JP3959488B2 - Guard interval length detection device, guard interval length detection method, and OFDM signal demodulation device - Google Patents

Description

  The present invention relates to an apparatus and method for detecting a guard interval length in an OFDM signal demodulating apparatus.

  As a conventional OFDM signal demodulator, a delay product calculation unit having a delay of an effective symbol length that may be transmitted obtains a correlation signal of the received signal and the delayed signal as many as the possibility, and this correlation signal The number of integrations in that case is performed by the integrator over the guard interval length that may be transmitted, the peak of this correlation integrated signal is detected and held by the peak holding unit, and then the magnitude of this peak signal Are compared by a comparison / determination unit, the transmission mode of the received signal is determined based on the detection pattern of the peak value, and the OFDM signal is demodulated based on the determination result (for example, see Patent Document 1).

  The correlation integrated signal is passed through a spectrum enhancement filter that passes only the reciprocal frequency of the transmission symbol length of the corresponding transmission mode and its harmonics, and after detecting this peak, the magnitude of the peak signal is determined by the comparison / determination unit. In some cases, the transmission mode of the received signal is determined based on the detection pattern of the peak value, and the OFDM signal is demodulated based on the determination result (see, for example, Patent Document 2).

JP-A-10-327122 JP-A-11-127131

  In the above conventional detection method, since the guard interval length is detected using the magnitude of the peak signal, there is a problem that the detection accuracy deteriorates when the fluctuation of the received signal level is large.

  Therefore, the present invention solves this problem, and even when the power of the received signal changes greatly, a guard interval length detection apparatus and method that can perform more accurate guard interval length detection, and An OFDM signal demodulating device including a guard interval length detecting device is provided.

  The present invention calculates a correlation between an OFDM signal and a signal obtained by delaying the OFDM signal by a certain time, and outputs a correlation maximum value position information output that outputs a position having the strongest correlation in a specific section as correlation maximum value position information. And a guard interval determination unit that detects a guard interval length based on the correlation maximum value position information.

  In the present invention, there is an effect that it is possible to detect the guard interval length with high accuracy even when the fluctuation of the received signal level is large.

Hereinafter, a guard interval length detection apparatus and method according to the present invention will be described.
Embodiment 1 FIG.
FIG. 1 is a block diagram illustrating a guard interval length detection apparatus according to the first embodiment. The guard interval length detection apparatus shown in FIG. 1 includes a correlation maximum value position information output unit 51 and a guard interval determination unit 52. The correlation maximum value position information output unit 51 calculates the correlation between the input signal and a signal obtained by delaying the input signal by a certain time, and calculates the correlation maximum value position information that is the position having the strongest correlation within a specific section. And output. Here, the input signal is a signal obtained by frequency-converting an OFDM signal into a baseband band (hereinafter also referred to as a baseband signal), and the OFDM signal is a signal having a plurality of carrier waves. The input signal may be the received OFDM signal itself.

  The correlation maximum value position information output unit 51 includes an effective symbol length delay unit 1, a complex multiplication unit 2, and a maximum value position information calculation unit 53 that calculates correlation maximum value position information based on the output from the complex multiplication unit 2. It consists of.

  The effective symbol length delay unit 1 and the complex multiplication unit 2 calculate a correlation between an input signal and a signal obtained by delaying the input signal by a certain time. The effective symbol length delay unit 1 outputs a signal delayed by one symbol length (hereinafter also referred to as “effective symbol length”) obtained by removing the guard interval length from the baseband signal. The complex multiplier 2 calculates the complex multiplication using the conjugate complex number output from the effective symbol length delay unit 1 and the baseband signal as the input signal, and outputs the result as a correlation value. Thus, originally, in order to detect the strongly correlated portion, it is necessary to obtain autocorrelation values at every time interval, but in the correlation maximum value position information output unit 51 in the present embodiment, The autocorrelation value may be obtained only for the time interval of the effective symbol length. That is, the correlation value between the baseband signal and the signal delayed by the symbol length excluding the guard interval length may be obtained. Accordingly, it is possible to detect a strongly correlated portion with a simpler circuit.

  FIG. 2 shows an output example of the complex multiplier 2. As shown in FIG. 2, in the output of the complex multiplier 2, the value corresponding to the guard interval length A out of one symbol length consisting of the guard interval length A and the effective symbol length B is large.

