JPS6135742B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a carrier wave regeneration method. In information transmission systems using PSK modulation (phase modulation) or QAM modulation (quadrature amplitude phase modulation that performs phase and amplitude modulation simultaneously),
A carrier recovery circuit is essential in the demodulator on the receiving side. This is because it is necessary to synchronously detect the received signal using the carrier wave regenerated by the carrier wave regeneration circuit and extract the original transmission information. Conventionally, various types of carrier wave regeneration circuits have been proposed, and a typical example is a so-called COSTAS type carrier wave regeneration circuit, which has become mainstream. This Costas-type carrier wave regeneration circuit uses an analog circuit to generate a signal of the sum and difference of the outputs of the phase shift detector, and based on this, a so-called phase error signal (Δ
θ) and form a reference carrier wave. However, carrier wave regeneration circuits of other types, including this Costas type, have the following characteristics: they are mainly composed of analog circuits, and they are basically based on obtaining information about the phase (Δθ). , often accompanied by inconvenient problems. For example, looking at the above characteristics, the fact that analog circuits are the main component has disadvantages such as the fact that it is not easy to integrate them into integrated circuits (ICs). On the other hand, regarding the above-mentioned characteristics, it is basically necessary to obtain information regarding the phase, and it is inconvenient that it is impossible to obtain information regarding the amplitude from the received signal unless a completely separate circuit is provided. In particular, this latter disadvantage is fatal when not only the phase but also the amplitude must be processed simultaneously, as in the QAM modulation method. Therefore, an object of the present invention is to propose a new carrier wave regeneration method based on a concept completely different from the conventional carrier wave regeneration method, and to solve the above-mentioned disadvantages and also make it possible to derive new technical effects. (Later). In accordance with the above object, the present invention provides the conventional X-axis and Y-axis
At least one small orthogonal coordinate system consisting of an x-axis and a y-axis is set in an orthogonal coordinate system consisting of axes, and a reference carrier wave is reproduced from information on the quadrant position of a signal point with respect to the small orthogonal coordinate system. It is characterized by the following. The present invention will be explained below with reference to the drawings. FIG. 1A is a graph illustrating a conventional carrier recovery method (eg, used in a Costas type carrier recovery circuit). In this figure, the X-axis and Y-axis are the axes that define the orthogonal coordinate system, and in the case of 4-phase PSK, signal points appear in any of the 1st, 2nd, 3rd, and 4th quadrants of the orthogonal coordinate system. . Each unique data is displayed depending on which quadrant the signal is in. Typically, the received carrier signal is input to an X-axis phase detector and a Y-axis phase detector, which detects the signal with a reference carrier, while
The x-axis phase detector detects the signal using a carrier wave whose phase is shifted by π/2 from the reference carrier wave.
Signal point S in FIG. 1A is an example of one signal point obtained by these detections. Here, the reference carrier wave is reproduced from the received carrier signal, but when performing the above-mentioned detection, complete phase synchronization must always be achieved between the reference carrier wave and the received carrier signal. If this phase synchronization is not perfect, a so-called phase error occurs and high-quality data transmission cannot be expected. Therefore, the reference carrier wave is usually obtained from a voltage controlled oscillator (VCO), and the control voltage thereof is appropriately adjusted to ensure the complete phase synchronization described above. That is, this control voltage corresponds to the Δθ regeneration circuit in this figure. Δθ is a so-called phase error, and is a deviation between the phase of the set reference carrier wave (the dotted straight line in the figure) and the phase of the received carrier signal that actually arrives (the solid line in the figure). However, since the reference carrier wave is not absolute and fixed, but is extracted and formed from the received carrier signal, the voltage control type is adjusted so that Δθ in this graph always converges to zero. Driving the oscillator equivalently reproduces the carrier wave. In other words, carrier wave regeneration boils down to the problem of how to extract the Δθ and converge it to zero. The matters described above are general and universal matters in carrier wave regeneration, and the present invention is also based on this idea. However, the method of extracting Δθ described above is completely different from the conventional method. FIG. 1B is a graph illustrating the carrier wave recovery method of the present invention. In this figure, the meanings of the X-axis, Y-axis, and signal points S (S1, S2, S3, S4) are the first
