JP4966121B2 - Received signal equalizer - Google Patents

Description

  The present invention relates to an equalizer that compensates for distortion of a received signal.

  Wireless communication systems using digitally modulated signals are widely used. In an environment where a wireless communication system is provided, a plurality of wireless signals transmitted from the same transmitting device and passing through propagation paths having different characteristics may be received by the receiving device. Under such a multipath environment, there arises a problem that the receiving apparatus cannot output a digital signal satisfactorily. In addition, when at least one of the receiving device and the transmitting device is a mobile communication terminal device, the propagation path varies with the movement of the mobile communication terminal device, and this problem becomes particularly significant.

  In order to avoid such a problem, an adaptive equalizer that compensates for distortion of a received signal based on a multipath environment is used for the receiving apparatus. The adaptive equalizer compensates for distortion of the received signal using a reference signal whose value changes in a predetermined pattern transmitted from the transmission apparatus.

  The adaptive equalizer stores in advance a reception-side reference signal for collation with a reference signal transmitted from the transmission device. The adaptive equalizer performs equalization processing on the received signal so as to reduce an error from the receiving side reference signal, and outputs a signal. The receiving device outputs a digital signal based on the signal output from the adaptive equalizer.

  According to such processing of the adaptive equalizer, it is possible to satisfactorily output a digital signal from the received signal even in a multipath environment.

Patent No. 2503726 Patent No. 2503715

  The reference signal is composed of a plurality of reference values, and the reference value changes at a predetermined time interval and a predetermined pattern. Here, when the number of reference values included in the reference signal is small, there is a possibility that the process of compensating for distortion of the received signal by the adaptive equalizer may be incomplete. Therefore, it is conceivable to increase the reference value included in the reference signal. However, the time interval of the reference value is determined by the digital modulation method adopted by the wireless communication system. Therefore, increasing the number of reference values included in the reference signal increases the time length of the reference signal. In a wireless communication system, a reception signal having a predetermined time length is assigned to a receiving device. Therefore, when the time length of the reference signal is increased, the time length of the data signal including the transmission target data is inevitably shortened, resulting in a problem that the information transmission capacity is reduced.

  The present invention has been made for such a problem. That is, an object of the present invention is to provide a received signal equalizer that can compensate for distortion of a received signal with a reference signal having a short time length.

The present invention includes a feedforward calculation unit on the received signal and outputs by performing an operation by the feed-forward tap coefficients, and the feedback arithmetic unit with respect to the reference signal output by performing an operation by the feedback tap coefficients, the feed-forward calculation unit facilities and signal output, and signal the feedback arithmetic unit outputs a and an adder summing unit for summing sum and outputs the equalization process signals the adder summing unit outputs to the received signal In the received signal equalizer that outputs the received reference signal, the reference signal included in the received signal is input to the received signal equalizer. When the data signal included in the received signal is input to the received signal equalizer, the addition is performed. A signal selection unit that outputs, as the reference signal, a signal obtained by performing symbol position determination processing on the signal output from the measuring unit, wherein the feedforward calculation unit is one feedforward delay to which the received signal is input Or a plurality of stages of feed-forward delays to which the received signal is input at the first stage, and feed-forward taps provided to the signals output from the corresponding feed-forward delays. A feedforward multiplier that multiplies a coefficient and outputs a multiplication result, and outputs a multiplication result of the feedforward multiplier, wherein the feedback calculation unit sets m to an integer of 2 or more, and m stages of feedback delays And a first feedback delay device for receiving the reference signal. The reference signal is input to the feedback delay unit of the first stage included in the feedback delay unit of the first stage included in the feedback delay group of the feedback stage, or a plurality of feedback delay groups each including m feedback delay units. A plurality of stages of feedback delay units, a feedback multiplier provided corresponding to the feedback delay group, multiplying a signal output from the corresponding feedback delay group by a feedback tap coefficient, and outputting a multiplication result; , And outputs a multiplication result of a feedback multiplier, and the received signal equalizer obtains a difference between the signal output from the summing unit and the reference signal and outputs it as an error signal. And the error signal multiplied by the feedforward tap coefficient by the feedforward multiplier A tap coefficient calculation unit for obtaining a feedforward tap coefficient and a feedback tap coefficient based on the signal before being multiplied by the feedback tap coefficient by the feedback multiplier, and the receiving side reference signal It has participated Telc at symbol time intervals of the received signal, further comprising a complement between reference values to interpolate between two of the reference value in the first time interval of m minutes symbol period, the feedback The delay time in the delay unit is 1 / m times the symbol period and is a natural number multiple of the delay time in the feedforward delay unit, and the tap coefficient calculation unit receives the reception from the signal selection unit. When a side reference signal is output as the reference signal, feedforward is performed at a time interval of 1 / m of the symbol period. Tsu Request up coefficients and feedback tap coefficients, characterized in that.

