JP2005229281A - Circuit and method for detecting frequency error - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To precisely detect a frequency error while avoiding the increase of a circuit scale. <P>SOLUTION: A frequency error detection circuit for receiving an OFDM signal and detecting the frequency error between its carrier frequency and a local oscillation frequency is provided with a storage part and a calculation part. The storage part stores the argument of a complex number belonging to one of the areas obtained by radially dividing a complex plane with its origin as the center and the tangent value. When a complex number obtained by performing complex multiplication of the complex conjugate of a signal obtained by delaying the prescribed part of the OFDM signal by a period corresponding to one symbol with the local oscillation frequency and the prescribed part belongs to the one area, the calculation part specifies the argument of a complex number based on the stored content of the storage part. When the complex number belongs to the other area, the calculation part specifies an angle fixed previously to the other area as the argument of the complex number. The calculation part calculates the frequency error based on the specified result. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

The present invention relates to a technique for detecting a carrier frequency error in OFDM data transmission.

In recent years, in digital audio broadcasting for mobile communication terminals such as mobile phones and digital terrestrial television broadcasting, Orthogonal Frequency Division Multiplex:
OFDM) data transmission has attracted attention. In OFDM data transmission, data is transmitted by modulating with data transmitting a plurality of carrier waves (hereinafter referred to as subcarriers) orthogonal to each other, and multiplexing and transmitting these modulated waves. For this reason, when the number of subcarriers increases to several hundred to several thousand, the symbol period of each modulated wave becomes extremely long, and it is difficult to be affected by multipath interference.

FIG. 6 is a diagram illustrating an example of a frame format of the OFDM signal. As shown in FIG. 6, the OFDM signal includes a preamble portion and a payload portion. The preamble part includes a training symbol for synchronizing a local oscillation frequency with the carrier frequency of the OFDM signal in a receiving apparatus that receives the OFDM signal, and a symbol corresponding to the data is included in the payload part. include. More specifically, in the preamble portion, a short training symbol (FIG. 6: t1 to t10, where t1 = t2... = T10) in which the same signal is repeated 10 times with a period of 0.8 .mu.s, and 3.2 .mu.s.
A long training symbol in which the same signal (FIG. 6: T1 and T2, where T1 = T2) is repeated twice is included. Thus, by utilizing the fact that the same signal is repeated at the head of the frame, it is possible to detect the difference between the local oscillation frequency and the carrier frequency (hereinafter referred to as frequency error) and correct the frequency error.

Specifically, when the signal P (t) corresponding to the preamble part and a predetermined frequency (local oscillation frequency f0 of the receiving apparatus) serving as a reference, one of the carriers
A complex number P (t) P obtained by complex multiplication of a complex conjugate P * (t−T) with a signal obtained by delaying this P (t) by the duration T of the portion where modulation for symbols is performed. It is known that there is a relationship represented by the following formula 1 between * (t−T) and the frequency error Δf. It should be noted that there is a relationship expressed by the following formula 2 between the local oscillation frequency f0 of the receiving apparatus and the above T.

(Expression 1) Δf = (1 / 2πT) × arctan (P (t) P * (t−T))
(Equation 2) f0 = 1 / T

Therefore, the arc tangent value (arctan (P (t) P * (t−T)) on the right side of Equation 1, that is,
The frequency error can be detected if the complex number P (t) P * (t−T) declination) can be obtained, but calculating the arctangent value from the real and imaginary parts of the complex number. It is difficult to implement with a logic circuit. Therefore, the declination of the complex number and the tangent value thereof (that is, the ratio between the real part and the imaginary part of the complex number) in the range of −π to π are stored using a look-up table stored in association with each other. It is generally done to seek.

However, if such a lookup table is stored in a ROM (Read Only Memory) or the like and included in a circuit for detecting a frequency error, there is a problem that the circuit scale becomes large. As a technique for solving such problems, Patent Document 1
Can be mentioned. The technique disclosed in Patent Document 1 discloses a technique in which the calculation of the inverse positive value of Equation 1 is replaced with an approximate calculation that can be realized by a logic circuit without using the lookup table.
JP 2001-285246 A

However, in the technique disclosed in Patent Document 1, the arc tangent value (that is, the complex number P
Since (t) P * (t−T) declination) is obtained by approximate calculation, there is a problem in that the exact value cannot be obtained and the frequency error cannot be detected accurately. .
The present invention has been made in view of the above problems, and an object of the present invention is to provide a technique capable of accurately detecting a frequency error while avoiding an increase in circuit scale.