Here, when the input signal sampled at time n is a [n] and the effective symbol length is N _s , the output R [n] of the complex multiplier 2 at time n is expressed by the following equation. In the formula, * represents a complex conjugate.

  In complex multiplier 2, complex multiplication may be calculated using the output of effective symbol length delay unit 1 and the conjugate complex number of the input signal as inputs. In this case, the output R [n] of the complex multiplier 2 at time n is expressed by the following equation.

  The output of the complex multiplication unit 2 is input to the maximum value position information calculation unit 53. The maximum value position information calculation unit 53 is a moving average unit 3, 4... 5 that calculates a moving average value of a complex multiplication value that is an output of the complex multiplication unit 2, and an absolute value 2 of the output of the moving average unit. .., 8 for calculating the power, and maximum position detectors 12, 13... For continuously outputting the position of the maximum value in a certain section from the output of the absolute value P power calculator. ..14. Here, the absolute value P-power calculation units 6, 7,... 8 calculate the square of the absolute value of the complex multiplication value, but the absolute value itself may be used instead of the square of the absolute value. Alternatively, a power of an absolute value of 3 or more may be used, that is, P is an integer of 1 or more. Thus, the maximum value can be detected with higher accuracy by comparing the absolute value of the complex multiplication value to the Pth power.

  When K types (K is an integer of 1 or more) of guard intervals can be added to an OFDM signal that is an input signal, the moving average units 3, 4,... .., 8 and the maximum position detectors 12, 13,... 14 exist for each of the K systems, and the correlation maximum value position information output unit 53 includes K types. The correlation maximum value position information is output at each guard interval. Thus, the guard interval length can be detected at high speed by comparing the K systems.

  For example, in Japanese terrestrial digital broadcasting, three transmission modes of mode 1, mode 2 and mode 3 are provided, and the effective symbol length is different for each mode. The guard interval length is set to be 1/4, 1/8, 1/16, or 1/32 of the effective symbol length. Therefore, in Japanese terrestrial digital broadcasting, the guard interval length is different, and four types of guard intervals can be added to the OFDM signal. Therefore, when receiving Japanese terrestrial digital broadcasting, by setting K = 4, the guard interval length can be detected at a high speed as described above.

  First, in the maximum value position information output unit 53, the first, second, and Kth moving average units 3, 4,... 5 are used in order to compare the complex multiplication values for all the guard interval lengths A that can be added. Input to, moving average. In this case, a moving average is taken to clarify the peak. At this time, moving average is performed with different guard interval length A for each system. The maximum value is clarified in a system in which moving average is performed with the same length as the guard interval length A added to the input signal. Here, the moving average is an operation for continuously calculating an average value while shifting an averaged interval by one sample.

  The moving average results output from the first, second, and Kth moving average units 3, 4,... 5 are respectively the first, second, and Kth absolute value P-power calculating units 6, 7,. 8 and the square of the absolute value is output. This output value is input to the first, second and Kth maximum position detectors 12, 13,... As shown in FIG. 3A, the maximum position detectors 12, 13,... And the temporal position information of the maximum correlation value is output as indexes D1 and D2. The indexes D1 and D2 are detected in N reference intervals (the interval C in the first embodiment). That is, the index is obtained from each of the N reference sections, and N indexes D1, D2,... DN from each system in the maximum value position information calculation unit 53 are input to the guard interval determination unit 52. As described above, the index is obtained in each of the N reference intervals. In the following description, the index is referred to as indexes D1 and D2.

  Next, the guard interval determination unit 52 outputs first, second, and Kth variance calculation units 15, 16,... That output variances of the correlation maximum value position information output from the maximum value position information calculation unit 53. 17 and a minimum value determination unit 18 that outputs a guard interval length of a system that detects the minimum value from the outputs of the first, second, and Kth variance calculation units 15, 16, and 17 and outputs the minimum value. With. Here, the correlation maximum value position information output from the maximum value position information calculation unit 53 is indexes D1 and D2.

  The outputs of the first, second, and Kth maximum position detectors 12, 13, and 14 are input to the variance calculators 15, 16,..., And the variances of these indexes D1, D2 are calculated. Is done. When the indices D1 and D2 are detected in N reference intervals, as shown in FIG. 3B, the guard interval length A of the baseband signal matches the guard interval length included in the symbol length in the system. Since the fluctuations of the indexes D1 and D2 are small, the detection position of the maximum correlation value appears in a concentrated manner, and the variance value of the index becomes small. The arrow in FIG. 3B shows an example of the temporal position within one symbol of the maximum value (maximum value obtained as a result of moving average) corresponding to each index.

