KR101233826B1 - Method for compensating of frequency offset - Google Patents

Abstract

The present invention relates to a data communication technology of an orthogonal frequency division multiplexing (OFDM) technique, and more particularly, to a frequency offset compensation method in a data communication technology of an orthogonal frequency division multiplexing technique. Estimating an offset, compensating for a frequency offset of the received signal using the estimated frequency offset value, estimating a finer frequency offset using a preamble of the signal whose frequency offset is compensated for, approximately Estimating a final frequency offset by adding the estimated frequency offset value and the finely estimated frequency offset value, and compensating for the received signal frequency offset using the estimated final frequency offset. The phase error can be minimized by the method.

Description

Frequency offset compensation method {Method for compensating of frequency offset}

The present invention relates to a data communication technique of an orthogonal frequency division multiplexing (OFDM) technique, and more particularly, to a frequency offset compensation method in a data communication technique of an orthogonal frequency division multiplexing technique.

The present invention is derived from a study conducted as part of the IT source technology development project of the Ministry of Knowledge Economy. [Task name: VMC technology development]

Orthogonal Frequency Division Multiplexing (OFDM) is a technique of transmitting data using orthogonal subcarriers with different integer cycles. In an orthogonal frequency division multiplexing-based system, the frequency offset is caused by the operating frequency mismatch of the oscillator between the transmitter and the receiver and the Doppler transition caused by the movement in the mobile environment.

The frequency offset changes the phase of the received signal to destroy orthogonality between subcarriers. This causes inter-channel interference (ICI) between the channels. Therefore, if the accurate estimation and compensation of the frequency offset is not made, it may cause a decrease in system performance.

SUMMARY OF THE INVENTION An object of the present invention is to provide an apparatus and method for performing frequency offset estimation and compensation in a communication environment between vehicles and between a vehicle and a base station, which are derived from such a background and are not affected by the Doppler transition due to the movement of a transceiver. do.

The technical problem includes estimating a frequency offset using a preamble included in a received signal, compensating a frequency offset of the received signal using an estimated frequency offset value, and a signal having a frequency offset compensated. Estimating a finer frequency offset using a preamble of N, estimating a final frequency offset by adding a roughly estimated frequency offset value and a finely estimated frequency offset value, and receiving using the estimated final frequency offset Compensating for the signal frequency offset is achieved by a frequency offset compensation method characterized in that it comprises.

According to the present invention, the phase error can be minimized because the frequency offset is estimated approximately stepwise and finely. That is, unlike the estimation of the frequency offset by arbitrarily designating the number of repetitions of the vector rotation, the precision of the phase estimation can be improved.

1 is a block diagram showing the configuration of a typical Orthogonal Frequency Division Multiplexing (OFDM) communication system;
2 illustrates an example of a transport packet according to an embodiment;
3 is a reference diagram for explaining a frequency offset compensation method according to an embodiment;
4 is a block diagram of a frequency offset compensator according to an embodiment;
5 is an exemplary view for explaining a general frequency compensation method;
6 and 7 are reference diagrams for describing a frequency offset estimation method, according to an exemplary embodiment.

The foregoing and further aspects of the present invention will become more apparent through the preferred embodiments described with reference to the accompanying drawings. Hereinafter, the present invention will be described in detail so that those skilled in the art can easily understand and reproduce the present invention through these embodiments.

1 is a block diagram showing the configuration of a typical Orthogonal Frequency Division Multiplexing (OFDM) communication system.

As shown in FIG. 1, in order to transmit data, a communication apparatus performs phase shift keying (PSK) or quadrature amplitude modulation (PSK) in a modulation unit after channel encoding data to be transmitted. Modulate using QAM: Quadrature Amplitude Modulation.

After modulation, serial data is converted into parallel data through an S / P converter. Inverse Fast Fourier Transform (IFFT) is performed to convert the data back into serial data and output it through the RF antenna.

