JPH06244880A - Frequency error detector - Google Patents

Abstract

PURPOSE:To exactly calculate frequency error between a carrier wave and a reference wave by removing the offset influence of an A/D converter and reducing the influence of noise in the case of orthogonally detecting a digital angle modulated wave. CONSTITUTION:A signal for frequency error detection is received (step 1) and a signal vector provided by orthogonally detecting this signal for detection with the reference wave is quantized as a sample at a sample rate equal to a symbol rate (step 2). Then, a synthetic vector is provided by rotating the phases of signal vectors as many as the double number of symbols, for which the phase change of this signal vector is totally turned to pi radian, and averaging them later (steps 3 and 4). By using the phase change of this synthetic vector, the frequency error between the carrier wave and a reference wave is calculated (steps 5-7).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a frequency error detector for detecting a frequency error between a carrier wave and a reference wave in a quadrature detector.

[0002]

2. Description of the Related Art Generally, a quadrature detector is known in which quadrature detection is performed using a reference wave from an oscillator without reproducing a carrier wave. That is, in this type of quadrature detector, the difference in frequency between the carrier wave and the reference wave is obtained, the reference wave frequency is corrected by this frequency error, and the quadrature detection is performed by the corrected reference wave (for example,
JP-A-3-128550).

In the above-mentioned frequency error detection, a modulation signal modulated by a signal sequence in which the phase change of the signal vector in one symbol time is always equal is received as a frequency error detection signal at the receiving side, and this frequency error is detected. The detection signal is quadrature-detected with the reference wave from the oscillator to obtain a signal vector (this signal vector will be referred to as a first signal vector here). Then, a frequency error is detected by comparing the phase change of the first signal vector with the phase change of the signal vector of a predetermined signal sequence (herein referred to as the second signal vector) (for example, As the second signal vector, a signal vector when it is assumed that the frequency error detection signal is quadrature-detected by the carrier wave is used).

Here, a conventional frequency error detecting device will be described with reference to FIGS.

In FIG. 2, the quadrature detector 31 is connected to a digital signal processor 38 via A / D converters 34 and 35. On the receiving side, the frequency error detection signal 30 described above is received, quadrature detected by the quadrature detector 31, the I (t) signal 32 is output as the in-phase component, and the Q (t) signal 33 is output as the quadrature component. (Step 101). These I (t) signals 32 and Q (t)
The signal 33 is sampled by the A / D converters 34 and 35, respectively, at a sampling rate equal to the symbol rate,
Signal vectors (I (kT), Q (kT)) 36 and 37
(Step 102, k = 1 to (N-1))
, Where N represents the number of symbols and T represents the symbol period). Then, the signal vectors (I (kT), Q (k
T)) is provided to the digital signal processor 38.

Here, referring to FIG. 3, in the digital signal processor 38, the signal vectors (I (kT), Q
(KT)) (collectively indicated by reference numeral 21 in FIG. 3) is a predetermined signal vector (I ₀ (kT), Q
₀ (kT)) the phase angle of 22 (kT) only rotates in the opposite direction the signal vector _{(I R (kT), Q} R (kT)) 2
3 (note that the signal vector (I ₀ (kT), Q
₀ (kT)) is a signal vector obtained when it is assumed that the frequency error detection signal is quadrature detected by the carrier wave).

Here, the signal vector (I (kT), Q
(KT)) phase angle φ (kT) 24, (I ₀ (kT),
Q ₀ (kT)) phase angle φ ₀ (kT) 25, and (I _R
(KT), the phase angle φ ₀ (kT) 26 is the number respective one of Q _R (kT)), is defined by the number 2, and the number 3.

[0008]

[Equation 1]

[0009]

[Equation 2]

[0010]

[Equation 3]

│r (kT) │, │r ₀ (kT)
|, | R _R (kT) | is the length of the vector, respectively, and | r (kT) | = | r _R (kT) |. Then, the signal vector _{(I R (kT), Q} R (kT)) is calculated by the number 4 (Step 103).

