HK1154131A1 - Receiver - Google Patents

Abstract

A receiver (10) is provided with a circuit (16) for discriminating multilevel modulation system, a circuit (17) for selecting phase correction system, a phase correction circuit (14), and a judgment circuit (15). The circuit (16) discriminates a multilevel modulation system used for a modulation signal on the basis of the modulation signal received from a transmitter. The circuit (17) selects a phase correction system to be used for the phase correction of the symbol of the modulation signal from a plurality of prepared phase correction systems on the basis of the multilevel degree of the multilevel modulation system discriminated by the circuit (16). The circuit (14) corrects the phase of the symbol using the phase correction system selected by the circuit (17). The circuit (15) judges the bit string of the symbol phase-corrected by the circuit (14) on the basis of the multilevel modulation system discriminated by the circuit (16).

Description

Receiver with a plurality of receivers

Technical Field

The present invention relates to a receiver, and more particularly to a receiver used in a communication system using an adaptive modulation scheme.

Background

Conventionally, there is a communication system using an adaptive modulation scheme. The communication system has a plurality of multilevel modulation schemes with different multilevel degrees (bit rates), and switches the multilevel modulation scheme to be used, for example, according to the environment (line quality) where the communication system is located. In this case, the best transmission efficiency can be obtained depending on the line quality. Here, examples of the multilevel Modulation scheme include BPSK (Binary Phase Shift Keying), QPSK (Quadrature Phase Shift Keying), 16QAM (Quadrature Amplitude Modulation), and 64QAM (e.g., refer to japanese laid-open patent publication 2007-150906 and IEEE 802.11a-1999) in order of the bit rate from low to high.

However, each of the transmitter and the receiver constituting the communication system includes a reference signal source. The reference signal source utilizes a crystal emitter. The oscillation frequency of each reference signal source of the transmitter and receiver (hereinafter referred to as "reference frequency") contains errors due to the accuracy of the crystal transmitter. As a result, an error of several parts per million is generated between the reference frequency of the transmitter side that performs the modulation process and the reference frequency of the receiver side that performs the demodulation process. This error in the reference frequency causes phase rotation of data received by the receiver. The occurrence of such phase rotation has a great influence on the Bit Error Rate (hereinafter referred to as "BER") after demodulation. Therefore, the receiver corrects the phase rotation of the received data at the time of demodulation. In particular, in the case of multi-value modulation using a multi-carrier modulation scheme such as OFDM (Orthogonal Frequency Division Multiplexing), the influence of phase rotation due to a Frequency error at the time of demodulation becomes large. This is because the occupation time of 1 symbol becomes longer and the interval of phase correction becomes longer.

Conventionally, as a phase correction method for performing phase rotation correction of a symbol, there are a phase correction method using a pilot subcarrier (for example, see japanese laid-open patent publication 2008-22339) and a phase correction method using a pilot symbol (for example, see japanese laid-open patent publication 2006-352746).

In the phase correction method using the pilot subcarriers, when there is a large frequency selectivity in the transmission characteristics of the signal due to the influence of multipath fading, the correction error becomes large.

Therefore, the phase correction method using the pilot symbols is effective in a transmission environment with high frequency selectivity.

On the other hand, the phase correction method using the pilot symbols is not a sequential correction for each OFDM symbol, unlike the phase correction method using the pilot subcarriers. Therefore, the larger the multilevel degree of the multilevel modulation scheme is, the narrower the interval between pilot symbols to be embedded in the modulated signal must be. Therefore, if phase correction is performed in the adaptive modulation scheme so as to satisfy all the multilevel modulation schemes, the transmission efficiency is lowered particularly in the multilevel modulation scheme having a low multilevel degree.

In addition, since phase correction using pilot symbols cannot be performed until pilot symbols are obtained, the interval for performing phase correction is relatively long. Therefore, in the case of a multi-value modulation system with a high multi-value degree, an error in phase rotation may exceed an allowable range during demodulation processing.

As described above, the phase correction method is suitable and unsuitable for the multilevel modulation method. However, the conventional receiver uses the same phase correction scheme for the modulated signal received from the transmitter. Therefore, the conventional receiver cannot perform optimum phase correction according to the multilevel modulation scheme used in the transmitter.

Disclosure of Invention

The present invention has been accomplished in view of the above-described reasons. An object of the present invention is to provide a receiver capable of performing optimum phase correction according to a multilevel modulation scheme used in a transmitter.

A receiver according to the present invention is used in an adaptive modulation communication system together with a transmitter that transmits a modulation signal generated using a multilevel modulation scheme selected according to a line quality from among a plurality of multilevel modulation schemes. The modulated signal has a symbol string representing data to be transmitted to the receiver. The correspondence between the symbol and the bit string is determined by the multivalued modulation scheme selected in the transmitter. A receiver according to the present invention includes a multi-value modulation scheme determination unit, a phase correction scheme selection unit, a phase correction unit, and a determination unit. The multilevel modulation scheme determination unit is configured to determine the multilevel modulation scheme used for the modulation signal based on the modulation signal received from the transmitter. The phase correction scheme selection unit is configured to select a phase correction scheme to be used for phase correction of the symbol of the modulation signal from among a plurality of phase correction schemes prepared in advance, based on the multilevel degree of the multilevel modulation scheme determined by the multilevel modulation scheme determination unit. The phase correction unit is configured to correct the phase of the symbol using the phase correction method selected by the phase correction method selection unit. The determination unit is configured to determine a bit string corresponding to the symbol whose phase has been corrected in the phase correction unit, based on the multilevel modulation scheme determined by the modulation scheme determination unit.

According to the present invention, the phase correction scheme is selected based on the multi-level degree of the multi-level modulation scheme used in the transmitter. Therefore, the optimum phase correction can be performed according to the multilevel modulation scheme used in the transmitter.

