KR100249227B1 - Method of receiving digital signal broadcasting - Google Patents

Abstract

The present invention relates to an operation principle of a digital broadcast receiver, and more particularly, to a method of demapping data by extracting a response value of a reception channel using a pilot signal and inferring a response value of a data carrier using the response value. will be. In the digital broadcast receiving method of the present invention, a pilot signal capable of predicting a value at a receiving side is added at predetermined intervals, and the digital broadcast receiving apparatus receives data transmitted using multiple carriers. Receiving through a channel, applying the received data to an I & Q generator, converting the combined signal into a composite signal of the I signal and the Q signal, generating a FFT signal by Fourier transforming the composite signal, and detecting the pilot signals in the FFT signal. Checking a value of each pilot signal to extract a response value of a channel through which the pilot signal has passed; predicting a response value of a data carrier adjacent to the pilot signal based on the response value of each channel; Soft-determining the data in the response to correct errors contained therein, and demapping the soft-determined data. It is characterized by also configured.

Description

Digital broadcast receiving method

The present invention relates to a broadcast receiving apparatus, and more particularly, to a receiving apparatus for receiving a broadcast using multiple carriers.

With the development of digital technology, compression and decompression technology for video and audio data, research and efforts are being continued to replace the conventional analog transmission TV with a digital transmission TV. Such digital terrestrial TV broadcasting and transmission methods can be classified into US and European methods. In the United States, a VSB transmission method using a single carrier is transmitted, and a European method uses a Coded Orthogonal Frequency Division Multiplexing (COFDM) transmission method using a multi-carrier.

In particular, since the COFDM method transmits using multiple carriers, it is resistant to interference and ghosting in a multipath channel, and a SFN (Single Freqency Network) capable of wideband transmission at a single frequency can be constructed. The COFDM uses two error correcting codes, RS codes and a convolutional code encoder, in concatenation of a channel. The reason for concatenating and encoding channels as described above is to increase data reliability by double correcting channel errors that may occur in data transmission.

In particular, the error correction effect of the convolutional encoder and the Viterbi decoder determines the performance of the entire data transmission and reception system. Therefore, the data input to the Viterbi decoder is decoded by applying de-mapped data to the transmission data received through the antenna by a 3-bit or 4-bit soft decision method. It is known that the demapping data of this soft decision has a coding gain of about 2 dB compared to the hard decision of judging transmitted data only by 0 and 1.

FIG. 1 illustrates a hard decision, and FIG. 2 illustrates a soft decision. Assuming that an area is set in the quadrant of the complex number and the data is included in the area, the hard decision method is to replace the value corresponding to each quadrant and demap the data. This hard decision method is simple and easy to demap data.

The soft decision method is similar to hard decision, but it divides the area more by increasing the number of bits than hard decision, and when data contained in the area 10 is detected, the data is replaced by a value corresponding to each area to demap the data. That's how. This soft decision method can demap data more accurately than hard decision, but there is a concern that the system becomes more complicated as the number of bits increases.

However, since COFDM includes multiple carriers, each carrier is mixed with different noises during equalization. For example, if a carrier is in the notch portion of the frequency response, the carrier will have a lot of noise, while at peak this will be much less. Thus, carriers with high signal to noise ratios are more reliable than carriers with low signal to noise ratios. That is, the COFDM system detects the channel state of the current carrier and performs Viterbi decoding with reference to the channel state. Therefore, if there is interference between channels or there is selective fading of frequency, the system performance is greatly affected. Go crazy.

Therefore, in the conventional COFDM, as shown in FIG. 3, a signal called pilot 12, which can predict the value at the transmitting side, is transmitted between the carriers 11 of the data, and the pilot signal is received. The data signal is recovered by using a synchronization operation, an equalization operation, and the like which are necessary for the operation. In addition, the transmitter of the conventional COFDM transmits TPS (Transmission Parameter Signaling) including information on various transmission modes, and the receiver decodes the TPS to use for demodulation of data. That is, the COFDM is characterized by transmitting using multiple carriers, but also by adding a pilot signal to be used in the reception process.