  According to the present invention, distortion of a received signal can be compensated for by a reference signal having a short time length.

  FIG. 1 shows a configuration of a receiving apparatus according to an embodiment of the present invention. The receiving apparatus includes a radio signal receiver 10, an A / D converter 12, a low-pass filter 14, a buffer 16, a timing detector 22, an adaptive equalizer 18, and a determiner 20. The receiving device is used in a wireless communication system that can transmit a signal from one transmitting device to a plurality of receiving devices. In a wireless communication system, digital modulation signals of modulation schemes such as QPSK, QAM, MSK, DQPSK are transmitted / received.

  The digital modulation signal is modulated by associating the value of the digital signal with a change in the amplitude or phase of the carrier wave. A set of amplitude and phase of a digital modulation signal associated with a digital signal value is called a symbol. A time interval for associating a symbol with a change in amplitude or phase of a carrier wave is called a symbol period.

  FIG. 2 shows the configuration of the frame Fr included in the signal received by the receiving device. A predetermined time position of the frame Fr includes a reference signal Ref for performing frame synchronization detection and the like, and a data signal Dt to be transmitted is included after the reference signal Ref. The reference signal Ref and the data signal Dt can be included at any time position in the frame Fr.

  FIG. 3 shows an example of a reference signal in the case where the QPSK modulation method is adopted, by an in-phase component signal I and a quadrature component signal Q. The angle formed by the vector on the IQ plane and the I axis, each component of which is the value of the in-phase component signal I and the quadrature component signal Q, indicates the phase angle of the digital modulation signal with reference to the carrier phase.

  The in-phase component signal I in FIG. 3 takes values of 1, 1, -1, 1, 1, -1, -1, -1, 1, and -1 for each symbol period Ts. Further, the orthogonal component signal Q takes values of -1, 1, -1, 1, 1, -1, 1, 1, 1, and -1 for each symbol period Ts. That is, the illustrated reference signal is composed of 10 symbols. Each value of the in-phase component signal I and the quadrature component signal Q of the reference signal can be arbitrarily determined and is determined in advance at the design stage of the wireless communication system.

  Next, processing in which the receiving apparatus receives a radio signal and acquires a digital signal will be described. The radio signal receiving unit 10 receives a radio signal transmitted from a transmission device. Then, the received signal is decomposed into an in-phase component signal I and a quadrature component signal Q, and output to the A / D converter 12 respectively. The A / D converter 12 converts the input signal into a digital complex signal having the in-phase component signal I as a real part and the quadrature component signal Q as an imaginary part. Then, the digital complex signal is output to the low-pass filter 14 as a complex reception signal R = I + jQ. The low-pass filter 14 outputs to the buffer 16 a complex reception signal y obtained by reducing unnecessary signals in the high frequency band included in the complex reception signal R.

  The buffer 16 stores a signal having a predetermined time length, delays the signal by the time length, and outputs the delayed signal to the adaptive equalizer 18. The timing detection unit 22 generates a reference timing signal indicating the timing at which the reference signal is input to the adaptive equalizer 18 based on the complex reception signal y stored in the buffer 16 and outputs the reference timing signal to the adaptive equalizer 18. To do. The reference timing signal can be generated based on a comparison between the time waveform of the signal stored in the buffer 16 and the known time waveform of the reference signal.