It is similar to that in Figure A. Therefore, what is particularly noteworthy is that a small rectangular or oblique coordinate system consisting of an x-axis and a y-axis is introduced at each signal point. Moreover, there may be a plurality of small orthogonal coordinate systems in each quadrant. For convenience of explanation, if only the first quadrant is considered, the signal point S2 in the small orthogonal coordinate system is (-Δ
x, +Δy), and the signal point S4
is at (+Δx, -Δy). This is the first
Corresponding to diagram A, in the case of signal point S2, +Δ
θ, and in the case of signal point S4, there is a phase error of −Δθ. Looking at this from the hardware perspective, the phase error is caused by the comparator (see Figure 3).
With the help of ADX, ADY), a mere “0”
This means that it is expressed by a digital signal of "1". In other words, there is no need to use an analog circuit here. Furthermore, since the digital signal is used as a medium for subsequent signal processing, the use of analog circuits is no longer necessary. In other words, according to the method of the present invention, a carrier wave regeneration circuit mainly composed of digital circuits is realized, and as a result, an integrated circuit (IC)
This makes it easier to Furthermore, since there is no drift inherent in analogue, accuracy can also be improved. Referring again to FIG. 1B, what if the signal point now arrives at position S1 or S3? In this case, these signal points are (+Δx, +Δ
y) and (-Δx, -Δy) regions (quadrants)
Therefore, it is not possible to detect the phase error. This is because these signal points may appear either above or below the dotted line in the figure, making it impossible to extract +Δθ or −Δθ. Therefore, is the method of the present invention completely ineffective against the arrival of these signal points S1 and S3? However, in reality, new benefits are brought about. In other words, these signal points represent the scalar amount, or amplitude, of the vector component in the received carrier signal. Regarding signal point S1 ((+Δx, +Δ
y) area), this is larger than the specified amplitude, and looking at the signal point S3 ((-Δx, -Δy) area), this is smaller than the specified amplitude. This provides at least two advantages. That is,
QAM is particularly advantageous for QAM modulation systems because it is possible to extract not only phase information but also amplitude information using the same hardware.
Regardless of the modulation method or PSK modulation method, the so-called
The combined use of AGC (Automatic Gain Control) control makes it possible to suppress jitter in the reproduced carrier wave due to AM component noise. FIG. 2 is a graph for specifically explaining the method of the present invention, taking the case of a four-phase PSK modulation method as an example. In this figure, each area (quadrant),
In and, the correct entrainment phase and amplitude of the signal point is indicated by a black circle "●".
These black circles are also the origin of each small orthogonal coordinate system. If a black circle (●) in a region has a counterclockwise phase shift and moves to a white circle (Γ), the same thing will occur in other regions and at the same time. First, threshold levels x1, x2, x3 and y1, y2, y3 are set along the X, Y, x, and y axes as shown in the figure. Regarding the threshold levels x1, x2, x3, when the signal point is on the right side (+ side), it is set as "1", and when it is on the left side (- side), it is set as "0". For threshold levels y1, y2, y3, when the signal point is on the upper side (+ side), it is "1", and when it is on the lower side (- side), it is "0".
shall be. Also, regarding Δx and Δy mentioned above, +Δ
x is "1", -Δx is "0", +Δy is "1", -Δy
is set to “0”. Furthermore, let the + side of the X-axis be "1" and the - side be "0". Then, for each of the regions, , and, signal points appearing in each quadrant of each small orthogonal coordinate system (total 4 × 4 =
16 modes are possible) are represented by digital signals shown in Table 1 below. Note that Δθ and ΔA on the right side of Table 1 are the phase error and amplitude error, respectively, and the counterclockwise phase error is expressed as "1" and the opposite as "0".On the other hand, the amplitude error is expressed as specified. A higher level is represented by "1", and the opposite is represented by "0". However, these Δθ and ΔA are only revealed through digital processing, which will be described later, and are the objects to be determined.

