KR100466544B1 - Method of correcting OFDM symbol error of digital audio broadcasting receiver and system of the same - Google Patents

Abstract

PURPOSE: An OFDM(Orthogonal Frequency Division Multiplex) symbol error correcting method of a digital audio broadcast receiver and a system thereof are provided to reduce bit-error probability by correcting a phase which is abnormally transited due to noise. CONSTITUTION: A transmission stage(100) transmits digital audio broadcast signal. A channel(200) transfers the digital audio broadcast signal. A receiving stage(300) receives the digital audio broadcast signal. An output signal of a transmission convolutional decoder(110) of the sending stage(100) is transmitted to the channel(200) through an interleaver(120), a QPSK(Quadrature Phase Shift Keying) symbol mapper(130), a differential encoder(140) and an IDFT(Inverse Discrete Fourier Transform) unit(150). A DFT(Discrete Fourier Transform) unit(310) performs DFT on the signal transmitted through the channel(200) and transmits it to a differential decoder(320). The differential decoder(320) decodes the discrete-Fourier-transformed signal to limited bits and transmits it to a symbol phase prediction transistor(330). The symbol phase prediction transistor(330) predicts a phase of a currently transmitted OFDM symbol on the basis of a symbol of a previously transmitted signal. If the OFDM symbol has been distorted, the symbol phase prediction transistor(330) transits the distorted signal to an estimate region of a predictable phase and transmits it to a QPSK symbol demapper(340). The QPSK symbol demapper(340) combines separately transmitted I and Q symbols and transmits it to a reception convolutional decoder(360) through a deinterleaver(350).

Description

Method of correcting OFDM symbol error of digital audio broadcasting receiver and system of the same}

The present invention relates to a method and system for error correction of an orthogonal frequency division multiplexed symbol of a digital audio broadcasting receiver, and in particular, π / 4-differential quadrature phase shift keying (DQPSK) of a digital audio broadcasting receiver. The present invention relates to a symbol correction method of an orthogonal frequency division multiplex (hereinafter referred to as 'OFDM') modulated signal.

In general, DQPSK is a standardized (IEEE 802.11) standard for use as a baseband modulation method in a wireless local area information network (LAN) and a phase modulation that performs a logical sum of binary codes on the transmitting side to shift four kinds of phase shifted waves. It is used for the direct sequence spectrum spreading method. Transmitting the digital signal of the logical sum of the code to be transmitted at the transmitting side corresponding to the four phases of the carrier and performing quadrature phase shift, and recovering the original pulse by converting the logic difference after the demodulation process at the receiving side.

OFDM was developed for digital audio broadcasting in Europe as a digital modulation scheme to improve transmission rate per bandwidth and prevent multipath interference, and commercial broadcasting has been started since 1996. The characteristics are that the multicarrier modulation method uses hundreds of carriers, and that each carrier is orthogonal to each other. Therefore, the frequency components of each carrier may overlap each other. The frequency utilization efficiency is high because much more carrier multiplexing is possible than normal frequency division multiplexing (FDM). Serially and parallel-converted encoded data is allocated to each carrier and digitally modulated. More carriers can increase the transmission rate per bandwidth.

However, a signal transmitted through the above-described modulation and multiplexing scheme may generate a lot of noise during transmission, between repeaters, and during decoding and conversion. If the transmission signal passes through the ideal transmission channel, the random phase that may occur in the transmission channel is canceled during the differential decoding process. Even so, when detecting symbols in the baseband, the error of the local oscillator, the bit limitation of the analog-to-digital converter, and the increase of noise depending on the environment of the transmission channel, the i-channel ), Unbalance of the Q-Channel, fast fading channel, if the random phase changes rapidly compared to the symbol rate, unwanted phase values can be obtained, which can lead to symbol errors or bit errors. An increase in the noise power density versus bit energy (E _b / N ₀ ) transmitted causes a problem that can result in power loss.

