KR19990070866A - Automatic frequency control method of digital broadcasting receiver - Google Patents

Abstract

The present invention is to provide a method of automatic frequency control of a digital broadcasting receiver that is simple in hardware or software and is easy to implement and robust against SNR environment or channel distortion.

The present invention extracts a continuous pilot signal in a carrier index from output data of a fast Fourier transform (FFT) block of a DVB-T receiver and simultaneously extracts two adjacent orthogonal frequency division multiplexing (OFDM) symbols (m, m + for each carrier index (l) by performing complex conjugate multiplication to the data d _l, step a, the extracted data by assuming the integer n of the carrier offset with respect to the predetermined range of detection for extracting _m each d _{l, m} The continuous pilot position vector P _{l, m} stored in the DVD-T receiver is multiplied with respect to all the carrier indices 1 and added to obtain the absolute value, the size of each of the assumed carrier offsets n is obtained, Estimating an offset and compensating a carrier offset n; setting a carrier offset n in the data d _{l, m} to zero and setting phase information of the extracted continuous pilot signal to Dividing the phase information into two pieces of phase information (R ₁ , R ₂ ) around a center frequency; adding the two pieces of phase information (R ₁ , R ₂ ) a step of estimating a partial compensation for the fractional part of the carrier offset, and a complex conjugate of the phase information (R ₁₎ and the rest of the phase information (R ₂₎ in the divided two-phase information _{(R 1, R 2) (} R 2 ^And calculating a sampling frequency error (offset) by detecting a phase from the resultant value, and compensating for the sampling frequency error using the estimated sampling frequency error (offset).

Description

Automatic frequency control method of digital broadcasting receiver

The present invention relates to an automatic frequency control method for a European terrestrial digital broadcasting (DVB-T) receiver, and more particularly, to an automatic frequency control method for a digital broadcasting receiver in which a carrier offset correction and a sampling frequency offset correction are simultaneously performed with a simple hardware structure .

The DVB-T system is the first European terrestrial digital broadcasting system to replace the PAL system, which is a European analog system. It is currently in trial broadcasting and parts development in various European countries. In addition, In order to determine the broadcasting standard, researches are being actively conducted through comparison with the American type.

The DVB-T standard is similar to DVB-S (Satellite), DVB-C (Cable) and channel coding standards (for example, Viterbi decoder and RS decoder).

However, considering that the modulation / demodulation method is terrestrial, it adopts Orthogonal Frequency Division Multiplex (OFDM) scheme.

The OFDM method differs from a general single frequency modulation / demodulation method in which information is continuously transmitted on a time axis, and the information is distributed over a plurality of frequencies. As it is difficult to accurately implement an oscillator, it is rarely used for purposes other than military purposes, and recently it has become possible to implement it digitally using an FFT chip (Fast Fourier Transform Chip). For example, ADSI, HDSL, and DAB and DVB-T in broadcasting.

Since the OFDM scheme divides the information into small portions on the frequency axis, the OFDM scheme is excellent in coping with multi-path, which is one of the factors of the largest signal distortion as compared with the single frequency modulation / demodulation scheme in the terrestrial wave.

However, due to the nature of the OFDM scheme, difficulties due to the frequency offset are much greater than in the single frequency modulation / demodulation scheme, so strict AFC is required at the receiving end.

1 is a schematic block diagram of a conventional DVB-T receiver.

The receiver is roughly divided into an RF demodulator, an OFDM demodulator and a source decoder. The RF demodulator is composed of a tuner 11. The OFDM demodulator includes an A / D converter 12, an I / Q signal generator 13), a frequency corrector (14). The AGC circuit 15, the timing synchronizer 16, the FFT (Fast Fourier Transform) block 17, the pilot extractor 18, the equalizer 19, the AFC 20, the VCO 21, (22). An inner deinterleaver 23. Viterbi decoder 24. An out-interleaver 25, an outer decoder 26 and a de-randomizer 27, and the other portions are omitted.

In the conventional DVB-T, a pilot signal is inserted between general data for automatic frequency control (AFC), and a comb of pilot system is selected for each OFDM symbol.

1, the pilot signal is extracted from the output data of the FFT block 17 of the receiver, and the AFC 21 is extracted from the extracted pilot signal by a carrier The offset and the sampling frequency offset information are extracted to feed back the double sampling frequency offset information to the A / D 12 in the VCO 21, and the carrier offset information is fed back to the frequency corrector 14 to digitally process the carrier offset And the sampling frequency offset.

