JP6150685B2 - OFDM wave measuring apparatus and program - Google Patents

Description

  The present invention relates to a technique for detecting and measuring signals such as terrestrial digital broadcasting, terrestrial digital audio broadcasting, and multimedia broadcasting, and in particular, reception power of an OFDM (Orthogonal Frequency Division Multiplexing) signal including a pilot signal, and the like. The present invention relates to an OFDM wave measuring apparatus and a program for measuring the frequency.

  2. Description of the Related Art Conventionally, in fields such as terrestrial digital broadcasting, terrestrial digital audio broadcasting, and multimedia broadcasting, apparatuses that extract a pilot signal from an OFDM signal and measure received power, spectrum, delay profile, and the like are known. For example, Patent Document 1 detects a transmission mode and GI (Guard Interval) from a received OFDM signal, performs local frequency correction and sampling frequency correction, and extracts a pilot signal without capturing frame synchronization. An OFDM signal analyzing apparatus for calculating a delay profile is described. In Patent Document 2, an SP (Scattered Pilot) signal is extracted from the received OFDM signal, and based on the SP signal, the number of FFTs (Fast Fourier Transforms) is larger than the conventional number. A delay profile measuring device for calculating a delay profile is described.

  On the other hand, institutionalization of services and systems utilizing white space is being promoted, and research and development are being actively conducted to promote business development. White space refers to radio waves in a frequency region that is not used geographically and temporally among frequency regions assigned to a specific radio wave service. In a service utilizing this white space, the allowable level of interference from the white space using station to the broadcast wave needs to be equal to or less than thermal noise. For example, the allowable level of interference is I (Interference) / N (Noise) = − 10 dB. As described above, there is a need for a technique for measuring a low level signal equal to or lower than thermal noise.

  Therefore, an OFDM wave measuring apparatus disclosed in Patent Document 3 is known as an apparatus for measuring a low level signal equal to or lower than thermal noise. This OFDM wave measuring device is obtained by performing loop filter processing or moving average processing in units of a predetermined number of symbols on the time axis, then detecting the position of an effective symbol by GI correlation, and then performing FFT A pilot signal is extracted by synchronously adding carrier symbols for each predetermined symbol, and received power and the like are calculated. On the other hand, since this OFDM wave measuring apparatus performs loop filter processing or moving average processing and synchronous addition processing on the time axis, if there is a clock error or frequency error between the transceivers, the received power etc. It cannot be calculated with high accuracy. Therefore, in Patent Document 3, the clock error is calculated from the value of the GI correlation as a rough correction, and in order to further improve the accuracy, the power of the SP signal after the synchronous addition is calculated from the number of calculated symbols that becomes a peak value on the time axis. An example is shown in which received power and the like are calculated using a table or calculation formula for converting a frequency error.

  By the way, there is area broadcasting as one of systems utilizing white space. The standard for area broadcasting in Non-Patent Document 1 indicates that the allowable frequency error of the transmission frequency is a maximum of ± 20 kHz. The above-mentioned Patent Document 3 shows an example of a method for accurately detecting a frequency error. This method is ± 1/2 of the OFDM carrier interval (in the case of mode 3, the carrier interval is about 992 Hz (= Only a frequency error in the range of 1 / 1.008e-3) (hereinafter referred to as narrowband frequency error) can be detected.

  On the other hand, as a technique for detecting a frequency error of ± 1/2 or more of the OFDM carrier interval, Non-Patent Document 2 discloses a frequency error (hereinafter referred to as a broadband frequency) in units of carrier intervals due to a power difference between a pilot signal and a data signal. A method for detecting an error) has been proposed.

  This method calculates the average power for each carrier with the set number of symbols and uses the fact that the power of the carrier allocated to CP (Continual Pilot) is larger than that of other carriers. A frequency error is detected by calculating a correlation of pilot arrangement information defined in advance.

JP 2002-335226 A JP 2007-28367 A JP 2012-253553 A

ARIB STD-B55, “Area Broadcast Transmission Method” Hayashi et al., "Study of wideband frequency synchronization method for OFDM demodulation", 1997 IEICE General Conference, B-5-244

  However, assuming that the method of Non-Patent Document 2 described above is applied to a low level signal whose signal power is equal to or lower than thermal noise, it becomes difficult to discriminate CP because thermal noise power is added to the received signal. For this reason, the method of Non-Patent Document 2 has a problem that it is impossible to detect a frequency error with high accuracy.

  Therefore, the present invention has been made to solve the above-described problems, and the object of the present invention is to detect a frequency error equal to or greater than the carrier interval even when the signal power is low, and to accurately convert the OFDM signal. An object is to provide a measurable OFDM wave measuring apparatus and program.

  To achieve the above object, an OFDM wave measuring apparatus according to the present invention receives an OFDM wave including a pilot signal, extracts the pilot signal, and measures the signal of the OFDM wave. The orthogonal demodulation unit that generates the baseband signal by performing orthogonal demodulation on the OFDM wave signal, and the baseband signal generated by the orthogonal demodulation unit, adds in units of a predetermined number of symbols on the time axis, An error detector that detects the symbol head position, narrowband frequency error and clock error by guard correlation, and a baseband signal generated by the quadrature demodulator corrects the narrowband frequency error detected by the error detector. A narrowband frequency error correction unit, and a baseband in which the narrowband frequency error is corrected by the narrowband frequency error correction unit A first symbol cutout unit that removes a GI and cuts out an effective symbol based on the symbol head position detected by the error detection unit, and an effective symbol cut out by the first symbol cutout unit First FFT section for generating a carrier symbol, and differentially demodulating the IQ signal of the carrier symbol generated by the first FFT section, between the differential demodulated signal and the I axis An angle is calculated, cumulative processing using the angle is performed for a predetermined number of symbols, and a carrier position of a pilot signal for wideband frequency error detection that is differentially modulated and arranged at the same carrier position for each symbol is specified in units of carriers. The carrier position of the specified broadband frequency error detection pilot signal and the carrier of the preset broadband frequency error detection pilot signal A wideband frequency error detector that detects a wideband frequency error based on the correlation value, and a baseband signal in which the narrowband frequency error is corrected by the narrowband frequency error corrector. On the other hand, a broadband frequency error correction unit that corrects the broadband frequency error detected by the broadband frequency error detection unit, and a baseband signal whose broadband frequency error is corrected by the broadband frequency error correction unit, by the error detection unit. Based on the detected symbol head position, a second symbol cutout unit that removes GI and cuts out an effective symbol, and an effective symbol cut out by the second symbol cutout unit are FFTed to generate a carrier symbol The second FFT unit and the carrier symbol generated by the second FFT unit are synchronized every predetermined number of symbols. A symbol addition unit that generates a synchronous addition result, calculates a correlation value between the synchronous addition result generated by the symbol addition unit and a plurality of preset patterns, and the correlation value is maximized A pattern detection unit that detects a pattern, and a pilot extraction unit that extracts a pilot signal from the carrier symbol of the synchronous addition result based on the pattern detected by the pattern detection unit.

  In the OFDM wave measuring apparatus according to the present invention, the wideband frequency error detection unit differentially demodulates the carrier symbol IQ signal generated by the first FFT unit for each carrier to generate a differential demodulated signal. Calculating the angle between the differential demodulator and the IQ signal of the differential demodulated signal generated by the differential demodulator for each carrier and subtracting the sine value from the cosine value of the calculated angle The result is accumulated for the predetermined number of symbols in the symbol direction and averaged, the coefficient calculation unit calculating the average result as a coefficient, the coefficient for each carrier calculated by the coefficient calculation unit, and the wideband frequency error A correlation value between the coefficient for each carrier and the arrangement information is calculated using arrangement information in which the carrier position of the detection pilot signal is set in advance, and a wideband is calculated based on the correlation value. A correlation calculating section for detecting the frequency error, characterized by comprising a.

  The OFDM wave measuring apparatus according to the present invention is characterized in that the broadband frequency error detection pilot signal is a pilot signal that is modulated by a DBPSK or DQPSK modulation method and is arranged at the same carrier position for each symbol. .