Therefore, in order to solve the above problem, the present invention focuses on the fact that the phase shift from the current symbol phase to the next symbol is determined due to the π / 4-DQPSK modulation characteristic of the OFDM symbol passing through the transmission channel. By using the previous phase to anticipate the next phase to come, correcting the symbol by correcting the phase that has been abnormally shifted by the noise reduces the probability of bit error, and lower noise power density versus bit energy environment and lower power system. It is an object of the present invention to provide a method and apparatus for orthogonal frequency division multiplexed symbol error correction of a digital audio broadcasting receiver capable of realizing.

1 is a conceptual diagram illustrating a structure of a transmission frame for digital audio broadcasting according to the present invention.

2 is a block diagram of a system for orthogonal frequency division multiplexed symbol error correction of digital audio broadcasting according to the present invention.

3 is a diagram illustrating eight positions in which a symbol transmitted in a π / 4-DQPSK system according to the present invention may appear in a phase diagram.

4 is four constellations of OFDM symbols that are normally receivable when the positions of the initial and previous OFDM symbol phases are (1, 0).

FIG. 5 is four constellations of OFDM symbols in a case where an abnormal phase shift may occur due to channel noise when the positions of initial and previous OFDM symbol phases are (1, 0), and FIG. 6 is abnormal due to channel noise. This is a channel phase diagram for explaining a process of compensating the phase to a predictable OFDM symbol receiving region by correcting the phase when the phase shift is performed.

7 is a block diagram illustrating a digital audio broadcast receiving end of the present invention.

FIG. 8 is a detailed diagram of an I-Q diagram and a quantized value to represent all phases in which a transmitted symbol can be located at a receiver according to the present invention.

<Explanation of symbols for the main parts of the drawings>

100: transmitting end 110: transmitting convolutional encoder

120: interleaver 130: QPSK symbol mapper

140: difference encoder 150: inverse discrete Fourier conversion means

200: channel 300: receiving end

310: discrete Fourier transform means 320: differential decoding means

330: symbol phase predictor 340: QPSK symbol demapper

350: deinterleaver 360: reception weaving encoder

322, 324: memory 326: differential decoder

328: reference phase 331: phase comparator

332: error identification detectors 410, 420: digital phase shifter

412, 422: bit converter 414, 424: phase correction signal output means

416, 426: addition and subtraction part

In the error correction method of an OFDM symbol of a digital audio broadcasting receiver using an OFDM symbol signal separated by an I-channel and a Q-channel and modulated in a π / 4-DQPSK demodulation method, the digital audio broadcasting Determining an expected region by estimating the phase of the currently received OFDM symbol from the phase of the OFDM symbol received one OFDM symbol before the currently received OFDM symbol, and detecting the phase of the currently transmitted OFDM symbol And when the detected phase is located in the expected region, the currently received OFDM symbol is transmitted as it is, and when the detected phase is not located in the expected region, the phase of the currently received OFDM symbol is estimated. The present invention provides a method for correcting an error of an OFDM symbol of a digital audio broadcasting receiver, which is transmitted by transferring to an area.

In addition, differential decoding means for delaying the OFDM symbol currently received through the I-channel and the Q-channel by one symbol period and then differentially decoding the received OFDM symbol, and receiving the decoded OFDM symbol by using the mapping table. A phase comparator for detecting a phase of a phase; extracts and stores an expected region in which the currently received OFDM symbol phase is to be located from a received symbol one OFDM symbol before the currently received OFDM symbol, and then detects the phase comparator If the phase is not located in the expected region, the error symbol detector for generating an error phase flag signal and the error phase flag signal pass immediately if the error phase flag signal is not generated. If the error phase flag signal is generated, the currently received OFDM symbol is generated. A first and second digital phase shifters for shifting the phase of the &lt; RTI ID = 0.0 &gt; I &lt; / RTI &gt; It provides an error correction system of the OFDM symbol of the digital audio broadcasting receiver according to claim.