The pilot signal generally has a continuous pilot inserted in a constant position of the carrier index (0 to 1704 in the 2k mode) and a scattered pilot in which the carrier index inserted in accordance with the OFDM symbol changes .

Taking the 2k mode as an example, there are 45 consecutive pilots and 142 distributed pilots among 1705 total carriers per OFDM symbol.

The continuous pilot is mainly used for the AFC, and the distributed pilot is used for the equalizer 19.

The automatic frequency control method based on the AFC carrier offset and the sampling offset correction of the DVB-T receiver according to the related art is complicated not only in terms of hardware and software but also has a problem in that it can not be robust to a low SNR environment or channel distortion have.

SUMMARY OF THE INVENTION The present invention has been made in view of the problems of the prior art, and it is an object of the present invention to provide a method of automatic frequency control of a digital broadcast receiver that is simple in hardware and software and is easy to implement and robust against SNR environment and channel distortion.

1 is a block diagram showing a main part of a conventional DVB-T receiver,

FIG. 2 is a block diagram of a circuit part implementing the automatic frequency control method of the digital broadcast receiver of the present invention

DESCRIPTION OF THE REFERENCE NUMERALS

11: Tuner block 12: A / D converter

13: I / Q generator 14: Frequency compensator

15: AGC 16: timing synchronization device

17: FFT block 18: Pilot extraction section

19: equalizer 20: AFC

21: VCO 22: demapper

23: Inner deinterleaver 24: Viterbi decoder

25: Outer deinterleaver 26: Outer decoder

27: di-randomizer 30: mode setting unit

31: Continuous pilot extraction and multiplication unit

32: Noise estimator 33: Coarse frequency multiplication setting unit

34: Unnecessary pilot removal and pilot division

35: Pilot position information section 36: Adder

37: multiplier 38, 40: phase detector

39, 41: Loop filter

In order to achieve the above object, an automatic frequency control method of a digital broadcast receiver is characterized by extracting a continuous pilot signal in a carrier index from output data of a fast Fourier transform (FFT) block of a DVB-T receiver, Extracting data d _{l, m} by performing a complex conjugate multiplication operation with respect to each carrier index (l) on a division multiplex (OFDM) symbol (m, m + 1) n is calculated by multiplying the extracted data d _{l, m} by the continuous pilot position vector P _{l, m} stored in the DVD-T receiver with respect to all the carrier indices l, and then the absolute value is taken to calculate the assumed carrier offset n after the step of obtaining a size of the respective estimates the maximum value as a carrier offset that compensates for the carrier offset n and a, the data d _l, the carry in _m Dividing an offset n in a series of two phase information (R _1, R ₂₎ around the center of the phase information of the pilot signal frequency extraction were fixed to zero, and the two-phase information (R _1, R ₂₎ Estimating a phase from the resultant value to estimate a fractional part of the carrier offset to compensate for a fractional part of the carrier offset; and calculating phase information R (R) from the divided two phase information R ₁ , R _{2 1} ) with the conjugate complex number (R ₂ ^* ) of the remaining phase information (R ₂ ), then estimating the phase from the resultant value as a sampling frequency error (offset) and compensating for the sampling frequency error .

Before describing an embodiment of implementing an automatic frequency control method of a digital broadcast receiver of the present invention, the principle of construction of the present invention will be described in detail.

Since the carrier offset is usually several times the carrier spacing of the OFDM signal, it can be divided into an integer part n and a fractional part δ when the interval between the frequencies is normalized to 1, It is also different.

On the other hand, the sampling frequency offset (error) σ is generally very small in ppm units.

In order to obtain n, delta, and sigma, mathematical analysis of the OFDM signal is first required.

The output of the FFT block 17 of the OFDM signal is mathematically expanded and summarized as shown in Equation (1).

In formula 1 m is the OFDM symbol index, l is the carrier index, γ is a ratio, R _{1 + n, m} is the data D _{1 + n,} the frequency response of the _m and a channel of the guard interval (Guide Interval) of the effective region H _{1 + n, m} .

In Equation (1), the l-th carrier Y _{l, m} of the m-th OFDM symbol is reduced in the form of a Sinc function by the carrier offset and the sampling frequency offset, the phase is shifted, and the ICI (Inter- Carrier Interference) .

And, as can be seen from Equation (1), the carrier offset? Has the same effect on all the frequency carriers, while the sampling frequency offset has a linear influence on the carrier index l .