  The adaptive equalizer 18 processes the complex received signal y according to the timing indicated by the reference timing signal, and outputs it to the determiner 20. The determiner 20 determines a symbol of the signal output from the adaptive equalizer 18 and outputs a digital value corresponding to the symbol as a digital signal.

  The adaptive equalizer 18 executes processing for compensating for distortion of the received signal based on the multipath environment. Here, the configuration of the adaptive equalizer 18 and the processing executed by the adaptive equalizer 18 will be described.

  FIG. 4 shows the configuration of the adaptive equalizer 18. The adaptive equalizer 18 includes feedforward delay units 24-1 to 24-4, feedforward multipliers 26-0 to 26-4, feedback delay units 28-1 to 28-m (m is a natural number of 2 or more), A feedback multiplier 30, an addition and summation unit 32, a tap coefficient calculation unit 34, a reception-side reference signal output unit 36, a signal selection unit 38, an error calculation unit 40, and a hard decision unit 42 are configured.

  The feedforward delay units 24-1 to 24-4, the feedforward multipliers 26-0 to 26-4, and the summing unit 32 are based on the time when the signal is output from the feedforward multiplier 26-0 or the current or Synthesize future signals. The feedback delay units 28-1 to 28-m, the feedback multiplier 30, and the summing unit 32 synthesize past signals based on the time when the signal is output from the feedforward multiplier 26-0.

  The complex reception signal y is input to the tap coefficient calculation unit 34, the feedforward delay unit 24-4, and the feedforward multiplier 26-4. The feedforward delay units 24-4, 24-3, and 24-2 delay the signal by a time TF and output the delayed signal to the tap coefficient calculation unit 34, the subsequent feedforward delay unit, and the subsequent feedforward multiplier. The feedforward delay unit 24-1 delays the signal by a time TF and outputs the delayed signal to the tap coefficient calculation unit 34 and the feedforward multiplier 26-0. Here, the delay time TF in the feedforward delay units 24-1 to 24-4 is set to 1 / n of the symbol period Ts of the complex reception signal y. However, n is a natural number of 2 or more.

The feedforward multipliers 26-4, 26-3, 26-2, 26-1, and 26-0 are respectively conjugate complex numbers F4 ^* , F3 ^* , of feedforward tap coefficients obtained by the tap coefficient calculating unit 34. F2 ^* , F1 ^* , and F0 ^* are multiplied by the input signal and output to the summing unit 32.

  The signal selection unit 38 selects either the reception-side reference signal output from the reception-side reference signal output unit 36 or the hard decision signal output from the hard decision unit 42, and uses the feedback delay device 28-1 and the error as the reference signal s. It outputs to the calculation part 40. Feedback delay units 28-1 to 28-m-1 delay the signal by time TB and output the delayed signal to the subsequent feedback delay unit. The feedback delay unit 28-m delays the signal by the time TB and outputs the delayed signal to the tap coefficient calculation unit 34 and the feedback multiplier 30.

  Here, the delay time TB in the feedback delay units 28-1 to 28-m is set to 1 / m of the natural number of the symbol period Ts of the complex reception signal y. The natural number n is a natural number k times the natural number m. That is, the delay time TB is a natural number k times the delay time TF, and there is a relationship of TB = Ts / m, n = k · m, TF = Ts / n, and TB = k · TF.

The feedback multiplier 30 multiplies the signal output from the feedback delay unit 28-m by the conjugate complex number B0 ^* of the feedback tap coefficient B0 obtained by the tap coefficient calculation unit 34, and outputs the result to the summing unit 32.

  The summing unit 32 adds and sums the signals output from the feedforward multipliers 26-0 to 26-4 and the feedback multiplier 30, and outputs the summed signal z to the error calculation unit 40 and the hard decision unit 42. .

  The error calculation unit 40 obtains a signal obtained by subtracting the sum total signal z from the reference signal s output from the signal selection unit 38, and outputs the signal to the tap coefficient calculation unit 34 as an error signal e.