Preferably, the transmitter has a primary modulation scheme and a secondary modulation scheme. The primary modulation scheme is a multi-value modulation scheme selected from a plurality of multi-value modulation schemes having different values according to a predetermined reference, and generates a primary modulation symbol indicating the symbol. The secondary modulation scheme is a multi-carrier modulation scheme. The secondary modulation scheme generates a secondary modulation symbol by superimposing a plurality of subcarriers constituting a complex amplitude on the primary modulation symbol. The secondary modulation scheme constitutes the modulated signal composed of a plurality of the secondary modulation symbols. The modulated signal has pilot symbols at predetermined intervals. The pilot symbols are secondary modulation symbols known to the receiver. The known secondary modulation symbols are composed of subcarriers having known complex amplitudes. The secondary modulation symbols include pilot subcarriers. The pilot subcarriers are subcarriers known to the receiver. The known subcarriers have known complex amplitudes. The phase correction method selection unit is configured to select a1 st phase correction method for correcting the phase of the symbol using the pilot symbol if the multilevel degree of the multilevel modulation method determined by the multilevel modulation method determination unit is smaller than a predetermined value, and to select a 2 nd phase correction method for correcting the phase of the symbol using the pilot subcarrier if the multilevel degree is greater than or equal to the predetermined value.

In this case, when the multi-level degree of the multi-level modulation scheme used in the transmitter is smaller than a predetermined value, the phase correction using the pilot symbol is performed with a large phase correction effect even in a transmission environment with high frequency selectivity. On the other hand, when the multi-level degree of the multi-level modulation scheme used in the transmitter is equal to or greater than a predetermined value, phase correction using the pilot subcarrier is performed, which facilitates high transmission efficiency. Therefore, the optimum phase correction can be performed according to the multilevel modulation scheme used in the transmitter.

More preferably, the phase correction method selection unit is configured to select only the 1 st phase correction method if the multilevel degree of the multilevel modulation method determined by the multilevel modulation method determination unit is smaller than a predetermined value, and to select both the 1 st phase correction method and the 2 nd phase correction method if the multilevel degree is equal to or greater than the predetermined value.

In this case, the 1 st phase correction using the pilot symbol is always performed. Therefore, the control of the phase correction unit is simplified. In the 1 st phase correction method, phase correction is performed for each of the subcarriers. Therefore, even in a propagation environment where frequency selectivity is strong, a good phase correction effect can always be achieved.

Further preferably, the phase correction unit is configured to correct the phase of the symbol in accordance with the 2 nd phase correction method and then correct the phase of the symbol in accordance with the 1 st phase correction method when both the 1 st phase correction method and the 2 nd phase correction method are selected by the phase correction method selection unit.

In this case, phase correction can be performed on a per symbol basis using the pilot subcarriers. Further, errors caused by phase correction using pilot subcarriers can be removed by phase correction using pilot symbols. Therefore, the optimum phase correction can be performed for the multilevel modulation scheme having a large multilevel degree.

Preferably, the predetermined value is set according to a transmission efficiency when the phase correction unit corrects the phase of the symbol by using the phase correction method selected by the phase correction method selection unit.

In this case, the optimum phase correction can be performed according to the multilevel modulation scheme used in the transmitter, and the transmission efficiency can be improved.

Drawings

Fig. 1 is a schematic diagram of a receiver according to embodiment 1.

Fig. 2 is a schematic diagram of a communication system including the receiver as described above.

Fig. 3 is an illustration of a 16QAM constellation.

Fig. 4 is an explanatory diagram of the arrangement of subcarriers and pilot subcarriers.

Fig. 5 is an explanatory diagram of the structure of the OFDM signal.

Fig. 6 is an explanatory diagram of an embedding structure of pilot symbols.

Fig. 7 is a schematic diagram of a receiver according to embodiment 2.

Detailed Description

(embodiment mode 1)

As shown in fig. 2, the receiver 10 of the present embodiment forms an adaptive modulation communication system (hereinafter referred to as "communication system") together with the transmitter 20. The communication system performs packet communication based on an OFDM signal modulated (OFDM modulated in the present embodiment) by the transmitter 20. In addition, the transmission path 30 for transmitting the OFDM modulated wave from the transmitter 20 to the receiver 10 may be wired or wireless.

The transmitter 20 uses a multivalued modulation scheme as primary modulation and OFDM modulation as secondary modulation. The transmitter 20 error-correction codes data (information bit string) to be transmitted to the receiver 10. In addition, the transmitter 20 performs serial-to-parallel conversion on the error-correction coded data. The transmitter 20 generates a complex symbol (primary modulation symbol) for modulating a subcarrier from the data after serial-parallel conversion based on the correspondence between the symbol and the bit string determined by the multilevel modulation scheme (symbol matching). The transmitter 20 sequentially performs inverse discrete fourier transform (secondary modulation) on the complex symbols, and then generates complex baseband OFDM signals (OFDM symbols or secondary modulation symbols) in digital form by parallel-serial conversion. The transmitter 20 performs digital/analog conversion (DA conversion) on the complex baseband OFDM signal. The transmitter 20 generates an OFDM modulated wave by filtering the DA-converted OFDM signal with a filter for removing an alias generated in the DA conversion, multiplying the filtered OFDM signal by a carrier wave (performing frequency conversion), and amplifying the resultant signal. The transmitter 20 transmits the OFDM modulated wave thus generated to the transmission path 30.

Here, the transmitter 20 has a plurality of multilevel modulation schemes with different multilevel degrees, for example, 16QAM and 64QAM, as primary modulation schemes. The transmitter 20 selects a multilevel modulation scheme from a plurality of multilevel modulation schemes having different multilevel degrees at the time of symbol matching with a predetermined reference (that is, the transmitter 20 performs adaptive modulation). For example, the transmitter 20 selects a multivalued modulation scheme with the highest multivalued degree to maximize the transmission speed of data. The transmitter 20 may select a multilevel modulation scheme so that the transmission rate is equal to or higher than a fixed value, depending on the state (channel quality) of the transmission path 30 and the capacity of data.