[Table] The interpretation of Table 1 above will be explained using one sample. For example, the data in the top column of Table 1 is data related to signal point S1 in the graph of Figure 2, which is in the region (X, Y) = (1, 1) (Δ
x, Δy) is (-Δx, +Δy), so it is (0, 1), and it is on the negative side of x1, so x1
= 0, and since it is on the + side of x2, x2 = 1. Since it is naturally on the + side of x3, x3 = 1, and because it is on the + side of y1, y2, and y3, respectively. , y1=1, y2=1, y3
=1 is given. Now, what we need to find is whether the phase error Δθ of the signal point appearing in any of the 4×4=16 modes is “1” or “0” (+Δθ or −Δθ),
Also, it is determined whether the amplitude error ΔA is "1" or "0" (whether the amplitude is larger or smaller than the specified value) as necessary. The logic for determining these Δθ and ΔA is the following equations (1) and (2). Δθ=1x2y1y2+2 x3 1y2 +2x3y2y3+x1x22y3...(1) ΔA=x1x2y1y2+23y1y2 +23y2y3+x1x223...(2) In the above formula, - represents logic inversion, and + represents logical sum (OR). By the way, it is necessary to determine the discrimination signal q before calculating the above equations (1) and (2). In other words, the second
In the graph shown in the figure, the data of the signal points in the second and fourth quadrants of each small orthogonal coordinate system are valid only for determining the phase information, while the data of the signal points in the first and third quadrants Point data is only useful for determining amplitude information (as explained in Figure 1B). Therefore, depending on whether it is in the second or fourth quadrant, or in the first or third quadrant, it is necessary to make the information of Δθ or the information of ΔA valid, respectively. The discrimination signal q is required to discriminate these quadrants, and is obtained by the logic of equation (3) below.
For example, if q=1, the information of Δθ is valid, and q
If =0, the information on ΔA is valid. q=1x2y1y2+x1x21y2 +2312+2x3y1y2 +2x323+232y3 +x1x22y3+1x22y3...(3) To explain equations (1), (2), and (3) above in more detail, the first, second, and third on the right side of equation (1) above
and the fourth term are the areas in Figure 2, respectively.
Corresponding to and, for example, regarding the first term, if the signal point is in the second quadrant or the fourth quadrant, the logic of x1 x2y1y3 will be "1" or "0", respectively. The same applies to other terms.
After all, equation (1) shows that the signal point is +Δθ in any region.
If it appears at , it becomes "1", and if it appears at -Δθ, it becomes "0". Next, looking at equation (2) above, the first, second, third, and fourth terms on the right side correspond to the areas, and in Figure 2, respectively, and for example, for the first term, the signal point is If it is in the first quadrant or the third quadrant, the logic of x1x2y1y2 will be "1" or "0", respectively. The same applies to other terms. After all, equation (2) becomes "1" if the signal point appears at +ΔA in any region, and "0" if it appears at -ΔA. Furthermore, in equation (3) above, the first and second terms, the third and fourth terms, the fifth and sixth terms, and the seventh,
The 8th term corresponds to the area, , and, respectively. For example, if the signal point is in the 2nd or 4th quadrant in the area (1st, 2nd term),
The logic of y2+x1x2y1y2) is “1” and Δθ
is valid, ΔA is invalid, and otherwise is "0", Δθ is invalid, and ΔA is valid. This also applies to other areas. After all, if a signal point appearing in any region is in the second or fourth quadrant, q=1, and if it is in the first or third quadrant, q=0. The data Δ of “1” or “0” thus obtained
θ controls the voltage-controlled oscillator to complete synchronous detection with the reference carrier wave. Furthermore, if necessary, the "1" or "0" data ΔA obtained as described above can be used for level fluctuation suppression by combined use of the AGC control described above. FIG. 3 is a block diagram showing an embodiment of a carrier wave recovery circuit to which the method of the present invention is applied. In this figure, the received carrier signal S _io is amplified by an amplifier 31 and then demultiplexed via a hybrid circuit 32 to an X-axis phase detector 33X and a Y-axis phase detector 33Y, respectively. Here, π/
2 via the phase shifter 34 or not via the phase shifter 34,
Quadrature phase detection is performed from the reference carrier C. This reference carrier wave C is output by the voltage controlled oscillator (VCO) 38 described above. The analog signal from the detector 33X is sent to an A/D composed of a comparator.
It is converted into a digital signal by the (analog/digital) converter ADX. Similarly, the Y-axis system is also converted into a digital signal by the A/D converter ADY. Here, these A/D converters (also a kind of comparator as mentioned above) ADX and
ADY is the threshold level (x1, x
2, x3) and (y1, y2, y3), and the data (x1, x
2, x3), (y1, y2, y3). These are on the other hand applied to a demodulator (not shown) as output D _put for data recovery.
However, this is not relevant to the subject matter of the present invention. On the other hand, these data (x1, x2, x3) (y
1, y2, y3) are applied to digital logic circuits 35-1, 35-2 and 35-3, respectively,
Logical operations are executed according to the aforementioned logical formulas (1), (2), and (3), respectively, and outputs related to the aforementioned Δθ, ΔA, and q are sent out, respectively. As mentioned above,
For example, Δθ is valid when q=1, and ΔA is valid when q=0, so AND gates 36 and 36A are provided to selectively take out Δθ or ΔA depending on q. In addition, the digital logic circuit 35-
1, 35-2 and 35-3 will not be described in detail since they can be easily constructed by those skilled in the art using conventional decoder means. Phase error signal (Δ
θ) is applied to an oscillator (VCO) 38 via a low pass filter (LPF) 37 to bring the phase shift to zero. Amplitude error signal (Δ
A) is applied to the AGC amplifier 31 via a low pass filter (LPF) 39 to suppress, for example, the previously mentioned jitter. Although the above explanation has been made using the four-phase PSK modulation method as an example, it can naturally be applied to the QAM modulation method as well.
In fact, it is even more effective in QAM modulation. This is because both phase and amplitude are required information at the same time. However, since the basic idea of the method of the present invention is not changed in any way, detailed explanations will be replaced with figures, tables, and logical formulas. FIG. 4 corresponds to FIG. 2 described above. However, 16QAM
Since it is a modulation method, the signal point ~ in the case of the four-phase PSK modulation method shown in FIG. 2 is expanded to ~. Accordingly, the threshold levels (x1, x2, x3) in Figure 2 are expanded to (x1 to x7), and (y1, y2, y3) in Figure 2 are expanded to (y
1 to y7). In this case, each region ~ and (Δx, Δy, x1 to x7, y1 to y7, Δθ,
The relationship with ΔθA) is as shown in Table 2 below.

