JP4044109B2 - Signal processing device - Google Patents

Description

  The present invention relates to a signal processing technique, and more particularly to a signal processing device that determines synchronization of radio signals.

In a wireless communication system, a frequency offset occurs due to instability of a signal oscillated from a crystal or the like used in a transmission device and a reception device. In order to reduce the influence of the generated frequency offset, the phase error of the known signal included in the transmitted signal is estimated, and the phase of the signal is shifted by the estimated phase error. However, when the frequency offset is greater than or equal to ± 1 / 2M of the symbol rate (M is the number of phases in PSK (Phase Shift Keying) modulation), normal synchronization is not established and ± There may be a so-called pseudo-synchronization state where the synchronization state is fixed at a position shifted by 2π / M × symbol rate / 2π. Conventionally, in order to escape from this pseudo-synchronized state, error correction is performed, and whether or not pseudo-synchronization is achieved is determined based on the result. Further, when it is determined that the pseudo-synchronization is detected, the frequency shift is repeated until the pseudo-synchronization state is lost. (For example, refer to Patent Document 1).
Japanese Patent Laid-Open No. 11-308288 (FIG. 1)

  Under such circumstances, the present inventor has come to recognize the following problems. That is, a state in which a signal with very weak reception power is received at a terminal, as in a satellite communication system, and the reference phase in the carrier recovery loop is unnecessarily shifted due to the influence of the fading phenomenon and the Doppler phenomenon, That is, when a phase slip occurs, it takes a long time to make an accurate determination based on the reliability after error correction decoding.

  The present invention has been made in view of such circumstances, and an object of the present invention is to provide a receiving apparatus that can reduce the processing time required to escape from the pseudo-synchronized state.

  In order to solve the above problems, a signal processing device according to an aspect of the present invention includes an input unit, a demodulation unit, a partial signal acquisition unit, a differential interval determination unit, and a synchronization detection unit. The input unit inputs a signal sequence in which a plurality of known signal sequences including a plurality of known signals are inserted at a predetermined period L. The demodulation unit reproduces a carrier wave from the signal sequence input from the input unit, and demodulates the signal sequence according to the reproduced carrier wave. The partial signal acquisition unit acquires a plurality of partial signal sequences to be associated with the known signal sequences included in the signal sequence demodulated by the demodulation unit at the same cycle as the predetermined cycle. The differential interval determination unit uses a rule relating to a plurality of differential intervals for a partial signal that should correspond to a known signal included in a sequence of partial signals before and after the partial signal sequence acquired by the partial signal acquisition unit. And associate them with each other. The synchronization detection unit calculates the relative phase difference between the partial signals associated with each other according to the first rule among the plurality of rules regarding the differential interval, thereby determining the position where the sequence of known signals is inserted. Estimate and detect temporary synchronization. After detecting the provisional synchronization, the synchronization detector detects the relative position between the partial signals associated according to the second rule having a differential interval different from the first rule among the plurality of rules regarding the differential interval. By calculating the respective phase differences, it is detected whether or not the frequency of the carrier wave reproduced by the demodulator is correct.

  Here, “the input unit inputs the signal sequence” includes transmitting the received signal sequence to the signal processing device. For example, in order to cause the signal processing device to process the signal sequence, It also includes storing in memory. Further, the “known signal sequence” includes a sequence composed of a plurality of known signals arranged according to a predetermined rule. Each known signal may be the same symbol or may be a different symbol. That is, any information that is known on the receiving side may be used. The “partial signal sequence” includes a portion that should correspond to a known signal in a signal sequence received through a wireless channel. In other words, the partial signal is a known signal at the time of reception. The “partial signal” includes a portion that should correspond to a known signal in a signal series received through a wireless line. In other words, the partial signal is a known signal at the time of reception. “Associating” includes specifying, for example, two pieces of partial information for calculating a differential in partial information existing in two consecutive rows. The “relative phase difference” includes a phase difference between a plurality of phases, and includes, for example, a difference obtained by calculating a differential of the phases of known signals before and after. Further, “determining whether or not the frequency of the regenerated carrier wave is correct” includes determining whether or not frequency synchronization has been successfully captured.

  According to this aspect, it is possible to reduce the processing time for determining the pseudo-synchronized state by establishing synchronization in two stages using two different relative phase differences. Also, synchronization can be efficiently detected by first establishing timing synchronization and then establishing frequency synchronization. Also, synchronization can be detected more efficiently by using different rules regarding the differential interval for synchronization determination at each stage.

  The synchronization detection unit includes a first differential calculation unit, a first correlation calculation unit, a temporary synchronization determination unit, a second differential calculation unit, a second correlation calculation unit, and a pseudo synchronization determination unit. The first differential calculation unit calculates a relative phase difference between the partial signals associated with each other according to the first rule among the rules regarding the plurality of differential intervals. The first correlation calculation unit calculates the phase difference between the partial signals calculated by the first differential calculation unit and the relative phase difference between the known signals to be matched with the phase difference between the partial signals. A first correlation value is calculated. The temporary synchronization determination unit determines a position where a partial signal exists as a position where a sequence of known signals is inserted when the first correlation value calculated by the first correlation calculation unit is larger than a threshold value related to temporary synchronization. . The second differential calculation unit responds according to the second rule among the rules regarding a plurality of differential intervals among the partial signals existing at the position where the known signal sequence determined by the temporary synchronization determination unit is inserted. The relative phase error between the attached partial signals is calculated. The second correlation calculation unit calculates the phase difference between the partial signals calculated by the second differential calculation unit and the relative phase difference between the known signals to be matched with the phase difference between the partial signals. A second correlation value is calculated. The pseudo-synchronization determination unit determines that the frequency of the carrier wave reproduced by the demodulation unit is correct normal synchronization when the second correlation value calculated by the second correlation calculation unit is larger than a threshold value related to pseudo-synchronization.

  Here, “the relative phase difference between the known signals to be associated with the phase difference between the respective partial signals” includes a phase difference in the known signal which is an ideal value of the partial signal. That is, it includes a value indicating the relationship between the actual phase difference and the ideal phase difference, and includes, for example, the difference between the actual phase difference and the ideal phase difference. According to this aspect, by calculating the phase difference between the partial signals, synchronization can be accurately acquired even when a phase slip occurs.

  The synchronization detection unit may further include a measurement unit that measures the signal strength of the signal series. The number of integrations in the correlation values calculated by the first correlation calculation unit and the second correlation calculation unit may be determined according to the signal intensity measured by the measurement unit. Here, “the number of accumulated correlation values” includes the number of partial signals for which correlation values are calculated. According to this aspect, the synchronization can be captured in a short time by limiting the range in which the correlation value is calculated based on the signal strength.

  The synchronization detection unit may further include a control unit and a frequency correction unit. When the first correlation value calculated by the first correlation calculation unit is smaller than the threshold value related to temporary synchronization, the control unit targets again a partial signal sequence different from the partial signal sequence in the signal sequence, You may make a 1st differential calculation part calculate a differential. The frequency correction unit may correct the frequency of the carrier by shifting the frequency of the carrier when the second correlation value calculated by the second correlation calculation unit is smaller than a threshold value related to pseudo synchronization.

  Here, “correcting the frequency” includes shifting the center frequency of the carrier wave. According to this aspect, the position of the known signal inserted on the transmission side can be accurately captured by repeating predetermined processing until provisional synchronization is established. Further, the pseudo-synchronized state can be escaped by correcting the frequency of the carrier wave.

  The differential interval determination unit is located at a position separated from one known signal of a plurality of known signals included in the partial series to be associated with the known signal series by the same length as the predetermined period L. The arranged partial signals may be associated with each other. The differential interval determination unit separates one partial signal among a plurality of partial signals included in the partial series that should correspond to the known signal series, and the partial signal and the predetermined period L are different in length. The partial signals arranged at different positions may be associated with each other. According to this aspect, it is possible to determine not only anti-slip measures but also pseudo-synchronization by calculating differentials for different objects.