Hereinafter, with reference to the accompanying drawings will be described an embodiment of the present invention in more detail. However, the present invention is not limited to the embodiments disclosed below, but will be implemented in various forms, and only the embodiments are intended to complete the disclosure of the present invention, and to those skilled in the art to fully understand the scope of the invention. It is provided to inform you. Like numbers refer to like elements in the figures.

1 is a conceptual diagram illustrating a structure of a transmission frame for digital audio broadcasting according to the present invention.

Referring to FIG. 1, the structure of a transmission frame for digital audio broadcasting includes a sync channel, a fast information channel, and a main information channel, and the main information channel is composed of 76 OFDM symbols. In addition, one OFDM symbol consists of 3072 bits. The π / 4-DQPSK modulation scheme can noncoherent detect bits transmitted without accurate phase information, and is particularly suitable for fading channels with multipath. In the digital audio broadcast frame transmission, the phase reference signal is transmitted every frame in the sync channel.

2 is a block diagram of a system for orthogonal frequency division multiplexed symbol error correction of digital audio broadcasting according to the present invention.

2, the orthogonal frequency division multiplexed symbol error correction system of the digital audio broadcasting of the present invention includes a transmitter 100 for transmitting a digital audio broadcast signal, a channel 200 for transmitting a digital audio broadcast signal, and It comprises a receiving end 300 for receiving a broadcast signal for digital audio. The transmitter 100 comprises a transmission convolutional encoder 110, an interleaver 120, a QPSK symbol mapper 130, a differential encoder 140 and an inverse discrete Fourier transform means 150. The receiving end 300 includes a discrete Fourier transform means 310, a differential decoding means 320, a symbol phase predictor 330, a QPSK symbol demapper 340, a deinterleaver 350 and a reception convolutional encoder 360 It is made, including.

The output signal of the transmission convolutional decoder 110 located in the transmitter 100 is transmitted to the channel 200 through the interleaver 120, the QPSK symbol mapper 130, the differential encoder 140, and the inverse discrete Fourier transform means 150. Is sent. The signal inputted to the receiver 300 through the channel 200 is a discrete Fourier transform means 310, a differential decoding means 320, a symbol phase predictor 330, a QPSK symbol demapper 340, a deinterleaver And the reception convolutional decoder 360.

In detail, the digital convolutional broadcast frame passed through the transmission convolutional decoder 110 and the interleaver 120 is transmitted to the QPSK symbol mapper 130. Since the bit size of one OFDM symbol in the transmission mode-eye is 3,072 bits, the QPSK symbol mapper 130 splits it into 1,536 bits in each of I and Q channels and remaps the QPSK symbols. The differential encoder 140 transitions by limiting the phase of each remapped symbol to four cases of -1/4 π, +1/4 π, -3/4 π, and +3/4 π. That is, π / 4-DQPSK modulation is performed. The inverse discrete Fourier transform means 150 performs inverse discrete Fourier transform on the π / 4-DQPSK modulated signal and transmits the signal to the receiver 300 through the channel.

The discrete Fourier transform means 310 performs discrete Fourier transform on the signal transmitted through the channel 200 and transmits the signal to the differential decoding means 320. The differential decoding means 320 decodes the discrete Fourier transformed signal into limited bits and transmits it to the symbol phase predictor 330. The symbol phase prediction shifter 330 predicts the phase of the currently transmitted OFDM symbol based on the symbol of the previously transmitted signal, and if the OFDM symbol is distorted, shifts the distorted signal to the predicted region of the predictable phase. And transmits it to the QPSK symbol demapper 340. The QPSK symbol demapper 340 combines the symbols transmitted to each of I 'and Q' and transmits them to the reception convolutional decoder 360 through the deinterleaver 350.