[0012]

[Equation 4]

Here, when the frequencies of the carrier wave and the reference wave are completely the same, the signal vectors (I (kT), Q (k
T)) and the signal vector (I ₀ (kT), Q ₀ (k
The phase of (T)) changes by the same amount for each sampling. Thus, the vector _{(I R (kT), Q} R (k
The phase of T)) does not change at all. On the other hand, if there is a difference in frequency between the carrier and the reference wave, the vector (I _R (k
_{T), Q R (kT)} ) will be phase changes. The amount of phase change θ _k when the time t changes from (k−1) T to kT is represented by Expression 5.

[0014]

[Equation 5]

Then, the sine sin θ _k of θ _k is expressed by equation 6.

[0016]

[Equation 6]

│r _R ((k-1) T) │ is │r _R (k
T) | regarded as substantially equal to, also, theta _k when theta _k is small because regarded as substantially equal to sin [theta _k, the phase variation theta _k can be expressed by the number 7. In the digital signal processor 38, the phase change amount θ
_k is calculated (step 104).

[0018]

[Equation 7]

The average θ _AVG of the rotation amount per symbol, which is the phase reference, is expressed by Equation 8, and the digital signal processor 38 obtains θ _AVG using _Equation 8 (step 105).

[0020]

[Equation 8]

Then, the digital signal processor 38
Then, θ _AVG / 2π [rad / 2π] is divided by T (1 symbol time) [sec] to obtain the frequency error between the carrier wave and the reference wave (step 106) and output as a frequency error signal (Fig. 2). .

As described above, as the frequency error detection signal, a modulation signal which is modulated by a signal series in which the phase change of the signal vector in one symbol time is always equal is used. By using such a signal sequence, it is not necessary to synchronize the sampling clock with the symbol. For example, in the case of the MSK modulated wave, the signal sequence is all “1” or “0”, and the phase change Δφ at this time is
₀ becomes π / 2 or −π / 2, respectively. The same applies to the GMSK modulated wave. The series of processes described above is performed by the digital signal processor 38 used when decoding the detection signal obtained by the quadrature detector.

[0023]

By the way, in the case of detecting the frequency error as described above, if the A / D converter for quantizing the signal vector has an offset, the signal sampled by the A / D converter is The phase change calculated based on the vector will include an error due to the offset. In addition, when there is noise in the transmission path, the modulated wave whose phase change is always constant on the transmission side for each symbol does not have the same phase change on the receiving side for each symbol. The phase change at will include an error. If the noise is extremely large, the phase change on the receiving side is 18
An error of about 0 degrees may occur.

For example, in the GMSK modulation system, the k-th signal vector (I (kT), Q (k) obtained by sampling at a sampling rate equal to the symbol rate.
T)) is the 0th signal vector (I (0), Q
(0)), the offset vector by the A / D converter is (I 1 ^, Q 2 ^, ), the noise vector is (n _Ii , n _Qi ), and the phase offset per symbol is θ. it can.

[0025]

[Equation 9]

That is, if the A / D converter has an offset and the noise in the transmission line becomes large, there is a problem that the frequency error between the carrier wave and the reference wave cannot be accurately detected.

An object of the present invention is to provide a frequency error detecting device which can always accurately detect a frequency error between a carrier wave and a reference wave.

[0028]

According to the present invention, a modulation signal modulated by digital angle modulation is received through a transmission line, and the modulation signal is used together with a quadrature detector for quadrature detection using a reference wave. A frequency error detecting device for detecting a frequency error between a carrier wave of the modulated signal and the reference wave, wherein the modulated signal is modulated by a signal sequence in which the phase change of a signal vector per symbol time is equal. A frequency difference detection signal is used, and a first signal vector obtained by orthogonally detecting the frequency difference detection signal with the reference wave is sampled and quantized at a predetermined sampling rate to obtain a quantized signal vector A / D conversion means, combining means for combining a predetermined number of the quantized signal vectors to obtain a combined vector, phase change of the combined vector and pre-composition. Obtains the difference between the phase change of the second signal vector defined, frequency error detection device can be obtained, characterized in that it comprises a frequency error detecting means for determining the frequency error based on a difference of the phase change.

[0029]

EXAMPLES The present invention will be described below with reference to examples. It should be noted that in this embodiment, the digital signal processor 38 has the same components as those shown in FIG.
Function is different.