  The signal sequence input by the input unit may be modulated by a modulation scheme having a predetermined modulation multilevel number. Further, the different lengths in the second rule may be determined according to the predetermined period and the modulation multi-level number.

  A signal processing method according to another aspect of the present invention includes an input step, a demodulation step, an acquisition step, a determination step, and a detection step. The inputting step inputs a signal sequence in which a plurality of known signal sequences composed of a plurality of known signals are inserted at a predetermined period L. The demodulating step reproduces a carrier wave from the signal sequence input in the inputting step, and demodulates the signal sequence according to the reproduced carrier wave. In the acquiring step, a plurality of partial signal sequences to be associated with the known signal sequences included in the signal sequence demodulated in the demodulating step are acquired in the same cycle as the predetermined cycle. In the determining step, partial signals that should correspond to the known signals included in the predetermined partial signal sequences among the partial signal sequences acquired in the acquiring step are respectively handled using a plurality of rules regarding differential intervals. Put it on. The detecting step calculates the position where the sequence of known signals is inserted by calculating the relative phase difference between the partial signals associated according to the first rule among the plurality of rules regarding the differential interval. Estimate and detect temporary synchronization. The detecting step calculates a relative phase difference between the partial signals associated with each other according to the second rule having a differential interval different from the first rule among the plurality of rules regarding the differential interval. Thus, it is detected whether or not the frequency of the carrier wave reproduced by the demodulator is correct.

  According to this aspect, it is possible to reduce the processing time for determining the pseudo-synchronized state by establishing synchronization in two stages using two different relative phase differences. Also, synchronization can be efficiently detected by first establishing coarse synchronization and then establishing strict synchronization. Also, synchronization can be detected more efficiently by using different rules regarding the differential interval for synchronization determination at each stage.

  It should be noted that any combination of the above-described constituent elements and a conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, etc. are also effective as an aspect of the present invention.

  According to the present invention, it is possible to reduce the processing time required to escape from the pseudo-synchronized state.

  Before describing the present invention in detail, an outline will be described. Embodiments described herein relate generally to a terminal device that receives a low CNR (Carrier to Noise power Ratio) radio signal. In general, in a terminal device, when receiving a signal modulated by the M-phase PSK method, if a frequency offset occurs more than ± 1 / 2M of a symbol rate, normal synchronization is not secured. There may be a so-called pseudo-synchronization state where the synchronization state is fixed at a position shifted from the normal synchronization by ± 2π / M × symbol rate / 2π. Furthermore, there may be a phase slip state in which the reference phase in the carrier recovery loop is unnecessarily shifted due to the influence of the fading phenomenon and the Doppler phenomenon. In such a case, it is difficult to identify a known signal and acquire accurate synchronization.

  Assuming the case as described above, the terminal apparatus according to the present embodiment executes the synchronization process in two stages in order to capture the synchronization of the received radio signal. First, as a first step, the terminal apparatus according to the present invention uses the relative phase difference between known signals to remove the influence of phase slip and identify the position where the known signal is inserted in the radio signal sequence. . That is, the timing synchronization of the known signal is captured. Next, as a second stage, a relative phase difference between known signals different from the first stage is used to determine whether the captured timing synchronization is normal synchronization or pseudo synchronization. Here, the pseudo-synchronization includes that the frequency of the carrier wave reproduced by the terminal device is not correct, that is, the frequency synchronization cannot be normally captured. In the present embodiment, in order to simplify the description, the radio signal modulation method is assumed to be BPSK (Binary Phase Shift Keying) as an example. Further, it is assumed that all known signals included in the radio signal sequence are composed of the same symbols. Details will be described later.

  FIG. 1 is a diagram illustrating a configuration example of a terminal device 200 according to an embodiment of the present invention. The terminal device 200 includes an antenna 1, an RF unit 2, an orthogonal detection unit 3, a signal processing unit 100, and a decoding unit 6. The antenna 1 receives a signal sequence transmitted from a partner that is performing communication. The RF unit 2 performs processing for converting a signal sequence received by the antenna 1 from an RF band to an intermediate frequency band signal using a filter or the like. The quadrature detection unit 3 performs frequency conversion processing from the intermediate frequency band to the baseband band on the signal sequence processed by the RF unit 2. As will be described in detail later, the signal processing unit 100 performs analog / digital conversion, synchronization processing, and the like on the signal sequence processed by the quadrature detection unit 3, and sends the signal sequence in which synchronization is captured to the decoding unit 6. Output.

  The decoding unit 6 performs a decoding process corresponding to the error correction encoding process executed on the transmission side on a sequence excluding a sequence of known signals among the signal sequences whose synchronization is captured by the signal processing unit 100. The For example, when an interleaving process for rearranging the order of signal sequences is performed on the transmission signal sequence on the transmission side, the decoding unit 6 executes a rearrangement process opposite to the interleaving process for the received signal sequence. The deinterleaving process is executed. On the transmission side, when convolutional coding is performed on the signal sequence, the decoding unit 6 performs Viterbi decoding on the received signal sequence to estimate the transmitted signal.

  FIG. 2 is a diagram illustrating a configuration example of the signal processing unit 100 of FIG. The signal processing unit 100 includes a demodulation unit 20, a partial signal acquisition unit 22, a differential interval determination unit 25, a synchronization detection unit 70, and a buffer 40.

  Here, the signal series 50 received by the signal processing unit 100 is configured as shown in FIG. FIG. 3 is a diagram illustrating a configuration example of a signal sequence received by the signal processing unit 100 in FIG. The horizontal axis represents time. This signal series includes a series 52 of known signals and a data signal 54. The known signal series 52 is inserted into the signal series for each period L. Each known signal series 52 includes M known signals 53. M is a positive integer. M is the same value as the number of phases in M-phase PSK applied to the signal sequence. The reason for setting the same value is that there are as many pseudo-synchronization modes as the number of phases, so that the number of rules is required to determine pseudo-synchronization in order to determine pseudo-synchronization. Details will be described later. In this embodiment, since the modulation method is BPSK, M = 2. Note that the number of symbols included in the known signal series 52 and the data signal 54 is L. A signal sequence composed of an integral multiple of L is expressed as a frame.

  In addition, the known signal sequence 52 received by the signal processing unit 100 is generally a sequence of known signals transmitted due to continuous errors occurring in the transmission path or thermal noise generated in the RF unit 2 or the like. Is different. Therefore, in the following description, the “known signal series 52” will be described as “partial signal series corresponding to the known signal series 52” or simply “partial signal series”. Further, “known signal 53” will be described as “partial signal corresponding to known signal 53” or simply “partial signal”.

  Returning to FIG. The demodulator 20 reproduces a carrier wave from the signal sequence 50 and demodulates the signal sequence 50 according to the reproduced carrier wave. Specifically, synchronous detection using a Costas loop or the like is performed on the signal sequence 50 input from the quadrature detection unit 3 to reproduce the carrier wave. Further, processing such as band limitation processing, automatic gain control processing, DC offset correction, phase estimation, and the like is executed on the signal sequence 50 and output to the partial signal acquisition unit 22.

  Here, with reference to FIGS. 4A to 4C, a symbol point arrangement of a signal sequence subjected to BPSK modulation after reproducing a carrier wave will be described. 4A to 4C are diagrams showing examples of signal point arrangement of the signal sequence output from the demodulator 20 of FIG. FIG. 4A is a diagram showing a signal point arrangement when ideal carrier wave reproduction is performed. FIG. 4B is a diagram showing the signal point arrangement when the phase error θ resulting from the frequency offset remains. FIG. 4C is a diagram showing the signal point arrangement when the signal points A and B are shifted from each other by π. In other words, it shows that the phase error θ due to the frequency offset remains in the carrier wave reproduced by an integral multiple of ± π.