In addition, the transmitter 20 is provided with a reference signal source (not shown) having a crystal transmitter. The transmitter 20 performs inverse discrete fourier transform (OFDM modulation) and frequency transform using a reference frequency transmitted from a reference signal source. The receiving apparatus 10 also includes a reference signal source.

In this way, the transmitter 20 transmits an OFDM modulated wave generated using a multilevel modulation scheme selected by a predetermined reference from among a plurality of multilevel modulation schemes (16QAM and 64 QAM).

Here, in a multilevel modulation scheme with a high multilevel degree such as 64QAM, an allowable error angle is small when a bit string of a symbol is determined using a complex symbol on a complex plane. Therefore, it is necessary to perform phase correction for each symbol. Next, an allowable error angle in QAM will be described using 16QAM as an example.

Fig. 3 shows symbol arrangement (signal point arrangement) on a complex plane corresponding to bit strings [0000] to [1111] of 16 QAM. Here, gray code is assumed. The partition line L1 in fig. 3 passes through the middle point of the line segment connecting the symbol points of the bit string [1110] and the bit string [1010] in the Q (Quadrature-Phase) axis direction. The partition line L2 passes through the midpoint of a line segment connecting the symbol points of the bit string [1010] and the bit string [1011] In the I (In-Phase) axis direction. When the complex symbol received by the receiver 10 exists in the region a12 surrounded by the partition lines L1 and L2, including the bit string [1010], it is estimated that the probability that the complex symbol represents the bit string [1010] is high. In practice, however, phase rotation (see arrow Q1) due to errors in the reference frequency may result in complex symbols present in the transmitter 20 in region a12 that may not be present in the receiver 10 in region a 12. In this case, an error occurs in the determination of the bit string.

If there is no error of the reference frequency, the complex symbols corresponding to the bit string [1010] received by the receiver 10 are distributed in a Gaussian distribution centered on the symbol point representing the bit string [1010 ]. Therefore, it is not contradictory to set the allowable error angle θ 1 for each symbol point on the condition that the symbol point does not deviate from the original region. For example, in the case of 16QAM, the allowable error angle θ 1[ degree ] is 16.88.

Table 1 shows the allowable error angle θ 1 of a typical QAM. It can be seen from table 1 that the allowable error angle θ 1 becomes smaller as the multi-value degree becomes larger.

[ Table 1]

Multiple value modulation system	16QAM	64QAM	256QAM	1024QAM
					Allowable error angle theta 1(degree)	16.88	10.55	7.69	6.06

Next, phase rotation of each OFDM symbol due to an error of a reference frequency in OFDM modulation will be discussed. As a main cause of a phase error due to phase rotation of an OFDM symbol, two types of errors (1 st error) in carrier frequency synchronization (frequency transform) and (2 nd error) in sampling frequency synchronization (fast fourier transform) required for demodulation processing of OFDM can be considered. Assuming that the sampling frequency of the fast fourier transform is fs, the size of the fast fourier transform (FFT size) is N points, and the time of the Guard Interval (Guard Interval) is Tgi, the occupied time Ta per OFDM symbol is expressed by the following expression (1).

[ numerical formula 1]

The phase error based on the 1 st error and the 2 nd error is additive. Therefore, when the carrier frequency is fc and the reference frequency error between the two processes of modulation and demodulation is e, the phase error angle θ 2[ degree ] per OFDM symbol is expressed by the following expression (2).

[ numerical formula 2]

For example, according to the IEEE 802.11a-1999 standard (IEEE (institute of electrical and electronics engineers)) standard IEEE 802.11a for wireless LANs, if the reference frequency error of each reference signal source is allowed to be 20ppm, the reference frequency error e of 40ppm is reached in both the modulation and demodulation processes. The error in the carrier frequency fc is generally converged to a frequency error of fs/2 by using an automatic frequency correction circuit of the receiver. Therefore, the above formula (2) can be modified to the following formula (3).

[ numerical formula 3]

In addition, according to the IEEE 802.11a-1999 standard, the sampling frequency fs of the fast Fourier transform is 20MHz, the occupied time Ta of the OFDM symbol is 4 μ sec (where the time Tgi of the guard interval is 0.8 μ sec), and the size N of the fast Fourier transform is 64 points.

If equation (3) above is calculated in accordance with the specification of IEEE 802.11a-1999, the phase error angle θ 2 is 2.88. Therefore, the allowable error angle θ 1 is exceeded with 4 symbols in 64 QAM. In addition, according to the IEEE 802.11a-1999 specification, 1 packet is required to be 1000 bytes at maximum. Therefore, if the redundancy bits based on error correction are not attached, the OFDM symbols that can be transmitted in 1 packet are about 27 symbols. Therefore, when the allowable error angle θ 1 is exceeded with 4 symbols, 1 packet cannot be correctly demodulated.

Thus, IEEE 802.11a-1999 specifies the following: as shown in fig. 4, 4 of all 52 subcarriers are set as pilot subcarriers PSC1 to PSC4 irrelevant to data transmission, and the remaining 48 are set as subcarriers SC0 to SC47 for data transmission. Therefore, phase correction can be performed for each OFDM symbol using the pilot subcarriers PSC1 to PSC 4.

However, when the transmission characteristics of the signal have large frequency selectivity due to the influence of multipath fading, the S/N ratio of the frequency of the embedded pilot subcarrier may be extremely deteriorated. In this case, a correction error of the phase correction method using the pilot subcarriers becomes large. Therefore, in particular, in the case of a multilevel modulation scheme with a high multilevel degree, BER may be deteriorated when phase correction is performed. For example, in the vicinity of pilot subcarrier PSC1 in fig. 4, since frequency characteristic 1000 deteriorates, the accuracy of phase correction using pilot subcarrier PSC1 decreases.