【table】

【table】

[Table] In this case, the logic that ΔQ, ΔA, and q should take is the following (4) and (5 )and
(6) becomes as follows. Δθ＝1x2y1y2＋1x22y3y4 ＋23x42y3y4 ＋23x4y1y2 ＋45x61y2 ＋45x623y4 ＋671y2＋6723y4 ＋4x5x645y6 ＋x45667 ＋6x745y6 ＋6x767＋x1x24y5y6 ＋x1x 26y7＋2x3x44y5y6 ＋2x3x46y7 …(4) ΔA＝x1x2y1y2＋x1x22y3y4 ＋2x3x42y3y4 ＋2x3x4y1y2 ＋45x6y1y2 ＋45x62y3y4 ＋67y1y2＋672y3y4 ＋45x6y4y5y6 ＋ 45x667 +6745y6 +6767+x1x245y6 +x1x267+2x3445y6 +2x3x467 …(5) q=x2y2 (1y1+x11) +x22y4 (1y3+x13) +2x42y4 (3y3+x33) +2x4y2 (3y1+x31) +4x6y2 (51+x5y1) +4x62y4 (53+x5y3) +x6y2 (71+x7y1) +12y4 (73+x7y3) +4x64y6 (x55+5y5) +4x66 (x57+5y7) +64y6 (x75+7y) 5) +6y6(x77+7y7) +x24y6(x1y5+15) +x26(x1y1+11) +2x44y6(x3y5+35) +2x46( x1y7+17) ...(6) As explained above, according to the present invention, it can be configured with a digital circuit suitable for integrated circuit (IC), and it is possible to output not only phase information but also amplitude information using the same hardware. This makes it possible to realize a carrier wave regeneration circuit that is not available in the prior art, and it can also be realized with the same configuration for PSK modulation systems and QAM systems, where signal points are not arranged at equal angles from the origin like conventional signal points.

[Brief explanation of the drawing]

FIG. 1A is a graph schematically showing a conventional carrier wave recovery method, FIG. 1B is a graph schematically showing a carrier wave recovery method of the present invention, and FIG. 2 is a graph specifically explaining the method of the present invention. , FIG. 3 is a block diagram showing an embodiment of a carrier regeneration circuit to which the method of the present invention is applied, and FIG. 4 is a diagram showing a case where the method of the present invention is applied to a QAM modulation system, in correspondence with FIG. This is a graph. In the figure, 33X is an X-axis phase detector, 33Y
is the Y-axis phase detector, 35-1, 35-2 and 3
5-3 are digital logic circuits, 38 is a voltage controlled oscillator, ADX and ADY are A/D converters, C is a reference carrier wave, and _Sio is a received carrier signal.

Claims (1)

[Claims] 1. An X-axis for defining a signal point (S) obtained by detecting a received carrier signal subjected to phase modulation or quadrature amplitude phase modulation using a reference carrier wave in a quadrature phase detector. and the Y-axis, at least one small orthogonal coordinate system consisting of the x-axis and the y-axis is set within the four quadrants of the orthogonal coordinate system, and the signal point (S)
is in the second or fourth quadrant of the corresponding small orthogonal coordinate system as a reference, detecting a phase shift on the + side or - side with respect to the reference carrier wave depending on whether it is in the second or fourth quadrant, and detecting the phase shift. A carrier wave regeneration method characterized by controlling the reference carrier wave so as to converge it to zero. 2. The carrier wave regeneration method according to claim 1, wherein the amplitude component of the received carrier signal is detected depending on whether the signal point (S) is in the first quadrant or the third quadrant of a small orthogonal coordinate system.