  Even in the case of the signal point arrangement as shown in FIG. 4B, it is possible to correct the state as shown in FIG. 4A by using error correction decoding or the like. However, when the signal point arrangement is as shown in FIG. 4C, it is difficult for the demodulator 20 itself to recover. Further, a long time is required for correction by the decoding unit 6 as well. In addition, the state of FIG.4 (c) is generally called a phase slip or cycle slip. The phase slip is a phenomenon in which the reference phase is rotated in a short time, and in the case of BPSK modulation, it is a phenomenon in which the reference phase is rotated by π. In such a case, even if error correction decoding is used, correct decoding cannot be performed, and it is necessary to relieve by differential processing described later.

  Returning to FIG. The partial signal acquisition unit 22 acquires a plurality of partial signal sequences included in the signal sequence 50 demodulated by the demodulation unit 20 at the same cycle as the cycle. Specifically, a signal sequence is acquired over one frame and written into the buffer 40. At this time, the data signal is also written into the buffer 40 together with the partial signal series. The partial signal acquisition unit 22 writes all the signal sequences to the buffer 40 and then notifies the differential interval determination unit 25 that the writing has been completed.

  The buffer 40 is a memory space that holds a signal sequence written by the partial signal acquisition unit 22 or the like. The buffer 40 is also used as a temporary memory when the synchronization detector 70 determines the synchronization of the partial signal series. The buffer 40 may have a different start address for writing for each target to be written. Further, the buffer 40 may not be configured from a single memory.

  Here, the buffer 40 into which the partial signal acquisition unit 22 writes the signal series has a data structure as shown in FIG. FIG. 5 is a diagram showing an example of a data structure stored in the buffer 40 of FIG. The partial signal acquisition unit 22 writes the signal sequence from the upper left of the buffer 40 to the right when writing the signal series into the buffer 40. The partial signal acquisition unit 22 writes to the end of a certain row of the buffer 40 and then writes to the next row in the same manner. The decoding unit 6 reads a signal sequence other than the partial signal sequence corresponding to the known signal sequence from the buffer 40 after receiving notification that the normal synchronization has been acquired by the synchronization detection unit 70 described later. When reading the signal sequence from the buffer 40, the decoding unit 6 reads the signal sequence from the upper left of the buffer 40 downward. The partial signal acquisition unit 22 reads the next column in the same manner after reading a certain column of the buffer 40 to the end. By using the buffer 40 in this way, deinterleaving processing can be realized efficiently.

  By setting the size of one row of the buffer 40 to the same L as the insertion period of the known signal sequence in the signal sequence, as shown in FIG. Can be located. In FIG. 5, since it is assumed that frame synchronization is achieved, a partial signal sequence corresponding to a known signal sequence appears at the head of the column. When frame synchronization is not achieved, the partial signal sequence does not exist in the first column as shown in FIG. 5, but does not change in that it is located in the same column. Here, the frame synchronization includes, for example, determining the position of the sequence column of the partial signal in the synchronization detection unit 70.

  Here, in order to simplify the description, the configuration example of the buffer 40 in FIG. 5 has been described as being two-dimensionally arranged, but the actual buffer 40 may have a one-dimensional configuration. . Each row may be stored in a physically separated memory as long as each row is associated with a pointer or the like.

  Returning to FIG. The differential interval determination unit 25 relates partial signals that should correspond to the known signal 53 included in a predetermined sequence of partial signals among the partial signal sequences acquired by the partial signal acquisition unit 22 to a plurality of differential intervals. Use rules to associate them with each other. Specifically, when the relative phase difference between the partial signals is calculated in the synchronization detection unit 70 described later, the two partial signals to be calculated are associated with each other. This will be described using an example. Assume that two partial signals in a row of the buffer 40 are A0 and B0. Also assume that partial signals in other partial signal sequences are A1 and B1. Assume that the association executed at this time is not performed in the same partial signal series. The association is executed in a one-to-one relationship.

  If it does so, the result of having performed matching will become a relation as shown in Drawing 6 (a)-(b). FIG. 6 is a diagram illustrating an example of correspondence in the differential interval determination unit 25 of FIG. In FIG. 6A, A0 is associated with B0 and A1 is associated with B1. In FIG. 6B, they are A0 and B1, and A1 and B0. Here, the association as shown in FIG. 6A is expressed as “straight”. Further, the association as shown in FIG. 6B is expressed as “cross”. In other words, the straight is a rule meaning that “a partial signal existing in a signal sequence is associated with a partial signal separated by a period L”. In other words, the straight is a rule meaning that “a partial signal existing in the buffer 40 is associated with a partial signal in a row immediately below it”.

  The cross is a rule meaning that “a partial signal existing in the signal sequence is associated with a partial signal separated by a length different from the period L”. The cross in the case of BPSK is a rule meaning that “a partial signal existing in a signal sequence is associated with a partial signal separated by L + 1 or L−1”. In the case of FIGS. 6A to 6B, there is only one cross, but as the number of partial signals in one partial signal series increases, a plurality of crosses exist. . In the following, it is assumed that one of the straight and the cross is a “first rule”, and the “second rule” is a rule different from the “first rule”. That is, if the first rule is straight, the second rule is cross. Conversely, if the first rule is a cross, the second rule is straight. If there are a plurality of crosses, different crosses may be used for the first rule and the second rule.

  Returning to FIG. The synchronization detection unit 70 inserts the sequence 52 of known signals by calculating the relative phase difference between the partial signals associated according to the first rule among the plurality of rules regarding the differential intervals. The position is estimated and provisional synchronization is detected. Specifically, it is estimated in which column of the buffer 40 the partial signal sequence exists among the signal sequences written in the buffer 40 by the partial signal acquisition unit 22. The estimation calculates a relative phase difference between the partial signals in the preceding and following rows that are associated according to the first rule.

For example, assume that the phases corresponding to two partial signals in a row are A0 and B0, and the phases of the partial signals in the next row of the row are A1 and B1. Also assume that the first rule is a straight relationship. That is, it is assumed that A0 and A1 and B0 and B1 are associated with each other. Then, in the first differential calculation unit 30 and the second differential calculation unit 72 described later, the phase difference is calculated as follows. Note that | X | on the right side represents a function for calculating the absolute value of X.
Phase difference between A0 and A1 = | A0−A1 | Equation (1)
Phase difference between B0 and B1 = | B0−B1 | (2)

As described above, there are M partial signals in each row. Here, if the number of symbols included in the signal sequence is N, the number of calculated phase differences is as follows.
Number of calculated phase differences = M × (N / L-1-α) Equation (3)
Here, α is a value determined by the signal strength of the signal series, and will be described later in detail. The larger the signal strength, the larger the value. The value in parentheses on the right side is a value corresponding to the number of lines. In the case of the above example, since the phase difference between the two rows is obtained, the number of calculated phase differences is 2 × (2-1) = 2.

  Further, the synchronization detection unit 70 similarly calculates the relative phase difference of the known signal that is the ideal value of the partial signal. Furthermore, the correlation value of the phase difference between the two is obtained and summed. Finally, the timing synchronization of the partial signal series is estimated by comparing the sum and the threshold value. Details will be described later. In addition, the synchronization detection unit 70 determines a relative phase difference between partial signals associated according to a second rule, that is, a differential interval different from the first rule among a plurality of rules related to the differential interval, that is, according to a cross. Respectively, it is detected whether or not the frequency of the carrier wave reproduced by the demodulator 20 is correct. Since the method is the same as the calculation of the phase difference based on the first rule except that the second rule is used instead of the first rule, the description is omitted.

  FIG. 7 is a diagram illustrating a configuration example of the partial signal acquisition unit 22 of FIG. The partial signal acquisition unit 22 includes a writing unit 62 and a reliability acquisition unit 64. The writing unit 62 writes the signal series demodulated by the demodulating unit 20 into the buffer 40.