First, the detection principle of the present invention will be described. In the modulation method in which the signal vector rotates by Δφ for each symbol, the k-th signal vector (I (I ( kT), Q (kT)) is 0-th signal vector (I (0), Q ( 0)), the offset vector by the a / D converter ^{(I,, Q,),} a noise vector (n _Ii ，
Q _Qi ), where θ is the phase offset per symbol, it is expressed by Eq.

[0031]

[Equation 10]

Then, the number M of symbols in which the phase change of the signal vector is π radian in total is expressed by equation 11.

[0033]

[Equation 11]

When rotated in the opposite direction by Δφ, the average of 2M continuous vectors (x _k (θ), y _k (θ))
Is obtained, it is expressed by Equations 12 to 14.

[0035]

[Equation 12]

[0036]

[Equation 13]

[0037]

[Equation 14]

As is clear from the equations 12 to 14, A
The effect of the offset due to the / D converter is removed, and the effect of noise is reduced from the relationship shown in Expression 15.

[0039]

[Equation 15]

Here, in order to facilitate understanding, GMSK
The modulation method will be described as an example.

The signal vector of the GMSK modulated wave rotates by π / 2 for each symbol. The k-th signal vector (I (kT), Q (kT)) obtained by sampling at the sampling rate equal to the symbol rate is the 0-th signal vector (I (0), Q (0)), A / an offset vector by D converter ^{(I,, Q,),} a noise vector (n _Ii, n _Qi), if the phase offset per symbol and theta, expressed by the number 16.

[0042]

[Equation 16]

Since the number of symbols M for which the total phase change of the signal vector is 2π radians is 4, four consecutive averages (x _k (θ), y in the vector rotated in the opposite direction by π / 2 are used. _{When k} (θ)) is obtained, it is expressed by Equations 17 to 19.

[0044]

[Equation 17]

[0045]

[Equation 18]

[0046]

[Formula 19]

As is clear from the equations 17 to 19, A /
The effect of offset by the D converter is eliminated, and in addition,
From the relationship shown in Expression 20, the effect of noise will be reduced.

[0048]

[Equation 20]

If there is a frequency error between the carrier wave and the reference wave, the phase of the combined vector (x _k (θ), y _k (θ)) changes. Therefore, the frequency error is estimated by estimating this phase change. Can be accurately detected.

Here, referring to FIGS. 1 and 5, on the receiving side, the above-mentioned frequency error detection signal 30 is received, quadrature detected by the quadrature detector 31, and the I (t) signal 32 is received.
And Q (t) signal 33 is output (step 1). These I (t) signal 32 and Q (t) signal 33 are sampled by the A / D converters 34 and 35, respectively, at a sampling rate equal to the symbol rate, and the signal vector (I
(KT), Q (kT)) 36 and 37 (step 2, where k = 1 to (N-1)). Then, the signal vector (I (kT), Q (kT)) is given to the digital signal processor 40.

The digital signal processor 40 comprises a vector rotation unit 41 and an average vector calculation unit 42. As shown in FIG. 3, in the vector rotation unit 41, a signal vector (I (kT), Q (kT)) (collectively denoted by reference numeral 21 in FIG. 3) is a predetermined signal vector (I ₀ ( kT), Q ₀ (kT)) 22 phase angle (k
The signal vector (I _R (k
_{T), Q R (kT)} ) 23 to (Note that the signal vector _{(I 0 (kT), Q} 0 (kT)) is the signal vector (I
(KT), Q (kT)) is a signal vector obtained when it is assumed that quadrature detection is performed on the carrier.

Here, the signal vector (I (kT), Q
(KT)) phase angle φ (kT) 24, (I ₀ (kT),
Q ₀ (kT)) phase angle φ ₀ (kT) 25, and (I _{R
(KT), Q R (kT} )) of the phase angle φ ₀ (kT) 26 numbers respectively 21, it is defined by the number 22, and number 23.

