JP2005079632A - Channel estimation apparatus and channel estimation method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a channel estimation apparatus for bringing efficiency to averaging effects while suppressing increase in a capacity of a RAM. <P>SOLUTION: A correlation unit 101 carries out multiplication of the spread codes being the same as at transmission to perform despreading to a pilot sequence and a correlation unit 102 performs despreading to a data sequence. A frequency error estimate section 107 estimates a phase rotation due to a frequency shift between transmission and reception, a circuit 104 eliminates the phase rotation component due to the estimated frequency shift between the transmission and the reception, an in-phase summating circuit 105 averages the channel estimate value, and a circuit 106 again adds the phase rotation component due to the frequency shift between the transmission and the reception to the channel estimate value after the averaging. Further, a TCXO 108 is controlled on the basis of the estimated value to regulate the reference frequency. On the basis of this procedure, after the phase rotation component due to the detected frequency shift between the transmission and the reception is subtracted, the in-phase summating circuit 105 carries out in-phase summation. Thus, even when a fading vector FV is asymmetrically averaged, deterioration in the characteristic is suppressed to reduce the capacity of the RAM and a high averaging effect can be obtained while suppressing the increase in the capacity of the RAM. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a channel estimation apparatus and a channel estimation method used in a digital wireless communication system, particularly in a CDMA (Code Division Multiple Access) system. More specifically, the channel estimation apparatus and the channel estimation device are designed to improve the efficiency of the averaging effect over a long period while suppressing the increase in the RAM capacity by performing the averaging after removing the influence due to the correction deviation of the AFC from the channel estimation value. The present invention relates to a channel estimation method.

  In channel estimation, the influence of a signal on the channel is estimated using a known signal (pilot signal). However, the accuracy of channel estimation deteriorates due to the effect of phase rotation and fading caused by the frequency shift between the transmission side and the reception side. Therefore, conventionally, for the purpose of eliminating the influence of a noise signal in order to improve channel estimation accuracy, a method is used in which channel estimation results obtained from a plurality of pilot symbols before and after are added in phase and averaged.

FIG. 4 is a block diagram illustrating a configuration example of a conventional channel estimation apparatus. 5 to 9 are diagrams for explaining a procedure example of the conventional channel estimation process and its problems.
In this conventional example, the correlator 401 performs despreading processing on the received signal using the same spreading code as the spreading code used at the time of transmission. The FV obtained by this despreading process is sent to the in-phase addition circuit 403. This FV is a fading vector (FV) obtained by multiplying the pilot sequence obtained by the correlator 401 by the conjugate of the ideal pilot sequence. As shown in FIG. 7, the FV obtained here is a vector f (i) that rotates in a certain direction resulting from a frequency shift between transmission and reception, and a relatively random vector F (i that occurs due to the influence of fading in the propagation path. ) And a multiplication value.

  In the in-phase addition circuit 403 in FIG. 4, in order to improve the channel estimation accuracy by removing the influence of noise and instantaneous fluctuations, a process of adding and averaging for a predetermined period FV is performed.

In the conventional technique, pilot symbols before and after a pilot symbol corresponding to a data symbol to be decoded are used. However, in this case, as shown in FIG. 5, a channel estimation value obtained from a pilot symbol received at a time later than the reception time of the data symbol to be decoded is used. As a result, data symbols received simultaneously with pilot symbols used for channel estimation have a time difference from when they are received until when decoding is actually started. For this reason, it is necessary to temporarily store the data in the RAM, and there is a problem that the size of the RAM increases in proportion to the length of the averaging time.
Further, in order to solve the above problems, as shown in FIG. 6, the averaging process is performed using only the FV obtained from the pilot symbol received at a time earlier than the data symbol to be decoded. Consider the case. FIG. 8 shows an example of processing results when the same number of front and rear pilot symbols are used in-phase addition with the data symbol position to be decoded as the center. FIG. 9 shows an example of processing results when in-phase addition is performed using only pilot symbols preceding the data symbol position to be decoded. For simplicity, it is assumed that F (i) is a constant value in FIGS.