  The reliability acquisition unit 64 acquires the reliability of the partial signal sequence corresponding to the known signal sequence included in the signal sequence demodulated by the demodulation unit 20, and sequentially writes it in the buffer 40. Here, the reliability includes a value indicating the probability of the partial signal sequence, and includes, for example, a soft decision value of the partial signal sequence, a value indicating a correlation between the partial signal sequence and the known signal sequence, and the like. Further, it may be a value indicating the likelihood of the phase of the partial signal series. When one partial signal sequence is composed of a plurality of symbols, the reliability of the partial signal sequence may be acquired by adding the reliability of a plurality of symbols included in the partial signal sequence.

  FIG. 8 is a diagram illustrating a configuration example of the synchronization detection unit 70 of FIG. The synchronization detection unit 70 includes a first differential calculation unit 30, a first correlation calculation unit 32, a temporary synchronization determination unit 24, a measurement unit 38, a second differential calculation unit 72, and a second correlation calculation unit 74. A pseudo-synchronization determination unit 26, a control unit 42, and a frequency correction unit 28. Note that each configuration in FIG. 8 is connected to the above-described buffer 40 by a signal bus (not shown).

The first differential calculation unit 30 calculates a relative phase difference between the partial signals associated according to the first rule among the rules regarding the plurality of differential intervals. The phase θ of each partial signal used for calculating the phase difference is calculated as follows using the I component and Q component of the symbol included in the partial signal series.
θ = tan ⁻¹ (Q / I) (4)

  The first correlation calculation unit 32 compares the phase difference between the partial signals calculated by the first differential calculation unit 30 and the relative position between the known signals 53 to be associated with the phase difference between the partial signals. A first correlation value with the phase difference is calculated. Specifically, the phase difference between the partial signals existing in the preceding and succeeding rows and between the known signals is calculated by the number indicated in the parentheses on the right side of Expression (3) in the buffer 40. Further, the correlation value is obtained by adding the calculated phase differences in all rows. When the partial signal series corresponding to the known signal series are all composed of the same symbol “0”, the calculated phase θ becomes the correlation value as it is.

  When the first correlation value calculated by the first correlation calculation unit 32 is larger than the threshold value related to temporary synchronization, the temporary synchronization determination unit 24 sets the position where the partial signal exists as the position where the series 52 of known signals is inserted. Judge as. That is, it determines with it being a temporary-synchronization state. Temporary synchronization corresponds to a state where at least a position in a signal sequence into which a known signal is inserted can be determined, although the synchronization state may be pseudo-synchronization.

  When the first correlation value calculated by the first correlation calculation unit 32 is smaller than the threshold value related to temporary synchronization, the control unit 42 targets a partial signal sequence different from the partial signal sequence in the signal sequence, The first differential calculation unit 30 again calculates the differential. Here, the “different partial signal series” includes other columns in the buffer 40. In other words, if the sequence determined by the provisional synchronization determination unit 24 is not a partial signal sequence that should correspond to a known signal sequence, the other column is a portion that should correspond to a known signal sequence. It is determined whether or not the signal series. The “different partial signal sequence” in the case where the partial signal sequence that should correspond to the known signal sequence is composed of a plurality of symbols is a part of the already determined partial signal sequence and the target of the determination. Signal that was not. For example, when it is determined that the first column and the second column are not partial signal sequences, the columns to be determined next are the second column and the third column. At this time, the second and subsequent processes can be shortened by holding the differential result for the second column in which the determination is made first.

  The measuring unit 38 measures the signal intensity of the signal series. In accordance with the signal intensity measured by the measurement unit 38, the range of correlation values calculated by the first correlation calculation unit 32 and the second correlation calculation unit 74 is determined. This is because when the signal strength is high, the reliability of each partial signal is considered to be high, and therefore the range for calculating the correlation value may be narrowed. That is, α on the right side of Equation (3) is a large value. Thereby, the time of correlation processing can be shortened.

  The second differential calculation unit 72 calculates a relative phase difference between the partial signals associated according to the second rule among the rules regarding the plurality of differential intervals. Since it is the same as that of the 1st differential calculation part 30 except the point which uses a 2nd rule instead of a 1st rule, description is abbreviate | omitted here.

  The second correlation calculation unit 74 calculates the phase difference between the partial signals calculated by the second differential calculation unit 72 and the relative phase difference between the known signals that should correspond to the phase difference between the partial signals. The second correlation value is calculated. Since the process here is the same as that of the 1st correlation calculation part 32, description is abbreviate | omitted.

  The pseudo synchronization determination unit 26 determines that the frequency of the carrier wave reproduced by the demodulation unit 20 is correct normal synchronization when the second correlation value calculated by the second correlation calculation unit 74 is larger than a threshold value related to pseudo synchronization. Then, it notifies the decoding unit 6 (not shown) that normal synchronization has been established. When the second correlation value is smaller than the threshold value related to pseudo synchronization, it is determined as pseudo synchronization, and the fact and the frequency shift amount are notified to the frequency correction unit 28.

  The frequency correction unit 28 shifts the frequency of the carrier wave when the second correlation value calculated by the second correlation calculation unit 74 is larger than the threshold value related to pseudo synchronization according to the notification of the pseudo synchronization determination unit 26. The frequency of the carrier wave is corrected. Specifically, a frequency signal corresponding to the amount of frequency to be corrected is generated and output to the demodulator 20. The demodulator 20 corrects the frequency of the carrier wave by inputting the frequency signal generated by the frequency corrector 28 to one of frequency converters such as a mixer (not shown).

  Here, the principle that the phase synchronization can be avoided and the pseudo-synchronization can be determined by obtaining the relative phase difference between the partial signals associated in the differential interval determination unit 25 will be described.

First, a QPSK reception signal r (t) including BPSK when a (t) and b (t) are transmission information signals and ω _c is a transmission carrier angular frequency is assumed as follows.
r (t) = a (t) cos (ω _c t) −b (t) sin (ω _c t) (5)

Here, when the signal shown in Equation (5) is received and converted into a baseband signal, the in-phase component I (t) and the quadrature component Q (t) are as shown in Equation (6).
I (t) = a (t) cos (Δωt + θ ₀ ) − b (t) sin (Δωt + θ ₀ )
Q (t) = b (t) cos (Δωt + θ ₀ ) + a (t) sin (Δωt + θ ₀ ) (6)

Here, Δω represents an error between the transmission carrier angular frequency ω _c and the reception local angular frequency ω ₁ . Θ ₀ indicates the initial phase difference between the two. The baseband signals I (t) and Q (t) are then demodulated. Further, each demodulation result is output for each symbol period (Ts). Here, as a demodulation process, a process in which both the frequency error Δω and the initial phase difference θ ₀ are zero with respect to the baseband signals I (t) and Q (t) by a technique such as a Costas loop, for example. Is executed. This demodulation processing is obtained on the phase plane as shown in FIG. 4A when ideally operated, that is, when Δω = 0 and θ ₀ = 0. Further, when only the phase error remains at the frequency error Δω = 0, that is, when θ ₀ ≠ 0, or when the phase error θ ₀ converges to zero but the frequency error Δω is Δω ≠ 0, FIG. As shown in b) and FIG. The demodulation result is generally a soft decision value for each of the in-phase component I (t) and the quadrature component Q (t), but I (t) and Q (t ) May be phase information. For example, in the case of BPSK, they are 0 and π.