  Further, the program according to the present invention extracts the pilot signal from the OFDM wave signal including the pilot signal, and orthogonally demodulates the OFDM wave signal to a computer that measures the OFDM wave signal to generate a baseband signal. The function of the quadrature demodulator and the baseband signal generated by the quadrature demodulator are added in units of a predetermined number of symbols on the time axis, and the symbol head position, narrowband frequency error, and clock are added by guard correlation. A function of an error detection unit that detects an error, and a function of a narrowband frequency error correction unit that corrects a narrowband frequency error detected by the error detection unit with respect to the baseband signal generated by the orthogonal demodulation unit; For the baseband signal whose narrowband frequency error is corrected by the narrowband frequency error correction unit, the error detection unit Based on the detected symbol head position, the function of the first symbol cutout unit that removes the GI and cuts out the effective symbol, and the effective symbol cut out by the first symbol cutout unit are subjected to FFT to obtain a carrier symbol The function of the first FFT unit that generates the signal and the IQ signal of the carrier symbol generated by the first FFT unit are differentially demodulated, and the angle between the differentially demodulated signal and the I axis is calculated. Then, accumulation processing using the angle is performed for a predetermined number of symbols, the carrier position of a pilot signal for wideband frequency error detection that is differentially modulated and arranged at the same carrier position for each symbol is specified in units of carriers, and the specified wideband Correlation between carrier position of pilot signal for detecting frequency error and carrier position of pilot signal for detecting broadband frequency error set in advance And a wideband frequency error detecting unit for detecting a wideband frequency error based on the correlation value, and a baseband signal whose narrowband frequency error is corrected by the narrowband frequency error correcting unit. A function of a broadband frequency error correction unit that corrects a broadband frequency error detected by the error detection unit, and a baseband signal in which the broadband frequency error is corrected by the broadband frequency error correction unit, are detected by the error detection unit. Based on the symbol head position, the function of the second symbol cutout unit that removes the GI and cuts out the effective symbol and the effective symbol cut out by the second symbol cutout unit are FFTed to generate a carrier symbol. 2 functions of the FFT unit and the carrier symbol generated by the second FFT unit as a predetermined symbol A symbol addition unit that generates a synchronization addition result every time, and calculates a correlation value between a synchronization addition result generated by the symbol addition unit and a plurality of preset patterns, A function of a pattern detection unit that detects a pattern having the maximum correlation value, and a function of a pilot extraction unit that extracts a pilot signal from the carrier symbol of the synchronous addition result based on the pattern detected by the pattern detection unit; It is characterized by realizing.

  As described above, according to the present invention, even when the signal power is at a low level, a frequency error greater than the carrier interval can be detected, and the OFDM signal can be measured with high accuracy.

It is a block diagram which shows the structure of the OFDM wave measuring apparatus by Example 1. FIG. It is a block diagram which shows the structure of a wideband frequency error detection part. It is a flowchart which shows the process of a wideband frequency error detection part. It is a figure which shows the transition image of SP signal which the error detection part input from SP extraction part. It is a figure which shows the relationship between the frequency error and clock error between transmitter / receivers, and the center frequency of a pilot carrier. (1) is a figure which shows the example of a transition on the IQ axis in SP signal (left end SP signal) with the lowest frequency of the left end of an OFDM wave, (2) is CP signal with the highest frequency of the right end of an OFDM wave. It is a figure which shows the transition example on IQ axis in (right end CP signal), (3) is a figure which shows the transition example of variation | change_quantity (theta) of the deflection angle in a left end SP signal. It is a figure explaining the left end SP signal and right end CP signal used as an observation object in Example 1. FIG. 6 is a flowchart illustrating processing of an error detection unit according to the first embodiment. It is a figure explaining SP1 signal and SP2 signal selected in Example 2. FIG. 10 is a flowchart illustrating processing of an error detection unit according to the second embodiment. Angle theta _{m, n} and _{(cosθ m, n -sinθ m,} n) is a diagram showing the relationship.

  Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings. In the first embodiment (example 1) and the second embodiment (example 2) described below, the IQ signal is differentially demodulated and detected between the I axis and the wideband frequency error. By calculating the angle and accumulating the wideband frequency error detection pilot signal that is differentially modulated and arranged at the same carrier position for each symbol, the carrier position is identified and a frequency error greater than the carrier interval is detected. It is characterized by.

  In the first embodiment, when a narrowband frequency error and a clock error are detected, a narrowband frequency error detection pilot signal is extracted by synchronous addition of symbols on the frequency axis, and the extracted narrowband frequency error detection pilot is extracted. The deviation angle of the signal is observed on the time axis, and a frequency error and a clock error including positive and negative directions are detected from the amount of change. In the second embodiment, when detecting a narrowband frequency error and a clock error, a pilot signal for detecting a narrowband frequency error is extracted by synchronous addition of symbols on the frequency axis. Select a pilot signal for narrowband frequency error detection with a high amplitude value from the inside, observe the declination of the selected pilot signal for narrowband frequency error detection on the time axis, and determine the frequency including the positive and negative directions from the amount of change. Detect errors and clock errors. Examples 1 and 2 will be described below.

[Example 1]
First, Example 1 will be described. As described above, in the first embodiment, the IQ signal is differentially demodulated to calculate the angle between the I axis and the wideband frequency error detection pilot signal that is differentially modulated and arranged at the same carrier position for each symbol. By accumulating, the carrier position is identified, a wideband frequency error that is greater than the carrier interval is detected, and a pilot signal for narrowband frequency error detection is extracted by synchronous symbol addition on the frequency axis. The declination of the frequency error detection pilot signal is observed on the time axis, and a narrowband frequency error and a clock error including positive and negative directions are detected from the amount of change.

  FIG. 1 is a block diagram illustrating a configuration of an OFDM wave measurement apparatus according to the first embodiment. The OFDM wave measuring apparatus 1 includes a frequency conversion unit 11, an A / D (Analog / Digital) conversion unit 12, an orthogonal demodulation unit 13, an error detection unit 14, narrowband frequency error correction units 15-1 and 15-2, symbols. Cutout sections 16-1, 16-2, 16-3, 16-4, FFT sections 17-1, 17-2, 17-3, 17-4, symbol addition sections 18-1, 18-2, 18- 3, SP pattern detection unit (pattern detection unit) 19, clock error correction units 20-1, 20-2, SP extraction units (pilot extraction units) 21-1, 21-2, error detection unit 22, received power calculation unit 23, a spectrum calculation unit 24, a delay profile calculation unit 25, a broadband frequency error correction unit 26, and a broadband frequency error detection unit 27.

  The frequency converter 11 receives an RF (Radio Frequency) signal of the OFDM signal received by the receiving antenna, generates a IF (Intermediate Frequency) signal by frequency conversion, and an A / D converter 12 is output. The A / D converter 12 receives the IF signal from the frequency converter 11, converts the analog IF signal into a digital IF signal, and outputs the digital IF signal to the quadrature demodulator 13. The quadrature demodulator 13 receives the digital IF signal from the A / D converter 12 and performs quadrature demodulation to generate baseband signals of I (In-phase) and Q (Quadrature), and generates an error. It outputs to the detection part 14 and the narrow band frequency error correction | amendment part 15-1.

  The error detection unit 14 receives the IQ baseband signal from the quadrature demodulation unit 13, sequentially performs addition in units of four symbols on the basis of a predetermined data head position on the time axis, and adds the interval between guard interval phases to the addition result. Then, a guard interval correlation value is calculated, and a symbol head position, a clock error, and a frequency error are detected based on the guard interval correlation value. Here, the predetermined data head position indicates an arbitrary position on the time axis of the IQ baseband signal. The range of the frequency error detected by the error detector 14 is ± 1/2 of the carrier interval.

  Then, the error detection unit 14 outputs the detected symbol head position to the symbol cutout units 16-1, 16-2, 16-3, 16-4, and detects the detected clock error (symbol addition on the time axis). The detected clock error) is output to the clock error correction unit 20-1, and the detected frequency error (frequency error detected by symbol addition on the time axis) is defined as a narrowband frequency error. Output to. Here, the detection processing of the symbol head position, the clock error, and the frequency error by the error detection unit 14 is already known. For details, refer to Patent Document 3 described above.

  The narrowband frequency error correction unit 15-1 receives the IQ baseband signal from the quadrature demodulation unit 13 and the narrowband frequency error from the error detection unit 14, and based on the input narrowband frequency error, The frequency error in the band signal (IQ signal) is corrected. Then, the narrowband frequency error correction unit 15-1 outputs the baseband signal obtained by correcting the narrowband frequency error to the wideband frequency error correction unit 26 and the symbol cutout unit 16-4.

  The symbol cutout unit 16-4 receives the baseband signal with the narrowband frequency error corrected from the narrowband frequency error correction unit 15-1, and also receives the symbol head position from the error detection unit. Then, the symbol cutout unit 16-4 removes the GI from the baseband signal based on the position shifted by the symbol head position from the data head position, which is the reference for addition in units of four symbols in the error detection unit 14, and is effective. Cut out the signal of the symbol. Then, the symbol cutout unit 16-4 outputs an effective symbol signal to the FFT unit 17-4.

  The FFT unit 17-4 receives the effective symbol signal from the symbol cutout unit 16-4, performs FFT to generate a carrier symbol, and outputs the carrier symbol FFT output signal to the wideband frequency error detection unit 27.