  The reception side reference signal output unit 36 generates a reception side reference signal and outputs it to the signal selection unit 38. Here, the reception side reference signal output by the reception side reference signal output unit 36 will be described. FIG. 5 shows an example of the receiving side reference signal corresponding to the reference signal shown in FIG. Here, an example is shown in which m = 2, that is, the delay time TB of the feedback delay units 28-1 and 28-2 is one half of the symbol period Ts. The reference signal shown in FIG. 3 has a predetermined reference value for each symbol period Ts, whereas the reception-side reference signal output from the reception-side reference signal output unit 36 is the delay time TB of the feedback delay device. It has a predetermined value for every same time. The reference value for each symbol period Ts of the reception side reference signal is the same as the reference value of the reference signal transmitted from the transmission apparatus, and the interpolation reference value is inserted at the interpolation time point in the symbol period. In FIG. 5, the reference value for each symbol period Ts is indicated by a solid white circle, and the interpolation reference value is indicated by a dashed white circle.

  This interpolation reference value can be determined as follows based on the value indicated by the interpolation time point of the reference signal included in the signal transmitted from the transmission device and received by the reception device.

  The reference signal generated by the transmission device is subjected to filter processing by a filter included in the transmission device and transmitted. In the receiving apparatus, processing is performed by the low-pass filter 14 and the reference signal is input to the adaptive equalizer 18. The characteristics of the filter included in the transmission device and the characteristics of the low-pass filter 14 are determined in advance in the design of the wireless communication system. Therefore, if the reference signal generated by the transmitter is known, an interpolation reference value for the reference signal input to the adaptive equalizer 18 can be determined based on the known characteristics of these filters.

  The reception-side reference signal output unit 36 stores a reception-side reference signal in which an interpolation time point within a symbol period is interpolated with an interpolation reference value. Then, the timing detection unit 22 outputs a reference timing signal and outputs a receiving side reference signal.

  Here, the reception-side reference signal in the case where the delay time TB of the feedback delay device is 1/2 of the symbol period Ts has been described. More generally, when the delay time TB of the feedback delay units 28-1 to 28-m is 1 / m of the symbol period Ts, m−1 interpolation reference values are included in the time interval during the symbol period. Inserted with TB.

  According to the processing of the adaptive equalizer 18 as described above, the tap coefficient calculation unit 34 includes feedforward multipliers 26-4, 26-3, 26-2, 26-1, 26-0, and a feedback multiplier 30. And a signal vector U = (u4, u3, u2, u1, u0, v0) whose components are signals u4, u3, u2, u1, u0, and v0 before being multiplied by tap coefficients, respectively, and an error signal e is entered.

  The tap coefficient calculation unit 34 obtains feedforward tap coefficients F4 to F0 and a feedback tap coefficient B0 based on the signal vector U and the error signal e. As an algorithm for obtaining the tap coefficient, an LMS algorithm, an RLS algorithm, or the like is preferable. The adaptive equalizer 18 executes one step of the algorithm for each delay time TB of the feedback delay units 28-1 to 28-m for the reference signal. For the data signal, one step of the algorithm is executed every symbol period Ts. Here, an example of the RLS algorithm will be described.

The feedforward tap coefficients F4 to F0 and the feedback tap coefficient B0 calculated in the i-th calculation step are expressed by the following tap coefficient vector W = (F4, F3, F2, F1, F0, B0). It is expressed as (Equation 1). However, i is an integer of 1 or more.
(Equation 1) W (i) = W (i-1) + K (i) e ^* (i)

Here, * with the upper right means taking a conjugate complex number. The vector K is a vector obtained through the matrix P based on the following (Equation 2) and (Equation 3).
(Equation 2) P (i) = (P (i−1) −K (i) U ^H (i) P (i−1)) / λ
(Equation 3) K (i) = P (i−1) U (i) / (λ + U ^H (i) P (i−1) U (i))

Here, H with the upper right means transposing the vector and taking the conjugate complex number. Further, λ is a forgetting factor indicating how much the value of the tap coefficient vector W obtained in the previous calculation step contributes to the current calculation step. By determining P (0) = δ ⁻¹ I (δ ⁻¹ is an arbitrary initial value and I is a unit matrix) as an initial value matrix P (0), a tap coefficient vector W is obtained for each calculation step. Can do.