In order to solve the above-mentioned problem, the present invention corresponds to a tangent value corresponding to a declination of a complex number belonging to one of the regions obtained by dividing a complex plane radially around its origin, or to the declination. A storage unit storing the declination in association with either a sine value or a cosine value, and when the OFDM signal is received, the carrier frequency of the OFDM signal is a predetermined frequency.
A complex conjugate of a signal obtained by delaying a predetermined part of the OFDM signal by the duration of a part corresponding to one symbol, and an input unit to which an input signal obtained by complex multiplication of the predetermined part is input; Whether the complex number corresponding to the input signal input to the input unit belongs to the region 1 or not is determined by the absolute value of the real part, the absolute value of the imaginary part, the sign of the real part, and the sign of the imaginary part When the determination unit determines that the complex number corresponding to the input signal belongs to the region of 1, the declination angle of the complex number is stored in the storage unit. On the other hand, if the determination unit determines that the complex number belongs to another region, the complex number belonging to the other region, and a predetermined complex argument is The complex argument corresponding to the input signal and A frequency error detection circuit comprising: a specifying unit that specifies a frequency error that is a difference between a carrier frequency of the OFDM signal and the predetermined frequency based on a declination specified by the specifying unit. provide.

In another aspect of the present invention, a tangent value corresponding to a declination of a complex number belonging to one of the regions obtained by dividing the complex plane radially about its origin, or corresponding to the declination When a carrier frequency of the received OFDM signal is a predetermined frequency, the frequency error detection circuit having a storage unit in which the deviation angle is stored in association with either the sine value or the cosine value is 1
A complex conjugate of a signal obtained by delaying a predetermined portion of the OFDM signal by the duration of a portion corresponding to one symbol and an input signal obtained by complex multiplication of the predetermined portion are input.
And whether the complex number corresponding to the input signal input in the first step belongs to the region of 1 or not, the absolute value of the real part, the absolute value of the imaginary part, and the real part When the second step determines that the complex number corresponding to the input signal belongs to the first region, the second step determines based on the sign of the imaginary part and the sign of the imaginary part. , While identifying the argument of the complex number based on the storage content of the storage unit, if it is determined in the second step that the complex number belongs to another region, A complex number to which
A third argument that specifies a predetermined complex argument as a complex argument corresponding to the input signal;
And a fourth step of detecting a frequency error, which is a difference between the carrier frequency of the OFDM signal and the predetermined frequency, based on the declination specified in the third step. A characteristic frequency error detection method is provided.

According to such a frequency error detection circuit and frequency error detection method, when a complex number corresponding to the input signal belongs to the area of 1, the frequency error is specified based on the storage contents of the storage unit, In the case of belonging to the other area, the frequency error is specified based on the declination predetermined for the other area. The range of the declination stored in the storage unit is narrower than the range corresponding to the entire complex plane (ie, −π to π), so that the storage capacity of the storage unit can be reduced and the circuit scale can be reduced. Become. In addition, when the complex number corresponding to the input signal belongs to the region 1, the frequency error is detected based on the storage content of the storage unit. Therefore, compared with the case of calculating the frequency error by approximation calculation. The frequency error can be accurately detected. In addition, when the complex number corresponding to the input signal belongs to another region different from the one region, a predetermined angle with respect to the other region is specified as the deviation angle of the complex number. If the correction of the frequency error is repeated based on the specified declination in this way, the complex number corresponding to the input signal after the correction belongs to the above-mentioned area 1, and the frequency error is accurate. Will be detected.

In a more preferred aspect, when the complex conjugate of the complex number corresponding to the input signal belongs to the region 1, the declination specified based on the stored content of the storage unit is inverted and the deviation of the complex number is determined. Provided is a frequency error detection circuit characterized in that a corner is specified. According to such an aspect, even when the complex number corresponding to the input signal does not belong to the above-described region 1, if the complex conjugate belongs to the above-mentioned region 1, the argument of the complex number is It is possible to accurately specify based on the storage contents of the storage unit and to accurately detect the frequency error.

In a more preferred aspect, there is provided a frequency error detection circuit characterized in that the value of the declination stored in the storage unit is normalized by the local oscillation frequency of the receiving device. By normalizing in this way, if the complex number corresponding to the input signal is within the predetermined region, the frequency error can be directly detected by referring to the storage content of the storage unit, There is an effect that it is possible to omit the above-described calculation of Equation 1.