On the other hand, in the system that does not coincide with the guard interval length A of the baseband signal (FIG. 4A), the fluctuations of the indexes D1 and D2 become large, and therefore, as shown in FIG. The detection positions appear in a dispersed manner, and the dispersion value of the index is large.
The minimum value determination unit 18 inputs the output K system variance, and outputs the guard interval length of the system from which the minimum value is output.

Hereinafter, the maximum position detection units 12, 13... 14 and the variance calculation units 15, 16.
FIG. 5 is a block diagram showing the configuration of the maximum position detectors 12, 13,... As shown in FIG. 6, each of the maximum position detectors 12, 13... 14 starts from the head of this one reference section C when a section corresponding to one symbol length (A + B in FIG. 2) is defined as one reference section C. Maximum position detector 40 before outputting the position D1 of the maximum correlation value, and the maximum position after output of the position D1 ′ of the maximum correlation value from the beginning of the section C ′ obtained by shifting the one reference section C by a half symbol length. The detection unit 41 is provided, and both indexes (D1 and D1 ′) from these are output. In FIG. 6, the index position after the beginning of each reference section is indicated as D1 or D1 ′. However, the intermediate point of the reference section is set to 0, and the index value after the intermediate point is set to be positive, The index value before the intermediate point may be expressed as negative.

  In this way, by outputting position information (indexes D1, D1 ′) of the maximum correlation value between the one reference section C and the section C ′ shifted by a half symbol length in the maximum position detection units 12, 13,... A reference section suitable for implementation can be selected. However, even if it is other than this, the position information of the correlation maximum value in two different reference sections may be used. Further, it is sufficient that there are at least two systems, and there may be three or more systems.

Next, an expression for calculating a dispersion value in the dispersion calculators 15, 16... 17 will be described. The variance values in two different reference intervals C and C ′ are x ₁ and x ₂ , and the indices D1 and D1 ′ for the k-th symbol in the _two different reference intervals C and C ′ are A _k and B _k . If the number of intervals in which the value is calculated is N, the variance value x output from the variance calculators 15, 16,...

FIG. 7 is a block diagram showing the above equation for calculating the variance value in the variance calculators 15, 16,... The variance calculation units 15, 16... 17 respectively include first and second sample value square units 28 and 33, first and second square value average calculation units 29 and 34, and first and second 2 sample value averaging units 30, 35, first and second average square units 31, 36, first and second variance output units 32, 37, and a variance comparison unit 38. The sample value squaring units 28 and 33 output information on the position of the maximum correlation value in the reference sections C and C ′ of two different systems, that is, the square value of the index for each system. The first and second square value average calculation units 29 and 34 calculate the sum of the square values output from the sample value square units 28 and 33, respectively, and calculate the average of N symbols. Also, the first and second sample value averaging units 30 and 35 calculate the sum of the indexes for each system, and calculate the average of N symbols. The first and second average square units 31 and 36 output the square value of the average value output from the sample value average units 30 and 35. The difference between the first and second mean square calculation units 29 and 34 and the first and second mean square units 31 and 36 is calculated by the first and second distributed output units 32 and 37, respectively. Ask. The variance comparison unit 38 compares the two variances x ₁ and x ₂ output from the first and second variance output units 32 and 37, and sets the smaller one as the variance x at this time.

  Here, the case of comparison by variance has been described. However, when using standard deviation, the variance output units 32 and 37 may calculate the standard deviation by calculating the square root of the variance value, and the variance comparison unit 38 may perform comparison.

  The variance calculation units 15, 16... 17 calculate the variance value or standard deviation using the two indexes D1 and D1 '. The reason for this will be described with reference to FIG. As shown in FIG. 8A, when the maximum correlation value is near the boundary of one reference section C, when the indexes D1 and D2 are detected in N reference sections, as shown in FIG. 8B. In addition, as a result of the appearance of indexes near both boundaries, the resulting index tends to be highly dispersed. Accordingly, as shown in FIG. 9, the index variance values in the two intervals C and C ′ are compared, and the smaller one is output as the index variance value (system C ′ in FIG. 9). It can be prevented that the dispersion value becomes large near the boundary and the dispersion value is erroneously detected.