[0053]

[Equation 21]

[0054]

[Equation 22]

[0055]

[Equation 23]

Note that | r (kT) | and | r ₀ (kT)
|, | R _R (kT) | is the length of the vector, respectively, and | r (kT) | = | r _R (kT) |. Then, the signal vector _{(I R (kT), Q} R (kT)) is calculated by the number 24 (step 103). And
The signal vector _{(I R (kT), Q} R (kT)) is given to the vector calculation unit 42 (in-phase component I _R (kT)
And quadrature given to the vector calculation unit 42 as the component Q _R (kT)).

[0057]

[Equation 24]

[0058] In the average vector calculating section 42, the average of twice the number of vectors of the signal vector _{(I R (kT), Q} R (kT)) symbol number phase change is π radian in total of M (an integer) Ask for. The number of symbols M is given by Equation 25.

[0059]

[Equation 25]

[0060] That is, 2M pieces of vectors successive in average vector calculating section _{42 (I R (kT),} Q R (kT))
(K = k~k + 2M-1 ) average and calculation of the resultant vector _{(x k (θ), y} k (θ)) (k = 0~N-2M)
(Step 4). This composite vector (x
_k (θ), y _k (θ)) is represented by the equation 26.

[0061]

[Equation 26]

Further, the composite vector (x _k (θ), y
_k (θ)) can also be expressed by Equation 27.

[0063]

[Equation 27]

By the way, if the frequencies of the carrier wave and the reference wave are completely the same, the combined vector (x _k (θ), y
The phase of _k (θ)) does not change at all, and if there is a phase difference between the carrier wave and the reference wave, the combined vector (x _k (θ),
The phase of y _k (θ) changes. Time t is (k-
1) Sine sin θ of the phase change when changing from T to kT
_k is expressed by Equation 28. Note that | r _k | is represented by the equation 29.

[0065]

[Equation 28]

[0066]

[Equation 29]

[0067] | r _k-1 | is | r _k | regarded as substantially equal to, the theta _k when theta _k is small because regarded as substantially equal to sin [theta _k, therefore, the phase change theta _k is represented by the number 30 .

[0068]

[Equation 30]

Referring again to FIG. 5, the above composite vector has an in-phase component x _k (θ) and a quadrature component y _k (θ), and the in-phase component x _k (θ) is the first delay unit 43. , The first squarer 44, and the first multiplier 45. On the other hand, the quadrature component y _k (θ) is given to the second delay device 46, the second squarer 47, and the second multiplier 48. The first and second delay units 43 and 46 give a delay of a predetermined time (for example, a delay of the symbol period T) to the input signal. As a result, the first and second delay devices 43 and 46
Respectively outputs a delayed in-phase component x _k-1 (θ) and a delayed quadrature component y _k-1 (θ). Then, the delay in-phase component x _k-1 (θ) and the delay quadrature component y _k-1 (θ) are given to the second multiplier 48 and the first multiplier 45, respectively.

In the first multiplier 45, the in-phase component x _k (θ)
And the delay orthogonal component y _k-1 (θ) are multiplied to obtain a first multiplication component x _k (θ) y _k-1 (θ). The second multiplier 48 multiplies the quadrature component y _k (θ) and the delayed in-phase component x _k-1 (θ) to obtain the second multiplication component x _k-1 (θ) y _k (θ). The first squarer 44 squares the in-phase component x _k (θ) to obtain the first squared component x _k ² (θ). Similarly, in the second squarer 47, the orthogonal component y _k (θ) is squared to generate the second component.
The squared component y _k ² (θ) of is obtained.

The subtractor 49 receives the second multiplication component x _k-1 (θ)
From y _k (θ) to the first multiplication component x _k (θ) y _k-1 (θ)
To subtract the subtraction component {x _k-1 (θ) y _k (θ) −x
_k (θ) y _k-1 (θ)} is obtained. On the other hand, in the adder 50, the first square component x _k ² (θ) and the second square component y _k ²
(Θ) and the addition component {x _k ² (θ) + y
_{Find k} ² (θ)}. Then, the subtraction component and the addition component are provided to the phase change calculation unit 51.

The phase change calculator 51 obtains the phase change θ _k based on the equation 30 (step 5). This phase change θ
_k is given to the average value calculation unit 52, and the average value calculation unit 52 obtains the average θ _AVG of the rotation amount per symbol of the phase reference based on the equation 31 (step 6).