  In this case, av {FV (n)} that is the FV after averaging should ideally face the same direction as FV (n) that is the decoding target symbol position. As shown in FIG. 8, when using the same number of pilot symbols before and after the data symbol position to be decoded, the f (i) component is just canceled and av {FV (n)} becomes FV (n) Facing the same direction. However, as shown in FIG. 9, when only in-phase addition is performed using pilot symbols preceding the decoding target data symbol position, the f (i) component does not cancel each other. For this reason, there is a problem that the estimated av {FV (n)} is directed in a direction different from the original direction FV (n) and the characteristics are deteriorated.

In the prior art example related to the above problem, more accurate channel estimation is performed by controlling the weighting coefficient at the time of channel estimation (Patent Document 1), or the weighting coefficient calculation circuit performs weighting according to the fD detection value. The coefficient is calculated to improve the channel estimation system (Patent Document 2), the channel estimation using the pilot channel is used (Patent Document 3), the frequency error variance of the received signal is calculated, and the channel estimation is performed. (Patent Document 4).
JP 2000-078111 A JP 2002-033780 A JP 2002-057601 A Japanese Patent Laid-Open No. 2002-217783 JP 2002-300084 A JP 2002-344354 A JP 2002-359607 A

  The present invention has been devised in view of the above problems, and removes the component f (n) of frequency deviation between transmission and reception before performing in-phase addition averaging, and again by frequency deviation after averaging. A channel estimation value is calculated by adding the phase rotation. Accordingly, it is an object of the present invention to provide a channel estimation device and a channel estimation method that improve the accuracy of channel estimation without being affected by a frequency shift between transmission and reception while suppressing an increase in RAM capacity.

  The channel estimation apparatus according to the first aspect of the present invention includes a first correlator (101) that performs a despreading process on a pilot sequence by multiplying the same spreading code as that at the time of transmission, and a despreading process on a data sequence. A second correlator (102) to perform, a frequency error estimator (107) for estimating a phase rotation due to a frequency shift between transmission and reception, and adjusting a reference frequency based on the estimated value, and a frequency between transmission and reception It is characterized by having a channel estimation unit (103) that removes the phase rotation component due to deviation and adds it again after averaging.

  According to a second aspect of the present invention, in the channel estimation device according to the first aspect, a phase rotation removal circuit (104) for removing a phase rotation due to a frequency shift between transmission and reception estimated by the frequency error estimation unit (107). It is characterized by further having.

  The invention according to claim 3 is the channel estimation device according to claim 1 or 2, further comprising a channel estimation unit (103) for estimating phase rotation due to frequency deviation between transmission and reception, and the channel estimation unit ( 103) is a circuit (104) for removing the phase rotation due to the frequency shift between transmission and reception estimated by the frequency error estimation unit (107), an in-phase addition circuit (105) for averaging, and a channel after averaging And a circuit (106) for re-adding a phase rotation due to a frequency shift between transmission and reception to the estimated value.

  According to a fourth aspect of the present invention, there is provided the channel estimation device according to the second or third aspect, wherein a synchronous detection circuit (110) that rotates the phase of the data symbol according to the obtained channel estimation result, and a data symbol in the received signal Is further provided with a RAM (109) that temporarily stores until the decoding is started.

  The channel estimation method according to the fifth aspect of the present invention includes a first correlation step for performing despreading processing of a pilot sequence by multiplying the same spreading code as at the time of transmission, and second processing for performing despreading processing of a data sequence. A correlation step, a frequency error estimation step for estimating a phase rotation due to a frequency shift between transmission and reception, and adjusting a reference frequency based on the estimated value, and a phase rotation component due to a frequency shift between transmission and reception are removed and averaged It is characterized by having a channel estimation step to be added later.

  A sixth aspect of the present invention is the channel estimation method according to the fifth aspect, further comprising a phase rotation removal step for removing a phase rotation due to a frequency shift between transmission and reception estimated in the frequency error estimation step. It is structured.