<Transmitter Signal Conversion>

In digital audio broadcasting, signals are transmitted after modulating differentially encoded OFDM symbols by a quadrature phase shifting method using π / 4-DQPSK modulation. That is, a signal to be transmitted before transmission is first encoded by a gray code, and then transmitted through a channel after phase modulation is performed.

First, the differential encoding equation formed by the differential encoder 140 of the transmitter 100 will be described.

Where u _k and v _k represent the amplitude encoded differentially during one symbol of the I-channel and the Q-channel. Since the initial value of phase should be given during π / 4-DQPSK modulation, if u ₀ = 0 and v ₀ = 1, then u _k and v _k depend on the values of I and Q. And It can have a value of.

The signal differentially encoded by the above equation is again outputted as a transmission signal through inverse discrete Fourier transform. The transmission signal 's (t)' is described as follows.

here, This depends on the encoded value.

The phase relationship between the phase of the currently output signal and the signal output immediately before is as follows.

here Denotes the difference in phase determined by the input data. Also, the change in phase (difference between the phase of the current transmission signal and the phase of the previously transmitted signal) can be obtained by using the absolute phase ( ), Not the difference in phase ( Is caused by).

Table 1 shows the values of I and Q. It is shown.

I _k Q _k 0 0 π / 4 0 One -π / 4 - One 0 -3π / 4 - One One 3π / 4 - -

Referring to Table 1, according to I, Q values (Difference in phase determined by the input data) is a phase shift at odd times of? / 4. In other words, the phase shift determined by the input data cannot be shifted by + π / 2, -π / 2, + π, and -π. Through this, it can be seen that the phase of each symbol is limited to four cases of -1/4 π, +1/4 π, -3/4 π, and +3/4 π through the difference encoder 140. .

3 is a diagram illustrating eight positions in which a symbol transmitted in a π / 4-DQPSK system according to the present invention may appear in a phase diagram.

Referring to FIG. 3, phases that may appear in signals transmitted and received in a π / 4-DQPSK system are A (+1, 0), C (0, +1), E (-1, 0), G ( 0, -1), B ( , ), D (- , ), F (- ,- ) And H ( ,- )to be. In addition, as shown in the constellation diagram, the case of being able to appear in the next phase from the initial and previous phases is reduced to four again.

Receiver Signal Conversion

The signal transmitted by the transmitting end 100 of the orthogonal frequency division multiplexed symbol error correction system of the digital audio broadcasting is transmitted to the receiving end 300 through the channel 200. When the non-coherent method detects a symbol, the received signal is expressed as an equation.

here Denotes the random phase generated by the channel.

The outputs of the I and Q channels passed through the lowpass filter are respectively

And according to the decoding rule (differentially decoding rule)

Becomes

From the above results It can be seen that.

this is By estimating the value of in Table 1, the transmitted symbol can be predicted. It is output in multiplexed bit format for symbol determination, and finally the transmission bit is expected through soft decision.

Using the π / 4-DQPSK system characteristic, the receiving end inverts the phase of the currently transmitted OFDM symbol based on an initial reference phase or an OFDM symbol transmitted one OFDM symbol, and is distorted by channel noise or fading channel. By shifting the phase of the OFDM symbol to the expected region, it is possible to correct the error by re-detecting the transmitted OFDM symbol.

<Apply constellation and phase diagram>

Hereinafter, error correction will be described with reference to constellations and channel phase diagrams.

As shown in FIG. 3, the phase of the signal received by the receiver 300 has eight phase states (A, B, C, D, E, F, G, H), and the phase of the initial or previous signal. According to the state, the phase of the currently transmitted signal may appear in four states.

4 is four constellations of OFDM symbols that are normally receivable when the positions of the initial and previous OFDM symbol phases are (1, 0).

Referring to FIG. 4, if the initial and previous OFDM symbol phases received at the receiving end 300 were A (1, 0), the currently received OFDM symbol phase is B ( , ), D (- , ), F (- ,- ) And H ( ,- ), Otherwise it indicates an error in the received signal.

