CN1312625A - Dynamic average length regulating method and device for channel estimation - Google Patents

Abstract

The method of the present invention comprises the following steps: (1) estimating the maximum Doppler frequency shift of the mobile channel; (2) calculating the optimal average length according to the maximum Doppler frequency shift; (3) dynamically adjusting according to the calculated optimal average length . The device of the present invention includes an instantaneous channel parameter estimator 1, a maximum Doppler frequency shift estimator 2, an optimal average length calculator 3 and M data accumulation averagers 4, wherein the instantaneous channel parameter estimator 1 is simultaneously synchronized with the maximum Doppler frequency shift The frequency shift estimator 2 is connected with M data accumulation averagers 4 . The invention is simple and easy to calculate, and can be used in any CDMA mobile communication system with continuous pilot frequency, including 3GPP WCDMA and 3GPP2 CDMA2000 systems. The estimation error caused by channel estimation with fixed average length is overcome, and the receiving performance of RAKE receiver is greatly improved.

Description

A method and device for dynamically adjusting average length for channel estimation

The invention belongs to the field of CDMA (Code Division Multiple Access) cellular communication system.

CDMA cellular communication technology shows great development potential due to its simple frequency planning, large system capacity, strong anti-multipath ability, good communication quality, and low electromagnetic interference. It is the mainstream technology of future mobile communication including the third generation. CDMA spread spectrum signal receivers are divided into coherent receivers and non-coherent receivers. The coherent receiver needs to know the phase information of the received signal, while the non-coherent receiver does not need the phase information of the received signal, but requires the transmitted signal to be in a quadrature modulation mode. The present invention mainly considers the coherent receiving mode that will dominate in the future CDMA cellular system.

There is multipath fading phenomenon in the mobile communication system, which will cause serious multipath interference. In the CDMA cellular mobile communication system using spread spectrum technology, by receiving the pilot (Pilot) signal with definite information, the amplitude and phase information of the multipath signal can be estimated, so that the multipath diversity and coherent reception become possible. The coherent spread spectrum receiver that performs diversity processing for multipath fading signals is called a RAKE receiver, which can perform phase correction and maximum ratio combining processing on multiple single-path signals that carry the same information and have independent fading characteristics, so as to achieve The purpose of overcoming multipath fading and improving the received signal-to-interference ratio.

In order to achieve coherent reception, it is necessary to estimate the time-varying parameters of the fading channel, that is, to perform channel estimation. The requirements for channel estimation have two aspects: timeliness and accuracy. Timeliness means that when the channel changes rapidly, channel estimation can quickly track the change of the channel; Improve the accuracy of channel estimation. In the CDMA system of the continuous pilot system, the classic method used for channel estimation is that each branch of the RAKE receiver uses the sampling values of multiple continuous channel data estimated by the pilot symbols to average to suppress the estimation noise, and the average interval The larger the value, the smaller the interference caused by the noise. But on the other hand, this method of averaging in a certain interval is based on the fact that the channel parameters change slowly in this range, and this relatively slow-changing estimation interval of the channel parameters varies with the moving speed of the mobile station (Doppler Le frequency shift), the greater the moving speed, the smaller this interval, and vice versa. Considering comprehensively, there is a relative balance point in the length of channel estimation, which can take into account both the time-varying characteristics of the channel and the suppression of noise. In the RAKE receiver of the traditional continuous pilot CDMA system, the average length of the channel estimator is set as a fixed value. For the third-generation integrated service mobile communication system, if a fixed average length is adopted, the greater the moving speed, the shorter the time for the channel to be relatively stable, and the shorter the average length of the channel estimator. As the length shortens, the noise component contained in the channel estimation value gradually increases, and the accuracy of the channel estimation decreases. In order to make the RAKE receiver achieve the speed adaptability from static to 500 km/h, the method of channel estimation using a fixed average length is not the best solution, so we must dynamically adjust the estimated length.

The purpose of the present invention is to determine the optimal average length for channel estimation according to the different moving speeds of mobile stations, thereby establishing a dynamic modulation for the average length of channel estimation, thereby overcoming the estimation error caused by performing channel estimation with a fixed average length, Greatly improve the receiving performance of RAKE receiver.

Method and device of the present invention are respectively realized according to the following scheme:

The present invention proposes an adjustment method for dynamically adjusting the estimated average length of the channel. The basic idea is: first estimate the maximum Doppler frequency shift of the channel, and then use the closed formula of the optimal channel estimated average length to dynamically adjust the average length.

The method of the present invention comprises the following steps: (1) estimating the maximum Doppler frequency shift of the mobile channel; (2) calculating the optimal average length according to the maximum Doppler frequency shift; (3) dynamically adjusting according to the calculated optimal average length .

The following combined formulas are introduced separately.

1. Estimation of channel maximum Doppler frequency shift: Pilot channel in CDMA system is used to transmit pre-known pilot sequence, which can be used for system timing and carrier extraction, channel estimation, handover, etc.