  Next, processing that the adaptive equalizer 18 performs on the complex reception signal y will be described. When the reference timing signal is output from the timing detection unit 22, the reception side reference signal output unit 36 outputs the reception side reference signal. Further, the signal selection unit 38 selects and outputs a signal output from the receiving side reference signal as the reference signal s.

The tap coefficient calculation unit 34 obtains a tap coefficient vector W = (F4, F3, F2, F1, F0, B0) for each time TB according to the above algorithm. The conjugate complex numbers F4 ^{* to} F0 ^* of the feedforward tap coefficients F4 to F0 are output to the feedforward multipliers 26-4 to 26-0, respectively, and the conjugate complex number B0 ^* of the feedback tap coefficient B0 is output to the feedback multiplier 30. To do.

  After the input of the reference signal is started and until a time corresponding to the time length of the reference signal elapses, the tap coefficient calculation unit 34 uses the tap coefficient vector based on the algorithm with the reception-side reference signal as the reference signal s. Update W.

  At this time, the time interval TB for executing the algorithm is the same as the delay time TB of the feedback delay units 28-1 to 28-m. Therefore, each time the algorithm is executed, one value of the receiving side reference signal is output to the feedback multiplier 30. That is, since the receiving side reference signal has a value for each time interval TB, the value included in the receiving side reference signal can be reflected in the algorithm without omission by executing the algorithm at the time interval TB.

  When the delay time TF of the feedforward delay units 24-1 to 24-4 is one half of the symbol period Ts and the receiving side reference signal shown in FIG. 5 is output, that is, n = m = 2 and A case where k = 1 will be described. In this case, the delay time TF of the feedforward delay devices 24-1 to 24-4 and the delay time TB of the feedback delay devices 28-1 to 28-m coincide, and the time is one half of the symbol period Ts. . The tap coefficient calculation unit 34 executes one step of the algorithm for obtaining the tap coefficient vector W every time Ts / 2. Thereby, 19 steps are executed by 10 reference values and 9 interpolation reference values that the receiving side reference signal has for each symbol period Ts, that is, 19 values included in the receiving side reference signal, In each step, the value of the tap coefficient vector W is updated.

  In the adaptive equalizer 18 according to the present embodiment, while the reference signal is input to the adaptive equalizer 18, the tap coefficient vector W is converted to the time TB based on the algorithm using the reception side reference signal as the reference signal s. Every one is required. As described above, the reception side reference signal is a signal obtained by interpolating interpolation time points within a symbol period with an interpolation reference value.

  As shown in FIG. 7, the reception side reference signal used in the conventional adaptive equalizer 18 has the same reference value as the transmitted reference signal for each symbol period, and has an interpolation reference value. It was not. Therefore, when the conventional reception side reference signal is used, while the reference signal is input to the adaptive equalizer 18, the tap coefficient vector is based on the 10 values of the reception side reference signal for each symbol period. Will be updated 10 times.

  In the receiving side reference signal in the present embodiment, the interpolation time point of the symbol period is interpolated by the interpolation reference value. Therefore, the number of updates of the tap coefficient vector W can be increased by the number of interpolated values under the condition that the time length of the receiving side reference signal is constant. For example, in the above example in which the delay time TB of the feedback delay units 28-1 to 28-m is half the symbol period Ts and the receiving side reference signal shown in FIG. 5 is output, the interpolation reference value is Since the number is nine, the number of steps of the algorithm can be increased by nine times compared to the case of using the conventional reception side reference signal. As a result, the tap coefficient vector W can be updated more times with the reference signal having the same time length as the conventional one, and the value of the tap coefficient vector can be quickly converged. Therefore, the process for compensating for the distortion of the received signal can be performed quickly.

  Next, processing that the adaptive equalizer 18 performs on the data signal will be described. The time corresponding to the time length of the reference signal elapses after the reference timing signal is output, and the signal selection unit 38 selects and outputs the hard decision signal as the reference signal s.