  FIG. 10A shows the case where the guard interval length is correct and FIG. 11A shows the case where the guard interval length is not correct for the output of the complex multiplier 2 when the fluctuation of the received signal level is large. Show. In such a case, the magnitude of the complex multiplication value depends on the magnitude of the received signal level, but in this embodiment, since the variance of the index in the N reference intervals is calculated, only the position of the index is calculated. Is important and does not depend on the received signal level.

  FIG. 10B shows a result of detecting an index for N reference intervals when the guard interval length is the same as shown in FIG. In FIG. 10B, the variance is small and the index position does not change. On the other hand, FIG. 11B shows a result of detecting an index for N reference intervals when the guard interval length does not match as shown in FIG. In FIG. 11B, the index position varies regardless of the received signal level. That is, even in the case of FIG. 10 and FIG. 11 in which the received signal level changes greatly, the guard interval length detection device in the present embodiment that makes the determination based on the position information accurately sets the guard interval length. It can be detected.

  As described above, according to the guard interval length detection device in the present embodiment, the guard interval length can be detected with high accuracy even when the received signal changes greatly.

Embodiment 2. FIG.
In the first embodiment, the moving average units 3, 4,..., 5 are shown to repeatedly calculate the average of the data corresponding to the guard interval length in each of the K systems while shifting one sample at a time. The number of moving average data may be thinned out. FIG. 12 is a block diagram showing the configuration of the moving average units 3, 4,... 5 when data is thinned out. When thinning out the data, the moving average units 3, 4,... 5 use the data obtained by sampling the output of the complex multiplication unit 2 in the sample extraction unit 19 and thinning out at a predetermined interval as shown in FIG. The average calculator 20 calculates a moving average in the guard interval length section and outputs a moving average result.
As a result, the calculation amount of the moving average unit is reduced, and the moving average can be realized with a simpler circuit.

Embodiment 3 FIG.
FIG. 13 is a block diagram showing a configuration of the guard interval length detection apparatus according to the third embodiment. In FIG. 13, the same reference numerals as those in FIG. 1 are the same as those in the first embodiment.

  In FIG. 13, the correlation maximum value position information output unit 51 is the same as in Embodiment 1, and when K types (K is an integer of 1 or more) of guard intervals can be added to the OFDM signal, the correlation maximum value position information From the output unit 51, the position information of the maximum correlation value is output as an index for each K system and is input to the guard interval determination unit 54. The guard interval determination unit 54 is different from the first embodiment in that the guard interval determination unit 54 includes first, second, and Kth frequency calculation units 42, 43... 44 and a maximum value determination unit 45.

  The position information of the correlation maximum value output as an index from the correlation maximum value position information output unit 51 is input to the frequency calculation units 42, 43... 44 over N symbols, and the frequency of the position information of the maximum correlation value is output. . And the frequency of the output K system | strain is input into the maximum value determination part 45, and the guard interval length of the system | strain which output the maximum value among this frequency is made into an output.

  FIG. 14 is a block diagram showing the configuration of the frequency calculation units 42, 43... The frequency calculation units 42, 43,... 44 each include an index determination unit 49, first, second, and Z frequency storage units 46, 47,... 48, and a frequency comparison unit 50. The frequency storage units 46, 47,... 48 are as many as the number Z of data in the N symbol section (Z is an integer of 1 or more), and correspond to the respective indexes. First, the outputs of the maximum position detection units 12, 13... 14 of the correlation maximum value position information output unit 51 are input to the index determination unit 49, and the corresponding frequency storage unit is selected. Next, 1 is added to the data of the selected frequency storage unit. After this operation is repeated for the indexes of N reference intervals, the data of the Z frequency storage units 46, 47... 48 are input to the frequency comparison unit 50 to detect the maximum value and output it as a frequency.

  The index of the system with the same guard interval is as shown in FIG. 15A, and the index of the system with the non-matched guard interval is as shown in FIG. As shown in FIG. 15B, when the guard intervals match, the index is concentrated, so the maximum value of the frequency becomes large. When the guard intervals do not match as shown in FIG. Since the index is distributed, the maximum frequency becomes small. For this reason, the guard interval length of the system having the largest index frequency is output as the desired guard interval length.

  In the first embodiment, the maximum position detectors 12, 13,... 14 and the variance calculators 15, 16,... 17 are used in the case of using two systems of indexes D1 and D1 ′. As for the form of, the output of the maximum position detectors 12, 13... 14 may calculate the frequency using only the index for one reference section or the frequency using the index for two different reference sections. And one of the two frequencies obtained as a result of the calculation may be selected.

