CN1213558C - Up-link multipath tracking method for WCDMA system - Google Patents

Abstract

The present invention provides an uplink multi-path tracing method for wide band code division multiple access (WCDMA) systems, which adjusts the sampling phase of a path according to the signal energy of an early path position, a middle path position and a late path position of the traced path. If the interval between the traced path and other paths is smaller than a setting value, a numerical value corrected according to the following steps is used as the signal energy of the early path position, the middle path position and the late path position of the traced path: (1) determining the energy of other paths in the early path position, the middle path position and the late path position of the traced path; (2) respectively subtracting the energy of determined in step (1) in the early path position, the middle path position and the late path position from the signal energy of the early path position, the middle path position and the late path position of the traced path.

Description

Uplink multipath tracking method for Wideband Code Division Multiple Access (WCDMA) system

Technical Field

The present invention relates to a multipath tracking method in a third generation mobile communication system, particularly to an uplink multipath tracking method for a Wideband Code Division Multiple Access (WCDMA) system.

Background

The propagation of radio waves in land mobile channels is characterized by multiple reflections, diffraction and attenuation of signal energy. This is caused by the inevitable presence of obstacles in space (such as buildings, mountain bags, etc.), resulting in what is known as multipath propagation, i.e. the signal transmitted from the transmitter reaches the receiver via a plurality of different transmission paths, and the receiver will receive the transmission signals of the plurality of paths with different fading degrees and different phases with different delays. The signal-to-noise ratio of the received signal can be improved if these multipath signals can be combined appropriately.

WCDMA is a wideband direct-spread code division multiple access system that spreads user information over a wide bandwidth by multiplying the user data with a CDMA spreading code. WCDMA supports two basic modes of operation: frequency Division Duplex (FDD) and Time Division Duplex (TDD). In FDD mode, the uplink (mobile to base station connection) and downlink (base to mobile connection) use two separate 5MHz carriers, respectively, and in TDD mode only one 5MHz carrier is used.

Fig. 1 is a block diagram of a WCDMA uplink baseband receiving system. As shown in fig. 1, the baseband I-branch and Q-branch signals subjected to the front-end demodulation process are input to a multipath searching unit 11, which determines the energy, relative time position or phase, and noise power of each path signal belonging to the same user. A finger management unit (12) assigns path positions to be tracked to the multipath tracking modules (1 to 4) in a RAKE reception unit (13) on the basis of the multipath signal positions outputted from a multipath search unit (11). The RAKE receiving unit 13 includes a plurality of multipath tracking modules 1 to 4 and a maximum ratio combining unit (MRC)14, the multipath tracking modules 1 to 4 track the path positions allocated by the finger management unit 12, further determine the accurate positions of the paths, and output coherent de-spread signals on the path positions after channel estimation to the MRC unit 14, and output data signals with a large signal-to-noise ratio after maximum ratio combining. A signal-to-interference ratio (SIR) unit 15 determines the SIR of the signal based on the baseband I and Q branch signals and the output of the finger management unit 12.

In the baseband receiving system shown in fig. 1, a two-stage synchronization method is actually adopted, that is, multipath searching is first performed to obtain signals of each path in a wide range, then multipath tracking processing is performed, and a coarse position of each path obtained by the multipath searching processing is adjusted to determine an accurate position. The multipath tracking modules 1 to 4 of fig. 1 are further described below with the aid of fig. 2.

As shown in fig. 2, each multipath tracking module includes a resampling unit 21, descrambling units 22a to 22c, noncoherent despreading units 23a to 23c, infinite impulse response filters (IIR)24a and 24b, a comparator 25, a channel estimation unit 26, and a coherent despreading unit 27.

The multipath tracking module performs multipath tracking according to the flow shown in fig. 3. In step 31, the resampling unit 21 interpolates the baseband I-branch and Q-branch signals at twice the chip rate to obtain 16 times the chip rate, and in step 32, the resampling unit 21 samples the 16 times the chip rate signals obtained after interpolation in three positions to obtain three signals, namely, early, middle and late, the initial sampling position being determined by the output of the finger management unit 12, and the sampling positions of the three signals are sequentially different by 1/4 chips in time or phase. In step 33, the descrambling units 22a to 22c descramble the intermediate path, early path, and late path signals, respectively, and the noncoherent despreading units 23a to 23c despread the real and imaginary parts of the descrambled intermediate path, early path, and late path signals with spreading codes of the DPCCH channel, respectively, and calculate the symbol energy (i.e., the energy value in one symbol interval) of each path of signals. In step 34, the infinite impulse response filter 24a performs an IIR filter process on the signal energy of the middle path to obtain a reference value, and the infinite impulse response filter 24b performs an IIR filter process on the difference between the symbol energies of the early path and the late path to obtain an error value. In step 35, the comparator 25 compares the error value with the reference value, and determines the adjustment direction of the tracking position according to the following rule:

if R is less than or equal to V, the middle path sampling position is the actual path position, and therefore no adjustment is made;

if R > V, it indicates that the mid-way sample position lags the actual path position, thus indicating that the next three sample positions are advanced in time by 1/16 chips;

if R < -V, it indicates that the mid-way sample position is ahead of the actual path position, thus indicating that the next three sample positions are shifted backward in time by 1/16 chips.

Here, R is a ratio of the error value to the reference value, and V is a preset threshold value greater than 0.

When the two paths are separated by a small distance (e.g., less than 1.5 chip intervals), there is some interference between the paths, i.e., the symbol energy at the early, middle and late sample positions of one of the paths includes the contribution of the other path. Because such inter-path interference is not considered, the method causes the sampling phases of two paths with smaller intervals to gradually approach to the same point, and finally a receiving path is lost at a receiving end, thereby reducing the system performance.

Disclosure of Invention

An object of the present invention is to provide an uplink multi-path tracking method for a Wideband Code Division Multiple Access (WCDMA) system, which can avoid path loss in tracking a path having a small interval.

The above object of the present invention is achieved by the following technical solutions:

an uplink multi-path tracking method for a wideband code division multiple access system adjusts the sampling phase of a tracked path according to the symbol energy of the positions of an early path, a middle path and a late path of the path, wherein if the interval between one tracked path and other paths is less than a set value, the value corrected according to the following steps is taken as the symbol energy of the positions of the early path, the middle path and the late path of the tracked path:

(1) determining the energy of the other paths at the early, mid and late positions of the tracked path; and

(2) and (3) respectively subtracting the energy at the positions of the early road, the middle road and the late road determined in the step (1) from the symbol energy of the positions of the early road, the middle road and the late road of the tracked path.

In the above uplink multi-path tracking method for a wideband code division multiple access system, the step (1) determines the energy of each of the other paths at the early, middle and late positions of the tracked path in the following manner:

(1a) determining the symbol energy of the path position in the other path and the weight factors of the path position at the early path, the middle path and the late path of the tracked path; and

(1b) multiplying the symbol energy of the mid-way position of the other path by its weighting factors at the early, mid and late way positions of the tracked path, respectively.

The weight factors are values of the normalized distribution function of the symbol energy of the middle road position of the other path on the time at the positions of the early road, the middle road and the late road of the tracked path.

The weighting factor is determined in the following manner:

(a) determining the time slot energy on a plurality of equally spaced sampling positions in a certain range at the left side and the right side of the other path by taking the middle path position of the other path as a center;

(b) performing multi-time slot smoothing filtering processing on the time slot energy at the plurality of equal-interval sampling positions; and

(c) and normalizing the time slot energy after the smoothing filtering processing on the plurality of equal-interval sampling positions to obtain a weight factor of the symbol energy of the other path on each equal-interval sampling position.

In the uplink multipath tracking method for the wideband code division multiple access system, the sampling position of the tracked path is adjusted according to the symbol energy of the early path, the middle path and the late path of the tracked path in the following way:

(3) respectively carrying out infinite impulse response filtering on the symbol energy of the middle path position of the tracked path and the difference between the symbol energy of the early path position and the symbol energy of the late path position;

(4) and (3) dividing the difference between the early road and the late road position symbol energy processed in the step (3) and the middle road position symbol energy, if the absolute value of the ratio is less than or equal to a preset threshold, not adjusting the sampling position of the tracked path, if the ratio is greater than the preset threshold, moving the sampling position of the tracked path forward in time, and if the ratio is less than the negative value of the preset threshold, moving the sampling position of the tracked path backward in time.

In the uplink multi-path tracking method for the wideband code division multiple access system, the set value of the path interval is 1.5 times the chip interval.

As can be seen from the above, in the multipath tracking method of the present invention, when the interval between the two paths is small, the signal energy at the sampling positions of the early path, the middle path and the late path of the tracked path is corrected by subtracting the contribution of the adjacent path from the signal energy at the sampling positions of the early path, the middle path and the late path of the tracked path, so that the existence of the two paths can be always ensured, and the loss of the receiving path is avoided.