The problem targeted in the embodiment of the present invention is the situation shown in FIG. Even if θ ₀ ≠ 0, the effect is a phase offset common to all demodulated outputs, and if the difference is applied, the effect becomes invisible. Therefore, there is no problem even if θ ₀ = 0 thereafter. Therefore, equation (6) can be changed to equation (7).
I (t) = a (t) cosΔωt − b (t) sinΔωt
Q (t) = b (t) cosΔωt + a (t) sinΔωt Equation (7)

Since the demodulated output by equation (7) is obtained at symbol period Ts intervals, equation (7) can be changed to equation (8). Here, k is an integer.
I (kTs) = a (kTs) cosΔωkTs−b (kTs) sinΔωkTs
Q (kTs) = b (kTs) cosΔωkTs + a (kTs) sinΔωkTs (8)

Further, Expression (8) is obtained by treating Expression I (kTs) and Q (kTs) as complex real numbers and imaginary numbers, respectively.
I (kTs) −jQ (kTs) = {a (kTs) −jb (kTs)} Exp (−jΔωkTs)
... Formula (9)

  In Equation (9), {a (kTs) −jb (kTs)} is a known signal in the transmission information signal, and corresponds to a partial signal when performing correlation processing with the demodulated output. Even in the case of differential correlation, if a differential interval is prepared as a reference signal, the same applies. Therefore, the problem in the correlation is the part of Exp (−jΔωkTs), that is, the phase due to the frequency error Δω and the time kTs. When performing differential processing, pay attention to the phase change due to (k + Δk). It ’s fine. Here, Δk is a differential interval.

In the case of BPSK, the phase change Δφ due to the differential interval of Δk is expressed by Equation (10).
Δφ = 0 (Δω = 0; normal synchronization)
Δk · π (Δω = 2π · fs / 2; pseudo-synchronization shifted by fs / 2)
Δk · −π (Δω = 2π · -fs / 2; pseudo-synchronization shifted by -fs / 2)
Δk · π (Δω = 2π · -fs / 2; pseudo-synchronization shifted by -fs / 2)
... Formula (10)
Here, fs is a symbol rate, and an assumed range of a frequency error in which pseudo-synchronization occurs is an LPF (Low Pass Filter) (not shown) after the quadrature detection unit 3 in FIG. 1, a route roll-off filter applied before demodulation processing, or the like In the case of a bandwidth of −fs / 2, it is sufficient to set the range from −fs / 2 to fs / 2.

From the equation (10), in the case of BPSK, the phase of the correlation output becomes 0 regardless of the differential interval during normal synchronization. In the case of pseudo synchronization, it becomes 0 or π according to the differential interval Δk. Using this property, the combinations in Table 1 are obtained.

  If only two known differential signals are executed for only known signals, the phase interval is 0 during normal synchronization and either is π during pseudo synchronization. That is, since mutually inverted phases are obtained, it is possible to determine whether the state is normal synchronization or pseudo synchronization by comparing the two. Further, when Δk is an even number and an odd number, when using absolute values of both correlation values, for example, 0 or π, in addition to the direction and interval for taking the differential, the size of one line of the interleave that is a known signal Whether it is even or odd also affects the correlation output phase. In other words, the correlation output phase affects the setting of whether Δk is even or odd. Therefore, it is necessary to determine the determination method in consideration of the size of the interleaving, that is, the period in which the sequence of known signals is inserted into the signal sequence.

  Note that a soft decision value is generally applied as a demodulated output. If the soft decision value is converted into binarization of (1, -1), that is, a hard decision is made, the result of Table 1 can be read as a positive or negative phase state such as 0 or π. Further, if the demodulated output phase (0 and π) is used separately from the soft decision value, the relationship in Table 1 can be applied as it is.

Here, an example will be described. It is assumed that the size L of one row of the buffer 40 is a multiple of “the number of modulation levels per symbol of the signal sequence (2 for BPSK, 4 for QPSK)”. Note that the number of known signals included in one known signal sequence is equal to the “number of modulation levels per symbol of the signal sequence”, and is assumed to be 2 here. Further, it is assumed that the partial signal included in the partial signal series is composed of two symbols A and B. Also assume that all known signals are zero. Then, when there is no noise, the phase of the partial signal sequence stored in the buffer 40 in the case of normal synchronization can be expressed as in Expression (11-1). Similarly, when there is no noise, the phase of the partial signal sequence stored in the buffer 40 in the case of pseudo-synchronization can be expressed as Equation (11-2). In addition, “A; B; C” in parentheses indicate that the partial signal sequences A, B, and C are arranged in different continuous rows. It was assumed that the partial signal included in the partial signal series represented by Equation (11-1) and Equation (11-2) is composed of two symbols A and B. Also assume that all known signals are zero.
{0, 0; 0, 0; 0, 0} Expression (11-1)
{0, π; 0, π; 0, π} Formula (11-2)

Furthermore, it is assumed that the relationship associated with the differential interval determination unit 25 is as shown in FIGS. Based on the above assumptions, when the differential between the straight signal and the cross is calculated for the partial signals represented by Expression (11-1), Expressions (12-1S) and (12-1C) are obtained. . Further, when the differential is calculated with respect to the straight and the cross between the partial signals represented by Expression (11-2), Expression (12-2S) and Expression (12-2C) are obtained.
{0, 0; 0, 0}: In the case of straight: Formula (12-1S)
{0, 0; 0, 0}: In the case of a cross ... Formula (12-1C)
{0, 0; 0, 0}: In the case of straight: Formula (12-2S)
{Π, π; π, π}: In the case of a cross: Formula (12-2C)

Here, in the equations (12-1S), (12-1C), (12-2S), and (12-2C), the calculated differential results are summed as follows.
The result of adding the expression (12-1S) = 0 = expression (13-1S)
The result of adding the formula (12-1C) = 0 = formula (13-1C)
Result of summing up formula (12-2S) = 0 = formula (13-2S)
Result of summing up formula (12-2C) = π... Formula (13-2C)

  According to the above example, in the case of Expression (11-1), that is, in the case of normal synchronization, both straight and cross become “0”. In the case of Expression (11-2), that is, in the case of pseudo synchronization, the straight is “0”, but the cross is “π”. Therefore, it is possible to determine the provisional synchronization state by taking the differential with respect to the straight, and further to determine whether the state is the pseudo-synchronization state by taking the differential with respect to the cross.

Here, an example in which it is assumed that the size L of one row of the buffer 40 is not a multiple of the number of known signals included in the known signal series will be described. Then, when there is no noise, the phase of the partial signal sequence stored in the buffer 40 in the case of normal synchronization can be expressed as Expression (14-1). Similarly, the phase of the partial signal sequence stored in the buffer 40 in the case of pseudo-synchronization can be expressed as shown in Expression (14-2) when there is no noise.
{0, 0; 0, 0; 0, 0} Expression (14-1)
{0, π; π, 0; 0, π} Formula (14-2)

When the differential is calculated for the straight and the cross between the partial signals represented by Expression (14-1), Expression (15-1S) and Expression (15-1C) are obtained. When the differential between the straight signal and the cross is calculated with respect to the partial signal represented by Expression (14-2), Expression (15-2S) and Expression (15-2C) are obtained.
{0, 0; 0, 0}: In the case of straight: Formula (15-1S)
{0, 0; 0, 0}: In the case of a cross ... Formula (15-1C)
{Π, π; π, π}: In the case of straight: Formula (15-2S)
{0, 0; 0, 0}: In the case of a cross: Formula (15-2C)

Furthermore, when the differential results calculated in the formula (15-1S), formula (15-1C), formula (15-2S), and formula (15-2C) are added together, the result is as follows.
The result of adding the formula (15-1S) = 0 = formula (16-1S)
The result of adding the formula (15-1C) = 0 = formula (16-1C)
Result of summing up formula (15-2S) = π... Formula (16-2S)
The result of adding the formula (15-2C) = 0 = formula (16-2C)

  According to the above example, in the case of Expression (14-1), that is, in the case of normal synchronization, both straight and cross are “0”. In the case of the expression (14-2), that is, in the case of pseudo synchronization, the straight is “π”, but the cross is “0”. Therefore, it is possible to determine the provisional synchronization state by taking the differential with respect to the cross, and further to determine whether the state is the pseudo-synchronization state by taking the differential with respect to the straight.