FIG. 5 is four constellations of OFDM symbols in a case where an abnormal phase shift may occur due to channel noise when the positions of initial and previous OFDM symbol phases are (1, 0), and FIG. 6 is abnormal due to channel noise. This is a channel phase diagram for explaining a process of compensating the phase to a predictable OFDM symbol receiving region by correcting the phase when the phase shift is performed.

5 and 6, the phase of the currently received OFDM symbol when the phase of the OFDM symbol transmitted before the initial reference phase or one OFDM symbol is A (1, 0) is the same as that of FIG. 4 (B, D). , F, H). However, the currently received OFDM symbol is affected by noise, such as channel noise, to have a value other than the expected phase (B, D, F, H). The present invention detects this and shifts the phase of the error OFDM symbol to the nearest expected area. If the phase of the error OFDM symbol is B 'of FIG. 6 is predicted and the phase shift to B, the phase closest to B' and then detects the OFDM symbol. In addition, when the phase of the OFDM symbol is D ', it predicts the phase shift to D and then detects the OFDM symbol. This can reduce the bit error rate of the OFDM symbol.

<Configuration and operation of receiver>

FIG. 7 is a block diagram illustrating a digital audio broadcasting receiver of FIG. 2.

Referring to FIG. 7, the receiver for digital audio broadcasting of the present invention includes differential decoding means, a symbol phase predictor, and a QPSK symbol demapper.

The differential decoding unit 320 includes first and second memories 322 and 324 for delaying and storing the discrete Fourier transformed signal for each channel (I-channel, Q-channel) by one symbol period, and an OFDM symbol. And a differential phase decoder 326 for differentially decoding the signal and a reference phase generator 328 for storing and outputting the reference phase received from the sync channel.

The symbol phase predictor 330 is a phase comparator for digitally determining a position of the currently received I and Q phases by using a mapping table (comparing the input value with a value defined in the mapping table). 331) Determining and storing an expected region in which the currently received OFDM symbol phase is located from the OFDM symbol transmitted one OFDM symbol before the currently received OFDM symbol, and comparing the phase of the previous symbol and the currently received symbol to receive the current. Determines whether the phase of the symbol is in the expected area, and generates an error phase flag signal when the phase of the currently received symbol is not in four normal positions (expected area) (when the determined phase is an error symbol). Error symbol detector 332 for outputting a flag signal, and if an error phase flag signal does not occur, the phase of the currently received OFDM symbol is not changed for each channel (I, Q channel). If it passes immediately, and an error phase flag signal is generated, bit-convert the value of the currently received OFDM symbol for each channel, add or subtract a phase detection signal to the currently received OFDM symbol, and process the overflow or underflow process. First and second digital phase shifters 410 and 420 for shifting the phase of the currently received OFDM symbol. The error symbol detector 332 includes a phase prediction register that stores an expected phase region that the current symbol predicted by the previous symbol may have. In addition, the first and second digital phase shifters 410 and 420 may include first and second bit converters 412 and 312 for converting the values of OFDM symbols into I-channel and Q-channel, respectively. Add or subtract the first and second phase detection signal output means 414 and 424 for outputting the phase detection signal by the error phase flag signal of 332 and the value of the error phase detection signal and the bit-converted OFDM symbol, First and second addition and subtraction units 416 and 426 for controlling the overflow and underflow according to the result thereof.

The differential decoding means 320 and the symbol phase predictor 330 perform a current OFDM symbol, which is a discrete Fourier transformed signal, on the I-channel and the Q-channel, respectively, through the first and second memories 322 and 324. After delaying by a symbol period, differential decoding is performed using the differential decoder 326. The differential decoded signal uses the mapping table in the phase comparator 331 to determine the phase of the symbol. The error symbol detector 332 generates an error phase flag signal when the phase of the determined currently transmitted symbol does not match the phase predicted by the phase before one symbol stored in the error symbol detector. The first and second digital phase shifters 410 and 420 operate by the error phase flag signal to shift the phase of the error symbol to an adjacent expected phase region and redetect the symbol. The QPSK symbol demapper 340 converts the symbol into bits to detect the last transmitted bit.

