JPS641982B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an adaptive equalizer used in polyphase multilevel modulation.

In an adaptive equalizer for multi-phase multilevel modulation, if there is a lot of initial intersymbol interference, it is almost impossible to make an accurate judgment even if the judgment is made at the equalizer output, and equalization using the judgment result is difficult. There is a problem that the device cannot be corrected correctly.

An object of the present invention is to provide an equalizer that can converge even if initial intersymbol interference is large, by using a new determination method instead of the conventional determination method corresponding to transmitted symbols.

The present invention will be explained below using the polyphase multilevel modulation recommended by CCITTV-29 as an example.

The two series of demodulated waveforms of multi-phase multi-level modulation are expressed as trajectories on a two-dimensional plane with both in-phase and orthogonal components as two axes. In the example used here, each symbol is represented by (a ¹ to ^{a 16} ) on the two-dimensional plane shown in FIG. 1 and has an arrangement. On the other hand, let us consider intersymbol interference b ¹ to b ¹⁶ that is distributed within a constant circumference.
In the vicinity of the center point, intersymbol interference from various directions intersect with each other, making it almost impossible to make a correct determination. However, if we look at the outer level, we can see that most of the eight-phase symbols can be correctly identified. Therefore, only when the received waveform is placed outside the circle C, that is, only when the amplitude value of the received waveform (distance from the origin in FIG. 1) is larger than the radius of the circle C in FIG. If 8-phase determination is performed, 8-phase determination with fewer errors is possible, so if the adaptive equalizer is modified using this determination result, convergence can be easily reached.

Here, the determination as to whether the signal is outside or inside the circle C is expressed by calculating the square sum of the orthogonal and in-phase signals and comparing this with an evaluation value expressed as the square value of the radius of the circle C.

The present invention has a feature that adaptive equalization with a wide pull-in range is possible with a configuration using the above-mentioned means.

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 2 is a block diagram showing the entire embodiment of the present invention. The waveforms coming in via terminals 1 and 2 enter equalizer 3. The two series of signals from the equalizer 3 are each branched into three and input into an evaluation value calculation circuit 4, and the calculated evaluation value is input into a comparison circuit 8. Comparison circuit 8 compares the constant amplitude value input from terminal 7, that is, the radius of circle C in Figure 1, and determines 1' if the input from calculation circuit 4 is large, and O' if small Output to 5. In the determination circuit 5, the input from the comparison circuit 8 is 1'
Only in this case, polyphase modulation is determined according to the output of the equalizer 3 and the output thereof is input to the correction circuit 6. The output of the equalizer 3 and its internal state are also input to the correction circuit 6, and from these signals the equalizer 3
A signal is output that changes the characteristics of. Adaptive equalization is performed using such means.

FIG. 3 is a block diagram showing a more detailed embodiment of the calculation circuit 4. Equalizer outputs enter from terminals 41 and 42 and are added together by an adder 45 via squarers 43 and 44, respectively. By comparing this addition result with a constant value, it is possible to determine whether the received waveform is inside or outside the circle C shown in FIG. Although the example here uses the sum of squares as the evaluation value, it is also possible to use other evaluation values such as the sum of absolute values.

FIG. 4 shows an implementation including the equalizer 3 and the correction circuit 6 in the case where a 3-tap two-dimensional transversal type equalizer is used as the equalizer 3, and correction is performed based on the minimum mean square sum criterion. FIG. 2 is a block diagram showing an example. Signals coming from terminals 1 and 2 are sequentially input to delay elements 31, 33 and 32, 34, respectively. Signals branched from the input and output of each delay element are input to complex multipliers 35, 36, and 37.
The complex multiplier performs multiplication with the tap gain input to obtain two series of outputs, and the results are input to accumulators 38 and 39.
9 and equalized outputs are obtained at terminals 40 and 50. Equalizer 3 includes delay elements 31, 32, 33, 3
4. Complex multipliers 35, 36, 37 accumulators 38, 3
Consists of 9. The equalizer outputs output to terminals 40 and 50 are simultaneously input to an error calculation circuit 63, and are output to terminal 6.
1 and 62 are calculated, and an error signal is input to complex multipliers 64, 65, and 66. The signals branched from the delay elements are also input to the complex multipliers 64, 65, and 66, respectively, and complex multiplication is performed, and the outputs are sent to the fixed gain circuits 67, 6.
Integrator 7 via 8, 69, 70, 71, 72
3, 74, 75, 76, 77, and 78 to obtain a new tap gain at each integrator output. The correction circuit 6 includes an error calculation circuit 63 and a complex multiplier 64, 6.
5, 66, fixed gain circuit 67, 68, 69, 7
0, 71, 72, integrator 73, 74, 75, 7
Consists of 6, 77, and 78. Also, the complex multipliers 35, 36, 37, 64, 6 used here
5 and 66 are all constructed of four multipliers 311, 312, 313, 314 and two adders 315, 316 as shown in FIG. The case described here is a case in which a three-tap two-dimensional transversal type is corrected using the minimum mean-squares criterion, but configurations using other shapes of equalizers and other correction circuits are also possible. It is possible.

With the configuration described above, the present invention provides an adaptive equalizer with a wide pull-in range.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing a two-dimensional plane shown to explain the locus of a received signal, FIG. 2 is a block diagram showing an embodiment of the present invention, and FIG. 3 is a block diagram showing an embodiment of an evaluation value calculation circuit. , FIG. 4 is a block diagram showing one embodiment of the equalizer and correction circuit, and FIG. 5 is a block diagram showing an example of the complex multiplier. In the figure, 3 is an equalizer, 4 is an evaluation value calculation circuit, 8 is a comparison circuit, 5 is a judgment circuit, 6 is a correction circuit, 43, 44, 311, 312, 313, 31
4 is a multiplier, 45, 315, 316 is an adder, 3
1, 32, 33, 34 are delay elements, 35, 36,
37, 64, 65, 66 are complex multipliers, 38, 3
9 is an accumulator, 63 is an error detection circuit, 67, 68,
69, 70, 71, 72 are fixed gain circuits, 7
3, 74, 75, 76, 77, and 78 are integrators.

Claims (1)

[Claims] 1 In a receiver for multi-phase multi-level modulation data transmission,
an equalizer that inputs demodulated two-series signals and outputs two-series signals; and calculates an evaluation value indicating the amplitude of the multiphase multilevel modulation signal from the two-series output signals of the equalizer. an evaluation value calculation circuit; a comparison circuit that compares the evaluation value with a value indicating a constant amplitude; and a correction circuit that adaptively changes the characteristics of the equalizer using the output of the determination circuit, the output of the equalizer, and the internal signal of the equalizer. Features an adaptive equalizer.