  The broadband frequency error detection unit 27 receives the FFT output signal from the FFT unit 17-4, differentially demodulates the IQ signal, which is the FFT output signal, for each carrier, and performs a differential demodulation between the signal and the I axis. The angle is calculated, and based on the calculated angle, the pilot signal modulated by DBPSK (Differential Binary Phase Shift Keying) and arranged at the same carrier position for each symbol is accumulated. The carrier position is specified in units of carrier, and the correlation value between the carrier position of the specified pilot signal and the correct carrier position of the pilot signal (the carrier position of the pilot signal arranged by the transmission device) is calculated, and the carrier interval or more The frequency error of is detected. Then, the broadband frequency error detection unit 27 outputs the detected frequency error to the broadband frequency error correction unit 26 as a broadband frequency error. For example, a TMCC (Transmission and Multiplexing Configuration Control) signal and an AC (Auxiliary Channel) signal are used as pilot signals that are arranged at the same carrier position for each symbol and used to detect a wideband frequency error. Details of the processing of the broadband frequency error detection unit 27 will be described later.

  Here, it is assumed that the OFDM wave received by the OFDM wave measuring apparatus 1 includes a pilot signal that is modulated by the DBPSK modulation method and arranged at the same carrier position for each symbol by the transmitting apparatus that transmits the OFDM wave.

  The wideband frequency error correction unit 26 inputs the baseband signal with the narrowband frequency error corrected from the narrowband frequency error correction unit 15-1 and also inputs the wideband frequency error from the wideband frequency error detection unit 27. Based on the wideband frequency error, a frequency error in the input baseband signal (an IQ signal in which the narrowband frequency error is corrected) is corrected. Then, the wideband frequency error correction unit 26 outputs the baseband signal obtained by correcting the wideband frequency error to the narrowband frequency error correction unit 15-2 and the symbol cutout unit 16-1.

  The symbol cutout unit 16-1 receives the baseband signal whose broadband frequency error has been corrected from the broadband frequency error correction unit 26 and also receives the symbol head position from the error detection unit 14. Then, the symbol cutout unit 16-1 performs the same processing as the symbol cutout unit 16-4, cuts out an effective symbol signal, and outputs the cut out effective symbol signal to the FFT unit 17-1.

  The FFT unit 17-1 receives the effective symbol signal from the symbol cutout unit 16-1, performs FFT to generate a carrier symbol, and outputs the carrier symbol FFT output signal to the symbol adder 18-1. In this case, the FFT unit 17-1 divides into four groups of a symbol number of 4nth carrier symbol, a symbol number of 4n + 1th carrier symbol, a symbol number of 4n + 2nd carrier symbol, and a symbol number of 4n + 3rd carrier symbol. The FFT output signal is output to the symbol adder 18-1 for each group. n is an integer of 0 or more.

  The symbol adder 18-1 includes four groups of carrier symbols (symbol number is 4nth carrier symbol, symbol number is 4n + 1th carrier symbol, symbol number is 4n + 2nd carrier) that is an FFT output signal from the FFT unit 17-1. Symbol and symbol number 4n + 3rd carrier symbol) are input, and carrier symbols are synchronously added for each group. That is, the symbol addition unit 18-1 performs vector addition for each subcarrier for each group. Then, the symbol addition unit 18-1 performs the synchronous addition result of the 4nth carrier symbol whose symbol number is the synchronous addition result for each group, the synchronous addition result of the 4n + 1th carrier symbol, and the symbol number of 4n + 2th. The synchronous addition result of the carrier symbol and the synchronous addition result of the 4n + 3rd carrier symbol with the symbol number are output to the SP pattern detection unit 19. Specifically, the symbol addition unit 18-1 performs an addition process using a loop filter or an addition process using a moving average as the synchronous addition process. Here, the addition processing by the symbol addition unit 18-1 is known, and for details, refer to Patent Document 3 described above.

  The SP pattern detection unit 19 obtains four groups of synchronous addition results from the symbol addition unit 18-1 (synchronization addition result of the 4nth carrier symbol with the symbol number, the synchronous addition result of the 4n + 1th carrier symbol with the symbol number of the symbol number). 4n + 2nd carrier symbol synchronous addition result and 4n + 3rd carrier symbol synchronous addition result). Then, the SP pattern detection unit 19 calculates correlation values between these synchronous addition results and four preset SP patterns, and whether or not SP can be extracted based on the four correlation values. Is generated, a signal indicating whether or not SP extraction is possible is generated, and an SP pattern having the maximum correlation value when SP extraction is possible is detected. If the SP pattern detection unit 19 determines that SP extraction is impossible, the SP pattern detection unit 19 outputs a signal indicating that SP extraction is impossible to the SP extraction units 21-1 and 21-2, and determines that SP extraction is possible. The SP extractable signal and the SP pattern are output to the SP extraction units 21-1 and 21-2. Here, the SP pattern detection processing by the SP pattern detection unit 19 is known, and for details, refer to Patent Document 3 described above.

  Here, in the SP pattern detection unit 19, when the power of the received signal is low, immediately after the start of the SP pattern detection process, there is not much difference in the correlation value between the synchronous addition result and the four SP patterns. This is because when the power of the received signal is low, four different types of SP signals (SP signals having different amplitudes and phases) existing in one symbol cannot be clearly distinguished. As the number of symbols to be synchronously added increases and the SP pattern detection process proceeds, one of the four correlation values becomes larger than the other three correlation values. That is, the synchronous addition result becomes closer to one of the four SP patterns as the synchronous addition process proceeds. This is because even if the power of the received signal is low, the four different types of SP signals existing in one symbol can be clearly distinguished by the progress of the synchronous addition process. Based on the difference between the correlation values, it is determined whether the SP can be extracted and the SP pattern, or the SP cannot be extracted.

  The symbol cutout unit 16-2 and the FFT unit 17-2 perform the same processes as those of the symbol cutout unit 16-1 and the FFT unit 17-1, respectively.

The clock error correction unit 20-1 receives four groups of carrier symbols as FFT output signals from the FFT unit 17-2, and receives a clock error (clock detected by symbol addition on the time axis) from the error detection unit 14. Error) is input, and for each group, a difference τ indicating the number of clocks corresponding to the time difference between the current symbol position and the ideal symbol position is calculated based on the clock error, and the phase of the SP signal is set to 2πf. _The clock error is corrected by reversely rotating kτ, and four groups of carrier symbols with the corrected clock error are output to the symbol adder 18-2. Here, f _k represents the center carrier frequency of the SP signal at subcarrier number k. The clock error correction processing by the clock error correction unit 20-1 is already known. For details, refer to Patent Document 3 described above.

  The symbol adder 18-2 receives four groups of carrier symbols in which the clock error (clock error detected by symbol addition on the time axis) is corrected from the clock error corrector 20-1, and the symbol adder 18 -1 is performed, the symbol number is the 4nth carrier symbol synchronous addition result, the symbol number is the 4n + 1 carrier symbol synchronous addition result, the symbol number is the 4n + 2nd carrier symbol synchronous addition result, and the symbol number Outputs the synchronous addition result of the 4n + 3rd carrier symbol to the SP extraction unit 21-1. The symbol adder 18-2 counts the number of times of synchronous addition, and outputs this to the error detector 22 as the number of calculated symbols.

  The SP extraction unit 21-1 receives the four groups of synchronous addition results from the symbol addition unit 18-2, and inputs the SP extraction impossibility or SP extraction enabled and the SP pattern from the SP pattern detection unit 19. And SP extraction part 21-1 does not perform SP extraction processing, when SP extraction impossible is inputted. On the other hand, when SP extraction possible is input, the SP extraction unit 21-1 extracts the SP signal from the four groups of synchronous addition results based on the SP pattern, extracts the CP signal, and extracts the extracted SP signal and CP signal. Is output to the error detection unit 22.

  The error detection unit 22 receives the SP signal and the CP signal from the SP extraction unit 21-1, and also receives the number of calculation symbols indicating the number of synchronous additions from the symbol addition unit 18-2. The SP signal having the lowest frequency and the CP signal having the highest frequency at the right end are observed, and when the amount of change in the deflection angle per symbol of these SP signal and CP signal is equal to or greater than the number of calculated symbols, these SP signals For the CP signal, the declination is calculated on the time axis, and the frequency error and the clock error are detected. The error detection unit 22 then converts the detected frequency error (frequency error detected by symbol addition on the frequency axis and calculation of the deviation angle on the time axis) to the narrowband frequency error correction unit 15-2 as a narrowband frequency error. In addition to the output, the detected clock error (the clock error detected by the symbol addition on the frequency axis and the deviation calculation on the time axis) is output to the clock error correction unit 20-2. Details of the processing of the error detection unit 22 will be described later.

  The narrowband frequency error correction unit 15-2 receives the baseband signal with the wideband frequency error corrected from the wideband frequency error correction unit 26, and also receives a narrowband frequency error (symbol addition and symbol addition on the frequency axis) from the error detection unit 22. Narrowband frequency error detected by calculating the deviation angle on the time axis) is input, and the same processing as that of the narrowband frequency error correction unit 15-1 is performed again based on the input narrowband frequency error. The frequency error in the baseband signal is corrected. Then, the narrowband frequency error correction unit 15-2 cuts out a baseband signal obtained by correcting the narrowband frequency error (frequency error detected by symbol addition on the frequency axis and calculation of deviation angle on the time axis). To the unit 16-3.