The mapping table will first be briefly described.

FIG. 8 is a detailed diagram of an I-Q diagram and a quantized value to represent all phases in which a transmitted symbol can be located at a receiver according to the present invention.

Table 2 digitizes the 8-bit I-Q channel according to the present invention in a tabular format.

level Differential decoder output value (8 bits) Absolute Symbol Detection 0 0 0000000 +1.0000000 One 0000001 2 0000010 . . . . 35 0100011 +0.7176 ( ) 36 . . . . 127 1111111 +0.0078125 ( 0 ⁺ ) 128 One 0000000 -0.0078125 ( ⁰⁾ . . . . 219 0100011 -0.7176 ( ) 220 . . . . 253 1111101 254 1111110 255 1111111 -1.0000000

Referring to FIG. 8 and Table 2, if I and Q are separated from the front end of the digital audio broadcasting receiver, and then go through an analog-to-digital converter (not shown), 8-bit values of I and Q are obtained. Differential demodulation is performed using the differential decoder 326.

Specifically, the digital audio broadcasting receiver of the present embodiment is implemented with 8 bits, and the symbol detection decision region is digitized into 8 bits (256 * 256 = 25536) for each of the I-channel and the Q-channel to digitize the detection of the symbol, and the phase shift is performed. It is also simplified by processing with digital bit conversion. This results in 256 digitized results per channel. If the digitized value is differentially decoded, all I and Q phases that can be obtained as a result of differential decoding are represented in the I-Q phase diagram as shown in FIG. 8. In addition, Table 2 arbitrarily digitally maps 256 differentially decoded values that can be included in each channel to the portions described as 'differential decoder output values (8 bits)' at each level (0 to 255). The absolute value of symbol detection 'means the actual difference decoder 326 calculated value. At this time, one interval of the fine interval of Figure 8 is 0.0078125. However, the application of the present invention is not limited thereto and may be variously converted or substituted.

<Symbol error detection and correction method>

When the received symbol is in the symbol prediction region based on the previous symbol, the OFDM symbol is detected according to the normal procedure. However, if the currently received symbol is not located in the symbol prediction region as shown in FIG.Phase shift is performed. (At this timeIndicates the degree of error due to channel distortion.) When the phase of the previous OFDM symbol is located at A (+1, 0), four regions of the currently received symbol that may be caused by the channel error are shown in FIG. In Table 2, {(0 + n, 1 + n), (-1 + n, 0 + n), (0 + n, -1 + n), (+1 + n, 0 + n)} (where n represents the degree of dispersion of the signal due to channel noise) and the phase shift in the digital domain adds or subtracts a specified value to the I-channel and Q-channel to make the phase shift. This is done by converting the upper bits 0-> 1, 1-> 0 bits in hardware or adding and subtracting a specific value. And n represent the degree of dispersion of symbols due to channel noise, respectively, where ε and n do not have a certain criterion because the channel environment is changed arbitrarily and may have any value within a predetermined range. That is, it can have all real values in the range of-(π / 4) <ε <(π / 4) and -45 <n <45 by various variables such as the strength of the received signal and the distortion of the signal due to the multipath. .

If the phases A, C, E, and G of FIG. 5 are respectively phase converted to the digital symbol crystal plane of FIG. 8 (overlapping FIGS. 5 and 8),

((127 + n, 0 + n)-> (127-90 + n, 0 + 35 + n) = (37 + n, 35 + n),

(255 + n, 127 + n)-> (255-35 + n, 127-90 + n) = (220 + n, 37 + n),

(127 + n, 255 + n)-> (127-90 + n, 255-35 + n) = (37 + n, 220 + n),

(0 + n, 127 + n)-> (0 + 35 + n, 127-90 + n) = (35 + n, 37 + n)}.