  The principle of two examples has been described above. What can be derived from these examples is that first, provisional synchronization is established by calculating a differential in correspondence with one of straight and cross. Furthermore, pseudo-synchronization is determined by calculating a differential with a correspondence relationship other than the correspondence relationship for which provisional synchronization has been established. At this time, the correspondence used for calculating the differential is determined by the period L in which the sequence of known signals is inserted and the number of modulation levels per symbol of the signal sequence.

  Here, the operation of the signal processing unit 100 will be described with reference to FIG. FIG. 9 is a flowchart showing an operation example of the signal processing unit 100 of FIG. First, the demodulation unit 20 performs demodulation processing such as carrier wave reproduction processing on the signal sequence sent from the quadrature detection unit 3 (S10). Next, the partial signal acquisition unit 22 writes the signal series in the buffer 40 (S12). Next, the differential interval determination unit 25 performs association with the partial signal acquired by the partial signal acquisition unit 22 using the first rule and the second rule regarding the differential interval (S13). The synchronization detection unit 70 calculates a differential for a certain partial signal in the buffer 40 (S14). The synchronization detection unit 70 similarly calculates the differential for the corresponding known signal. Further, the synchronization detection unit 70 calculates the calculated correlation between the two differentials (S16). Next, the synchronization detector 70 adds the correlation results by the number of columns for which the differential is calculated. Next, the synchronization detection unit 70 compares the correlation result with the threshold value related to temporary synchronization, and determines whether or not the sequence for which the differential is calculated is a partial signal sequence corresponding to a known signal sequence, that is, frame synchronization. Is determined (S20).

  As a result of the determination, if the threshold value is not exceeded (N in S20), the synchronization detection unit 70 shifts the column in the buffer 40 (S22), associates again (S13), and performs differential calculation 1 (S14). ) Etc. are repeated. As a result of the determination, when the threshold value is exceeded (Y in S20), it can be regarded as temporary synchronization. In this case, the synchronization detection unit 70 calculates the differential between the partial signals associated with the differential interval different from S14 (S23). Further, the synchronization detection unit 70 calculates the differential at the same differential interval for the corresponding known signal. Further, the synchronization detection unit 70 calculates the correlation between the differential result between the partial signals and the differential result between the known signals, similarly to S16 (S25). Next, when the correlation result is larger than the threshold value related to pseudo synchronization, the synchronization detection unit 70 determines that the frequency of the carrier wave reproduced by the demodulation unit 20 is correct normal synchronization (Y in S26). If the second correlation value is smaller than the threshold value related to pseudo-synchronization, it is determined to be pseudo-synchronization (N in S26), and frequency correction is executed using the frequency shift amount determined together. Although not shown in FIG. 9, after normal synchronization is established, the decoding unit 6 performs decoding processing such as interleaving processing and Viterbi decoding processing.

  According to the present embodiment, it is possible to reduce the determination processing time of the pseudo-synchronized state by establishing synchronization in two stages using two different relative phase differences. Also, synchronization can be efficiently detected by first establishing coarse synchronization and then establishing strict synchronization. Also, synchronization can be detected more efficiently by using different rules regarding the differential interval for synchronization determination at each stage. Also, by calculating the phase differential between the partial signals, synchronization can be accurately captured even when a phase slip occurs. Further, the synchronization can be captured in a short time by limiting the range in which the correlation value is calculated based on the signal strength. In addition, by repeating the predetermined process until provisional synchronization is established, the position of the known signal inserted on the transmission side can be accurately captured. Further, the pseudo-synchronized state can be escaped by correcting the frequency of the carrier wave. In addition, by calculating the differential for different objects, not only anti-slip measures but also pseudo-synchronization can be determined.

  Next, a modification of the embodiment of the present invention will be shown. The modification of the Example of this invention is related with the terminal device 200 containing the signal processing part 100 which determines the synchronization of the radio signal of a low CNR. The difference from the embodiment of the present invention is that the signal sequence received by the terminal device 200 is not BPSK modulated but QPSK (Quadrature Phase Shift Keying) modulated. Since the configuration is the same as that of the above-described embodiment, the same reference numerals are used to simplify the description. The terminal device 200 according to the modification of the embodiment of the present invention includes the antenna 1, the RF unit 2, the quadrature detection unit 3, the signal processing unit 100, and the decoding unit 6, similarly to the configuration shown in FIG. Including. Similarly to the configuration shown in FIG. 2, the signal processing unit 100 includes a demodulation unit 20, a partial signal acquisition unit 22, a differential interval determination unit 25, a synchronization detection unit 70, and a buffer 40. Similarly to the configuration shown in FIG. 7, the partial signal acquisition unit 22 includes a writing unit 62 and a reliability acquisition unit 64. Similarly to the configuration shown in FIG. 8, the synchronization detection unit 70 includes a first differential calculation unit 30, a first correlation calculation unit 32, a temporary synchronization determination unit 24, a control unit 42, and a measurement unit 38. The second differential calculation unit 72, the second correlation calculation unit 74, the pseudo synchronization determination unit 26, and the frequency correction unit 28 are included.

  Here, in the case of the QPSK modulation method, the principle that the phase difference can be avoided and the pseudo-synchronization can be determined by obtaining the relative phase difference between the partial signals associated in the differential interval determination unit 25 will be described. To do.

In the case of QPSK, the phase change Δφ due to the differential interval of Δk is expressed by Equation (17).
Δφ = 0 (Δω = 0; normal synchronization)
Δk · π / 2 (Δω = 2π · fs / 4; pseudo-synchronization shifted by fs / 4)
Δk ・ −π / 2 (Δω = 2π ・ -fs / 4; pseudo-synchronization shifted by -fs / 4)
Δk · π (Δω = 2π · fs / 2; pseudo-synchronization shifted by -fs / 2)
Δk · −π (Δω = 2π · -fs / 2; pseudo-synchronization shifted by -fs / 2)
Δk · π (Δω = 2π · -fs / 2; pseudo-synchronization shifted by -fs / 2)
... Formula (17)

  Here, in the case of normal synchronization, it is always 0 regardless of Δk, whereas in pseudo synchronization shifted by fs / 4, Δk = 4, 8, 12,. Further, even with pseudo-synchronization shifted by −fs / 4, Δk = 4, 8, 12,. Further, in pseudo synchronization shifted by ± fs / 2, Δk = 2, 4, 6, 8, 10,. From the above, even with normal synchronization and any pseudo-synchronization, 0 (rad) is obtained when Δk = 4, 8, 12,. This pattern is referred to as pattern A. Further, when Δk = 2, 6, 10,..., 0 (rad) is obtained in normal synchronization and pseudo synchronization shifted by ± fs / 2. This pattern is referred to as pattern B. Further, for any other Δk (odd number), one of π / 2, −π / 2, and π is calculated. This pattern is referred to as pattern C. Since Δk = 0 is an autocorrelation, it is excluded here.

In other words, the above patterns A, B, C are
Pattern A: Differential interval with 2nπ = 0 (rad) for normal synchronization and any pseudo-synchronization Pattern B: Differential interval with 2nπ = 0 (rad) for normal synchronization and some pseudo-synchronization Pattern C: 2nπ for other than normal synchronization It can also be said that the differential interval does not become = 0 (rad). Accordingly, it is possible to determine whether the synchronization is normal synchronization or pseudo-synchronization by comparing “differential correlation due to Δk as pattern A” and “differential correlation due to Δk as pattern C”. This is referred to as Comparative Method 1. Alternatively, it is possible to determine whether the synchronization is normal synchronization or pseudo-synchronization by comparing “differential correlation by Δk as pattern A” and “combination of two or three types of differential correlation other than pattern A”. This is referred to as Comparative Method 2.