4 and 5 show a receiving block diagram of a COFDM. The receiving operation principle of this COFDM is as follows. First, the signal s (t) received through the antenna is amplified by a necessary signal through the transmission channel 21 and the amplification stage 22. This amplification gain is adjusted via an Auto Gain Controller (AGC) 26. The amplified signal is converted into a digital signal by the A / D converter 23, and applied to the I & Q generator 24 to be converted into a composite signal of the I signal and the Q signal. The composite signal of the I signal and the Q signal is frequency and intensity adjusted through a fast fourie transformer (FFT) 25 and an equalizer 41, and a demapper 42 and a deinterleaver. 43, 45, and FEC (Foward Error Corrector) (not shown) to restore the original data.

The received data is first applied to a course timing block 27 and divided into valid data intervals and guard intervals. This time section obtains each correlation by shifting the data by one section, and the correlation is applied to the FFT start window generator 28 so that an appropriate FFT implementation section is obtained. Is set. After the I & Q composite signal is converted by the FFT unit 25, the TPS information is demodulated. At this time, the TPS decoder 29 analyzes the pilot signal, and corrects the distortion of the frequency and time of the received data by using the analyzed result to decode the correct TPS signal.

The FFT signal is then passed through a scattered pilot extractor 30 to detect a pilot signal, and the pilot signal is divided into an IFFF block 31 and a frequency controller (Narrow & Fine Freq. Ctrl). Analyzed at 32, the correct frequency band of the signal is corrected. In addition, the signal passing through the FFT is applied to a continuous pilot extractor 33 to perform frequency synchronization to correct an error of a received frequency.

The fine timing block 34 receives the data subjected to the FFT, and decodes the pilot signal included in the data to find the exact time of conversion of the FFT. The signal whose frequency synchronization and frequency band are corrected through the IFFF block 31, the frequency controller 32, and the proper time unit 34 is applied to the FFT unit 25 again.

The signal applied to the FFT corrects the error of the data carrier by hard decision or soft decision method.

Conventionally, evaluation values of data carriers have been detected by the method of soft decision and hard decision. However, the hard decision method simplifies the evaluation value of the carrier, and there is a high possibility that an error occurs. This is because the contrast is neglected for errors that may occur at the boundary 13 of the quadrant used in the complex model as shown in FIG. 1 or FIG.

The soft decision method may be somewhat more accurate than the hard decision method, but has a disadvantage in that the system becomes more complicated as the number of bits increases. In addition, there is a limitation that the performance of the system cannot be improved beyond a certain number of bits.

An object of the present invention is to reduce an error occurring in an evaluation value of a carrier and to improve the performance of a system in detecting an evaluation value of a data carrier.

1 shows a hard decision method such as QPSK decoding.

2 shows a soft decision method such as 16-QAM decoding.

3 is a diagram illustrating a pilot signal added to data in COFDM.

4 is a diagram illustrating a part of a reception block diagram of a COFDM.

5 is a diagram illustrating a part of a reception block diagram of a COFDM.

6 is a diagram illustrating a signal applied to a channel evaluator and an equalizer after passing through a channel.

* Explanation of symbols for main parts of the drawings

11, 11 ', 11 ", 11'": carrier 12, 12 ', 12 ": pilot signal

21: receiving channel 22: variable amplifier

23: A / D converter 24: I & Q generator

25: FFT 26: automatic gain controller

27: time section 28: start signal generator

29: TPS decoder 30, 33: pilot detection unit

31: IFFF Block 32: Frequency Controller

34: titration part 35: TPS extractor

40: channel evaluation unit 41: equalizer

42: demapper 43, 45: deinterleaver

44: Viterbi decoder 46: tuning detector

47: RS decoder 48: drandomizer

51: signal via COFDM converter 52: signal with noise

53: Signal with noise correction

The present invention is characterized in that the soft decision is made by extracting the response value of the pilot signal and predicting the response value of the carrier adjacent to the pilot signal.