In a transmission environment where frequency selectivity is high, a phase correction method using pilot symbols is effective. The pilot symbols are composed of symbols known in the reception apparatus 10 and the transmitter 20. In addition, pilot symbols are embedded at regular time intervals in a modulated signal (packet) composed of OFDM symbols. Therefore, phase correction per subcarrier can be performed using the pilot symbols.

However, the phase correction method using the pilot symbols is not sequential correction for each OFDM symbol. Therefore, the larger the multilevel degree of the multilevel modulation scheme is, the narrower the interval between pilot symbols to be embedded in the modulated signal must be. For example, in 16QAM, it is sufficient to embed pilot symbols every 5 symbols in the modulated signal. In contrast, in 64QAM, the interval at which pilot symbols are embedded in the modulated signal must be set to every 3 symbols. Therefore, if the adaptive modulation scheme is configured to perform phase correction that can satisfy all the multilevel modulation schemes, the transmission efficiency is reduced particularly in the multilevel modulation scheme having a low multilevel degree.

In addition, since phase correction using pilot symbols cannot be performed until pilot symbols are obtained, the interval for performing phase correction is relatively long. Therefore, even in the case of the multilevel modulation scheme having a high multilevel degree, the allowable error angle θ 1 may be exceeded during the demodulation processing.

Therefore, the transmitter 20 generates a modulation signal so that the receiver 10 can selectively perform phase correction using the pilot symbol and phase correction using the pilot subcarrier.

For example, the modulated signal (packet) is composed of a short preamble SP, a long preamble LP, and a data portion D as shown in fig. 5.

In order to establish symbol timing synchronization, a synchronization pattern (specific pattern) X known to both the transmitter 20 and the receiver 10 is repeated 10 times (X1 to X10) every basic period T1(═ 0.8 μ sec) to constitute a short preamble SP. That is, the short preamble SP is constituted by the repeated signal of the basic period T1.

For channel estimation, the synchronization pattern Y known to both the transmitter 20 and the receiver 10 is repeated 2 times (Y1, Y2) every basic period T2(═ 3.2 μ sec) to constitute a long preamble LP.

The data portion D is an area for transmitting data such as data bits and information on a modulation scheme.

In the modulated signal, the short preamble SP, the long preamble LP, and the data section D are arranged in this order.

Further, guard intervals GI1 and GI2 obtained by copying a part of the second half of each region are added to the front of each region of the long preamble LP and the data section D. The effect of multipath can be reduced by the guard intervals GI1, GI 2.

As shown in fig. 1, the receiver 10 includes an automatic frequency correction circuit (AFC)11, a guard interval removal circuit 12, a fast fourier transform circuit (FFT)13, a phase correction circuit (phase correction unit) 14, a determination circuit (determination unit) 15, a multilevel modulation scheme determination circuit (multilevel modulation scheme determination unit) 16, and a phase correction scheme selection circuit (phase correction scheme selection unit) 17. In the figure, analog signal processing circuits such as signal amplification, frequency conversion (down-conversion), interference wave removing filter, analog-to-digital conversion (AD conversion) and the like in the analog section are omitted.

The automatic frequency correction circuit 11 performs analog/digital conversion (AD conversion) on the baseband signal, and then corrects the phase rotation of each OFDM signal using the short preamble SP and the long preamble LP.

The automatic frequency correction circuit 11 initially detects a comparatively large frequency error between the reference frequency of the transmitter 20 and the reference frequency of the receiver 10 using the short preamble SP. The detection of the frequency error is performed, for example, by multiplying the complex conjugate of the modulated signal delayed by the basic period T1 by the modulated signal after the basic period T1.

Next, the automatic frequency correction circuit 11 detects a frequency error using the long preamble LP. The detection of the frequency error using the long preamble LP is performed in the same order as the detection of the frequency error using the short preamble SP. If a long preamble LP is used, a relatively small frequency error of 1/{2 · T2} (═ fs/{2 · 64}) can be detected.

The automatic frequency correction circuit 11 multiplies the received modulation signal by the phase reversal of the frequency error detected using the short preamble SP and the long preamble LP. Thereby, the automatic frequency correction circuit 11 performs phase correction (frequency correction).

The guard interval removing circuit 12 removes guard intervals GI1 and GI2 added to the modulated signal by the transmitter 20.

The fast fourier transform circuit 13 performs discrete fourier transform on the OFDM signal at a sampling frequency based on a reference frequency. Thus, the fast fourier transform circuit 13 performs multicarrier demodulation for demultiplexing into a plurality of subcarrier signals. Thereby extracting the component of the complex symbol for each subcarrier.

The phase correction circuit 14 corrects the phase rotation of the primary modulation symbol based on the frequency error. The phase correction circuit 14 includes an estimation unit 141, an equalization unit 142, and a phase error removal unit 143.

The estimation unit 141 estimates the impulse response of the frequency region of the transmission channel 30 for each subcarrier using the pilot symbol. The impulse response indicates the transmission characteristics of each subcarrier. The estimation unit 141 estimates the phase rotation and amplitude error (impulse response per subcarrier) of the known data of the preamble (the synchronization pattern X of the short preamble SP or the long preamble LP synchronization pattern Y) by regarding the known data as pilot symbols. The following modulated signals have pilot symbols at predetermined time intervals, and phase rotation and amplitude error are estimated from the known data. The estimated phase rotation and amplitude error may be valid before correction using the next pilot symbol, or values obtained by appropriately weighting the correction amount of the next pilot symbol and the correction amount of the current pilot symbol may be used as correction values based on the next pilot symbol. In addition, although the complex amplitudes of all subcarriers may be used as known data for the pilot symbols, the known data may be embedded only in subcarriers in a frequency region where the frequency selectivity is high. In this case, the transmission characteristics of the subcarriers in which the known data is not embedded may be derived from the transmission characteristics of the subcarriers in which the known data is embedded.