In order to detect an AFC signal, i.e., a carrier offset and a sampling frequency offset signal, the phase of the random signal H _{l + n, m} must be removed.

However, since H _{l + n, m} can be assumed to be a slowly varying amount as compared to during the OFDM symbol, it can be seen that there is little change during two adjacent OFDM symbols.

Assuming that the values of two adjacent OFDM symbols (m, m + 1) are complex conjugate multiplication for each carrier index l under the above assumption, Equation 2 is obtained.

here alpha Is a conjugate complex multiplication of the data D _{1 + n, m} and D _{1 + n, m + 1} , and is a constant when it is a pilot.

Therefore, the integer part n, the fractional part delta, and the sampling frequency offset? Of the above-described carrier offset are detected in the following manner.

Detection of carrier offset integer part n

Assuming n for the determined detection range, add the result dl _{, m} of Equation (2 ₎ to all consecutive pilot indexes l for each n _, and then calculate an absolute value to calculate an index indicating the greatest amplitude, Is an estimated value n ', and mathematically expressed as Equation (3).

In Equation (3), P _{l, m} is a vector indicating continuous pilot positions.

The integer part n of the carrier offset can be detected in the manner described above.

This is because the absolute value of the correction value of Equation (3) has the largest peak value at the desired position n and has a very small value at the remaining positions because the position of the continuous pilot is random.

Detection of carrier offset δ and sampling frequency error (offset) σ

After compensating for the carrier offset n, the carrier offset [delta] and the sampling frequency error [sigma] can be set to zero if n is compensated in equation (2), where the phase is a function of the carrier offset [ have.

The above-described carrier offset [delta] has the same effect on all the frequency indices l .

On the other hand, the sampling frequency error sigma contains information in its slope. Therefore,? And? Can be calculated as follows. That is, since? Is generally a very small value, the carrier offset? Can be obtained from the average value of the phase of the carrier index l in Equation (2).

On the other hand, the sampling frequency error sigma can be obtained in a very simple manner by hardware by estimating the slope of the position.

The automatic frequency control method of the digital broadcast receiver will be described with reference to FIG.

FIG. 2 is a block diagram for implementing an automatic frequency control method of a digital broadcast receiver according to the present invention. First, a mode setting unit 30 sets a mode to a coarse mode to detect an integer part n of a carrier offset .

Subsequently, a continuous pilot index (45 in the 2k mode) inserted at a predetermined position of the carrier index m of each symbol is extracted from the output data of the fast Fourier transform (FFT) block 17 of the digital broadcast receiver (M, m + 1) for each of the carrier indexes to obtain OFDM signals d _{l, m} whose phases are removed from the channel frequency response H _{l + n, m} ). This process is performed in the pilot extraction and multiplication unit 31 of Fig.

Next, n is assumed for a predetermined detection range, and the continuous pilot position information vector Pl, n (supplied from the pilot position information section 36) already stored in the receiver is multiplied by dl _{, m} , (n) of the carrier offset n is calculated by taking the next absolute value added for all consecutive pilot indices l for n, and choosing the maximum value among the values for each assumed n. This process is roughly performed in the frequency setting unit 33. [ And then feeds the estimated value n 'back to the frequency corrector 14.

When the carrier offset n is obtained in the above-described approximate mode, the mode is switched to the fine mode. This process is performed in the setting mode (30).

In the fine mode, n is fixed to 0 in the OFDM signal dl.m, and at the same time, data of extremely small or large size among the extracted continuous pilots is removed to remove unnecessary pilot signals, and then the remaining extracted continuous pilots are collected (R ₁ , R ₂ ) based on the intermediate value of the frequency phase of the continuous pilot.

This process is performed in the unnecessary pilot removal and pilot division section 34.

Since the SNR can be estimated to be a relatively low value by removing the data having a small size (amplitude) among the continuous pilots extracted as described above, the data having such a small size deteriorates the entire SNR rather in the averaging process. To remove too large a data, the algorithm of the present invention is not influenced by the phase of the channel response (Hl + n, m in Equation (2)) but slightly influenced by its size. Value can be biased.

The divided pilot frequency phase information R _1, R ₂ is a complex number, for example, _{R 1 = a + jb, R} 2 = c can be represented as + jd, this complex phase information of R _1, R ₂ are then added As a result, the phase, that is, the carrier offset? Is calculated based on R ₁ + R ₂ = a + c + j (b + d).

This process is performed by using the adder 36 and the phase detector 37. [

In general, it is very difficult to extract the phase information strictly from a complex number to a hardware.