The best mode for carrying out the present invention will be described below with reference to the drawings.
[A. Constitution]
FIG. 1 is a block diagram illustrating a configuration example of a receiving apparatus according to an embodiment of the present invention. In the receiving apparatus shown in FIG. 1, a signal P (t) received by a tuner (not shown) and passed through an orthogonal demodulator (not shown) and an AD converter (not shown) in order is a frequency error corrector 10.
Is input. The frequency error corrector 10 corrects the carrier frequency of the signal P (t) by a frequency error Δf detected by a frequency error detector 20 to be described later, and distributes and outputs the signal as shown in FIG. It is. One of the signals P (t) output after being divided and distributed by the frequency error corrector 10 is input to a demodulator (not shown), and the other is input to the frequency error detector 20. Although not shown in detail, the signal P (t) includes an in-phase detection axis signal (hereinafter I signal) and a quadrature detection axis signal (hereinafter Q signal).

The signal P (t) input to the frequency error detector 20 is further divided into two. One is input to the multiplication circuit 210 as it is, and the other is input to the delay circuit 220. The delay circuit 220
The input signal P (t) is a predetermined frequency (in this embodiment, the local oscillation frequency f of the receiving device).
0), a delay signal P (t−T) is generated by delaying by a time corresponding to one symbol (hereinafter referred to as symbol period) T, and the delayed signal P (t−T) is output to the complex conjugate circuit 230. The complex conjugate circuit 230 inverts the Q signal included in the delay signal P (t−T) input from the delay circuit 220 on the complex plane to generate and multiply the complex conjugate signal P * (t−T). Output to the circuit 210. The multiplication circuit 210 performs complex multiplication on the input signal P (t) and the complex conjugate signal P ^* (t−T) input from the complex conjugate circuit 230, and the multiplication result (that is, P (t) P *).
(T−T)) as an input signal is output to the frequency error detection circuit 240.

FIG. 2 is a block diagram illustrating a configuration example of the frequency error detection circuit 240. As shown in FIG. 2, the frequency error detection circuit 240 includes an input terminal 240a, a frequency error specifying unit 240b, and a storage unit 240c. The input terminal 240a is the multiplication circuit 2 described above.
10 functions as an input unit that delivers the input signal P (t) P * (t−T) received from 10 to the frequency error specifying unit 240b. The frequency error specifying unit 240b will be described in detail later.
Based on the I signal and Q signal of the input signal P (t) P * (t−T) delivered from the input terminal 240a, the complex number corresponding to the input signal is divided into 8 equal parts as shown in FIG. It is determined which one of the areas belongs, and the frequency error is specified based on the determination result. The storage unit 240c is a ROM, for example, and stores a lookup table shown in FIG. As shown in FIG. 4, this lookup table includes values calculated based on the real part and imaginary part of the complex number included in the area A shown in FIG. 3 (in this embodiment, the declination of the complex number). Correlation with the tangent value, that is, the imaginary part / real part), the complex argument θ (0 ≦ θ ≦ π / 4)
Is stored. Thus, the deflection angle θ stored in the lookup table shown in FIG.
Since the range of 0 is in the range of 0 to π / 4, the data size is smaller than that of the conventional lookup table corresponding to the range of −π to π and can be stored in a ROM having a smaller circuit scale than the conventional one. It becomes possible. Thus, according to the frequency error detection circuit according to the present embodiment, −π
Compared to a conventional frequency error detection circuit having a look-up table corresponding to a range of ˜π, the circuit scale is reduced. In the present embodiment, the case where the tangent value of the declination is stored in the lookup table in association with the declination of the complex number included in the A region has been described. A set of a real part (ie, cosine value corresponding to the argument) and an imaginary part (ie, a sine value corresponding to the argument) normalized by the magnitude of the complex number is associated with the argument of the complex number. Of course it is good.

[B. Operation]
Next, the frequency error specifying operation performed by the frequency error specifying unit 240b will be described. FIG.
These are flowcharts showing the flow of the frequency error specifying operation performed by the frequency error specifying unit 240b. As shown in FIG. 5, when the input signal P (t) P * (t−T) is delivered from the input terminal 240a (step SA1), the frequency error specifying unit 240b responds to the input signal. It is determined based on the I signal and Q signal of the input signal which of the eight regions shown in FIG. 3 the complex number to belong to belongs to (step SA2). Specifically, the frequency error specifying unit 24
0b is based on the absolute value of the I signal (hereinafter expressed as | I |), the absolute value of the Q signal (hereinafter expressed as | Q |), and the sign of the I signal and the sign of the Q signal. Then, it is determined as shown in Table 1 below to which of the above eight regions (see FIG. 3) the complex number corresponding to the input signal belongs.

