and/or severe noise.

(48) It is to be recognised that various alterations, modifications, and/or additions may be introduced into the constructions and arrangements of parts described above without departing from the scope of the present invention as defined in the following claims.