Therefore, in the present invention, the estimated value? Of the carrier open set is generated by dividing -180 degrees to 180 degrees into only a few sets. For example, the phase can be separated by 90 degrees by only the sign of the real part and the imaginary part of the complex number, and phase separation can be performed at 45 degrees by adding the magnitude comparison of the imaginary part and the imaginary part.

At the loop filter 39, the gain according to the phase information noise devel is multiplied and filtered to produce a carrier offset estimate, and the frequency is automatically compensated by pifbacking with the frequency corrector 14, such as the carrier offset n described above.

The sampling frequency error (offset)? Is also obtained by using the continuous pilot division phase information R ₁ and R ₂ described above.

In other words, the phase information multiplied by R using a commitment complex ₂ R ₁ R ₂ and the phase information to the phase detector 40 from the result value of a complex number is obtained estimation value σ 'of σ sampling frequency error (offset).

In this case as well, in the same manner as the method of obtaining the carrier offset?, -180 degrees to 180 degrees are divided into several sets, and the estimation value? 'Of the sampling frequency error is obtained using the loop filter 41 based on the result.

Of course, at this time, the gain according to the noise level is multiplied by the phase information, and then filtered to estimate? '.

This estimated value? 'Is fed back to the VCO 21 to control the frequency.

The noise level is obtained using the noise estimator 32. [

The noise estimator 32 calculates the variance of the output at the pilot position on the time axis at the output of the FFT 17 block, roughly estimates the magnitude of the noise therefrom, Thereby generating a signal noise level.

This noise level is used to estimate the phase information? And? As described above. If the noise level is high, the error of the steady state of the automatic frequency control device increases due to the catch, When the noise level is low, the gain is controlled to be slightly higher than the normal value in order to raise the tracking speed of the automatic frequency controller.

According to the automatic frequency control method of the digital broadcast receiver of the present invention as described above, the automatic frequency control of the DVB-T receiving end can be implemented effectively and stably, and in actual implementation, both the software and the hardware have a simple structure , The technique of the present invention can be applied not only to DVB-T but also to any OFDM receiver having a pilot insertion structure similar to DVB-T.

Claims (6)

A continuous pilot signal in a carrier index is extracted from an output data of a fast Fourier transform block of a digital broadcasting receiver and simultaneously two adjacent orthogonal frequency division multiplex symbols are subjected to a conjugate complex multiplication operation for each carrier index Extracting data, Assuming the integer part n of the carrier offset of the carrier offset with respect to the determined detection range, the continuous data pile position vector stored in the digital broadcast receiver is multiplied by the extracted data and all the carrier indexes are added to the extracted data, Calculating a magnitude in each of the carrier offsets and then compensating for a carrier offset by estimating a maximum value as a carrier offset; Dividing the phase information of the extracted continuous pilot lot signal into two phase information about the center frequency after fixing the carrier offset at 0 in the data, Adding the two phase information, detecting a phase from the resultant value to estimate a fractional part of the carrier offset, and compensating a fractional part of the carrier offset; Estimating a phase from the resultant value by multiplying a conjugate complex number of the divided two phase information and estimating the phase as a sampling frequency and compensating for a sampling frequency error using the estimated sampling frequency; Control method. The method according to claim 1, Wherein phase detection of the carrier offset delta is performed separately by 90 degrees using the sign of the real part and the imaginary part of the sum of the divided two pieces of phase information R ₁ and R ₂ . Control method. The method according to claim 1, The phase detection of the sampling frequency open set sigma is performed using the sign of the real part and the imaginary part of the complex number formed by the conjugate complex number R ₂ ^* of the divided two phase information R ₁ and R ₂ and the product R ₁ x R ₂ ^* Wherein the first and second frequency bands are detected separately by 90 degrees. 3. The method of claim 2, The phase detection of the carrier offset δ is performed by detecting the real and imaginary parts of the sum R ₁ + R ₂ of the synthesized R ₁ and R ₂ as well as the real part and the imaginary part, Of the digital broadcasting receiver. The method of claim 3, The phase detection of the sampling frequency offset? Is separately detected at 45 degrees by using not only the sign of the real part and the imaginary part of the complex number calculated as a result of the multiplication of the complex conjugate of the phase information R ₁ and R ₂ but also the size thereof Of the digital broadcasting receiver. The method according to claim 1, Wherein the estimated value of the carrier offset? And the sampling frequency offset? Is multiplied by a gain according to each noise devel through each loop filter, and then filtered.