When receiving data, the signal received by the RF antenna is processed in the reverse order of the transmitting side. That is, Fourier transform (FFT) converts a received signal into parallel data through an S / P converter, and converts parallel data into serial data through a P / S converter. The demodulator demodulates the demodulator and outputs the original signal through channel decoding.

2 is an exemplary view of a transport packet according to an embodiment.

The transport packet shown in FIG. 2 is a form suggested in WLAN (IEEE 802.11a) and WAVE (IEEE 802.11p) standards. As shown, one packet is largely composed of a preamble, a signal field, and a data field. The preamble is transmitted at the front of the packet and includes signals used for signal detection, auto gain control (AGC), and frequency offset estimation at the receiving end.

The preamble consists of ten short training symbols and two long training symbols. A short training symbol is used to estimate the approximate initial frequency offset. The second frequency offset is finely estimated using the long training symbol.

An algorithm for estimating the frequency offset using the preamble signal will be briefly described below.

The carrier frequency offset is estimated by arranging the result of the cross-correlation function of the received training signal.

In Equation 1, y (k) represents a received signal affected by carrier frequency offset, x (k) represents a transmitted signal, and x (k) represents additional white gaussian noise (AWGN). ε is the frequency offset normalized according to the interval between subcarriers, N is the number of points of the Fourier transform and the Inverse Fourier transform. The frequency offset is estimated through an argument calculation of the cross correlation function result of the received training signal.

Where v is the number of samples used for cross-correlation. At this time, ν is 16 for the approximate frequency offset estimation and 64 for the fine frequency offset estimation.

는 추정된 주파수 옵셋, ∠(·)는 인수(argument)연산을 나타낸다. 추정된 주파수 옵셋에 의해 주파수 옵셋을 보상할 수 있다. At this time

Denotes an estimated frequency offset, and ∠ (·) denotes an argument operation. The frequency offset may be compensated for by the estimated frequency offset.

3 is a reference diagram for explaining a frequency offset compensation method according to an embodiment.

As described above, the first frequency offset is roughly estimated by using a short training symbol included in the received signal, and the fine second frequency offset is estimated by using a long training symbol.

)을 추정한다(300). 그리고 추정된 주파수 옵셋(

)값을 이용하여 긴 훈련 심볼의 주파수 옵셋을 보상한다(320). 그리고 주파수 옵셋이 보상된 긴 훈련 심볼을 이용하여 보다 세밀한 제 2 주파수 옵셋(

)을 추정한다(330). First, the approximate first frequency offset for the short training symbol (

E) (300). And estimated frequency offset (

Value is used to compensate for the frequency offset of the long training symbol (320). And using a longer training symbol with a compensated frequency offset,

) Is estimated (330).

)값과 긴 훈련 심볼을 이용하여 보다 세밀하게 추정된 제 2 주파수 옵셋(

)을 합하여 최종 주파수 옵셋(

)을 추정한다(340). 그리고 최종 주파수 옵셋(

)을 이용하여 수신 신호(Receive signal) 즉, 데이터 심볼의 주파수 옵셋을 보상한다(350). 이 후에 퓨리에 변환과 같은 수신 데이터 처리(360) 과정은 일반적인 사항으로 그 상세한 설명은 생략한다. After this, the first estimated frequency offset for the short training symbol (

Value and the second frequency offset (

) To sum the final frequency offset (

) Is estimated (340). And the final frequency offset (

In operation 350, the frequency offset of the received signal, that is, the data symbol is compensated for. Thereafter, the process of receiving data processing 360, such as Fourier transform, is a general matter and a detailed description thereof will be omitted.

4 is a block diagram of a frequency offset compensator according to an embodiment.

As shown, the frequency offset compensator includes a frequency offset estimator 400 and a frequency offset compensator 410.

The frequency offset estimator 400 performs approximate frequency offset estimation and fine frequency offset estimation. The frequency offset estimator 400 includes a signal delay unit 402, an integrator 404, and a phase determining unit 406.