  Here, with reference to FIGS. 10A to 10D, a rule in which the differential interval determination unit 25 associates the eight partial signals with each other will be described. FIGS. 10A to 10D are diagrams showing an example of correspondence in the differential interval determination unit 25 according to the modification of the embodiment of the present invention. FIG. 10A is a diagram showing an association pattern related to a straight in the differential interval determination unit 25 of FIG. FIGS. 10B to 10D are diagrams showing examples of association patterns related to crosses in the differential interval determination unit 25 of FIG. Note that the number of known signals included in one known signal series is equal to the “number of modulation levels per symbol of a signal series”, and is assumed to be four here. Unlike FIG. 6 shown in the case of BPSK, it is characterized in that there are many patterns for crosses.

  Here, the rules in the differential interval determination unit 25 in FIG. 2 will be described with reference to FIGS. FIGS. 11A to 11D are diagrams illustrating examples of differential results according to a modification of the embodiment of the present invention. FIG. 11A is an example of a correlation value when the length L per row of the buffer 40 is a multiple of “the number of modulation multi-values per symbol of a signal sequence”. In this example, a total of 16 correlation values are shown for the combination of the case of either normal synchronization or three pseudo-synchronizations and the case of the correspondence being either straight or three crosses. Note that these examples correspond to the differential result formula (16-1S) shown in BPSK. Cross 1, cross 2, and cross 3 refer to the association patterns shown in FIGS. 10 (b) to 10 (d), respectively. Further, a pattern other than the illustrated patterns A and B is referred to as a pattern C.

  As shown in FIG. 11A, temporary synchronization can be determined using the correlation value of straight (Δk = 4, 8, 12,...). Further, by comparing the correlation values of any one of the patterns B (cross 1, cross 3) (comparison method 1), it can be determined whether the synchronization is normal synchronization or pseudo synchronization. At this time, the polarity of the frequency error can also be grasped only in the pseudo-synchronization of ± fs / 4, that is, ± π / 2. In such a case, the pseudo-synchronization determination unit 26 notifies the frequency correction unit 28 of information about the polarity.

  In the case of the comparison method 2, the correlation outputs of crosses 1 and 2 and 3 are converted into absolute values, and then the phase components are added in the state of being decomposed into sin and cos. Finally, the phase is again calculated from the addition result sin and cos. It may be calculated (if the absolute value is not taken, there may be a phase 0 at the time of synthesis). For example, when quasi-synchronized at π / 2, the phase of π / 2, π, | −π / 2 | is decomposed into sin and cos as a combination of patterns B and C, When cos is added and finally the phase is calculated, the phase result asymptotically approaches tan-1 (2/1) ≈116 (deg) ≈2π / 3 (rad).

  FIGS. 11B to 11D are examples of correlation values in each case where the length L per row of the buffer 40 is not a multiple of “the number of modulation multi-levels per symbol of a signal sequence”. Sixteen correlation values are shown for the case of synchronization or pseudo-synchronization and the case of correspondence being straight or cross.

  As shown in FIG. 11B, when the interleaved line size is mod (N, 4) = 1, that is, the length L per line of the buffer 40 is expressed as “modulation multivalue per symbol of the signal sequence. When the remainder is calculated by “number”, the temporary synchronization can be determined using the correlation value of the pattern A (cross 3). Further, by comparing the correlation value of any one of the patterns B (straight, cross 2) (comparison method 1), it can be determined whether the synchronization is normal synchronization or pseudo-synchronization. At this time, the polarity of the frequency error can also be grasped only in the pseudo-synchronization of ± π / 2.

  As shown in FIG. 11C, when the length L per line of the buffer 40 is 2 when the remainder is calculated by “the number of modulation multi-values per symbol of the signal sequence”, the pattern A (cross 2 ) Can be used to determine provisional synchronization. Further, by comparing the correlation values of any one of the patterns B (cross 1, cross 3) (comparison method 1), it can be determined whether the synchronization is normal synchronization or pseudo synchronization. At this time, the polarity of the frequency error can also be grasped only in the pseudo-synchronization of ± π / 2.

  As shown in FIG. 11D, when the length L per line of the buffer 40 is 3 when the remainder is calculated by “the number of modulation multi-values per symbol of the signal sequence”, the pattern A (cross 1 ) Can be used to determine provisional synchronization. Further, by comparing the correlation value of any one of the patterns B (straight, cross 2) (comparison method 1), it can be determined whether the synchronization is normal synchronization or pseudo synchronization. At this time, the polarity of the frequency error can also be grasped only in the pseudo-synchronization of ± π / 2.

  In the case of the comparison method 2, the concept is the same as that in the case of FIG. 11A in any of the cases shown in FIGS. When differential processing is actually performed, the demodulated output for each symbol may be converted into four phases (0, ± π / 2, π).

  The principle has been described above for the case of QPSK. As can be derived from these examples, even in the case of QPSK, as in the case of BPSK, temporary synchronization can be established first by calculating the differential with either the straight or cross correspondence. Furthermore, pseudo-synchronization can be determined by calculating a differential with a correspondence relationship other than the correspondence relationship for which provisional synchronization is established. At this time, the correspondence used for calculating the differential is determined by the period L in which the sequence of known signals is inserted and the number of modulation levels per symbol of the signal sequence. Note that the modulation multi-level number per symbol of a signal sequence can be regarded as the number of known signals included in one known signal sequence.

  According to the modification of the present embodiment, the synchronization processing time can be reduced by establishing two stages of synchronization using two different relative phase differences. Further, by determining the correspondence relationship for calculating the correlation value based on the period L in which the sequence of the known signal is inserted and the number of modulation multi-values per symbol of the signal sequence, it is possible to simulate in any case. The synchronization status can be determined. As in the case of BPSK, the synchronization can be efficiently detected by first establishing the synchronization roughly and then establishing the strict synchronization. Also, synchronization can be detected more efficiently by using different rules regarding the differential interval for synchronization determination at each stage. The polarity of frequency error can also be grasped.

  In the above, this invention was demonstrated based on the Example. This embodiment is an exemplification, and it will be understood by those skilled in the art that various modifications can be made to the combination of each component and each processing process, and such modifications are also within the scope of the present invention. .

  In the present embodiment, it has been described that the signal sequence is written in the buffer 40 on one side. However, the present invention is not limited to this, and a two-surface buffer 40 may be used. Further, it has been described that the upper and lower sides written in the buffer 40 are executed in the differential processing. However, the present invention is not limited to this, and every second and third rows may be differentially executed.

Further, in the present embodiment and the modified example of the present embodiment, the signal sequence has been described as being BPSK or QPSK modulated. However, the present invention is not limited to this, and OQPSK (Offset QPSK), DBPSK (Differential BPSK), DQPSK (Differential QPSK), π / 4 shift QPSK, M-PSK, M-DPSK (Differential PSK), M-APSKde (Amplu PPS), It may be M-ADPSK (Amplitude DPSK) or 2 ^L -QAM (Quadrature Amplitude Modulation). In that case, the same effect as described above can be obtained by operating according to the following principle.

  In the case of OQPSK, at the time of demodulation, either I (t) or Q (T) is generally delayed by Ts / 2, and converted to normal QPSK and demodulated. At this time, pseudo-synchronization may occur as in normal QPSK, but it can be considered in the same manner as in the case of QPSK described above. However, since a Ts / 2 symbol delay is applied, there may occur a situation unique to OQPSK that does not result in a uniform correlation phase according to the correlation direction as in QPSK (at ± fs / 4). (If pseudo-synchronized). For example, when the interleave 1 row size is mod (N, 4) = 1 and fs / 4 is quasi-synchronized (the phase change is π / 2), the differential correlation in the straight direction with 1 row spacing is performed. It changes as 0 → 3π / 4, 3π / 4 → π, π → −π / 4, and −π / 4 → 0. However, if each phase change amount is decomposed and combined into sin and cos, the resultant value finally approaches π / 2. Therefore, compared with the case of QPSK, there is a problem that “the phase decomposition process increases” and “the accuracy of the correlation composite value is low”, but basically the same determination as in QPSK is possible.