  Here, the process executed by the hard decision unit 42 will be described. The hard decision unit 42 corrects the value of the addition sum signal z output from the addition summation unit 32 by the following symbol position determination process, and outputs it to the signal selection unit 38 as a hard decision signal.

  The symbol position determination process is a process of correcting and outputting the position on the IQ plane indicated by the signal to be processed to the nearest one of the symbol positions predetermined in the wireless communication system. For example, symbols defined in the QPSK modulation scheme are (I, Q) = (1, 1), (−1, 1), (−1, −1), or (1, 1) on the IQ plane. -1). In this case, when the sum total signal z indicates a position in the first quadrant, the second quadrant, the third quadrant, or the fourth quadrant, the hard summation unit z is (1, 1), ( −1, 1), (−1, −1), or (1, −1) is corrected so as to indicate the position, and is output as a hard decision signal. According to the symbol position determination process, an error component included in the sum total signal z can be removed.

The tap coefficient calculation unit 34 calculates a tap coefficient vector W = (F4, F3, F2, F1, F0, B0) for each symbol period Ts according to an algorithm for calculating a tap coefficient vector, and conjugates the feedforward tap coefficients F4 to F0. Complex numbers F4 ^{* to} F0 ^* are output to feedforward multipliers 26-4 to 26-0, and conjugate complex number B0 ^* of feedback tap coefficient B0 is output to feedback multiplier 30, respectively.

  According to such processing, after the input of the reference signal from the buffer 16 to the adaptive equalizer 18 is completed and the input of the data signal is started, based on the algorithm using the hard decision signal as the reference signal s. A tap coefficient vector W is obtained. Then, the sum total signal z obtained based on the tap coefficient vector W is output to the determiner 20. As a result, the tap coefficient vector W that follows the fluctuation of the propagation path at the time of receiving the data signal is obtained using the value of the tap coefficient vector W that has converged while the reference signal is input to the adaptive equalizer 18 as an initial value. be able to.

  In the above description, the adaptive equalizer 18 including five feedforward multipliers, four feedforward delay units, one feedback multiplier, and m feedback delay units has been described. By applying this configuration, the number of processing stages can be increased. FIG. 6 shows an adaptive equalizer 18 having N feedforward multipliers, N−1 feedforward delay units, M feedback multipliers, and M · (m−1) feedback delay units. (N is a natural number of 2 or more, and M is a natural number). The same components as those in the adaptive equalizer of FIG. 4 are denoted by the same reference numerals. The feed forward multiplier is given a number indicating the number of stages after the sign of “26−”. The feedforward delay unit is given a number indicating the number of stages after the sign “24-”. A number indicating the number of stages is added to the feedback multiplier after the sign “30−”. A number indicating the number of stages is attached to each of the feedback delay devices after each code of “28-1-” to “28-m-”. In this configuration, m feedback delay devices are sandwiched between two adjacent feedback multipliers.

In this case, the tap coefficient calculation unit 34 obtains W = (FN-1, FN-2,..., F0, B0,..., BM-1) as the tap coefficient vector. Conjugate complex numbers F0 ^{* to} FN-1 ^* of tap coefficients F0 to FN-1 are output to Ford forward multipliers 26-0 to 26-N-1. Further, conjugate complex numbers B0 ^{* to} BM-1 ^* of tap coefficients B0 to BM-1 are output to feedback multipliers 30-0 to 30-M-1. By increasing the number of processing stages, the effect of compensating for distortion of the received signal can be enhanced. Further, by increasing the number of processing stages and increasing the number of interpolation reference values for interpolating the reception-side reference signal, the tap coefficient vector W is updated more times with the reference signal having the same time length. And the value of the tap coefficient vector can be quickly converged.

  The receiving apparatus according to the embodiment of the present invention is preferably used in a prefectural / municipal digital mobile communication system. As a standard for prefectural / municipal digital mobile communication systems, for example, ARIB STD-T79 2.0 is defined. In this mobile communication system, a plurality of base station apparatuses that can communicate with each other form a communication cell. The mobile communication terminal apparatus in the cell communicates with the base station in the cell. A mobile communication terminal apparatus can communicate with another mobile communication terminal apparatus existing in the same cell or in another cell via a base station in the cell in which the mobile communication terminal apparatus exists. Each base station also has a function as a communication terminal. Therefore, it is possible to communicate between a person who operates the mobile communication terminal and a person who operates the base station.