  The symbol cutout unit 16-3 and the FFT unit 17-3 perform the same processing as the symbol cutout unit 16-1 and the FFT unit 17-1. The clock error correction unit 20-2 receives four groups of carrier symbols, which are FFT output signals, from the FFT unit 17-3, and receives a clock error (symbol addition on the frequency axis and deviation on the time axis) from the error detection unit 22. The clock error detected by the angle calculation) is input, and the same processing as that of the clock error correction unit 20-1 is performed to correct the carrier symbol clock error for each group. Then, the clock error correction unit 20-2 corrects the clock error (clock error detected by adding the symbol on the frequency axis and calculating the deviation angle on the time axis) to the four groups of carrier symbols as the symbol addition unit 18-3. Output to.

  The symbol adder 18-3 and the SP extractor 21-2 perform the same processing as the symbol adder 18-1 and the SP extractor 21-1. The received power calculation unit 23, the spectrum calculation unit 24, and the delay profile calculation unit 25 receive the SP signal from the SP extraction unit 21-2, and calculate the received power, spectrum, and delay profile of the OFDM signal, respectively. In addition, since the calculation method of the received power, spectrum, and delay profile of the OFDM signal is known, detailed description is omitted here. Further, the SP extraction unit 21-2 extracts a pilot signal other than the SP signal, and the reception power calculation unit 23, the spectrum calculation unit 24, and the delay profile calculation unit 25 use the pilot signal to receive the reception power, the spectrum, and Each delay profile may be calculated.

(Broadband frequency error detector)
Next, the broadband frequency error detector 27 shown in FIG. 1 will be described in detail. As described above, the wideband frequency error detector 27 differentially demodulates the IQ signal that is the FFT output signal for each carrier, calculates the angle between the differentially demodulated signal and the I axis, and sets the calculated angle to the calculated angle. Based on the cumulative processing of the pilot signals modulated by DBPSK and arranged at the same carrier position for each symbol, the carrier position is specified in units of carriers, and the carrier position of the specified pilot signal and the correct carrier of the pilot signal are specified. A correlation value between the position (carrier position of the pilot signal arranged by the transmitter) is calculated, and a frequency error equal to or greater than the carrier interval is detected.

  FIG. 2 is a block diagram illustrating a configuration of the wideband frequency error detection unit 27, and FIG. 3 is a flowchart illustrating processing of the wideband frequency error detection unit 27. With reference to FIG. 2, the wideband frequency error detection unit 27 includes a differential demodulation unit 28, a coefficient calculation unit 29, an arrangement information memory 30, and a correlation calculation unit 31.

  Referring to FIG. 3, the differential demodulation unit 28 of the wideband frequency error detection unit 27 differentially demodulates the FFT output signal input from the FFT unit 17-4, so that the IQ signal of the current symbol for each carrier. A differential demodulated signal is generated using the IQ signal of the previous symbol (step S301). Then, the differential demodulator 28 outputs the differential demodulated signal to the coefficient calculator 29.

Here, the number of symbols is M, the number of carriers is N, the symbol number is m (0 ≦ m <M), and the carrier number is n (0 ≦ n <N). Also, an IQ signal that is an FFT signal of symbol number m and carrier number _n is represented as a _{m, n} + jb _{m, n} . When the current symbol number is m, the differential demodulator 28 determines, for each carrier, the IQ signal a _{m, n} + jb _{m, n} of the current symbol and the IQ signal a _{m−1, n of the} previous symbol. Using + jb _{m−1, n} , a differential demodulated signal a ′ _{m, n} + jb ′ _{m, n} is obtained by the following equation.

  The coefficient calculator 29 receives the differential demodulated signal from the differential demodulator 28 and calculates the angle between the IQ signal, which is a differential demodulated signal, and the I axis for each carrier (step S302). For the pilot signal carrier modulated by the DBPSK modulation method and arranged at the same carrier position for each symbol, the angle is close to 0 degrees.

  Then, the coefficient calculation unit 29 calculates a coefficient RxTMCC (n) accumulated and averaged in the symbol direction using the result of subtracting the sine value from the cosine value of the calculated angle (step S303). The coefficient RxTMCC (n) is close to 1 for the carrier of the pilot signal modulated by the DBPSK modulation method and arranged at the same carrier position every symbol.

角度θ_m,nは、０〜９０ｄｅｇ（度）の値となる。 Specifically, the coefficient calculation unit 29 calculates an angle θ _{m, n} between the differential demodulated signal a ′ _{m, n} + jb ′ _{m, n} and the I axis by the following equation.

The angle θ _{m, n} has a value of 0 to 90 deg (degrees).

係数RxTMCC（ｎ）は、０〜１の値となる。 Then, the coefficient calculation unit 29 calculates the coefficient RxTMCC (n) by the following formula using the calculated angle θ _{m, n} .

The coefficient RxTMCC (n) takes a value between 0 and 1.

FIG. 11 is a diagram illustrating the relationship between the angle θ _{m, n} and (cos θ _{m, n} −sin θ _{m, n} ). As shown in FIG. 11, the angle theta _{m, n} and _{(cosθ m, n -sinθ m,} n) and the relative angle _{θ m, n = 0~90deg (cosθ} m, n -sinθ m, n) = 1 to −1, which is a substantially linear relationship. For the pilot signal carrier modulated by the DBPSK modulation method and arranged at the same carrier position for each symbol, the angle θ _{m, n} takes a value close to 0 deg, and the coefficient RxTMCC (n) as shown in FIG. Is close to 1. On the other hand, for a carrier other than the pilot signal, the coefficient RxTMCC (n) is close to 0 because a random angle θ _{m, n} is taken.

  Thereby, the value of n in the coefficient RxTMCC (n) close to 1 can be specified as the carrier position of the pilot signal. Further, by increasing the number M of symbols to be measured, it becomes easier to specify the carrier position of the pilot signal. Therefore, the carrier position of the pilot signal is specified even when the signal power is at a low level below thermal noise. be able to. Then, the correlation calculation unit 31 described later can detect a frequency error greater than the carrier interval.

  The arrangement information memory 30 stores arrangement information TxTMCC (n) in which the carrier position (carrier number) of a pilot signal that is modulated by the DBPSK modulation method and is arranged at the same carrier position for each symbol is defined. In TxTMCC (n), it is assumed that 1 (TxTMCC (i) = 1) is set when the carrier number i of the pilot signal is 0, and 0 is set for other carrier numbers.

The correlation calculation unit 31 inputs the coefficient RxTMCC (n) for each carrier from the coefficient calculation unit 29 and reads the arrangement information TxTMCC (n) from the arrangement information memory 30 to obtain the coefficient RxTMCC (n) and the arrangement information TxTMCC (n). Correlation value Corr (k) is calculated (step S304). Here, assuming that the maximum frequency error to be detected is ± K times (integer multiple) of the carrier interval, the correlation value Corr (k) (−K ≦ k ≦ K) is calculated by the following equation.

  Then, the correlation calculation unit 31 specifies k where the calculated correlation value Corr (k) is maximum in the range of −K ≦ k ≦ K, and multiplies the specified k by the carrier interval, thereby reducing the wideband frequency error. Detection is performed (step S305). The detected broadband frequency error is output to the broadband frequency error correction unit 26.

  As described above, the processing of steps S301 to S305 in the broadband frequency error detection unit 27 performs the accumulation process on the pilot signals modulated by the DBPSK modulation method and arranged at the same carrier position for each symbol, By obtaining the correlation, the carrier position of the pilot signal can be specified in units of carriers, and a wideband frequency error that is greater than the carrier interval can be detected.

(Error detection unit)
Next, the error detection unit 22 shown in FIG. 1 will be described in detail. As described above, the error detection unit 22 receives the SP signal and the CP signal from the SP extraction unit 21-1, and also receives the number of calculation symbols indicating the number of synchronous additions from the symbol addition unit 18-2. When the SP signal with the lowest frequency at the left end and the CP signal with the highest frequency at the right end are observed, and the amount of change in deflection angle per symbol of these SP signal and CP signal is equal to or greater than the number of calculated symbols, Deviations are calculated on the time axis for these SP and CP signals, and narrowband frequency errors and clock errors are detected.

  FIG. 4 shows a transition image of the SP signal input from the SP extraction unit 21-1 by the error detection unit 22. Specifically, for the SP signal input from the SP extraction unit 21-1, the error detection unit 22 observes the SP signal at the same carrier position in time series and plots the image on the IQ axis (SP in the Mth symbol). The position of the signal and the position of the SP signal in the Nth symbol) are shown. Let N> M. (1) shows a case where there is no narrowband frequency error and clock error (hereinafter collectively referred to as an error), and (2) shows a case where there is an error.