Let the number of finger peaks used by the RAKE receiver be L, and consider the lth finger peak. The received pilot channel data is despread in correlation with the scrambling code and channel code of the transmitted signal pilot channel, and then the received pilot symbol sequence rl(n) affected by the fading channel is obtained, namely

r _l (n)=d _p (n)c _l (n)+v(n)

c _l (n)=c _lI (n)+jc _lQ (n) [Formula 1]

是c_l(n)的符号瞬时估值，z(n)是v(n)引入的估计白噪声，其方差为σ² _n。d _p (n)=d _pI (n)+jd _pQ (n) Among them, d _p (n) is the transmission symbol of the pilot channel, c _l (n) is the channel parameter of the nth symbol interval of the lth path, v (n) is compound additive white Gaussian noise. The sampling time interval of r _l (n) and v(n) is the symbol interval _Δt . The channel parameter sequence estimated from the pilot symbols is: in,

is the symbolic instantaneous estimate of c _l (n), z(n) is the estimated white noise introduced by v(n), and its variance is σ ² _n .

ω_d的估计为： Let x _l,T (n) be The envelope of , we use the value of x _l,T (n) in the interval 1≤n≤N to estimate the maximum Doppler shift ω _d of the channel. The RAKE receiver can distinguish very small multipath time delay, and can search for the strongest path among them, and set the strongest path as the lth path. Construct the variable G as follows:

The estimate of _ωd is:

2. Calculate the best average length

When using continuous pilots to estimate the channel parameters of the nth point, the average method is adopted, that is, Input into an average system model with an average length of 2N+1, as shown in Figure 1.

其中，ε_s(n)是由于平均长度所带来的误差，ε_z(n)是由于噪声所带来的误差。ε_Total(n)是两种误差的总和。From this, the estimated sequence of new channel parameters is obtained Consists of three parts on the right-hand side of the following equation:

Among them, ε _s (n) is an error caused by the average length, and ε _z (n) is an error caused by noise. ε _Total (n) is the sum of the two errors.

The present invention provides a complete expression of error average power. ω _d is the maximum Doppler frequency shift of the channel. Therefore, the mean square error due to averaging is:

When ω _d and symbol interval Δ _t are known, the optimal average time 2N+1 at different speeds can be obtained from the above formula. Assuming M=2N+1, it can be seen from [Formula 6] that the best M is the integer that makes the following formula the smallest where a=6.51×10 ^-4 (ω _d Δt) ⁴ σ ² _l , b=σ ² _n [Formula 8] Using the basic inequality, it is easy to get $m=\sqrt[5]{b/4a}$ When [Formula 7] takes the minimum value, the average length shown in Figure 1 can be taken as

3. After the optimal average length is obtained, the instantaneous channel parameters are dynamically adjusted by the data accumulative averager, that is, the instantaneous channel parameter estimates are accumulated, summed and averaged.

The device of the present invention is shown in Fig. 2, and it comprises instantaneous channel parameter estimator 1, maximum Doppler frequency shift estimator 2, optimal average length calculator 3 and M data accumulative averager 4, wherein instantaneous channel parameter estimator 1 is connected with the maximum Doppler frequency shift estimator 2 and the M data accumulation averager 4 at the same time.

Advantages of the present invention:

(1) The calculation of the present invention is simple and easy, and can be used for any CDMA mobile communication system with continuous pilots, including 3GPP WCDMA and 3GPP2 CDMA2000 systems.

(2) The estimation error caused by channel estimation with fixed average length is overcome, and the receiving performance of the RAKE receiver is greatly improved.

Description of drawings:

Figure 1 is a schematic diagram of estimating the channel parameters of the nth point by the continuous pilot averaging method

Figure 2 is a general block diagram of a device for dynamically adjusting the optimal average length for channel estimation

Figure 3 shows the instantaneous channel parameter estimator

Figure 4 shows the maximum Doppler shift estimator

Figure 5 is the best average length calculator for channel estimation

Figure 6 is M data accumulative averager

Figure 7 is an implementation example of a device for channel estimation by dynamically adjusting the optimal average length

Embodiments are described in detail below in conjunction with the device of the present invention:

The device of the present invention is shown in Fig. 2, and it comprises instantaneous channel parameter estimator 1, maximum Doppler frequency shift estimator 2, optimal average length calculator 3 and M data accumulative averager 4, wherein instantaneous channel parameter estimator 1 is connected with the maximum Doppler frequency shift estimator 2 and the M data accumulation averager 4 at the same time.