The equalizer 142 multiplies the inverse characteristic of the impulse response of each subcarrier estimated by the estimator 141 by the complex symbol of each subcarrier following the preamble. In this way, the equalization unit 142 corrects distortion in the frequency domain for each subcarrier, and corrects phase rotation based on the frequency error. In addition, for a transmission path with large amplitude variation, not only phase rotation but also amplitude error is corrected.

In this way, the estimation unit 141 and the equalization unit 142 perform the 1 st phase correction method for correcting the phase of the symbol using the pilot symbol. Since the phase correction using the pilot symbol is performed for each subcarrier, a good phase correction effect is obtained even in a transmission environment with high frequency selectivity.

In the present embodiment, as shown in fig. 4, 4 of all 52 subcarriers are pilot subcarriers PSC1 to PSC4 irrelevant to data transmission, and the remaining 48 are subcarriers SC0 to SC47 for data transmission. The symbols on the pilot subcarriers PSC 1-PSC 4 are known data (known symbols).

The phase error removal unit 143 performs phase correction for each OFDM symbol using the 4 pilot subcarriers PSC1 to PSC 4. The phase error removal unit 143 detects a frequency error in each pilot subcarrier from the known symbols of the pilot subcarriers PSC1 to PSC 4. The phase error removing unit 143 calculates a phase error of each complex symbol obtained by performing discrete fourier transform on the same OFDM symbol using the detected frequency error. The phase error removing unit 143 multiplies each complex symbol by the phase opposite to the calculated phase error. Thus, the phase error removal unit 143 corrects the phase rotation of the sign based on the frequency error.

In this way, the phase error removal unit 143 performs the 2 nd phase correction scheme for correcting the phase of the symbol using the pilot subcarrier.

The modulation scheme discrimination circuit 16 discriminates a multilevel modulation scheme used for modulating a signal from the modulated signal received from the transmitter 20. In the present embodiment, it is determined whether the multilevel modulation scheme (of each complex symbol obtained by performing discrete fourier transform on the same OFDM symbol) for each OFDM symbol is 16QAM or 64QAM, based on the information of the modulation scheme contained in the data portion D of the modulated signal received from the transmitter 20.

The correction scheme selection circuit 17 selects a phase correction scheme for correcting the phase of the symbol of the modulation signal from a plurality of phase correction schemes prepared in advance, based on the multilevel degree of the multilevel modulation scheme determined by the modulation scheme determination circuit 16. In the present embodiment, the correction scheme selection circuit 17 selects the 1 st phase correction scheme when the determination result of the modulation scheme determination circuit 16 is 16QAM, and selects the 2 nd phase correction scheme when the determination result is 64 QAM.

The phase correction circuit 14 corrects the phase of the symbol by using the phase correction method selected by the correction method selection circuit 17.

The determination circuit 15 determines a bit string of data based on the symbol of which the phase is corrected by the phase correction circuit 14, according to the multilevel modulation scheme determined by the multilevel modulation scheme determination circuit 16. More specifically, the determination circuit 15 converts each complex symbol whose phase has been corrected by the correction circuit 14 into a soft determination value by demapping, based on the multilevel modulation scheme discriminated by the multilevel modulation scheme discrimination circuit 16. Thus, the decision circuit 15 outputs the bit string of the data received from the transmitter 20 to a data processing circuit, not shown, inside the receiver 10 or outside the receiver 10.

The phase error angle θ 2 of each OFDM symbol is 2.88 ° as described above. Table 2 shows the values of θ 1/θ 2 in each of QPSK, 16QAM, and 64QAM multilevel modulation schemes. Table 2 shows a minimum symbol interval M (maximum positive integer of θ 1/θ 2 or less) in which pilot symbols are embedded in the modulated signal so as not to exceed the allowable error angle θ 1 during the demodulation process. Table 2 also shows the transmission efficiency P1 of the case where pilot symbols are embedded in the modulated signal at the minimum symbol interval M, which is M/(M + 1).

[ Table 2]

Multiple value modulation system	QPSK	16QAM	64QAM
				θ1/θ2	15.63	5.86	3.66
M	15	5	3
				P1＝M/M+1	0.94	0.83	0.75

As shown in fig. 6, in each multilevel modulation scheme, a pilot symbol PS is embedded in a modulated signal every M symbols. Thereby, the phase error is prevented from exceeding the allowable error angle θ 1 during the demodulation processing. In addition, the transmission efficiency also becomes highest.

In the 1 st phase correction scheme using pilot symbols, the correction interval is longer than in the case of using pilot subcarriers. This is because the equalization parameters of the equalization unit 142 are updated every M symbols. Therefore, in the case of a multi-value modulation scheme with a high multi-value degree, that is, 64QAM (minimum symbol interval M is 3), if the pilot symbol interval is 5, the phase error may exceed the allowable error angle θ 1 during the demodulation process.

On the other hand, the 2 nd phase correction scheme using pilot subcarriers PSC1 to PSC4 is sequential correction for each OFDM symbol. Therefore, optimal phase correction can be performed for each OFDM symbol (for each complex symbol obtained by performing discrete fourier transform on the same OFDM symbol). Therefore, even in the multi-value modulation system with a high multi-value degree, it is possible to prevent the phase error from exceeding the allowable error angle θ 1 during the demodulation processing.