  As shown in FIG. 4A, when there is no error, the SP signal is plotted on a straight line passing through the origin of the IQ axis, and the deflection angle α1 of the SP signal changes at any time (symbol). Absent. On the other hand, as shown in FIG. 4B, when there is an error, the SP signal is not plotted on a straight line, and the deflection angles α1 and α2 of the SP signal change with the passage of time. In this case, the amount of change in the declination of the SP signal (the amount of declination changing between one OFDM symbol) varies depending on the carrier symbol.

  FIG. 5 is a diagram showing the relationship between the frequency error and clock error between the transceiver and the center frequency of the pilot carrier. The vertical solid line in FIG. 5 (1) and the vertical dotted line in FIGS. 5 (2) to (4) indicate the position of the center frequency in each OFDM carrier. As shown in FIG. 5 (2), when the OFDM carrier includes only the frequency error Δfc, the center frequencies of all the carrier symbols are uniformly shifted by Δfc. For this reason, the amount of change in declination is the same for all carrier symbols. As shown in FIG. 5 (3), when the OFDM carrier includes only the clock error Δfclk, the carrier symbol clock error Δfclk increases as the frequency position is further away from the carrier symbol at the center position (0). Therefore, as shown in FIG. 5 (4), when the OFDM carrier includes a frequency error and a clock error, the error is the sum of the frequency error Δfc and the clock error Δfclk. For this reason, the amount of change in the deviation angle of the carrier symbol differs depending on the frequency position of the carrier symbol. Therefore, the error detection unit 22 calculates a change amount of each declination for a plurality of carrier symbols, and detects a narrowband frequency error and a clock error from the change amount.

  FIG. 6 (1) is a diagram showing a transition example on the IQ axis in the SP signal having the lowest frequency at the left end of the OFDM wave (left end SP signal), and FIG. 6 (2) is the frequency at the right end of the OFDM wave. FIG. 6C is a diagram showing a transition example on the IQ axis of a CP signal having a high value (right end CP signal), and FIG. 6C is a diagram showing a transition example of the change amount θ of the deflection angle in the left end SP signal. As shown in FIGS. 6 (1) and (2), the position of both the left end SP signal and the right end CP signal is rotated by an error as the symbol advances. It rotates on the circumference around the origin on the axis. Also, as shown in FIG. 6 (3), when the left end SP signal becomes a calculation symbol of a predetermined symbol or more, the variation amount θ of the declination per symbol becomes stable. The same applies to the right end CP signal. Therefore, the error detection unit 22 determines that the amount of change in the deflection angle used for error detection is 1 symbol when the variation amount θ of the deflection angle per symbol is stable (when the number of calculation symbols is equal to or greater than the predetermined number of calculation symbols). What is necessary is just to obtain | require the variation | change_quantity (theta) of a perclination.

  FIG. 7 is a diagram for explaining the left end SP signal and the right end CP signal to be observed in the first embodiment. As shown in FIG. 7, the leftmost SP signal is a signal at the 2808th carrier position in the direction of lower frequency from the position of the center carrier on the frequency axis, and is the SP signal with the lowest frequency. The rightmost CP signal is a signal at the 2808th carrier position in the direction of higher frequency from the position of the center carrier on the frequency axis, and is the CP signal with the highest frequency. The error detection unit 22 observes the left end SP signal and the right end CP signal at the positions shown in FIG. 7 among the SP signal and the CP signal input from the SP extraction unit 21-1, and as shown in FIG. When the amount of change in the deflection angle per symbol of the SP signal and the right end CP signal is stabilized, the amount of change in the deflection angle is calculated, and the narrowband frequency error and the clock error are detected from these changes.

FIG. 8 is a flowchart illustrating the processing of the error detection unit 22 according to the first embodiment. First, the error detection unit 22 receives the SP signal and the CP signal from the SP extraction unit 21-1, and also receives the number of calculation symbols indicating the number of synchronous additions from the symbol addition unit 18-2 (step S801). When the error detection unit 22 determines the M-th symbol (M-th symbol) set in advance from the number of input calculation symbols, the IQ value in the left end SP signal of the SP signal and CP signal of the M-th symbol ( I _SP (M), Q _SP (M)) and the IQ value (I _CP (M), Q _CP (M)) in the right end CP signal are stored in the memory (step S802). Here, M is the number of symbols at the stage where the amount of deviation change per symbol is equal to or greater than the number of calculated symbols in the left end SP signal and the right end CP signal, and the left end SP signal and the right end CP signal are arranged. A preset value for designating a symbol is used. The same applies to Example 2 described later.

When the error detection unit 22 determines a preset N (> M) th symbol (Nth symbol) from the number of input calculation symbols, the IQ in the left end SP signal of the Nth symbol SP signal and CP signal is determined. The value (I _SP (N), Q _SP (N)) and the IQ value (I _CP (N), Q _CP (N)) in the right end CP signal are stored in the memory (step S803). Here, N is the number of symbols larger than M at the stage where the amount of change in the deflection angle per symbol is equal to or more than the number of calculated symbols in the left end SP signal and the right end CP signal, and the left end SP signal and the right end CP A preset value for designating the symbol in which the signal is arranged is used. The same applies to Example 2 described later.

  The error detection unit 22 determines whether the deviation change amount per symbol of the left end SP signal and the right end CP signal is equal to or greater than the number of calculation symbols that are stable in steps S802 and S803. Although the Mth and Nth symbols set in advance are determined, as shown in FIG. 6, it is determined whether the deflection change amount per symbol of the left end SP signal and the right end CP signal is stable, and thereafter The symbol in which the left end SP signal and the right end CP signal are arranged may be determined as the Mth symbol, and the symbol that is larger than M and in which the left end SP signal and the right end CP signal are arranged may be determined as the Nth symbol. For example, the error detection unit 22 calculates a deviation change amount per symbol of the left end SP signal and the right end CP signal for each predetermined number of symbols, and the change in the deviation change amount is equal to or less than a certain threshold value. When it becomes, it may be determined that the above-mentioned stable stage is reached. Further, the error detection unit 22 calculates the amplitude values of the left end SP signal and the right end CP signal for each predetermined number of symbols using the input number of calculated symbols, and the change in the average value of the amplitude values is constant. You may determine with the above-mentioned stable stage when it becomes below a threshold. The same applies to Example 2 described later.

The error detection unit 22 reads the IQ value (I _SP (M), Q _SP (M)) of the left end SP signal of the Mth symbol and the IQ value (I _CP (M), Q _CP (M) of the right end CP signal from the memory. )), And the IQ value (I _SP (N), Q _SP (N)) of the left end SP signal of the Nth symbol and the IQ value (I _CP (N), Q _CP (N)) of the right end CP signal are read out. Using these IQ values, the deviation change amount (rotation amount) θ _SP per symbol in the left end SP signal and the deviation change amount θ _CP per symbol in the right end CP signal are calculated using the IQ values below. (Step S804).
[Equation 5]
θ _SP = {atan (Q _SP (N) / I _SP (N)) − atan (Q _SP (M) / I _SP (M))} / (N−M) (5)
[Equation 6]
θ _CP = {atan (Q _CP (N) / I _CP (N)) − atan (Q _CP (M) / I _CP (M))} / (N−M) (6)

Error detecting unit 22, a polarization angle variation theta _CP polarization angle variation theta _SP and right CP signal left SP signals calculated in step S804, the by the following equation, the deviation amount ΔF of the center frequency of the left SP signals The center frequency of the _SP and right end CP signal is converted into a shift amount ΔF _CP (step S805).
[Equation 7]
ΔF _SP = θ _SP /1.008e ⁻³ / 2 / π (7)
[Equation 8]
ΔF _CP = θ _CP /1.008e ⁻³ / 2 / π (8)
Here, 1.008e ⁻³ indicates the OFDM symbol length (sec).

The error detector 22 uses the following equation to calculate the frequency error Δfc and the clock error using the shift amount ΔF _SP of the center frequency of the left end SP signal and the shift amount ΔF _CP of the center frequency of the right end CP signal converted in step S805. Δfclk is calculated, the frequency error Δfc is output to the narrowband frequency error correction unit 15-2 as a narrowband frequency error, and the clock error Δfclk is output to the clock error correction unit 20-2 (step S806).
[Equation 9]
Δfc = (ΔF _CP + ΔF _SP ) / 2 (9)
[Equation 10]
Δfclk = (ΔF _CP −ΔF _SP ) × 8192/5616 (10)
Here, the FFT size is 8192, and the number of subcarriers is 5617. Thereby, the frequency error Δfc and the clock error Δfclk are calculated simultaneously.