  Also in the present embodiment, as in the second embodiment, a moving average may be calculated using data thinned out at a predetermined interval to obtain a moving average result.

  As described above, in the present embodiment, by outputting the guard interval length of the system having the largest index frequency, the guard interval length can be accurately detected even when the received signal changes greatly.

Embodiment 4 FIG.
FIG. 17 is a block diagram showing a configuration of the guard interval length detection apparatus according to the fourth embodiment. In FIG. 17, the same reference numerals as those in FIG. 1 are used in the same manner as in the first embodiment.
In FIG. 17, the guard interval determination unit 52 is the same as that of the first embodiment, but the first absolute value P-power calculation unit 6 that constitutes the maximum value position information calculation unit 56 of the correlation maximum value position information output unit 55. , 7... 8 and the maximum position detectors 12, 13... 14, a moving average between symbols for the outputs of the K system absolute value P-power calculators 6, 7. The difference is that the first, second, and Kth inter-symbol moving average units 9, 10,.

  The operation will be described. In FIG. 17, the square of the absolute value output from the absolute value P-power calculation units 6, 7... 8 is input to the inter-symbol moving average units 9, 10. In the inter-symbol moving average units 9, 10,... 11, the symbols are averaged and the influence of noise components is reduced. This output value is input to the K system maximum position detectors 12, 13,...

  FIG. 18 shows a block configuration of the inter-symbol moving average units 9, 10... 11 used here. The inter-symbol moving average units 9, 10,... 11 and the storage unit 27 that stores the outputs of the absolute value P-power arithmetic units 6, 7. It is composed of a symbol average calculation unit 21 that calculates an average between certain data points. Next, the same operation is performed on data shifted by one symbol. By repeating this operation, an inter-symbol moving average output value is output from the symbol average calculation unit 21 for each symbol.

  When a moving average is taken between symbols, the maximum value of the inter-symbol moving average output value is clarified even when the correlation value is buried in noise as shown in FIG. 19 (a) (FIG. 19 (b)). . Therefore, even if it is buried in noise, the index output from the maximum position detectors 12, 13,... From this, the guard interval can be detected with high accuracy.

  Also in the present embodiment, as in the second embodiment, a moving average may be calculated using data thinned out at a predetermined interval to obtain a moving average result. Moreover, it is also possible to apply to the guard interval determination part 54 which has the frequency calculation parts 42, 43 ... 44 shown in Embodiment 3.

  As described above, since the inter-symbol moving average is taken in the present embodiment, the index becomes clear and the guard interval length can be detected accurately even when the received signal level changes greatly.

Embodiment 5 FIG.
FIG. 20 is a block diagram showing a configuration of the guard interval length detection apparatus according to the fifth embodiment. In FIG. 20, the same reference numerals as those in FIG. 1 are used in the same manner as in the first embodiment.

  In FIG. 20, the guard interval determination unit 52 is the same as that in the first embodiment, but the first absolute value P-power calculation unit 6 constituting the maximum value position information calculation unit 58 of the correlation maximum value position information output unit 57. , 7... 8 and the maximum position detectors 12, 13... 14, a predetermined number of symbols between the symbols for the outputs of the K system absolute value P-power arithmetic units 6, 7. The first, second, and Kth block averaging units 22, 23... 24 that output the average of each are different.

  The operation will be described. 20, the squares of the absolute values output from the absolute value P-power calculation units 6, 7... 8 are input to the block averaging units 22, 23. In the block averaging units 22, 23... 24, block averaging is performed with a plurality of symbols, and the influence of noise components is reduced. This output value is input to the K system maximum position detectors 12, 13,...

  FIG. 21 shows a block configuration of the block average units 22, 23... 24 used here. The block averaging units 22, 23... 24 store the outputs of the absolute value P-power calculating units 6, 7,... 8 and add the outputs from the 1-symbol delay unit 26, and 1 A symbol delay unit 26 that delays symbols and a block average calculation unit 39 that averages the outputs of the block addition unit 25 are configured. That is, the outputs of the absolute value P-power calculators 6, 7... 8 are first input and stored in the block adder 25, and then the data stored in the block adder 25 is converted by the 1-symbol delay unit 26. After the delay, the operation of adding these data at the same position in the symbol is repeated. This operation is repeated for a predetermined symbol, and the block average calculation unit 39 outputs an average for each position in the symbol. In this way, the block average output value is output from the block average calculation unit 39 for each predetermined symbol.