The signal delay unit 402 delays the input value by a certain number of samples. A complex power form value is obtained by multiplying the first input value by a signal delayed by a certain number of samples with respect to the input by the signal delay unit 402. After integrating through the integrator 404, the phase grasping unit 406 grasps a phase from the integration result. In this case, the accuracy of frequency estimation can be improved by using an average value of a plurality of phases by using a preamble repeated rather than simply selecting one value to estimate a frequency.

The frequency offset compensator 410 includes a numerically controlled oscillator 412 (NCO: Numerically Controlled Oscillator). The numerically controlled oscillator 412 generates a cosine and a sin value with the estimated frequency offset provided from the frequency offset estimator 400. The cosine and sine values are multiplied by the received signal to compensate for the frequency offset of the received signal.

5 is an exemplary diagram for describing a general frequency compensation method.

In general, frequency offset synchronization blocks using a Coordinate Rotation Digital Computer (CORDIC) use a CORDIC vector mode for phase estimation and a CORDIC rotation mode for compensation.

The tan ^{- 1} operation is used to estimate the value of θ, and the cosθ and sinθ values to be compensated for are calculated according to the estimated value of θ. The calculated cosθ and sinθ values are used to multiply the I / Q values of the input signal in order to compensate for the frequency offset. The CORDIC algorithm consists of the rotation of the vector as shown below.

,

값은 i번째 반복에서의 복소수의 실수부 값, 허수부 값을 나타낸다.

는 i번째 반복에서 선택되는 부호를 나타낸다.

,

의 초기값과 회전 방향에 따라 tan^-1, cos/sin 연산이 수행된다. here,

,

The value represents a complex real value and an imaginary value in the i-th iteration.

Denotes the symbol selected in the i-th iteration.

,

The tan ^-1 and cos / sin operations are performed according to the initial value and rotation direction of.

To determine the final phase angle, the number of vector rotations (repeats) must be determined. The number of rotations affects the accuracy of the frequency offset value estimation. Therefore, if the frequency offset is estimated by arbitrarily setting the number of rotations as in the conventional method, there is a limitation that precision cannot be guaranteed. In order to solve this problem, the present patent intends to provide a method using a binary search method. Therefore, according to the present invention, resolution, that is, precision can be improved.

6 and 7 are reference diagrams for describing a frequency offset estimation method, according to an exemplary embodiment.

The angle of one point (I, Q) on the complex plane can be represented by an angle between 0 and 45 °. In FIG. 6, the points (x, y) may be represented by 90-α °. That is, if an angle of α ° is formed from the Q axis, it is equal to the angle α ° of one point (x ', y') between 0 ° and 45 °. The angle of the point can be obtained by considering the position in the quadrant after matching the angle of any one point on the plane with 0 to 45 °. As a result, the capacity of the phase memory can be reduced.

In FIG. 6, when the input is located in each region, the position transformation relation is as follows.

Location area Relation Location area Relation B (45 ° to 90 °) 90 ° -deg (A) F (-90 ° to -45 °) -90 ° + deg (A) C (90 °-135 °) 90 ° + deg (A) G (-135 ° to -90 °) -90 °-deg (A) D (135 ° to 180 °) 180 ° -deg (A) H (-180 ° to -135 °) -180 ° + deg (A) E (-45 ° to 0 °) -deg (A)

In addition, according to one embodiment, as shown in Figure 7, by dividing the region between 0 ° ~ 45 ° by a straight line having a slope k / 2 ^N , k = 1, 2, ..., 2 ^N -1 The divided regions may be used to find an angle θ of the input I and Q values. In this case, the method of finding an angle using the divided region uses a binary search method. First, in order to ensure the phase accuracy of the inputs I and Q, the number of rotations, that is, how many parts to divide should be determined. In other words, a value of 2 ^N is determined. And the accuracy of the angle to be approximated can be found as 45 ° / 2 ^N. And the value of N is initialized to 0. Then, it is determined whether I> Q for the input (I, Q). In the case of I> Q, the inputs I and Q are present between 0 ° and 45 °. On the contrary, if I <Q, it exists between 45 degrees and 90 degrees. In this case, when I and Q values are interchanged, that is, when I value is applied to Q and Q value is applied to I value, it is switched to 0 ° to 45 ° region.