  A seventh aspect of the present invention is the channel estimation method according to the fifth or sixth aspect, further comprising a channel estimation step of estimating a phase rotation due to a frequency shift between transmission and reception, wherein the channel estimation step includes a frequency error. Phase rotation removal step for removing phase rotation due to frequency deviation between transmission and reception estimated in estimation step, In-phase addition step for averaging, and phase rotation due to frequency deviation between transmission and reception for the channel estimation value after averaging And a step of adding the phase rotation again for adding the minute again.

  According to an eighth aspect of the present invention, in the channel estimation method according to the sixth or seventh aspect, the synchronous detection step of rotating the phase of the data symbol according to the obtained channel estimation result, and the decoding of the data symbol in the received signal are performed. And a data symbol temporary storage step for temporarily storing data until it is started.

  The channel estimation apparatus and the channel estimation method of the present invention can obtain a high averaging effect by suppressing deterioration of characteristics while suppressing an increase in RAM capacity.

  Embodiments of a channel estimation apparatus and a channel estimation method according to the present invention will be described in detail with reference to the accompanying drawings. 1 to 4, an embodiment of a channel estimation apparatus and a channel estimation method according to the present invention is shown. FIG. 1 is a block diagram illustrating a configuration example of a channel estimation apparatus applied to the first embodiment. FIG. 2 is a block diagram showing a more detailed circuit configuration example of the frequency error estimation unit 107 in FIG. FIG. 3 is a block diagram of a channel estimation apparatus showing a variation / application example corresponding to FIG.

  In FIG. 1, phase rotation due to a frequency shift between transmission and reception detected by the frequency error estimation unit 107 from a fading vector (FV) obtained by multiplying the pilot sequence obtained by the correlator 101 by the conjugate of the ideal pilot sequence. After subtracting the minutes, the in-phase addition circuit 105 performs in-phase addition. Accordingly, even if FV is asymmetrically averaged as shown in FIG. 6 in order to reduce the RAM capacity, a high averaging effect can be obtained while suppressing an increase in RAM capacity while suppressing deterioration of characteristics.

(Configuration of Example)
Example 1 is shown in FIG. The channel estimation apparatus shown in FIG. 1 includes correlators 101 and 102, a channel estimation unit 103, a circuit 104 (removing the phase rotation), an in-phase addition circuit 105, a circuit 106 (adding the phase rotation again), A frequency error estimation unit 107, a TCXO (temperature compensated crystal oscillator) 108, a RAM 109, and a synchronous detection circuit 110 are configured.
In the channel estimation apparatus configured by each of these components, in-phase addition is performed using the in-phase addition circuit 105. Details of these configurations will be described below.

In the channel estimation apparatus of FIG. 1, correlator 101 performs despreading processing of a pilot sequence by multiplying the same spreading code as that at the time of transmission.
Correlator 102 performs despreading processing on the data series.
The channel estimation unit 103 removes a phase rotation due to a frequency shift between transmission and reception estimated by the frequency error estimation unit 107, an in-phase addition circuit 105 that performs averaging, and a channel estimation value after averaging. And a circuit 106 for re-adding the amount of phase rotation due to the frequency deviation between them.
The frequency error estimator 107 estimates phase rotation due to frequency deviation between transmission and reception. Further, the TCXO 108 is controlled based on the estimated value, and the reference frequency is adjusted.
The RAM 109 temporarily stores data symbols in the received signal until decoding is started.
The synchronous detection circuit 110 rotates the phase of the data symbol based on the obtained channel estimation result.

  FIG. 2 shows a more detailed circuit configuration example of the frequency error estimation unit 107. In FIG. 2, the frequency error estimation unit 107 includes a frequency error estimation circuit 201, a phase correction value calculation circuit (1) 202 that calculates the amount of phase rotation to be removed from the FV based on the estimated frequency error, and an in-phase addition. A phase correction value calculation circuit (2) 203 for calculating the rotation amount of the phase to be added again to the subsequent FV, and a TCXO correction value calculation circuit 204 for calculating the correction value of TCXO from the frequency error. .