  When the block average is taken between symbols, the maximum value of the obtained block average output values becomes clear even when the correlation value is buried in noise as shown in FIG. 19A (FIG. 19B). Therefore, even if it is buried in noise, the index output from the maximum position detectors 12, 13,... From this, the guard interval can be detected with high accuracy.

  Also in the present embodiment, as in the second embodiment, a moving average may be calculated using data thinned out at a predetermined interval to obtain a moving average result. Moreover, it is also possible to apply to the guard interval determination part 54 which has the frequency calculation parts 42, 43 ... 44 shown in Embodiment 3. Note that the number of symbols used for the average calculation can be appropriately determined depending on the required circuit scale and detection time.

  As described above, since the block average is taken in the present embodiment, the index becomes clear and the guard interval length can be detected with high accuracy even when the received signal level changes greatly.

1 is a block diagram illustrating a guard interval length detection device according to Embodiment 1. FIG. It is an example of an output of a complex multiplication part. It is an example of the system | strain index with which guard interval length is suitable in Embodiment 1. FIG. It is an example of the index | index of the system | strain in which guard interval length does not match in Embodiment 1. FIG. It is a block diagram of a maximum position detector. It is a figure which shows the example of the reference area in a maximum position detection part. It is a block diagram of a dispersion | distribution calculation part. It is a figure for demonstrating the reason for using a 2 system | strain index in a dispersion | distribution calculation part. It is a figure for demonstrating the reason for using a 2 system | strain index in a dispersion | distribution calculation part. It is an example of the index | index of the system | strain with which guard interval length is suitable when a received signal level changes a lot. It is an example of the index | index of the system | strain with which guard interval length does not match when a received signal level changes a lot. FIG. 10 is a block diagram illustrating a configuration of a moving average unit when data is thinned out in a second embodiment. FIG. 10 is a block diagram illustrating a guard interval length detection device according to a third embodiment. It is a block diagram of a frequency calculation part. It is an example of the system | strain index with which guard interval length is suitable in Embodiment 3. FIG. It is an example of the index | index of the system | strain in which guard interval length does not match in Embodiment 3. FIG. FIG. 10 is a block diagram illustrating a guard interval length detection device according to a fourth embodiment. It is a block diagram which shows the moving average part between symbols. It is an example of an output when the inter-symbol moving average or block average according to the fourth and fifth embodiments is performed. FIG. 10 is a block diagram illustrating a guard interval length detection device according to a fifth embodiment. It is a block diagram which shows a block average part.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Effective symbol length delay part, 2 Complex multiplication part, 3 1st moving average part, 4 2nd moving average part, 5th K moving average part, 6 1st absolute value P power calculation part, 7 2nd Absolute value P-power operation unit, 8 Kth absolute value P-power operation unit, 9 first inter-symbol moving average unit, 10 second inter-symbol moving average unit, 11 Kth inter-symbol moving average unit, 12 1st maximum position detector, 13 2nd maximum position detector, 14th K maximum position detector, 15 1st variance calculator, 16 2nd variance calculator, 17th K variance calculator, 18 minimum value determination unit, 19 sample extraction unit, 20 moving average calculation unit, 21 symbol average calculation unit, 22 first block average unit, 23 second block average unit, 24 Kth block average unit, 25 block addition , 26 1-symbol delay unit, 27 storage unit, 28 first sample value Square part, 29 1st square value average calculating part, 30 1st sample value average part, 31 1st average square part, 32 1st dispersion | distribution output part, 33 2nd sample value square part , 34 Second square value average calculation unit, 35 Second sample value average unit, 36 Second average square unit, 37 Second variance output unit, 38 Variance comparison unit, 39 Block average calculation unit, 40 Previous maximum position detection unit, 41 Rear maximum position detection unit, 42 First frequency calculation unit, 43 Second frequency calculation unit, 44th K frequency calculation unit, 45 Maximum value determination unit, 46 First frequency storage unit 47 Second frequency storage unit, 48 Zth frequency storage unit, 49 Index determination unit, 50 Frequency comparison unit, 51, 55, 57 Maximum correlation position information output unit, 52, 54 Guard interval determination unit, 53, 56, 58 Maximum value position information calculation unit.