If the position of the previous symbol is B ( , ), The four phase regions of the received OFDM symbol, which may be caused by , ), D (- , ), F (- ,- ), H ( ,- ), And the phase shift in the digital domain is performed by adding or subtracting a predetermined value to the I-channel and Q-channel in the same manner as before.

Phase shifting B, D, F, and G into the digital symbol crystal plane of FIG. 8,

((35 + n, 35 + n)-> (35-35 + n, 35 + 90 + n) = (0 + n, 125 + n),

(219 + n, 35 + n)-> (219-90 + n, 35-35 + n) = (129 + n, 0 + n),

(219 + n, 219 + n)-> (219 + 35 + n, 219-90 + n) = (254 + n, 129 + n),

(35 + n, 219 + n)-> (35-35 + n, 219-90 + n) = (0 + n, 129 + n)}, and the same method for other symbol phase positions do.

As shown in the above phase conversion result, the values of the decoded I-channel and Q-channel are added and subtracted according to the previous phase state as shown in the following equation, and the error phase is corrected to the expected area and then soft-decided to OFDM Correct it.

ego,

At this time, Is, And n denote the degree of dispersion of symbols due to channel noise, respectively.

With reference to FIG. 8 and Table 2, it demonstrates with a specific example based on the above-mentioned description. If the previous phase is (1, 0), it is assumed that the next phase (phase of the currently transmitted symbol) is (0.2968750, 08984375) rather than the four phase prediction regions in which the next phase may appear.

Converting the current phase to level values in the digital planes of Figures 8 and 2 results in (I, Q) = (89, 12), and converting it to bit values (I, Q) = (1011001, 00001100) Becomes Since the previous phase ((I, Q) = (1, 0)) is not the four phases expected in the mapping table, the current phase ((I, Q) = (89, 12)) is determined to be an incorrect phase . To correct this, (I, Q) = (89-90, 12 + 35) = (01011001-01011010, 00001100 + 00100011) using the above equation. At this time, since underflow has occurred in the I-channel side, the I value is mapped to 0, which is the minimum value in the IQ phase diagram of FIG. 8. Eventually we convert the current phase to the position of (0, 47), , The current phase is determined by the phase of) and the phase of the currently transmitted symbol is corrected.

The symbol phase prediction shifter 330 of FIG. 7 configures the above-described phase comparison, detection of error symbols, and correction of errors in hardware. This is the X output from the differential decoder 320_kAnd Y_kThe phase is determined by comparing the values to the values defined in Table 2, and then the signal is determined to be an error symbol and a flag signal is generated. In case of normal phase, it passes directly and if error flag occurs, bit-convert the value of the 8-bit input OFDM symbol or bit-convert as shown in Equation 7. Also, the error phase correction signal (), Subtract, underflow, and overflow to produce a phase shift, then for soft decision I '_k, Q '_kGenerates a value.

As described above, the present invention reduces the bit-error probability by correcting a phase that is abnormally shifted by noise by predicting the currently received OFDM symbol in advance using a previous phase. It can achieve the desired bit error rate in low E _b / N ₀ environments.

In addition, the use of π / 4-DQPSK modulation compared to the QPSK modulation reduces the bit error rate lost without the need for extra bits or devices for additional error correction. It is advantageous.