The digital broadcasting apparatus inserts a pilot signal for each predetermined carrier into data at the transmitting side so that the receiving side can predict the value. Hereinafter, the digital broadcast reception method of the present invention will be described with reference to FIGS. 4 and 5.

First, the data including the pilot signal is received through the channel 21. The received data is converted into digital data through a predetermined amplification process and conversion process. This digital data is applied to the I & Q generator 24 and converted into a composite signal of the I signal and the Q signal. This composite signal is applied to a Fast Fourie Transformer (FFT) or FFT unit 25 to generate an FFT signal.

After synchronizing the signal passed through the FFT, the pilot included in the signal is extracted and the characteristics of the channel on which the signal is transmitted are detected. FIG. 6 shows graphs of signals 51, 52, and 53 applied to the channel estimator 40 and the equalizer 41 after the synchronization process.

Data transmitted from the COFDM transmitter is that x (l, k) is applied to channel H (l, k). At this time, noise n (l, k) is added to become an error signal y (l, k). At this time, l represents a symbol of COFDM, and k represents each carrier. The data transmitted from the transmitter is corrupted by the noise of the channel. This error is corrected by predicting the frequency response of the channel using the pilot in the channel evaluator and applying the predicted value to the data in the equalizer.

That is, knowing the predicted value of the channel state, it is possible to compensate for the distortion of the channel. The predicted value of this channel can be known by using the pilot previously added to the data. 3 shows an example of a pilot signal added to data. The horizontal axis of the data shown in FIG. 3 represents the direction of the carrier 11 of the COFDM, and the vertical axis represents the direction of the symbol of each COFDM. That is, the carrier is the data listed in the direction of the frequency axis, the symbol is the data is listed in the direction of the time axis. Each pilot is inserted every symbol of four COFDMs, and a pilot is inserted every 12 carriers in each symbol. Therefore, data received in the k-th carrier of the l-th symbol may be expressed as follows.

Y _lk = X _lk × H _lk + N _lk

Where X _lk is the value of the corresponding transmitted carrier, H _lk is the frequency response of the channel, and N _lk is the Gaussian noise.

In the transmitter, a pilot signal is inserted at a predetermined position. Referring to FIG. 3, the horizontal axis has a repeating frequency every 12 spaces, and the pilot 12 is inserted, and the vertical time axis has a repeating time every 4 spaces. In order to correctly obtain the transmission data X _lk , the prediction value of the channel is obtained, and the received signal is divided by the prediction value of the channel to obtain an evaluation value of the transmission data. The predicted value of the channel is obtained from the evaluation value of the frequency response value H _lk of the channel.

However, since the received signal is a pilot carrier, the evaluation value obtained through the prediction value of the channel is not the evaluation value of the actually transmitted data carrier but the evaluation value of the pilot carrier. Therefore, by using interlolation on the time axis and the frequency axis, respectively, from the evaluation values of the pilot, an approximation of the evaluation values of the actually transmitted data carriers can be obtained. Interpolation on the frequency axis only needs to store one COFDM symbol in the memory, and interpolation on the time axis may be performed after storing up to four COFDM symbols. At this time, since the interpolation of the time axis should have a memory for storing up to four COFDM symbols, it is possible to use only the interpolation of the frequency axis when there is no space in the memory.

The FFT signal is stored in the memory and the pilot signal inserted at the transmitting side is detected. When data is received through the value of this pilot signal, the response value (or gain value) of the channel which has passed is obtained. The system can predict the response of the carrier adjacent to the pilot signal by the response of this channel. For example, as shown in FIG. 3, when the pilot signal is inserted into each carrier on the time axis and the frequency axis at regular intervals, the response value of the first pilot signal 12 'is 1, and the second pilot signal 12 ". If the response value of? Is 0.5, the response values of the carriers 11 ', 11 ", 11'" in between can be predicted to be approximately 0.625, 0.75, 0.875, and the like.