Claims (20)

A correlation maximum value position information output means for calculating a correlation between the OFDM signal and a signal obtained by delaying the OFDM signal by a certain time, and outputting a position having the strongest correlation in a specific section as correlation maximum value position information;
Guard interval determination means for detecting a guard interval length based on the correlation maximum value position information ,
The OFDM signal is added with any one of a plurality of types of guard intervals,
The correlation maximum value position information output means generates the correlation maximum value position information for each guard interval of a plurality of types of guard intervals,
The guard interval determining means is a variance calculating means for outputting a variance or standard deviation of the correlation maximum value position information generated for each guard interval;
A guard interval length detection apparatus comprising: a minimum value determination unit that detects a minimum value from an output of the variance calculation unit and outputs a guard interval length corresponding to the minimum value. The correlation maximum value position information output means includes:
An effective symbol length delay means for outputting a received OFDM signal or a signal obtained by frequency-converting the received OFDM signal as an input signal and outputting a signal delayed by an effective symbol length;
Complex multiplication means for outputting a complex multiplication result of the conjugate complex number of the output of the effective symbol length delay means and the input signal, or a complex multiplication result of the output of the effective symbol length delay means and the conjugate complex number of the input signal;
The guard interval length detection device according to claim 1, further comprising: a maximum value position information calculation unit that calculates the correlation maximum value position information based on an output of the complex multiplication unit. The maximum value position information calculating means includes:
A plurality of moving average means for outputting a moving average value from the output of the complex multiplication means;
Absolute value P-power calculating means for outputting an absolute value P to the power of each of the plurality of moving average means (P is an integer of 1 or more);
And a maximum position detecting means for outputting the correlation maximum value position information within one symbol length interval for each of the plurality of P values of the absolute value output from the absolute value P power calculating means. The guard interval length detection device according to claim 2. The maximum position detecting means includes
When a section corresponding to one symbol length composed of the guard interval length and the effective symbol length is set as one reference section, the maximum previous position detecting means for outputting the position of the maximum correlation value in the one reference section;
4. The guard interval length detection apparatus according to claim 3, further comprising a post maximum position detection unit that outputs a position of the maximum correlation value in a section obtained by shifting the one reference section by a half symbol length. The maximum value position information calculating means further includes:
Inter-symbol moving average means for outputting a moving average between symbols for the output of the absolute value P-power calculating means,
5. The guard interval length detecting device according to claim 3, wherein the output of the inter-symbol moving average means is inputted to the maximum position detecting means. The maximum value position information calculating means further includes:
Block average means for outputting an average of a predetermined number of symbols between symbols for the output of the absolute value P power calculation means;
5. The guard interval length detection apparatus according to claim 3, wherein the output of the block averaging means is input to the maximum position detection means. The guard according to any one of claims 3 to 6, wherein the moving average means is configured to calculate the moving average value by using a length of a guard interval that can be added to the OFDM signal as a length of an average section. Interval length detector. The moving average means includes
Sample extraction means for outputting data obtained by thinning out the output of the complex multiplication means at a predetermined interval;
The guard interval length detector according to any one of claims 3 to 6, characterized in that it comprises a moving average calculation means for calculating the moving average value from the output of said sampling means. The correlation maximum value position information output by the correlation maximum value position information output means is information on the position of the correlation maximum value in two different reference sections when one symbol length section is defined as one reference section. ,
The variance calculating means includes
Sample value squaring means for outputting a square value for each piece of information of the position of the maximum correlation value in the two different reference sections;
A mean square value computing means for computing the mean square value output from the sample value square means for each of the two systems;
Sample value averaging means for outputting an average for each piece of information of the position of the maximum correlation value in the reference sections of the two different systems;
Mean square means for outputting the square value of the average value output from the sample value averaging means;
A variance output means for outputting either a variance value or a standard deviation from the output of the square value average calculation means and the output of the mean square means;
The distributed printing unit guard interval length detection apparatus according to claim 1, characterized in that it comprises a dispersion comparing means for outputting a smaller one compares the output from. A correlation maximum value position information output means for calculating a correlation between the OFDM signal and a signal obtained by delaying the OFDM signal by a certain time, and outputting a position having the strongest correlation in a specific section as correlation maximum value position information;
Guard interval determination means for detecting a guard interval length based on the correlation maximum value position information,
The OFDM signal is added with any one of a plurality of types of guard intervals ,