In the 1 st phase correction scheme, the interval of pilot symbols to be embedded in a modulated signal needs to be shorter as the multi-value degree is larger. Therefore, the transmission efficiency easily becomes low. As shown in table 2, the transmission efficiency P1 is 0.83 in 16QAM and the transmission efficiency P1 is 0.75 in 64 QAM. On the other hand, in the 2 nd phase correction scheme, 4 out of all 52 subcarriers are used for pilot subcarriers PSC1 to PSC 4. Therefore, the transmission efficiency P2 when the pilot subcarriers are used is 0.92(═ 48/52). Therefore, the 2 nd phase correction scheme is easier to achieve high transmission efficiency than the 1 st phase correction scheme.

As described above, according to the receiver 10 of the present embodiment, the phase correction scheme is selected based on the multi-level degree of the multi-level modulation scheme used in the transmitter 20. Therefore, the phase correction can be optimally performed according to the multilevel modulation scheme used in the transmitter 20.

In particular, when the multi-level degree of the multi-level modulation scheme used in the transmitter 20 is smaller than a predetermined value (multi-level degree corresponding to 64QAM), phase correction using the pilot symbol is performed with a large phase correction effect even in a transmission environment with strong frequency selectivity. On the other hand, when the multi-level degree of the multi-level modulation scheme used in the transmitter 20 is equal to or greater than a predetermined value (multi-level degree corresponding to 64QAM), phase correction using the pilot subcarriers is performed, which facilitates high transmission efficiency. Therefore, the phase correction can be optimally performed according to the multilevel modulation scheme used in the transmitter 20.

However, the phase correction method selection circuit 17 may be configured as follows. That is, the phase correction scheme selection circuit 17 selects only the 1 st phase correction scheme if the multi-level degree of the multi-level modulation scheme discriminated in the multi-level modulation scheme discrimination circuit 16 is smaller than a predetermined value (multi-level degree corresponding to 64QAM), and the phase correction scheme selection circuit 17 selects both the 1 st phase correction scheme and the 2 nd phase correction scheme if the multi-level degree is equal to or more than the predetermined value (multi-level degree corresponding to 64 QAM).

In this case, phase correction using pilot symbols is always performed. Therefore, the control of the phase correction circuit 14 becomes simple. In the 1 st phase correction scheme, phase correction is performed for each subcarrier. Therefore, even in a transmission environment where frequency selectivity is strong, the effect of phase correction can be always made large.

However, the estimating section 141 estimates the impulse response of the transmission path 30 using the known symbols of the preamble and the pilot symbols embedded at fixed intervals in the modulated signal. For pilot symbols, known data is embedded for all subcarriers. Therefore, if the pilot symbol is used, highly accurate phase correction can be performed for all subcarriers. However, the impulse response of the transmission channel 30 cannot be estimated until the pilot symbols are obtained. That is, in the phase correction using the pilot symbol, the correction interval becomes long. Therefore, in the case of a multi-level modulation scheme with a high multi-level, the complex amplitude of the subcarrier, that is, the allowable error angle θ 1 of the primary modulation symbol may be exceeded during the demodulation process.

On the other hand, the phase error removing unit 143 estimates the impulse response of the transmission channel 30 using the pilot subcarriers. Therefore, the impulse response of the transmission channel 30 can be updated every OFDM symbol. However, the estimated values of the impulse responses for subcarriers other than the pilot subcarrier (that is, subcarriers used for data transmission) are calculated by extrapolation and interpolation from the impulse responses of the pilot subcarriers. Therefore, the estimated value of the impulse response of the subcarrier used for data transmission contains an error.

Therefore, if the impulse response is updated only by the phase error removal unit 143, the equalization is performed by the equalization unit 142 using the inverse characteristic calculated from the updated impulse response, and therefore an error is accumulated every time the impulse response is updated. Therefore, if a certain time elapses, the accumulated amount of errors exceeds the allowable amount (allowable error angle θ 1), and the effect of phase correction disappears. In particular, the multilevel modulation method having a larger multilevel degree has a higher possibility of losing the effect of phase correction because the allowable error angle θ 1 is smaller.

Therefore, when the multilevel degree of the multilevel modulation scheme discriminated by the multilevel modulation scheme discrimination circuit 16 is equal to or more than a predetermined value (multilevel degree corresponding to 64QAM) (when both the 1 st phase correction scheme and the 2 nd phase correction scheme are selected by the phase correction scheme selection circuit 17), the phase correction circuit 14 is preferably configured as follows. That is, the phase correction circuit 14 corrects the phase of the symbol in accordance with the 2 nd phase correction method and then corrects the phase of the symbol in accordance with the 1 st phase correction method.

By doing so, phase correction can be performed on a per OFDM symbol basis using pilot subcarriers. Further, since an error caused by phase correction using the pilot subcarrier can be removed by phase correction using the pilot symbol, it is possible to perform optimum phase correction for a multilevel modulation scheme having a large multilevel degree.

However, in the above example, the predetermined value of the phase correction scheme selection circuit 17 is set to a multi-value degree corresponding to 64 QAM. However, the predetermined value may be set according to the transmission efficiency when the phase correction circuit 14 corrects the phase of the symbol by using the phase correction method selected by the phase correction method selection circuit 17.

For example, consider a case where the transmitter 20 has QPSK, 16QAM, and 64QAM as the multivalued modulation schemes.

In this case, the multilevel modulation scheme discrimination circuit 16 discriminates which of QPSK, 16QAM, and 64QAM the multilevel modulation scheme used for modulating the signal is based on the information of the modulation scheme contained in the data section D of the modulated signal received from the transmitter 20.

Here, in the case of the 1 st phase correction scheme, the transmission efficiencies P1 of QPSK, 16QAM, and 64QAM are 0.97, 0.83, and 0.75, respectively (refer to table 2).

On the other hand, in the case of the 2 nd phase correction method, the transmission efficiency P2 is 0.92.