Drawings

The objects, features and advantages of the present invention will be further understood from the following description of the preferred embodiments of the invention, taken in conjunction with the accompanying drawings, in which:

fig. 1 is a block diagram of a WCDMA uplink baseband receiving system.

Fig. 2 is a schematic block diagram of a multipath tracking module in the WCDMA uplink baseband receiving system shown in fig. 1.

Fig. 3 is a flow chart of a prior art WCDMA uplink multipath tracking process.

Fig. 4 is a diagram illustrating a frame structure of a dedicated physical channel in a WCDMA uplink.

Fig. 5 is a flow chart of a multipath tracking process in accordance with a preferred embodiment of the present invention.

FIG. 6 is a diagram illustrating the input-output relationship of the interpolation step in the preferred embodiment shown in FIG. 5.

FIG. 7 is a block diagram of the interpolation algorithm used in the preferred embodiment shown in FIG. 5.

Fig. 8a and 8b show the results of comparing the bit error rates obtained with the multipath tracking method of the present invention with those obtained with other prior art multipath tracking methods.

Detailed Description

The basic idea of the multi-path tracking method of the invention is that when the distance between two paths is small, the energy of the tracked path is corrected by deducting the contribution of the adjacent path from the energy of the tracked path, thereby achieving the purpose of eliminating the energy interference between the paths.

Preferred embodiments of the present invention are described below with the aid of the accompanying drawings. There are three physical channels in the uplink in WCDMA systems: a dedicated physical channel, a physical random access channel, and a common packet channel, since multipath tracking is generally performed on the dedicated physical channel, in this embodiment, a specific implementation of the present invention is described by taking the dedicated physical channel as an example.

In WCDMA uplink, the Dedicated Physical Data Channel (DPDCH) and Dedicated Physical Control Channel (DPCCH) are coded multiplexed in real/imaginary parts within one radio frame, the frame structure of which is shown in fig. 4. For simplicity, it is assumed that there is only one DPDCH per user. Let N denote the number of active users, L_nThe multipath number of the nth user (N is more than or equal to 1 and less than or equal to N) is represented, and the data bit streams (the value is +/-1) of the DPDCH and the DPCCH of the nth user are respectively

$\left\{d_{n,i}^{\left(I\right)}\right\}=\{\&CenterDot;\&CenterDot;\&CenterDot;,d_{n,0}^{\left(I\right)},d_{n,1}^{\left(I\right)},\&CenterDot;\&CenterDot;\&CenterDot;\}---\left(1\right)$

$\left\{d_{n,i}^{\left(Q\right)}\right\}=\{\&CenterDot;\&CenterDot;\&CenterDot;,d_{n,0}^{\left(Q\right)},d_{n,1}^{\left(Q\right)},\&CenterDot;\&CenterDot;\&CenterDot;\}---\left(2\right)$

The spreading sequences (with the value of +/-1) of the DPDCH and the DPCCH of the nth user are respectively

$\left\{c_{n,i}^{\left(I\right)}\right\}=\{\&CenterDot;\&CenterDot;\&CenterDot;,c_{n,0}^{\left(I\right)},c_{n,1}^{\left(I\right)},\&CenterDot;\&CenterDot;\&CenterDot;\}---\left(3\right)$

$\left\{c_{n,i}^{\left(Q\right)}\right\}=\{\&CenterDot;\&CenterDot;\&CenterDot;,c_{n,0}^{\left(Q\right)},c_{n,1}^{\left(Q\right)},\&CenterDot;\&CenterDot;\&CenterDot;\}---\left(4\right)$

Wherein $c_{n,i}^{\left(I\right)}=c_{n,i+M_n^{\left(I\right)}}^{\left(I\right)},$ $c_{n,i}^{\left(Q\right)}=c_{n,i+M^{\left(Q\right)}}^{\left(Q\right)},$ M_n ^(I)And M^(Q)Respectively representing the spreading factors of the DPDCH and DPCCH of the nth user. It should be noted that the spreading factor M of the DPDCHs of different users_n ^(I)Spreading factor M of DPCCH for different, but all users^(Q)Are all the same. Let d_n ^(I)(t)、d_n ^(Q)(t) narrow-band modulated signal waveforms of the n-th users DPDCH and DPCCH, respectively, c_n ^(I)(t)、c_n ^(Q)(t) indicates the spreading waveforms of the n-th users DPDCH and DPCCH, respectively, if there are