The correlation maximum value position information output means generates the correlation maximum value position information for each guard interval of the plurality of types of guard intervals,
The guard interval determining means outputs a frequency calculating means for outputting the frequency of the correlation maximum value position information generated for each guard interval ;
Wherein by detecting a maximum value from the output of the frequency calculation means, the maximum value determination unit, features and be Ruga over de interval length detector further comprising an outputting a guard interval length corresponding to the maximum value. A correlation maximum value position information output step of calculating a correlation between the OFDM signal and a signal obtained by delaying the OFDM signal by a certain time, and outputting a position having the strongest correlation in a specific section as correlation maximum value position information;
A guard interval determination step of detecting a guard interval length based on the correlation maximum value position information ,
The OFDM signal is added with any one of a plurality of types of guard intervals,
In the correlation maximum value position information output step, generating the correlation maximum value position information for each guard interval of a plurality of types of guard intervals,
The guard interval determination step includes
A variance calculating step of calculating a variance or standard deviation of the correlation maximum value position information generated for each guard interval;
A guard interval length detection method comprising: detecting a minimum value from a calculation result in the variance calculation step and outputting a guard interval length corresponding to the minimum value . A correlation maximum value position information output step of calculating a correlation between the OFDM signal and a signal obtained by delaying the OFDM signal by a certain time, and outputting a position having the strongest correlation in a specific section as correlation maximum value position information;
A guard interval determination step of detecting a guard interval length based on the correlation maximum value position information,
The OFDM signal is added with any one of a plurality of types of guard intervals,
In the correlation maximum value position information output step, the correlation maximum value position information is generated for each guard interval of the plurality of types of guard intervals,
The guard interval determination means includes
A frequency calculating step for calculating the frequency of the correlation maximum value position information generated for each guard interval;
A guard interval length detection method comprising: detecting a maximum value from a calculation result in the frequency calculation step, and outputting a guard interval length corresponding to the maximum value. The correlation maximum value position information output means includes:
An effective symbol length delay means for outputting a received OFDM signal or a signal obtained by frequency-converting the received OFDM signal as an input signal and outputting a signal delayed by an effective symbol length;
Complex multiplication means for outputting a complex multiplication result of the conjugate complex number of the output of the effective symbol length delay means and the input signal, or a complex multiplication result of the output of the effective symbol length delay means and the conjugate complex number of the input signal;
The guard interval length detection device according to claim 10, further comprising: a maximum value position information calculation unit that calculates the correlation maximum value position information based on an output of the complex multiplication unit. The maximum value position information calculating means includes:
A plurality of moving average means for outputting a moving average value from the output of the complex multiplication means;
Absolute value P-power calculating means for outputting an absolute value P to the power of each of the plurality of moving average means (P is an integer of 1 or more);
And a maximum position detecting means for outputting the correlation maximum value position information within one symbol length interval for each of the plurality of P values of the absolute value output from the absolute value P power calculating means. The guard interval length detection apparatus according to claim 13. The maximum position detecting means includes
When a section corresponding to one symbol length composed of the guard interval length and the effective symbol length is set as one reference section, the maximum previous position detecting means for outputting the position of the maximum correlation value in the one reference section;
15. The guard interval length detection device according to claim 14, further comprising a post maximum position detection unit that outputs a position of the maximum correlation value in a section obtained by shifting the one reference section by a half symbol length. The maximum value position information calculating means further includes:
Inter-symbol moving average means for outputting a moving average between symbols for the output of the absolute value P-power calculating means,
16. The guard interval length detection device according to claim 14, wherein the output of the inter-symbol moving average means is input to the maximum position detection means. The maximum value position information calculating means further includes:
Block average means for outputting an average of a predetermined number of symbols between symbols for the output of the absolute value P power calculation means;
16. The guard interval length detection apparatus according to claim 14, wherein the output of the block averaging means is input to the maximum position detection means. The guard according to any one of claims 14 to 17, wherein the moving average means is configured to calculate the moving average value using a length of a guard interval that can be added to the OFDM signal as a length of an average section. Interval length detection device. The moving average means includes
Sample extraction means for outputting data obtained by thinning out the output of the complex multiplication means at a predetermined interval;
The guard interval length detection device according to claim 14, further comprising a moving average calculation unit that calculates the moving average value from an output of the sample extraction unit. An OFDM signal demodulator comprising the guard interval length detector according to any one of claims 1 to 10 or 13 to 19.

Images