In this case, the phase correction scheme selection circuit 17 selects the phase correction using the pilot symbol for the multilevel modulation scheme in which the transmission efficiency P1 is higher than the transmission efficiency P2(═ 0.92). That is, the phase correction scheme selection circuit 17 selects the 1 st phase correction scheme when the transmission efficiency P1 is QPSK of 0.97. On the other hand, for a multilevel modulation scheme having a transmission efficiency P1 lower than the transmission efficiency P2, phase correction using a pilot subcarrier is selected. That is, the phase correction scheme selection circuit 17 selects the 2 nd phase correction scheme when the transmission efficiency P1 is 16QAM of 0.83. Also, the phase correction manner selection circuit 17 selects the 2 nd phase correction manner in the case where the transmission efficiency P1 is 64QAM of 0.75.

That is, the phase correction scheme selection circuit 17 performs phase correction using the pilot subcarrier when the multi-value degree of the multi-value modulation scheme (QPSK, 16QAM, or 64QAM) used in the transmitter 20 is lower than a predetermined value based on the transmission efficiency P2. On the other hand, the phase correction scheme selection circuit 17 performs phase correction using the pilot symbol when the multi-level degree of the multi-level modulation scheme used in the transmitter 20 is equal to or greater than a predetermined value based on the transmission efficiency P2.

In this case, the optimum phase correction can be performed according to the multilevel modulation scheme used in the transmitter 20, and the transmission efficiency can be improved.

(embodiment mode 2)

The receiver 40 of the present embodiment is used in a single carrier communication system.

The transmitter 20 for a single carrier communication system error correction encodes data for transmission to the receiver 40. The transmitter 20 generates complex symbols (symbol matching) from the error correction coded data based on the correspondence between the symbols and bit strings determined by the multilevel modulation scheme. The transmitter 20 performs DA conversion, which is an appropriate waveform formation process for complex symbols, and then performs frequency conversion by multiplying a baseband signal, which is generated using a symbol sequence obtained by filtering with a filter that removes an alias signal generated in the DA conversion, by a carrier wave, performs displacement in a necessary frequency band, and then performs predetermined signal amplification to generate a modulated wave. The transmitter 20 transmits the generated modulated wave to the transmission path 20.

Here, the transmitter 20 has a plurality of multivalued modulation schemes with different multivalued degrees, for example, QPSK and 16 QAM. When the transmitter 20 performs symbol matching, a multilevel modulation scheme having the highest transmission rate is selected from a plurality of multilevel modulation schemes having different multilevel degrees according to the state of the transmission line 30 (that is, the transmitter 20 performs adaptive modulation).

The transmitter 20 is also provided with a reference signal source (not shown) having a crystal transmitter. The transmitter 20 performs the above-described frequency conversion and the like using the reference frequency transmitted from the reference signal source. The receiver 40 also includes a reference signal source.

The transmitter 20 transmits a modulated wave generated by using a multilevel modulation scheme selected according to the line quality from among a plurality of multilevel modulation schemes (QPSK, 64 QAM). The modulated wave has a symbol string representing data to be transmitted to the receiver 40. The correspondence between the symbol and the bit string is determined by the multivalued modulation scheme selected in the transmitter 20.

As shown in fig. 7, the receiver 40 of the present embodiment includes an a/D converter 41, an FIR filter 42, a down-sampling circuit 43, a phase correction circuit 44, a determination circuit 45, a multilevel modulation scheme determination circuit (multilevel modulation scheme determination unit) 46, and a phase correction scheme selection circuit (phase correction scheme selection unit) 47. In the figure, analog signal processing circuits such as signal amplification and interference wave removal filters in the analog section are omitted.

The a/D conversion circuit 41 generates a carrier wave using a reference frequency transmitted from a reference signal source (not shown) of the transmitter 40. The a/D conversion circuit 41 down-converts the modulated signal by multiplying the carrier wave by the modulated signal received via the transmission path 30, thereby generating a baseband signal. The a/D conversion circuit 41 performs analog/digital conversion on the baseband signal and outputs the baseband signal to the FIR filter 42.

The down-sampling circuit 43 down-samples the baseband signal received via the FIR filter 42. The down-sampling circuit 43 outputs the down-sampled baseband signal to the phase correction circuit 44.

The phase correction circuit 44 includes a phase error removal unit 441, a modulator 442, and a phase estimation unit 443. The phase correction circuit 44 selectively corrects the phase rotation based on the frequency error by using two phase correction methods, i.e., a phase correction method using remodulation and a phase correction method using pilot symbols.

The phase error removal unit 441, the modulator 442, and the phase estimation unit 443 perform a phase correction method using remodulation. In the case of the phase correction method using remodulation, the phase error removal unit 441 transmits the output of the down-sampling circuit 43 to the determination circuit 45 as it is. The modulator 442 converts the bit string determined by the determination circuit 45 into an IQ signal on the complex plane, and performs remodulation in which the IQ signal is signed by a complex number. The phase estimation unit 443 calculates the product of the remodulated signal output from the modulator 442 and the output of the down-sampling circuit 43. Thus, the phase estimation unit 443 calculates a phase error. The phase error removing unit 441 multiplies each complex symbol by the phase (phase coefficient) opposite to the phase error calculated by the phase estimating unit 443. Thus, the phase error removing unit 441 corrects the phase rotation based on the frequency error. That is, the phase correction circuit 44 performs phase correction using remodulation. Phase correction using remodulation is performed periodically.

The phase error removing unit 441 performs a phase correction method using pilot symbols. Here, the pilot symbol in the present embodiment is a known symbol having a known phase, but is not a pilot symbol after multicarrier modulation as in embodiment 1. However, the phase correction method using the known sign is the same as that of embodiment 1, and therefore, the description thereof is omitted.