And set the value of N to 1. In this step, it is determined whether the inputs I and Q are in the area larger than the straight line y = (1/2) x or in the small area. To determine this, first find the half value of input I and compare it with the Q value. At this time, half of the value of I is a value in which the value of I is right-shifted by one bit. And if Q> 1/2, then (I, Q) is between [tan ^-1 (1/2), π / 4], otherwise it is at [0, tan ^-1 (1/2)] do. Then increase the value of N by 1. Then repeat this process several times to achieve the desired precision.

On the other hand, the above-described frequency offset compensation method can be created by a computer program. The program may also be embodied by being stored in a computer readable media and being read and executed by a computer. The storage medium includes a magnetic recording medium, an optical recording medium and the like.

So far I looked at the preferred embodiments of the present invention. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. The scope of the present invention is shown in the claims, and all differences within the scope equivalent thereto should be construed as being included in the present invention.

400: frequency offset estimator
402: signal delay unit
404: integrator
406: phase grasping unit
410: frequency offset compensation unit

Claims (10)

A frequency offset compensation method in a mobile communication receiver,
Estimating a final frequency offset by summing two or more frequency offset estimates using a preamble included in the received signal;
Compensating for the received signal frequency offset using the frequency offset estimated by the frequency offset estimator.
The method of claim 1, wherein the estimating
Estimating a first frequency offset using a preamble included in the received signal;
Compensating for the frequency offset of the received signal using the estimated first frequency offset value;
Estimating a second frequency offset using the preamble of the signal whose frequency offset is compensated for;
Estimating a final frequency offset by adding the estimated first frequency offset value and the estimated second frequency offset value.
The method of claim 1, wherein the received signal is
Frequency offset compensation method characterized in that the signal of an orthogonal frequency division multiplexing (OFDM) communication system.
The method of claim 1, wherein the first frequency offset is
Frequency offset compensation method characterized in that estimated using the short training symbol of the preamble.
The method of claim 1, wherein the second frequency offset is
Frequency offset compensation method, characterized in that estimated using the long training symbol of the preamble.
A frequency offset estimator for estimating a final frequency offset by summing two or more frequency offset estimates using a preamble included in the received signal;
And a frequency offset compensator for compensating for the received signal frequency offset using the final frequency offset estimated by the frequency offset estimator.
The method of claim 6, wherein the frequency offset estimator
A first frequency offset is estimated using a short training symbol included in a received signal, a frequency offset of the received signal is compensated using the estimated first frequency offset value, and a preamble of the signal whose frequency offset is compensated for. And estimate a second frequency offset using a sum, and estimate a final frequency offset by adding the estimated first frequency offset value and the estimated second frequency offset value.
The method of claim 6, wherein the frequency offset estimator
A signal delay unit for delaying the input value by a certain number of samples;
A multiplier for multiplying the first input value by a signal delayed by a predetermined number of samples from the signal delay unit;
An integrator for integrating and outputting a signal output from the multiplier;
And a phase grasp for grasping a phase from the signal output from the integrator and outputting the phase.
The method of claim 8, wherein the phase grasping unit
Frequency offset estimation apparatus characterized in that the phase is identified by the average value of two or more phases.
The method of claim 8, wherein the frequency offset compensation unit
A frequency offset oscillator for generating a cosine and a sin value with an estimated frequency offset provided from the frequency offset estimator,
And a multiplier that multiplies the generated cosine and sine by the received signal.

Images