The mathematical formulas (9) and (10) in step S806 will be described in detail. The shift amount ΔF _CP of the center frequency of the right end CP signal is determined by a constant frequency error Δfc in all carriers and a clock error component at the time of FFT. At the time of FFT, the clock is aligned with the center carrier, and the right end CP signal is located at the 2808th carrier position in the direction of higher frequency from the position of the center carrier. Therefore, the clock error component of the right end CP signal is Δfclk × 2808 / 8192. Therefore, the shift amount ΔF _CP of the center frequency of the right end CP signal is expressed by the following equation.
[Equation 11]
ΔF _CP = Δfc + Δfclk × 2808/8192 (11)

Similarly to the right end CP signal, the shift amount ΔF _SP of the center frequency of the left end SP signal is also determined by a constant frequency error Δfc in all carriers and a clock error component at the time of FFT. Further, since the left end SP signal is located at the 2808th carrier position in the direction of lower frequency from the position of the center carrier, the clock error component of the left end SP signal is a negative value unlike the right end CP signal, and Δfclk × ( -2808) / 8192. Therefore, the shift amount ΔF _SP of the center frequency of the left end SP signal is expressed by the following equation.
[Equation 12]
ΔF _SP = Δfc + Δfclk × (−2808) / 8192 (12)
Thereby, the formulas (9) and (10) are derived from the formulas (11) and (12).

  As described above, the left end SP signal and the right end CP signal are observed among the SP signal and the CP signal extracted by the SP extraction unit 21-1 by the processing in steps S801 to S806 in the error detection unit 22 of the first embodiment. When the number of calculation symbols exceeds a predetermined number, the declination angle is calculated on the time axis for these SP signal and CP signal to detect a narrowband frequency error and a clock error.

  Conventionally, in order to determine the direction of the frequency error after detecting the frequency error of the absolute value, complicated processing such as obtaining the number of calculation symbols at which the SP signal power reaches the peak value has been required. On the other hand, the error detection unit 22 observes the declination of the SP signal or the like on the time axis, and calculates the narrowband frequency error including the positive and negative directions from the change amount. It is possible to calculate a narrow band frequency error with high accuracy by a simple method. Conventionally, the clock error is calculated by the addition processing on the time axis and the guard correlation processing, whereas the error detection unit 22 extracts the pilot signal by symbol addition on the frequency axis, and the extracted pilot signal Was observed on the time axis, and the clock error was calculated from the amount of change. This makes it possible to calculate a clock error with higher accuracy than in the past. Therefore, even if the received power is at a low level, highly accurate narrowband frequency error and clock error are calculated, and these errors are corrected. Therefore, it is possible to measure an OFDM wave signal with high accuracy. That is, the received power, spectrum, and delay profile can be measured with high accuracy.

  As described above, according to the OFDM wave measuring apparatus 1 of the first embodiment, the wideband frequency error detection unit 27 differentially demodulates the IQ signal that is the FFT output signal for each carrier, By calculating the angle between the axis and accumulating pilot signals that are modulated by DBPSK and arranged at the same carrier position for each symbol based on the calculated angle, the carrier position is specified in units of carriers, A correlation value between the carrier position of the identified pilot signal and the correct carrier position of the pilot signal is calculated, and a wideband frequency error equal to or greater than the carrier interval is detected.

  Thereby, even if the signal power is a low level below thermal noise, it is modulated by the DBPSK modulation method, and the carrier position of the pilot signal arranged at the same carrier position for each symbol can be specified in units of carriers. Wideband frequency error greater than the carrier interval can be detected. Therefore, it is possible to measure the received power, spectrum, and delay profile with high accuracy even for a low level signal having a frequency error equal to or greater than the carrier interval and less than thermal noise. Further, the error detection unit 22 of the first embodiment detects a narrowband frequency error and a clock error, and the narrowband frequency error correction unit 15-2 and the clock error correction unit 20-2 correct these so that the received power, It becomes possible to measure the spectrum and the delay profile with higher accuracy.

[Example 2]
Next, Example 2 will be described. As described above, in the second embodiment, the IQ signal is differentially demodulated to calculate the angle with respect to the I axis, and the wideband frequency error detection pilot signal that is differentially modulated and arranged at the same carrier position for each symbol is obtained. By accumulating, the carrier position is identified, a wideband frequency error that is greater than the carrier interval is detected, and a pilot signal for narrowband frequency error detection is extracted by synchronous symbol addition on the frequency axis. Select a narrowband frequency error detection pilot signal with a high amplitude value from the frequency error detection pilot signal, observe the declination of the selected narrowband frequency error detection pilot signal on the time axis, and use the amount of change. , Narrowband frequency errors including positive and negative directions and clock errors are detected.

  The OFDM wave measuring apparatus 1 according to the second embodiment has the same configuration as the OFDM wave measuring apparatus 1 according to the first embodiment shown in FIG. 1 and detects a wideband frequency error by the same processing as that of the first embodiment. Processing different from that of the error detection unit 22 is performed. Since the components other than the error detector 22 are the same as those in the first embodiment, the description thereof is omitted here.

  The error detection unit 22 according to the second embodiment inputs the SP signal and the CP signal from the SP extraction unit 21-1, and also receives the number of calculation symbols indicating the number of synchronous additions from the symbol addition unit 18-2. SP1 signal and SP2 signal having the highest amplitude values in the ranges A and B are selected, respectively, and the selection is made when the amount of deviation change per symbol of the SP1 signal and SP2 signal is equal to or greater than the number of calculated symbols. The declination angle on the time axis is calculated for the SP1 signal and SP2 signal, and a narrowband frequency error and a clock error are detected.

(Error detection unit)
Next, the error detection unit 22 according to the second embodiment will be described in detail. FIG. 9 is a diagram illustrating the SP1 signal and the SP2 signal selected in the second embodiment. As shown in FIG. 9, a range A in which the SP1 signal is selected and a range B in which the SP2 signal is selected are set in advance on the frequency axis. Here, it is assumed that the center frequency of the SP1 signal is lower than the center frequency of the SP2 signal. The SP1 signal is selected as the SP signal having the highest amplitude value among the plurality of SP signals within the range A, and the SP2 signal is selected as the SP signal having the highest amplitude value among the plurality of SP signals within the range B. . The selected SP1 signal is the signal at the a-th carrier position in the direction of lower frequency from the position of the center carrier on the frequency axis, and the SP2 signal is b in the direction of higher frequency from the position of the center carrier on the frequency axis. Assuming that the signal is at the first carrier position, the SP1 signal is at a position away from the center carrier by a × Δf (Hz) in the direction of lower frequency, and the SP2 signal is at the position of b × Δf in the direction of higher frequency from the center carrier. (Hz) away. Here, Δf represents the frequency of the carrier interval.

  For example, in the frequency region in which carriers are arranged on the frequency axis, the range A in which the SP1 signal is selected is a range including the SP signal having the lowest frequency and is ¼ from the carrier position having the lowest frequency. As the range B in which the range is preset and the SP2 signal is selected, a range including the SP signal with the highest frequency and a range of ¼ from the carrier position with the highest frequency is preset. As a result, the center frequency of the SP1 signal and the center frequency of the SP2 signal are separated from each other as compared with the case where the ranges A and B where the frequencies are close to each other, and the difference in the declination becomes large. Band frequency error and clock error can be calculated.

  FIG. 10 is a flowchart illustrating the processing of the error detection unit 22 according to the second embodiment. First, the error detection unit 22 receives the SP signal and the CP signal from the SP extraction unit 21-1, and calculates the number of synchronous additions from the symbol addition unit 18-2, similarly to step S801 shown in FIG. The number of symbols is input (step S1001).

  When the error detection unit 22 determines the Mth symbol set in advance from the number of input calculation symbols, the error detection unit 22 calculates the amplitude value for each of the plurality of SP signals in the preset range A at the Mth symbol, The SP1 signal (a carrier from the center carrier, see FIG. 9) is selected as the SP signal having a high value, and amplitude values are calculated for a plurality of SP signals within a preset range B, respectively. The SP2 signal (the bth carrier from the center carrier) is selected as the SP signal having a high (step S1002).

  Note that the error detection unit 22 may select the SP1 signal and the SP2 signal by a process similar to the above for the symbols before the preset Mth symbol.

In the Mth symbol, the error detection unit 22 performs the IQ value (I _SP1 (M), Q _SP1 (M)) of the SP1 signal selected in step S1002 and the IQ value (I _SP2 (M), Q _SP2 ) of the SP2 signal. (M)) is stored in the memory (step S1003). When the error detection unit 22 determines the Nth (> M) symbol set in advance from the number of input calculation symbols, the IQ value (I _SP1 (N) of the SP1 signal at the same carrier position selected in step S1002 is determined. , Q _SP1 (N)) and the IQ value (I _SP2 (N), Q _SP2 (N)) of the SP2 signal at the same carrier position are stored in the memory (step S1004).