The multilevel modulation scheme discrimination circuit 46 discriminates whether the multilevel modulation scheme of each packet is QPSK or 16QAM based on the modulation scheme information included in the modulation signal received from the transmitter 20.

The phase correction scheme selection circuit 47 selects the phase correction scheme to be executed in the phase correction circuit 44 based on the multi-level degree of the multi-level modulation scheme discriminated in the multi-level modulation scheme discrimination circuit 46.

Here, in the case of 16QAM, since the inter-symbol distance on the complex plane is short, the allowable error angle θ 1 of each symbol point is small compared to QPSK. Therefore, when the multilevel modulation scheme is 16QAM, if the same phase correction scheme as QPSK is used, there is a possibility that a large number of errors are included in the demodulated bit string, and the complex symbol after remodulation is not always correct.

Therefore, the phase correction scheme selection circuit 47 selects the phase correction scheme using remodulation when the discrimination result of the multilevel modulation scheme discrimination circuit 46 is QPSK. The phase correction scheme selection circuit 47 selects the phase correction scheme using the pilot symbol when the discrimination result of the multilevel modulation scheme discrimination circuit 46 is 16 QAM.

The determination circuit 45 determines a bit string corresponding to the symbol whose phase is corrected in the phase correction circuit 44, based on the multilevel modulation scheme determined in the multilevel modulation scheme determination circuit 46. More specifically, the decision circuit 45 converts each complex symbol whose phase is corrected in the phase correction circuit 44 into a soft decision value by demapping according to the multilevel modulation scheme discriminated in the multilevel modulation scheme discrimination circuit 46. Thus, the decision circuit 45 outputs the bit string of the data received from the transmitter 20 to a data processing circuit, not shown, inside the receiver 40 or outside the receiver 40.

As described above, the receiver 40 of the present embodiment performs phase correction using remodulation when the multilevel degree of the multilevel modulation scheme (QPSK or 16QAM) used in the transmitter 20 is smaller than a predetermined value (multilevel degree corresponding to 16 QAM). When the multi-level degree of the multi-level modulation scheme used in the transmitter 20 is equal to or greater than a predetermined value (multi-level degree corresponding to 16QAM), the receiver 40 performs phase correction using the pilot symbol.

In this way, according to the receiver 40, the phase correction scheme is selected based on the multi-level degree of the multi-level modulation scheme used in the transmitter 20. Therefore, the phase correction can be optimally performed according to the multilevel modulation scheme used in the transmitter 20.

Claims (5)

1. A receiver, characterized in that,

used in an adaptive modulation communication system together with a transmitter which transmits a modulation signal generated using a multilevel modulation scheme selected from a plurality of multilevel modulation schemes based on a predetermined reference,

the modulated signal having a symbol string representing data to be transmitted to the receiver;

determining a correspondence between the symbol and the bit string according to a multilevel modulation scheme selected by the transmitter;

the receiver includes a multi-value modulation scheme determination unit, a phase correction scheme selection unit, a phase correction unit, and a determination unit;

the multilevel modulation scheme determining section is configured to determine the multilevel modulation scheme used for the modulation signal based on the modulation signal received from the transmitter;

the phase correction scheme selection unit is configured to select a phase correction scheme to be used for phase correction of a symbol of the modulation signal from among a plurality of phase correction schemes prepared in advance, based on the multilevel degree of the multilevel modulation scheme determined by the multilevel modulation scheme determination unit;

the phase correction unit is configured to correct the phase of the symbol using the phase correction method selected by the phase correction method selection unit;

the determination unit is configured to determine a bit string corresponding to the symbol whose phase has been corrected in the phase correction unit, based on the multilevel modulation scheme determined by the modulation scheme determination unit.

2. The receiver of claim 1,

the transmitter has a primary modulation mode and a secondary modulation mode;

a primary modulation scheme for generating a primary modulation symbol indicating the symbol, the primary modulation scheme being a multi-value modulation scheme selected from a plurality of multi-value modulation schemes having different values in accordance with a predetermined reference;

a secondary modulation scheme which is a multi-carrier modulation scheme and which generates a secondary modulation symbol by superimposing a plurality of subcarriers constituting a complex amplitude on the primary modulation symbol, thereby constituting the modulated signal composed of the plurality of secondary modulation symbols;

the modulation signal has a pilot symbol at every predetermined time;

a secondary modulation symbol for which the pilot symbol is known to the receiver;

the known secondary modulation symbols are formed by subcarriers having known complex amplitudes;

the secondary modulation symbol comprises a pilot frequency subcarrier;

the pilot subcarriers are subcarriers known to the receiver;

the known subcarriers have known complex amplitudes;

the phase correction method selection unit is configured to select a1 st phase correction method for correcting the phase of the symbol using the pilot symbol if the multilevel degree of the multilevel modulation method determined by the multilevel modulation method determination unit is smaller than a predetermined value, and to select a 2 nd phase correction method for correcting the phase of the symbol using the pilot subcarrier if the multilevel degree is greater than or equal to the predetermined value.

3. The receiver of claim 2,

the phase correction method selection unit is configured to select only the 1 st phase correction method if the multilevel degree of the multilevel modulation method determined by the multilevel modulation method determination unit is smaller than a predetermined value, and to select both the 1 st phase correction method and the 2 nd phase correction method if the multilevel degree is equal to or greater than the predetermined value.

4. The receiver of claim 3,

the phase correction unit is configured to correct the phase of the symbol in accordance with the 2 nd phase correction method and then correct the phase of the symbol in accordance with the 1 st phase correction method when both the 1 st phase correction method and the 2 nd phase correction method are selected by the phase correction method selection unit.

5. The receiver of claim 2,

the predetermined value is set according to a transmission efficiency when the phase correction unit corrects the phase of the symbol using the phase correction method selected by the phase correction method selection unit.