(Description of operation of the first embodiment)
An actual operation example will be described below with reference to the drawings.
In FIG. 1, the received signal is first input to correlators 101 and 102. Correlators 101 and 102 perform despreading processing by multiplying the same spreading code as at the time of transmission.
In the CDMA system, a signal is spread by a specific spreading code, and signals spread by a plurality of different spreading codes are multiplexed (code multiplexing). By multiplying again the same spreading code as that used at the time of transmission on the receiving side, the signal before spreading can be taken out. In this embodiment, it is assumed that one pilot signal sequence and one data sequence are code-multiplexed with the received signal.

  The received signal is received with the phase rotated due to the frequency shift between the transmitting side and the receiving side and the influence of fading received on the propagation path. For this reason, channel estimation is required. In channel estimation, a known signal sequence (pilot sequence) is multiplexed with a data sequence in advance on both the transmission side and the reception side, and the fluctuation is estimated on the reception side. This estimates how the data sequence is affected by the channel.

  The pilot signal sequence extracted by the despreading process by which the correlator 101 multiplies the same spreading sequence as that used in the spreading process on the transmission side is an ideal pilot sequence before being affected by the channel for each symbol. The phase variation received by the channel is calculated as a fading vector (FV). Here, as shown in FIG. 7, FV obtained from the i-th symbol is generated by f (i), which is a constant phase change caused by a frequency shift between the transmission side and the reception side, and fading in the propagation path. It is detected as a multiplication value with F (i) which is a relatively random phase change. The obtained FV is output to the circuit 104 and the frequency error estimation unit 107.

The frequency error estimator 107 adds the difference between the FV and the previous FV in phase over a relatively long period of time, and estimates f (i), which is a phase change in a certain direction. The estimated phase change f (i) is output to the circuit 104.
In the circuit 104, f (i) that is a phase rotation change due to a frequency shift from FV is removed by reverse rotation, and F (i) that is an influence of fading in the propagation path is calculated.

The FV, which is the influence of the channel estimated by the change of the pilot symbol, includes instantaneous fluctuations and noise as it is. Such instantaneous fluctuations and noise cause deterioration of characteristics. For this reason, in order to reduce the influence, a method of averaging a plurality of FVs over a certain period is used.
The obtained phase rotation change F (i) is output to the in-phase addition circuit 105, averaged as shown in FIG. 6 for a certain period, for example, between 5 symbols, and output to the circuit 106.

The circuit 106 adds f (n) estimated by the frequency error estimator to the averaged F (n) by phase rotation again to calculate a channel estimation value, and outputs it to the synchronous detection circuit 110.
The above channel estimation value represents the influence of the pilot sequence on the channel, and it is considered that the data sequence multiplexed and transmitted is also subjected to the same phase fluctuation. Therefore, in the synchronous detection circuit 110, the phase of the data symbol to be decoded once stored in the RAM 109 is reversely rotated by the phase of the channel estimation value.

(Effect of Example 1)
The first effect is that in the process of averaging FV, an increase in RAM capacity can be suppressed and an increase in circuit scale can be prevented. This is because the phase rotation component caused by the frequency shift between the transmitting side and the receiving side is once removed from the FV estimated from the pilot symbols and then averaged.
Based on the above procedure, by removing the phase rotation component in advance, it is not necessary to use the same number of pilot symbols before and after in time with reference to the data symbol to be decoded as shown in FIG. As a result, since averaging is performed using only the pilot symbols that are temporally previous as shown in FIG. 6, it is not necessary to temporarily store the data symbols in the RAM, and the required RAM capacity can be reduced.

  In the first embodiment of the present invention, channel estimation in the CDMA system has been described. However, the general wireless method is not limited to the CDMA method, and may be applied to other wireless methods. In the first embodiment, the case where one pilot sequence and one data sequence are multiplexed on the received signal has been described for the sake of simplicity. However, the number of data series in the received signal is not limited to this, and a plurality of data series may be multiplexed as shown in FIG. A more specific example of a change example and an application example will be described below.

In FIGS. 5 and 6 showing the first embodiment, the symbol rate of the data series and the pilot series are the same. However, the symbol rates need not be the same. Specific examples are as follows.
Although the averaging period in the in-phase addition circuit 105 is 14 symbols, the averaging period is not limited.
The averaging in the in-phase addition circuit 105 is not limited to simple averaging but may be averaging including weighting.