  In the case of DBPSK or DQPSK, that is, when differential encoding is performed on the modulation side, the reception side converts to a baseband signal by a quasi-synchronous detection configuration as shown in FIG. By performing differential correlation on the demodulated output without decoding the normalization part, frame synchronization is established regardless of normal synchronization and pseudo-synchronization, and then the differential encoding part is not decoded. What is necessary is just to perform a correlation with the differential output of the above-mentioned space | interval of (DELTA) k with a state. Note that data portions other than the known signal sequence included in the signal sequence may be decoded at the normal synchronization stage. If the differential encoding portion is decoded, error propagation frequently occurs in pseudo synchronization, and frame synchronization cannot be achieved. Even in the determination of normal synchronization and pseudo-synchronization, if the differentially encoded portion is decoded, the above relationship cannot be obtained. In addition, the method of converting IF signals into baseband signals by the delay detection method can also be applied to differentially encoded signals as a demodulation method, but the differential encoding part is decoded at the stage of delay detection execution. Therefore, the present invention cannot be applied to such a circuit configuration.

  In the case of π / 4 shift QPSK, the shift operation portion of π / 4 radians performed on the modulation side is canceled for the in-phase component and the quadrature component of the detection output by the offset of the frequency component of 1 / (8Ts). After that, it can be demodulated by the same method as QPSK or DQPSK. That is, the same method as in the case of QPSK or DQPSK described above can be applied.

  In the case of M-PSK and M-DPSK, for example, even in the case of 8PSK or 16PSK, the generation mechanism of pseudo synchronization is the same. In addition, it is easily conceivable for those skilled in the art that the present method can be applied as an extension in the case of QPSK by grasping the correlation phase relationship by the differential interval Δk. Furthermore, even if differential encoding such as 8DPSK or 16DPSK is performed, the application method can be easily considered by the relevant trader in consideration of the above-mentioned DBPSK and DQPSK.

  In the case of M-APSK and M-ADPSK, the pseudo-synchronization generation mechanism is the same. In the case of 8APSK or 16APSK, as long as a known signal is arranged in symbols of the same radius, the application method can be easily considered from the above description. Even if symbols with different amplitudes are used, the output at the time of differential correlation is If phase information is used, an application method can be easily considered. Moreover, even if differential encoding is performed, the application method can be easily considered by a person skilled in the art by following the conventional method.

In the case of 2 ^L -QAM, pseudo-synchronization occurs even if a so-called multi-value quadrature amplitude modulation method such as 16QAM is applied as a carrier reproduction method similar to the PSK method. Even in this case, the same idea as QPSK can be applied if a known signal to be applied to frame synchronization is arranged in the outer symbol (four corners). In addition, even if it is arranged in symbols other than outer, considering the correspondence between the differential interval Δk and all the phase change patterns between one symbol generated in the pseudo-synchronization occurring within the assumed frequency error range, Since the same judgment standard as the above-mentioned QPSK can be obtained, even such a modulation scheme can be applied by those skilled in the art.

It is a figure which shows the structural example of the terminal device concerning the Example of this invention. It is a figure which shows the structural example of the signal processing part of FIG. It is a figure which shows the structural example of the signal series which the signal processing part of FIG. 2 receives. 4A to 4C are diagrams showing examples of signal point arrangement of the signal sequence output from the demodulator in FIG. It is a figure which shows the example of the data structure memorize | stored in the buffer 40 of FIG. 6A and 6B are diagrams illustrating an example of association in the differential interval determination unit in FIG. It is a figure which shows the structural example of the partial signal acquisition part of FIG. It is a figure which shows the structural example of the synchronous detection part of FIG. It is a flowchart which shows the operation example of the signal processing part of FIG. FIGS. 10A to 10D are diagrams illustrating an example of correspondence in the differential interval determination unit according to the modification of the embodiment of the present invention. FIGS. 11A to 11D are diagrams illustrating examples of correlation values according to a modification of the embodiment of the present invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Antenna, 2 RF part, 3 Quadrature detection part, 6 Decoding part, 20 Demodulation part, 22 Partial signal acquisition part, 24 Temporary synchronization determination part, 25 Differential space | interval determination part, 26 Pseudo-synchronization determination part, 28 Frequency correction part, 30 first differential calculation unit, 32 first correlation calculation unit, 38 measurement unit, 40 buffer, 42 control unit, 50 signal series, 52 known signal series, 53 known signal, 54 data signal, 62 writing unit, 64 A reliability acquisition unit, 70 synchronization detection unit, 72 second differential calculation unit, 74 second correlation calculation unit, 100 signal processing unit, 200 terminal device.

Claims (5)

An input unit for inputting a signal sequence in which a plurality of known signal sequences including a plurality of known signals are inserted at a predetermined period L;
A demodulator that reproduces a carrier wave from the signal sequence input from the input unit, and demodulates the signal sequence according to the reproduced carrier wave;
A partial signal acquisition unit for acquiring a plurality of partial signal sequences to be associated with the sequence of known signals included in the signal sequence demodulated in the demodulation unit;
A difference in which partial signals to be associated with the known signal included in a predetermined partial signal sequence among the partial signal sequences acquired by the partial signal acquisition unit are respectively associated using a plurality of rules regarding differential intervals. A movement interval determination unit;
By calculating the relative phase difference between the partial signals associated according to the first rule among the plurality of rules relating to the differential interval, the position where the known signal sequence is inserted is estimated. After detecting the provisional synchronization, the relative phase difference between the partial signals associated with each other according to the second rule, which is a differential interval different from the first rule, among the plurality of rules regarding the differential interval, respectively. A synchronization detection unit that detects whether the frequency of the carrier wave reproduced by the demodulation unit is correct by calculating;
A signal processing apparatus comprising: The synchronization detector
A first differential calculation unit that calculates a relative phase difference between the partial signals associated according to a first rule among the rules regarding the plurality of differential intervals;
A first correlation value between a phase difference between the respective partial signals calculated by the first differential calculation unit and a relative phase difference between the known signals that should correspond to the phase difference between the respective partial signals. A first correlation calculation unit for calculating;
Temporary synchronization determination that determines the position where the partial signal exists as the position where the sequence of known signals is inserted when the first correlation value calculated by the first correlation calculation unit is larger than a threshold value related to temporary synchronization And
Among the partial signals existing at the position where the sequence of known signals determined by the temporary synchronization determination unit is inserted, between the partial signals associated according to the second rule among the rules regarding the plurality of differential intervals A second differential calculation unit for calculating a relative phase difference;
A second correlation value between a phase difference between the respective partial signals calculated by the second differential calculation unit and a relative phase difference between the known signals to be matched with the phase difference between the respective partial signals; A second correlation calculator for calculating;
A pseudo-synchronization determination unit that determines that the frequency of the carrier wave reproduced by the demodulation unit is correct normal synchronization when the second correlation value calculated by the second correlation calculation unit is larger than a threshold value related to pseudo-synchronization;
The signal processing apparatus according to claim 1, comprising: The synchronization detector
When the first correlation value calculated by the first correlation calculation unit is smaller than a threshold value related to provisional synchronization, the signal sequence is again set to a partial signal sequence different from the partial signal sequence. A control unit that causes a differential calculation unit to calculate a differential;
A frequency correction unit that corrects the frequency of the carrier wave by shifting the frequency of the carrier wave when the second correlation value calculated by the second correlation calculation unit is smaller than a threshold value related to pseudo synchronization; The signal processing apparatus according to claim 2.   The differential interval determining unit is separated from one partial signal of a plurality of partial signals included in the partial sequence to be associated with the known signal sequence, by the same length as the partial signal and the predetermined period L. The signal processing apparatus according to claim 1, wherein the signal is associated with a partial signal arranged at a different position.   The differential interval determination unit has one partial signal of a plurality of partial signals included in a partial sequence that should correspond to the known signal sequence, and the partial signal and the predetermined period L have different lengths. The signal processing apparatus according to claim 1, wherein partial signals arranged at distant positions are associated with each other.

Images