The error detection unit 22 reads the IQ value (I _SP1 (M), Q _SP1 (M)) of the _SP1 signal and the IQ value (I _SP2 (M), Q _SP2 (M)) of the SP2 signal from the memory. And the IQ value (I _SP1 (N), Q _SP1 (N)) of the SP1 signal of the Nth symbol and the IQ value (I _SP2 (N), Q _SP2 (N)) of the SP2 signal are read out, and Thus, using these IQ values, the deflection change amount (rotation amount) θ _SP1 per symbol in the _SP1 signal and the deflection change amount θ _SP2 per symbol in the SP2 signal are calculated (step S1005).
[Equation 13]
θ _SP1 = {atan (Q _SP1 (N) / I _SP1 (N)) − atan (Q _SP1 (M) / I _SP1 (M))} / (N−M) (13)
[Formula 14]
θ _SP2 = {atan (Q _SP2 (N) / I _SP2 (N)) − atan (Q _SP2 (M) / I _SP2 (M))} / (N−M) (14)

The error detection unit 22 calculates the deviation angle variation θ _{SP1 of the SP1} signal and the deviation variation θ _SP2 of the SP2 signal calculated in step S1005 by using the following expressions, and the deviation amounts ΔF _SP1 and SP2 of the center frequency of the SP1 signal. Conversion to a shift amount ΔF _SP2 of the center frequency of the signal is performed (step S1006).
[Equation 15]
ΔF _SP1 = θ _SP1 /1.008e ⁻³ / 2 / π (15)
[Equation 16]
ΔF _SP2 = θ _SP2 /1.008e ⁻³ / 2 / π (16)
Here, 1.008e ⁻³ indicates the OFDM symbol length (sec).

The error detection unit 22 calculates the frequency error Δfc and the clock error Δfclk using the following equation using the shift amount ΔF _SP1 of the center frequency of the SP1 signal and the shift amount ΔF _SP2 of the center frequency of the SP2 signal converted in step S1006. The frequency error Δfc is calculated and output to the narrowband frequency error correction unit 15-2 as a narrowband frequency error, and the clock error Δfclk is output to the clock error correction unit 20-2 (step S1007).
[Equation 17]
Δfc = (a × ΔF _SP2 + b × ΔF _SP1 ) / (a + b) (17)
[Equation 18]
Δfclk = (ΔF _SP2 −ΔF _SP1 ) × 8192 / (a + b) (18)
Here, the FFT size is 8192, and the number of subcarriers is 5617. Thereby, the frequency error Δfc and the clock error Δfclk are calculated simultaneously.

The mathematical formulas (17) and (18) in step S1007 will be described in detail. The shift amount ΔF _SP1 of the center frequency of the SP1 signal is determined by a constant frequency error Δfc in all carriers and a clock error component at the time of FFT. At the time of FFT, the clock is aligned with the center carrier, and since the SP1 signal is located at the a-th carrier position in the direction of lower frequency from the position of the center carrier, the clock error component of the SP1 signal is Δfclk × (−a). / 8192. Therefore, the shift amount ΔF _SP1 of the center frequency of the SP1 signal is expressed by the following equation.
[Equation 19]
ΔF _SP1 = Δfc + Δfclk × (−a) / 8192 (19)

Similarly to the SP1 signal, the shift amount ΔF _SP2 of the center frequency of the SP2 signal is also determined by a constant frequency error Δfc in all carriers and a clock error component at the time of FFT. Further, since the SP2 signal is located at the b-th carrier position in the direction of higher frequency from the position of the center carrier, the component of the clock error of the SP2 signal is a positive value unlike the SP1 signal, and Δfclk × b / 8192. Become. Therefore, the shift amount ΔF _SP2 of the center frequency of the SP2 signal is expressed by the following equation.
[Equation 20]
ΔF _SP2 = Δfc + Δfclk × b / 8192 (20)
As a result, the equations (17) and (18) are derived from the equations (19) and (20).

  As described above, the SP signal extracted by the SP extraction unit 21-1 by the processing of steps S1001 to S1007 in the error detection unit 22 of the second embodiment, which is the most SP signal within the predetermined ranges A and B. The SP1 signal and SP2 signal having high amplitude values are selected, respectively, and the selected SP1 signal and SP2 signal are observed. When the number of calculation symbols exceeds a predetermined number, the SP1 signal and SP2 signal are deviated on the time axis. The angle was calculated and the narrowband frequency error and clock error were detected. As a result, the narrowband frequency error and the clock error are calculated using the SP1 signal and the SP2 signal in good reception state, so that in addition to the effects of the first embodiment, the received OFDM signal is transmitted at a specific frequency of the carrier. Even in the case of including a so-called multipath having a low amplitude value, an OFDM wave signal can be measured with high accuracy. That is, the received power, spectrum, and delay profile can be measured with high accuracy.

  As described above, according to the OFDM wave measuring apparatus 1 of the second embodiment, as in the first embodiment, the carrier position of the pilot signal can be specified in units of carriers, and a wideband frequency error greater than the carrier interval is detected. be able to. Therefore, it is possible to measure the received power, spectrum, and delay profile with high accuracy even for a low level signal having a frequency error equal to or greater than the carrier interval and less than thermal noise. Further, the error detection unit 22 of the second embodiment detects a narrowband frequency error and a clock error, and the narrowband frequency error correction unit 15-2 and the clock error correction unit 20-2 correct these so that the received power, It becomes possible to measure the spectrum and the delay profile with higher accuracy.

  The present invention has been described with reference to the first and second embodiments. However, the present invention is not limited to the first and second embodiments, and various modifications can be made without departing from the technical idea thereof. In the first and second embodiments, the wideband frequency error detection unit 27 of the OFDM wave measuring apparatus 1 specifies the carrier position of the pilot signal modulated by DBPSK and arranged at the same carrier position for each symbol for each carrier. However, the carrier position of the pilot signal modulated by DQPSK (Differential Quadrature Phase Shift Keying) and arranged at the same carrier position for each symbol may be specified. In this case, the OFDM wave received by the OFDM wave measuring apparatus 1 includes a pilot signal that is modulated by the DQPSK modulation method and arranged at the same carrier position for each symbol by the transmitting apparatus that transmits the OFDM wave. . In the present invention, the pilot signal only needs to be differentially modulated in order to identify the carrier position of the pilot signal in units of carriers based on the angle between the differentially demodulated signal and the I axis.

  In addition to the first and second embodiments, the present invention can be applied to any OFDM wave measuring apparatus that measures an OFDM signal based on a pilot signal. For example, the present invention is also applicable to the OFDM wave measuring apparatus described in FIGS. Specifically, in FIGS. 1 to 4 of Patent Document 3, the OFDM wave measuring apparatus includes a wideband frequency error correction unit 26 in the subsequent stage of the frequency error correction unit (narrowband frequency error correction unit) as in the present invention. , A symbol cutout unit 16-4, an FFT unit 17-4, and a broadband frequency error detection unit 27.

  Further, in the error detection unit 22 according to the first and second embodiments, the narrowband frequency error and the clock error are detected using the SP signal and the CP signal or the SP signal, but pilots other than the SP signal and the CP signal are used. A signal may be used. For example, the error detection unit 22 according to the second embodiment selects an SP signal having the highest amplitude value from a plurality of SP signals within a preset range A (low frequency region), and sets the preset range B (frequency). The SP signal having the highest amplitude value is selected from a plurality of SP signals within the range (B), but the SP signal or CP signal having the highest amplitude value is selected from the plurality of SP signals and CP signals within the range B. You may make it do.

  Further, the error detection unit 22 according to the first embodiment detects the narrowband frequency error and the clock error using two different SP signals and CP signals, but may use two different SP signals. However, three or more different SP signals may be used. Further, the error detection unit 22 according to the second embodiment detects the narrowband frequency error and the clock error using two different SP signals and the like, but may use three or more different SP signals and the like. Good. For example, the error detection unit 22 according to the first and second embodiments detects a first frequency error and a clock error by a combination of two SP signals, and a second frequency error and a clock error by a combination of the other two SP signals. A clock error is detected, a median value is obtained from the detected first frequency error, clock error, second frequency error, and clock error, and a narrowband frequency error correction unit using the median value of the frequency error as a narrowband frequency error. 15-2, and the median value of the clock error may be output to the clock error correction unit 20-2. Further, for each of three or more combinations of two SP signals, the respective frequency errors and clock errors are detected, and an average value thereof is obtained to obtain a narrowband frequency error correction unit 15-2 and a clock error correction unit 20-. 2 may be output respectively.

  As a hardware configuration of the OFDM wave measuring apparatus 1, a normal computer can be used. The OFDM wave measuring apparatus 1 is configured by a computer including a volatile storage medium such as a CPU and a RAM, a non-volatile storage medium such as a ROM, an interface, and the like. Orthogonal demodulation unit 13, error detection unit 14, narrowband frequency error correction units 15-1 and 15-2, symbol cutout units 16-1, 16-2, 16-3, and 16-included in OFDM wave measuring apparatus 1. 4, FFT units 17-1, 17-2, 17-3, 17-4, symbol addition units 18-1, 18-2, 18-3, SP pattern detection unit 19, clock error correction units 20-1, 20 -2, SP extraction units 21-1, 21-2, error detection unit 22, received power calculation unit 23, spectrum calculation unit 24, delay profile calculation unit 25, broadband frequency error correction unit 26, and broadband frequency error detection unit 27 Each function is realized by causing a CPU to execute a program describing these functions.