[0073]

[Equation 31]

This average θ _AVG is given to the frequency error calculation unit 53, and in the frequency error calculation unit 53, θ _AVG / 2π [r
The frequency error between the carrier wave and the reference wave can be obtained by dividing [ad / 2π] by T (1 symbol time) [sec] (step 7). Then, this frequency error is output as a frequency error signal.

In the above embodiment, when the combined vector is obtained, the rotated vector components are averaged to obtain the combined vector, but only the rotated vector components are added (not divided by the number of samples. Alternatively, the combined vector may be obtained.

[0076]

As described above, in the present invention, the signal vector obtained by receiving the frequency error detection signal on the receiving side and orthogonally detecting the frequency error detection signal with the reference wave is equal to the symbol rate. Sampling and quantizing at the sampling rate, the total phase change of the signal vector is 2
Since the composite vector is obtained by using the signal vector of the number of symbols of π radian and the frequency error between the carrier wave and the reference wave is detected by using the composite vector, the offset of the A / D converter can be removed only. It is possible to reduce the influence of noise. As a result, there is an effect that the frequency error can be accurately detected.

[Brief description of drawings]

FIG. 1 is a flow chart for explaining an embodiment of a frequency error detecting device according to the present invention.

FIG. 2 is a block diagram showing an apparatus used for frequency error detection.

FIG. 3 is a diagram for explaining rotation of a signal vector.

FIG. 4 is a flow chart for explaining a conventional frequency error detection device.

FIG. 5 is a block diagram showing an embodiment of a frequency error detection device according to the present invention.

[Explanation of symbols]

 31 quadrature detector 34, 35 A / D converter 38 digital signal processor

Claims (8)

[Claims] 1. A quadrature detector for receiving a modulated signal modulated by digital angle modulation through a transmission line and performing quadrature detection of the modulated signal using a reference wave, and the carrier of the modulated signal and the reference. A frequency error detection device for detecting a frequency error with a wave, wherein the modulation signal is a frequency difference detection signal that is modulated by a signal series in which the phase change of a signal vector per symbol time is equal, A / D conversion means for obtaining a quantized signal vector by sampling and quantizing a first signal vector obtained by quadrature detection of the frequency difference detection signal with a reference wave, and the quantized signal Combining means for obtaining a combined vector by combining a predetermined number of vectors, and a phase change of the combined vector and a predetermined second signal vector A frequency error detecting device, comprising: a frequency error detecting unit that obtains a difference from the phase change and obtains the frequency error based on the difference between the phase changes. 2. The frequency error detecting device according to claim 1, wherein the predetermined sampling rate is equal to the symbol rate of the frequency difference detecting signal. 3. The frequency error detection device according to claim 2, wherein the second signal vector is a signal obtained by assuming that the frequency difference detection signal is orthogonally detected by the carrier wave. And a frequency error detector. 4. The frequency error detecting apparatus according to claim 2, wherein the predetermined number is the number of symbols such that the phase change of the first signal vector becomes 2π radians. Detection device. 5. The frequency error detecting device according to claim 4, wherein the synthesizing means rotates the quantized signal vector clockwise by the phase of the second signal vector to obtain a rotation vector. And an averaging means for averaging the rotation vectors by the predetermined number to obtain the composite vector. 6. The frequency error detecting device according to claim 5, wherein the frequency error detecting means detects a phase change difference between the combined vector and the second signal vector and outputs a phase change difference detection signal. Phase change difference detection means for sending, phase difference calculation means for calculating the phase difference between the combined vector and the second signal vector based on the phase change difference detection signal, and the frequency error based on the phase difference. A frequency error detecting device comprising: a frequency error calculating means for calculating. 7. The frequency error detection device according to claim 5, wherein the frequency difference detection signal is a signal modulated with a signal sequence in which a phase change of 1 symbol time is π / 2 or −π / 2. A frequency error detecting device used, wherein the predetermined number is every 4 symbols. 8. The frequency error detecting device according to claim 7, wherein GMSK modulation is used as the digital angle modulation.