Then, the frequency error specifying unit 240b specifies the complex argument θ corresponding to the input signal P (t) P * (t−T) as shown in Table 1 based on the specifying result of step SA2 (
In step SA3), the deviation angle θ is substituted into the right side of Equation 1 described above to calculate the frequency error Δf (step SA4). In the present embodiment, a region other than the region corresponding to the lookup table stored in the storage unit 240c (that is, the region A) and the region that is axisymmetric to the region with respect to the real axis (that is, the H region). Has been described for the case where the median value of the range of the declination corresponding to each region is associated with each region, but the angle associated with each region in advance is not limited to the median value. . In short, any angle within the range of the declination corresponding to each region may be used.

For example, the complex number corresponding to the input signal P (t) P * (t−T) is in the A region (see FIG. 3).
The frequency error specifying unit 240b specifies the declination by referring to the stored contents of the lookup table. If the frequency error is included in the H region, the frequency error specifying unit 240b determines the stored contents of the lookup table. A declination is specified by reversing the sign of the declination identified with reference. In this way, when the complex number corresponding to the input signal P (t) P * (t−T) belongs to the A region or the H region shown in FIG. 3, the frequency error specifying unit 240b stores the lookup table. Since the declination is specified based on the contents, the frequency error Δ is compared with the case where the declination is calculated by approximate calculation.
It becomes possible to detect f accurately.

When the complex number belongs to any of the B, C, D, E, F, and G regions shown in FIG. 3, the frequency error specifying unit 240b belongs to the complex number as the argument of the complex number. A predetermined angle (see Table 1 above) is specified for the region, and the deviation error is substituted into the right side of Equation 1 to calculate the frequency error Δf. In such a case, the frequency error is not always accurately detected by one frequency error specifying operation, but the carrier frequency of the signal P (t) is set to the frequency error by the frequency error detected by the frequency error specifying operation. It is possible to accurately detect and correct the frequency error by repeatedly performing the above-described frequency error specifying operation after the correction device 10 corrects the frequency error. For example, when the argument of the complex number corresponding to the input signal is π / 3 (
In other words, in the case of belonging to the B region), 3π / 8 is specified as a declination corresponding to the frequency error by the frequency error detection circuit, and a value obtained by substituting the specification result into the right side of Equation 1 is a frequency error. Detected as When the frequency error detected in this way is corrected by the frequency error corrector 10 described above, the complex argument corresponding to the corrected input signal is π /
3-3π / 8 = −π / 24. Since the complex number corresponding to the corrected input signal belongs to the H region in this way, the frequency error is accurately detected based on the stored contents of the lookup table in the second frequency error detection. It will be corrected. Thus, according to the frequency error detection circuit according to the present embodiment, while avoiding an increase in circuit scale,
There is an effect that the frequency error can be accurately detected.

[C. Deformation]
Although the embodiment of the present invention has been described above, it is needless to say that the embodiment may be modified as described below.

(C-1: Modification 1)
In the embodiment described above, a case is described in which a tangent value of a declination is stored in a lookup table in association with a declination of a complex number belonging to a specific region on the complex plane (A region in the above embodiment). However, as a matter of course, a value obtained by normalizing the declination corresponding to the tangent value with the local oscillation frequency of the receiving apparatus (that is, a value obtained by multiplying by f0 / 2π) may be stored in the tangent value. In this way, the frequency error Δf is referred to by referring to the stored contents of the lookup table.
Can be obtained directly, and the processing speed when calculating the frequency error is increased.

(C-2: Modification 2)
In the above-described embodiment, the case where the frequency error is detected using the training symbol included in the preamble portion of the OFDM signal has been described, but it is also possible to detect the frequency error using the payload portion. In the OFDM system, in order to prevent interference between symbols,
A part of the symbol to be transmitted (for example, the last quarter) is written as a guard interval in front of the symbol and written in the payload portion. In other words, the waveform of the signal corresponding to the guard interval matches a part of the waveform of the signal of the symbol that follows the guard interval. Using this, it is possible to correct the frequency error for each data superimposed on the OFDM signal. Thus, if it is a predetermined part including the same signal, it is possible to detect a frequency error using either the preamble part or the payload part.