When estimating the response value of the carrier, it is better to obtain the response value of the time axis first and the response value of the frequency axis later. This is because there are 12 carriers between the pilot signals on the frequency axis, but since there are 4 carriers between the pilot signals on the time axis, it is more reliable to predict carriers on the time axis.

When the response value of the carrier is predicted, a soft decision of the data is used using the response value of each carrier to correct an error of data generated during transmission. The soft decision method uses a 3-bit system. The 3-bit soft decision has an advantage of increasing data reliability than hard decisions such as QPSK, which simply determine 0 and 1. Of course, the 4-bit soft decision is more reliable than the 3-bit system, but as a result of various experiments, it has been found that the soft decision exceeding 3 bit has a limitation in increasing the reliability of the data due to the complicated system configuration.

If the response value of a carrier is predicted to be 100% or more, the bit value of the soft decision result is 11, and if it is predicted to be between 66% (2/3) and 100%, the bit value of the soft decision result is 10. If the value is predicted to be between 33% (1/3) and 66% (2/3), the bit value of the soft decision result is 01, and if it is predicted to be between 0 and 33% (1/3), the bit value of the soft decision result Is set to 00. This bit value can be arbitrarily changed as required by the system.

The signal power is calculated from the data carriers obtained as described above, and the signal-to-noise ratio of the data carrier is obtained by averaging the difference between the evaluation values of the data carriers calculated as an approximation according to the soft decision. Calculate the power.

The state information of the data carrier calculated in this way is de-mapped through the array of data transmitted from the TPS extractor 35. COFDM is transmitted in the array of QPSK, 16QAM, 64QAM, etc. This TPS extractor detects the information on the arrangement of such data. The transmitted data is multiplied by the channel state information or the signal-to-noise ratio, and then converted into the appropriate number of bits for processing by the system and applied to the inner deinterleaver 43. The data applied to the internal deinterleaver is restored to COFDM symbols and bit values corresponding to the respective carriers, and errors generated in transmission by the Viterbi decoder 44 are corrected and output.

Subsequently, the data of which the error is first corrected in the Viterbi decoder is detected by the synchronization detector 46 that detects packet sync of the transmission data, and then the external interleaver after the synchronization signal sync is detected. At 45). The data restored by the deinterleaver is further corrected by the RS decoder 47 to improve reliability.

The de-randomize 48 plays a role of restoring the signal rand-randomized by the transmitting side again. The reason for randizing the signal at the transmitting side is to remove DC components that may be included in the signal. The DC component included in the signal occurs due to the repetition of the same signal.

The present invention provides a system with better performance than using a conventional single carrier. This is because various data can be transmitted by using multiple carriers. Further, in obtaining an evaluation value of each data carrier in the process of demapping the data carrier, by inferring the response value of the channel for the carrier using the response value of the channel for the pilot signal, it is more than the conventional soft decision method. The advantage is that accurate estimates can be obtained.

Claims (3)

A digital broadcast receiver for receiving data transmitted using a multicarrier, wherein a pilot signal capable of predicting a value at a transmitter side is added at predetermined intervals, the method comprising: receiving the data through a predetermined channel; Applying the received data to an I & Q generator and converting the received signal into a composite signal of an I signal and a Q signal; Generating an FFT signal by Fourier transforming the composite signal; Detecting pilot signals in the FFT signal; Checking a value of each pilot signal and extracting a response value of a channel through which the pilot signal has passed; Predicting a response value of a data carrier adjacent to the pilot signal based on the response value of each channel; And softly determining the data in the response value of the predicted data carrier to correct errors included therein; And demapping the soft-determined data. The method of claim 1, wherein the pilot signal is inserted with the same value every four symbols in the data, and a constant value is inserted every twelve carriers in each symbol. The method of claim 1, wherein predicting the response value of the data comprises: first predicting a carrier between each pilot signal of the data in a time axis direction; And predicting a carrier between carriers predicted in the time axis direction in the frequency axis direction.

Images