It is a block diagram which shows the structural example of the channel estimation apparatus applied to Example 1. FIG. 3 is a block diagram illustrating a more detailed circuit configuration example of a frequency error estimation unit 107. FIG. It is a block diagram of the channel estimation apparatus which shows the example of a change and an application example. It is a block diagram which shows the structural example of the conventional channel estimation apparatus. It is a figure for demonstrating the example of a channel estimation procedure of the conventional channel estimation apparatus. It is a figure for demonstrating the example of a channel estimation procedure of the conventional channel estimation apparatus. It is a figure which shows the example of distribution of the general fading vector (FV) of a received signal. An example of processing results when the same number of pilot symbols before and after is used is shown. An example of processing results when only the previous pilot symbol is used is shown.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101,102 Correlator 103 Channel estimation part 104 (The phase rotation part is removed) circuit 105 In-phase addition circuit 106 (Phase rotation part is added again) circuit 107 Frequency error estimation part 108 TCXO (temperature compensated crystal oscillator / temperature compensation type) Crystal oscillator)
109 RAM
DESCRIPTION OF SYMBOLS 110 Synchronous detection circuit 201 Frequency error estimation circuit 202 Phase correction value calculation circuit 203 Phase correction value calculation circuit 204 TCXO correction value calculation circuit

Claims (8)

A first correlator for performing a despreading process of a pilot sequence by multiplying the same spreading code as at the time of transmission;
A second correlator for performing despreading processing of the data sequence;
A frequency error estimator that estimates phase rotation due to frequency deviation between transmission and reception, and adjusts a reference frequency based on the estimated value;
A channel estimation apparatus comprising: a channel estimation unit that removes a phase rotation component due to a frequency shift between transmission and reception and adds it again after averaging.   The channel estimation apparatus according to claim 1, further comprising a phase rotation removal circuit for removing a phase rotation due to a frequency shift between transmission and reception estimated by the frequency error estimation unit.   A channel estimator for estimating phase rotation due to frequency shift between the transmission and reception; and the channel estimation unit; a circuit for removing phase rotation due to frequency shift between transmission and reception estimated by the frequency error estimation unit; 2. A circuit comprising: an in-phase addition circuit that performs averaging; and a circuit that re-adds a phase rotation due to a frequency shift between transmission and reception to an averaged channel estimation value. 2. The channel estimation apparatus according to 2.   A synchronous detection circuit that rotates the phase of the data symbol based on the obtained channel estimation result; and a RAM that temporarily stores the data symbol in the received signal until decoding is started. The channel estimation apparatus according to claim 2 or 3, characterized in that A first correlation step of multiplying the same spreading code as at the time of transmission and performing despreading processing of the pilot sequence;
A second correlation step for performing despreading processing of the data series;
A frequency error estimation step of estimating a phase rotation due to a frequency shift between transmission and reception, and adjusting a reference frequency based on the estimated value;
A channel estimation method comprising: a channel estimation step of removing a phase rotation component due to a frequency shift between transmission and reception and adding it again after averaging.   6. The channel estimation method according to claim 5, further comprising a phase rotation removal step for removing a phase rotation due to a frequency shift between transmission and reception estimated in the frequency error estimation step.   A channel estimation step for estimating a phase rotation due to a frequency shift between the transmission and reception, wherein the channel estimation step removes a phase rotation due to a frequency shift between transmission and reception estimated in the frequency error estimation step 107; A rotation removal step, an in-phase addition step for performing averaging, and a phase rotation re-addition step for re-adding a phase rotation due to a frequency shift between transmission and reception to the averaged channel estimation value. The channel estimation method according to claim 5 or 6, wherein   A synchronous detection step of rotating the phase of the data symbol according to the obtained channel estimation result, and a data symbol temporary storage step of temporarily storing the data symbol in the received signal until decoding is started, The channel estimation method according to claim 6 or 7, further comprising:

Images