Claims (10)

In the OFDM symbol error correction method of a digital audio broadcasting receiver separated by an I-channel and a Q-channel, and using an OFDM symbol signal modulated by π / 4-DQPSK demodulation method, (a) determining an expected region by estimating the phase of the currently received OFDM symbol from the phase of the OFDM symbol received one OFDM symbol before the OFDM symbol currently received by the digital audio broadcasting receiver; (b) detecting a phase of the currently transmitted OFDM symbol; And (c) If the detected phase is located in the expected region, the currently received OFDM symbol is transmitted as it is; if the detected phase is not located in the expected region, the phase of the currently received OFDM symbol is changed. And an OFDM symbol of an OFDM symbol of the digital audio broadcasting receiver, which is transmitted by transitioning to the expected region. The method of claim 1, wherein the method for shifting the phase of the currently received OFDM symbol to the expected region comprises: bit-converting the currently received OFDM symbol, or adding or subtracting a phase correction signal value to the currently received OFDM symbol. And shifting the phase of the currently received OFDM symbol by performing an overflow or underflow process. The method of claim 2, wherein the phase correction signal has a value of 0, or Error correction method of an OFDM symbol of a digital audio broadcasting receiver. The method of claim 1, wherein in step (a), When the phase of the OFDM symbol received before the one OFDM symbol is (1, 0), (0, 1), (-1, 0) or (0, -1), the currently received OFDM for each Symbol phase ( , ), (- , ), (- ,- ) or ( ,- And the phase of the OFDM symbol received before the one OFDM symbol is ( , ), (- , ), (- ,- ) or ( ,- For each of them, the phase of the currently received OFDM symbol is estimated to be (1, 0), (0, 1), (-1, 0) or (0, -1). Error correction method of OFDM symbol in audio broadcasting receiver. Differential decoding means for delaying the OFDM symbols currently received through the I-channel and the Q-channel by one symbol period and then differentially decoding them; A phase comparator that receives the decoded OFDM symbol and detects a phase of the currently received OFDM symbol using a mapping table; After extracting and storing an expected region in which the currently received OFDM symbol phase is to be located from a symbol received one OFDM symbol before the currently received OFDM symbol, the phase detected by the phase comparator is located in the expected region. An error symbol detector that otherwise generates an error phase flag signal; And If the error phase flag signal is not generated, the signal passes immediately. If the error phase flag signal is generated, the first and second digital signals shift the phase of the currently received OFDM symbol for each of the I-channel and the Q-channel. And a phase shifter. An OFDM symbol error correction system of a digital audio broadcasting receiver. The method of claim 5, wherein the phase shift of the OFDM symbol, Bit-converting the value of the currently received OFDM symbol, adding or subtracting a phase correction signal to the currently received OFDM symbol, and performing an overflow or underflow process to perform the OFDM symbol of the digital audio broadcasting receiver. Error correction system. The method of claim 5, wherein the difference decoding means, First and second memories for delaying and storing an OFDM symbol input through each of the I-channel and the Q-channel by one symbol period; And And a differential phase decoder for differentially decoding the OFDM symbol signal and a reference phase generator for generating a reference phase. The method of claim 5, wherein the first and second digital phase shifter, First and second bit converters operating by the error phase flag signal and bit converting a value of an OFDM symbol received through each of the I-channel and the Q-channel; First and second phase correction signal output means for outputting the phase correction signal by the error phase flag signal of the error symbol detector; And And adding and subtracting the error phase correction signal and the value of the bit-converted OFDM symbol, and controlling the overflow and the underflow according to the result. Error correction system of the OFDM symbol of the receiver. The method of claim 5, wherein the expected area, When the phase of an OFDM symbol received before the one OFDM symbol is (1, 0), (0, 1), (-1, 0) or (0, -1), for each of them, the expected area is ( , ), (- , ), (- ,- ) or ( ,- ) And the phase of the OFDM symbol received before the one OFDM symbol is ( , ), (- , ), (- ,- ) or ( ,- ), For each of them, the expected region is (1, 0), (0, 1), (-1, 0) or (0, -1) of the OFDM symbol of the digital audio broadcasting receiver. Error correction system. The method of claim 5, wherein the phase correction signal has a value of 0, or Error correction system of the OFDM symbol of the digital audio broadcasting receiver.