  Further, components other than the frequency conversion unit 11 and the A / D conversion unit 12 included in the OFDM wave measuring apparatus 1 of FIG. 1, that is, an orthogonal demodulation unit 13, an error detection unit 14, a narrowband frequency error correction unit 15-1, 15-2, symbol cutout units 16-1, 16-2, 16-3, 16-4, FFT units 17-1, 17-2, 17-3, 17-4, symbol addition units 18-1, 18 -2, 18-3, SP pattern detection unit 19, clock error correction units 20-1, 20-2, SP extraction units 21-1, 21-2, error detection unit 22, received power calculation unit 23, spectrum calculation unit 24, a measurement apparatus including a delay profile calculation unit 25, a broadband frequency error correction unit 26, and a broadband frequency error detection unit 27 can be configured. This measuring apparatus can also use a normal computer. Like the OFDM wave measuring apparatus 1 described above, a volatile storage medium such as a CPU and a RAM, a non-volatile storage medium such as a ROM, an interface, etc. Consists of a computer equipped. Each function of the quadrature demodulator 13 and the like provided in this measuring apparatus is realized by causing the CPU to execute a program describing these functions.

  In this case, the reception apparatus including the frequency conversion unit 11 and the A / D conversion unit 12 illustrated in FIG. 1 receives the OFDM signal by the reception antenna, and converts the digital IF signal converted by the A / D conversion unit 12. The received OFDM signal data is stored in the memory. Then, the measurement device stores the received OFDM signal data in the memory by downloading the received OFDM signal data stored in the memory of the receiving device or reading it through the storage device. Then, the measurement apparatus reads the received OFDM data from the memory, and executes a program describing the functions of the orthogonal demodulation unit 13 and the like. As a result, the OFDM signal received by the receiving device can be processed by the measuring device, and the received power and the like can be calculated.

  These programs can be stored and distributed on a storage medium such as a magnetic disk (floppy (registered trademark) disk, hard disk, etc.), optical disk (CD-ROM, DVD, etc.), semiconductor memory, etc., and sent and received via a network. You can also

DESCRIPTION OF SYMBOLS 1 OFDM wave measuring device 11 Frequency conversion part 12 A / D conversion part 13 Orthogonal demodulation part 14,22 Error detection part 15 Narrow band frequency error correction part 16 Symbol extraction part 17 FFT part 18 Symbol addition part 19 SP pattern detection part 20 Clock error correction unit 21 SP extraction unit 23 Received power calculation unit 24 Spectrum calculation unit 25 Delay profile calculation unit 26 Wideband frequency error correction unit 27 Wideband frequency error detection unit 28 Differential demodulation unit 29 Coefficient calculation unit 30 Arrangement information memory 31 Correlation calculation Part

Claims (4)

In an OFDM wave measuring apparatus that receives an OFDM wave including a pilot signal, extracts the pilot signal, and measures the signal of the OFDM wave,
A quadrature demodulator for demodulating the received OFDM wave signal to generate a baseband signal;
An error detection unit that adds a predetermined number of symbols on the time axis to the baseband signal generated by the orthogonal demodulation unit, and detects a symbol head position, a narrowband frequency error, and a clock error by guard correlation; ,
A narrowband frequency error correction unit that corrects a narrowband frequency error detected by the error detection unit with respect to the baseband signal generated by the orthogonal demodulation unit;
Based on the symbol head position detected by the error detection unit, the first symbol cut-out for removing the GI and cutting out the effective symbol is performed on the baseband signal whose narrowband frequency error is corrected by the narrowband frequency error correction unit. And outing,
A first FFT unit that performs FFT on the effective symbols cut out by the first symbol cutout unit to generate carrier symbols;
The carrier symbol IQ signal generated by the first FFT unit is differentially demodulated, an angle between the differentially demodulated signal and the I axis is calculated, and an accumulation process using the angle for a predetermined number of symbols The carrier position of the pilot signal for wideband frequency error detection that is differentially modulated and arranged at the same carrier position for each symbol is specified in units of carriers, and the carrier position of the pilot signal for broadband frequency error detection that has been specified is set in advance. A broadband frequency error detection unit that calculates a correlation value between the carrier position of the pilot signal for broadband frequency error detection that is performed, and detects a broadband frequency error based on the correlation value;
A broadband frequency error correction unit that corrects a broadband frequency error detected by the broadband frequency error detection unit with respect to a baseband signal whose narrowband frequency error is corrected by the narrowband frequency error correction unit;
A second symbol cutout unit that removes the GI and cuts out an effective symbol based on the symbol head position detected by the error detection unit for the baseband signal whose broadband frequency error has been corrected by the wideband frequency error correction unit. When,
A second FFT unit that performs FFT on the effective symbols cut out by the second symbol cutout unit to generate carrier symbols;
A symbol addition unit that synchronously adds the carrier symbols generated by the second FFT unit for each predetermined symbol, and generates a synchronous addition result;
A pattern detection unit that calculates a correlation value between a synchronous addition result generated by the symbol addition unit and a plurality of preset patterns, and detects a pattern having the maximum correlation value;
A pilot extraction unit that extracts a pilot signal from the carrier symbol of the synchronous addition result based on the pattern detected by the pattern detection unit;
An OFDM wave measuring apparatus comprising: In the OFDM wave measuring device according to claim 1,
The broadband frequency error detector is
A differential demodulator that differentially demodulates the IQ signal of the carrier symbol generated by the first FFT unit for each carrier to generate a differential demodulated signal;
The angle between the IQ signal of the differential demodulated signal generated by the differential demodulator and the I axis is calculated for each carrier, and the result of subtracting the sine value from the cosine value of the calculated angle is displayed in the symbol direction. A coefficient calculation unit that accumulates and averages the predetermined number of symbols, and calculates the average result as a coefficient;
The correlation between the coefficient for each carrier and the arrangement information using the coefficient for each carrier calculated by the coefficient calculation unit and the arrangement information in which the carrier position of the pilot signal for wideband frequency error detection is preset. A correlation calculation unit that calculates a value and detects a broadband frequency error based on the correlation value;
An OFDM wave measuring apparatus comprising: In the OFDM wave measuring device according to claim 1 or 2,
2. The OFDM wave measuring apparatus according to claim 1, wherein the wideband frequency error detection pilot signal is a pilot signal that is modulated by a DBPSK or DQPSK modulation method and arranged at the same carrier position for each symbol. Extracting the pilot signal from an OFDM wave signal including a pilot signal and measuring the OFDM wave signal;
A function of an orthogonal demodulator that orthogonally demodulates the signal of the OFDM wave and generates a baseband signal;
An error detection unit that adds a predetermined number of symbols on the time axis to the baseband signal generated by the quadrature demodulation unit and detects a symbol head position, a narrowband frequency error, and a clock error by guard correlation. Function and
A function of a narrowband frequency error correction unit that corrects a narrowband frequency error detected by the error detection unit with respect to the baseband signal generated by the orthogonal demodulation unit;
Based on the symbol head position detected by the error detection unit, the first symbol cut-out for removing the GI and cutting out the effective symbol is performed on the baseband signal whose narrowband frequency error is corrected by the narrowband frequency error correction unit. The function of the department,
A function of a first FFT unit that performs an FFT on the effective symbol cut out by the first symbol cutout unit and generates a carrier symbol;
The carrier symbol IQ signal generated by the first FFT unit is differentially demodulated, an angle between the differentially demodulated signal and the I axis is calculated, and an accumulation process using the angle for a predetermined number of symbols The carrier position of the pilot signal for wideband frequency error detection that is differentially modulated and arranged at the same carrier position for each symbol is specified in units of carriers, and the carrier position of the pilot signal for broadband frequency error detection that has been specified is set in advance. A broadband frequency error detection unit that calculates a correlation value between the carrier position of the pilot signal for broadband frequency error detection performed and detects a broadband frequency error based on the correlation value;
A function of a wideband frequency error correction unit that corrects a wideband frequency error detected by the wideband frequency error detection unit with respect to a baseband signal whose narrowband frequency error is corrected by the narrowband frequency error correction unit;
A second symbol cutout unit that removes the GI and cuts out an effective symbol based on the symbol head position detected by the error detection unit for the baseband signal whose broadband frequency error has been corrected by the wideband frequency error correction unit. Functions and
A function of a second FFT unit that performs FFT on the effective symbols cut out by the second symbol cutout unit to generate carrier symbols;
A function of a symbol addition unit that synchronously adds the carrier symbols generated by the second FFT unit for each predetermined symbol and generates a synchronous addition result;
A function of a pattern detection unit for calculating a correlation value between a synchronous addition result generated by the symbol addition unit and a plurality of preset patterns, and detecting a pattern having the maximum correlation value;
Based on the pattern detected by the pattern detection unit, the function of a pilot extraction unit that extracts a pilot signal from the carrier symbol of the synchronous addition result;
A program to realize

Images