The instantaneous channel parameter estimator 1 includes: a complex multiplication device 101, as shown in FIG. 3 ;

The maximum Doppler frequency shift estimator 2 includes: a complex number modulo device 201, a delay device 202, a subtractor 203, a squarer 204, a squarer 205, N-1 data accumulation and averaging devices 206, and N data accumulation and averaging devices 207, divider 208, find quadratic root device 209, multiplier 210, and subtractor (203) is connected with delay device (202), complex number modulus device (201) and squarer (204) simultaneously, divider (208) is simultaneously connected with N data accumulative average devices (207), N-1 data accumulative average devices (206) and seeking quadratic root device (209), as shown in Figure 4;

Channel estimation optimum average length calculator 3 comprises: multiplier 301, quartic device 302, multiplier 303, finds 5th root device 304, and these four devices are connected in series as shown in Figure 5;

The M data accumulative averagers 4 include: M data accumulative averagers 401 , as shown in FIG. 6 .

Devices 1, 2, 3, and 4 are introduced respectively below

For device 1, its input is the complex channel parameters sampled at the symbol rate by the pilot channel after despreading, and the complex multiplier of 101 completes the operation shown in [Formula 2]. The reason why the multiplier is used is because the pilot Data is known. The result data of device 1 is sent to device 2 and device 4 respectively.

For device 2, its input is the estimated instantaneous channel parameters from device 1, and its output is the estimated channel maximum Doppler shift, which is sent to device 3. Among them, 201 completes the modulo operation for the input complex number, 202 and 203 complete the difference operation between two adjacent values, the result is completed by 204 for square operation, and by 206 for average operation. 205 and 207 complete the squaring and cumulative average calculation of the denominator part shown in [Formula 3]. 208 completes the division operation shown in [formula 3], and the result is sent to 209 to find the quadratic root, and at 210 completes the multiplication operation with the known data 2/Δt and obtains the maximum Doppler frequency shift and sends it to device 3 .

For device 3, its input comes from the maximum Doppler frequency shift calculated by device 2, it completes the operations shown in [Formula 8] and [Formula 9] and finds the best average length M and sends it to device 4.

For device 4, according to the optimal M value calculated by device 3, the M complex numbers sent by device 1 are summed and averaged by M data accumulative averagers 401, which is dynamic adjustment, and the result is Channel parameters estimated after dynamically adjusting the average length.

In Fig. 7, it is the implementation example that the device of the present invention is applied in the RAKE receiver, except the device of the present invention, all the other are general devices in the RAKE receiver, and its function and connection relationship are described as follows:

●Tapped delay line

The tapped delay line completes the sampling of the received baseband data and delays tap output, and the result is sent to the data channel related despreader and the pilot channel related despreader through time division multiplexing.

● Data channel correlation despreader

The data channel correlation despreader completes the correlation despreading of the data channel in the sampling data output by the tapped delay line, and the result is sent to the multiplier.

●Pilot channel correlation despreader

The pilot channel correlation despreader completes the correlation despreading of the pilot channel in the sampling data output by the tapped delay line, and the result is sent to the instantaneous channel parameter estimator.

●Multiplier

The multiplier completes the complex multiplication of the data channel data output by the data channel correlation despreader and the channel parameter estimation value output by the data averager with length M, that is, performs phase correction on the received signal. The result is sent to the maximum ratio combiner of the RAKE receiver.

Claims (4)

1. A method for dynamically adjusting an average length for channel estimation, comprising the steps of:

(1) estimating the maximum Doppler frequency shift of a mobile channel;

(2) calculating the optimal average length according to the maximum Doppler frequency shift;

(3) and dynamically adjusting according to the calculated optimal average length.

2. A method of dynamically adjusting the average length for channel estimation according to claim 1, characterized in that the dynamic adjustment is accumulating, summing and averaging the instantaneous channel parameter estimates.

3. A device for dynamically adjusting average length to estimate channel is characterized by comprising an instantaneous channel parameter estimator (1), a maximum Doppler shift estimator (2), an optimal average length calculator (3) and M data accumulation averagers (4), wherein the instantaneous channel parameter estimator (1) is simultaneously connected with the maximum Doppler shift estimator (2) and the M data accumulation averagers (4).

4. An apparatus for dynamically adjusting average length for channel estimation according to claim 3, wherein:

(1) the instantaneous channel parameter estimator (1) comprises: a complex multiplication device (101);

(2) the maximum Doppler shift estimator (2) comprises: the device comprises a complex modulus calculating device (201), a time delay device (202), a subtracter (203), a squarer (204), a squarer (205), N-1 data accumulation averaging devices (206), N data accumulation averaging devices (207), a divider (208), a quadratic root calculating device (209) and a multiplier (210), wherein the subtracter (203) is simultaneously connected with the time delay device (202), the complex modulus calculating device (201) and the squarer (204), and the divider (208) is simultaneously connected with the N data accumulation averaging devices (207), the N-1 data accumulation averaging devices (206) and the quadratic root calculating device (209);

(3) the channel estimation optimal average length calculator (3) comprises: a multiplier (301), a fourth power device (302), a multiplier (303) and a 5 th power root solving device (304), wherein the four devices are connected in series;

(4) the M data accumulation averagers (4) comprise: m data accumulation averagers (401).

Images