It is a figure which shows the structure of the receiver which concerns on embodiment of this invention. It is a figure which shows the structure of the signal received with the receiver which concerns on embodiment of this invention. It is a figure which shows the example of the reference signal at the time of employ | adopting a QPSK modulation system. It is a figure which shows the structure of the adaptive equalizer which concerns on embodiment of this invention. It is a figure which shows the example of the receiving side reference signal which concerns on embodiment of this invention. It is a figure which shows the structure of the adaptive equalizer which made the number of process steps arbitrary. It is a figure which shows the conventional receiving side reference signal.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 Radio signal receiver, 12 A / D converter, 14 Low-pass filter, 16 buffer, 18 Adaptive equalizer, 20 Determinator, 22 Timing detector, 24-1-24-4 Feedforward delay device, 26 −0 to 26-4 Feedforward multiplier, 28-1 to 28-m Feedback delay unit, 30 Feedback multiplier, 32 Addition summation unit, 34 Tap coefficient calculation unit, 36 Reception side reference signal output unit, 38 Signal selection unit , 40 error calculation unit, 42 hard decision unit.

Claims (1)

A feedforward calculation unit for outputting by performing calculation by the feed-forward tap coefficient to the received signal,
A feedback computation unit that performs computation using a feedback tap coefficient on the reference signal and outputs the result,
An addition summing unit that adds and sums the signal output by the feedforward computing unit and the signal output by the feedback computing unit ;
In the received signal equalizer that outputs the signal output from the summing unit as a signal obtained by performing equalization processing on the received signal,
When a reference signal included in the reception signal is input to the reception signal equalizer, a reception-side reference signal is output as the reference signal at a timing according to the reference signal,
When a data signal included in the reception signal is input to the reception signal equalizer, a signal obtained by performing a symbol position determination process on the signal output by the addition summing unit is output as the reference signal. With a signal selector,
The feedforward calculation unit is
One feedforward delay device to which the received signal is input, or a plurality of feedforward delay devices to which the received signal is input in the first stage;
A feedforward multiplier provided corresponding to the feedforward delay unit, multiplying a signal output from the corresponding feedforward delay unit by a feedforward tap coefficient, and outputting a multiplication result, and feedforward Outputs the multiplication result of the multiplier,
The feedback calculation unit includes:
a feedback delay group including m stages of feedback delays, where m is an integer greater than or equal to 2, and the reference signal is input to the first stage feedback delays; or
Each of a plurality of stages of feedback delay elements including m stages of feedback delay elements, and a plurality of stages of feedback delay elements in which the reference signal is input to the first stage feedback delay elements included in the first stage feedback delay elements When,
A feedback multiplier provided corresponding to the feedback delay group, multiplying a signal output from the corresponding feedback delay group by a feedback tap coefficient, and outputting a multiplication result. Output the multiplication result,
The received signal equalizer is
An error calculation unit that obtains a difference between the signal output by the summing unit and the reference signal and outputs the difference as an error signal;
Based on the error signal, the signal before being multiplied by the feedforward tap coefficient by the feedforward multiplier, and the signal before being multiplied by the feedback tap coefficient by the feedback multiplier, the feedforward tap coefficient and the feedback tap coefficient are A tap coefficient calculation unit to be obtained,
The receiving side reference signal is:
Has participated Telc at symbol time intervals of the received signal, and further,
Has Interpolation reference value to interpolate between the two of the reference value in the first time interval of m minutes symbol period,
The delay time in the feedback delay device is 1 / m times the symbol period, and is a natural number multiple of the delay time in the feed forward delay device,
The tap coefficient calculation unit
When the receiving side reference signal is output as the reference signal from the signal selection unit, a feedforward tap coefficient and a feedback tap coefficient are obtained at a time interval of 1 / m of the symbol period. Receive signal equalizer.

Images