(C-3: Modification 3)
In the above-described embodiment, the lookup table corresponding to the angle range of 0 to π / 4 is RO.
Although the case of storing in M has been described, the range of angles stored in the lookup table is not limited to such a range. For example, it may be an angle range of −π / 4 to 0 (that is, a range corresponding to the H region in FIG. 3), or may be a range corresponding to the B region, the C region, or the like. Further, it may be a range corresponding to 0 to π / 2 (that is, a range corresponding to the A region and the B region). In short, any lookup table corresponding to a range corresponding to one of the regions obtained by radially dividing the complex plane around its origin (that is, an angle range of −π to π) may be used. good. However, in carrying out the present invention, among the angle ranges stored in the look-up table, the most preferable range is 0 to 0 described above.
The range is π / 4. The reason is that, for example, when a lookup table corresponding to the range of 0 to π / 2 is stored in the ROM, the tangent value diverges infinitely in the vicinity of π / 2.
This is because it becomes difficult to accurately specify the declination in the vicinity.

It is a block diagram which shows the structural example of the receiver which has a frequency error detection circuit based on this invention. It is a block diagram which shows the structural example of the same frequency error detection circuit. It is a figure which shows the relationship between the input signal input into the same frequency error detection circuit, and the deflection angle specified with respect to the input signal. It is a figure which shows an example of the lookup table stored in the memory | storage part. It is a flowchart which shows the flow of the frequency error specific operation | movement which the frequency error specific part performs. It is a figure which shows an example of the frame format of an OFDM signal.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Frequency error corrector, 20 ... Frequency error detector, 210 ... Multiplication circuit, 220 ... Delay circuit, 230 ... Complex conjugate circuit, 240 ... Frequency error detection circuit, 240a ... Input terminal, 240
b: Frequency error specifying unit, 240c: Storage unit.

Claims (4)

Corresponds to either a tangent value corresponding to a declination of a complex number belonging to one of the regions obtained by radially dividing the complex plane around its origin, or a sine value and cosine value corresponding to the declination A storage unit storing the declination angle;
When an OFDM signal is received, when the carrier frequency of the OFDM signal is a predetermined frequency, the complex conjugate of the signal obtained by delaying the predetermined portion of the OFDM signal by the duration of the portion corresponding to one symbol An input unit to which an input signal obtained by complex multiplication of the predetermined part is input;
Whether the complex number corresponding to the input signal input to the input unit belongs to the region 1 or not is determined by the absolute value of the real part, the absolute value of the imaginary part, the sign of the real part, and the imaginary part of the imaginary part. A determination unit for determining based on the code;
When the determination unit determines that the complex number corresponding to the input signal belongs to the region of 1, the declination of the complex number is specified based on the storage content of the storage unit, while the complex number is When the determination unit determines that it belongs to another region, it is a complex number belonging to the other region, and a predetermined complex number declination is used as a complex number declination corresponding to the input signal. A specific part to identify;
A frequency error detection circuit comprising: a calculation unit that calculates a frequency error, which is a difference between a carrier frequency of the OFDM signal and the predetermined frequency, based on the declination specified by the specification unit. When the complex conjugate of a complex number corresponding to the input signal belongs to the region 1, the specifying unit calculates an angle obtained by sign-inverting the angle specified based on the storage content of the storage unit. The frequency error detection circuit according to claim 1, wherein the frequency error detection circuit is specified as an argument of the complex number. The frequency error detection circuit according to claim 1, wherein the declination angle stored in the storage unit is standardized at the predetermined frequency. Corresponds to either a tangent value corresponding to a declination of a complex number belonging to one of the regions obtained by radially dividing the complex plane around its origin, or a sine value and cosine value corresponding to the declination When the carrier frequency of the received OFDM signal is a predetermined frequency, the frequency error detection circuit having the storage unit storing the declination stores the predetermined OFDM signal for the duration of the portion corresponding to one symbol. A first step in which a complex conjugate of a signal obtained by delaying a part and an input signal obtained by complex multiplication of the predetermined part are input;
Whether or not the complex number corresponding to the input signal input in the first step belongs to the region of 1, whether or not the absolute value of the real part, the absolute value of the imaginary part, the sign of the real part, and the code A second step of determining based on the sign of the imaginary part;
When it is determined in the second step that the complex number corresponding to the input signal belongs to the first region, the argument of the complex number is specified based on the storage content of the storage unit,
If it is determined in the second step that the complex number belongs to another area, the complex number belonging to the other area corresponds to the input signal with a predetermined complex angle. A third step of identifying the complex angle as
And a fourth step of sequentially detecting a frequency error, which is a difference between the carrier frequency of the OFDM signal and the predetermined frequency, based on the declination specified in the third step. Error detection method.

Images