$d_n^{\left(I\right)}\left(t\right)=\underset{i=-\&infin;}{\overset{\&infin;}{\mathrm{\&Sigma;}}}{d}_{n,i}^{\left(I\right)}{P}_{{\mathrm{Td}}_n^{\left(I\right)}}(t-{\mathrm{iTd}}_n^{\left(I\right)})---\left(5\right)$

$d_n^{\left(Q\right)}\left(t\right)=\underset{i=-\&infin;}{\overset{\&infin;}{\mathrm{\&Sigma;}}}{d}_{n,i}^{\left(Q\right)}{P}_{{\mathrm{Td}}^{\left(Q\right)}}(t-{\mathrm{iTd}}^{\left(Q\right)})---\left(6\right)$

$c_n^{\left(I\right)}\left(t\right)=\underset{i=-\&infin;}{\overset{\&infin;}{\mathrm{\&Sigma;}}}{c}_{n,i}^{\left(I\right)}{P}_{\mathrm{Tc}}(t-\mathrm{iTc})---\left(7\right)$

$c_n^{\left(Q\right)}\left(t\right)=\underset{i=-\&infin;}{\overset{\&infin;}{\mathrm{\&Sigma;}}}{c}_{n,i}^{\left(Q\right)}{P}_{\mathrm{Tc}}(t-\mathrm{iTc})---\left(8\right)$

Wherein P is_T(T) represents a unit pulse signal having a width T:

Td_n ^(I)、Td^(Q)respectively representing the data bit periods of the n-th users DPDCH and DPCCH, Tc is the chip period and has

$M_n^{\left(I\right)}\&CenterDot;\mathrm{Tc}={\mathrm{Td}}_n^{\left(I\right)}---\left(10\right)$

M^(Q)·Tc＝Td^(Q) (11)

Xi (xi)_n，l(f)＝x_n，l(t)+jy_n，l(t)、τ_n，lRespectively representing the complex value channel impact response and the transmission time delay of the ith path of the nth user signal, and assuming that the transmission power of the DPDCH and the DPCCH is the same, the received baseband signal can be represented as

$r\left(t\right)=\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{l=1}{\overset{L_n}{\mathrm{\&Sigma;}}}\sqrt{{2P}_n}{\mathrm{\&xi;}}_{n,l}\left(t\right)\&CenterDot;[c_n^{\left(I\right)}(t-{\mathrm{\&tau;}}_{n,l})d_n^{\left(I\right)}(t-{\mathrm{\&tau;}}_{n,l})$

$+{\mathrm{jc}}_n^{\left(Q\right)}(t-{\mathrm{\&tau;}}_{n,l})d_n^{\left(Q\right)}(t-{\mathrm{\&tau;}}_{n,l})]+n\left(t\right)---\left(12\right)$

Wherein P is_nIndicating the transmission power of the DPDCH/DPCCH for the nth user. If the influence of the additive white Gaussian noise n (t) is not considered, the above formula can be expressed as

$r\left(t\right)=r_R\left(t\right)+{\mathrm{jr}}_I\left(t\right)$

$=\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{l=1}{\overset{L_n}{\mathrm{\&Sigma;}}}\sqrt{{2P}_n}[(x_{n,l}\left(t\right)c_n^{\left(I\right)}(t-{\mathrm{\&tau;}}_{n,l})d_n^{\left(I\right)}(t-{\mathrm{\&tau;}}_{n,l})-y_{n,l}\left(t\right)c_n^{\left(Q\right)}(t-{\mathrm{\&tau;}}_{n,l})d_n^{\left(Q\right)}(t-{\mathrm{\&tau;}}_{n,l}))$

$+j(x_{n,l}\left(t\right)c_n^{\left(Q\right)}(t-{\mathrm{\&tau;}}_{n,l})d_n^{\left(Q\right)}(t-{\mathrm{\&tau;}}_{n,l})+y_{n,l}\left(t\right)c_n^{\left(I\right)}(t-{\mathrm{\&tau;}}_{n,l})d_n^{\left(I\right)}(t-{\mathrm{\&tau;}}_{n,l}))]---\left(13\right)$

In the present embodiment, multipath tracking is performed according to the flow shown in fig. 5. In step 51, as in the prior art, the signals of the I branch and the Q branch of the baseband with twice chip rate are interpolated to obtain signals with 16 times chip rate.

As shown in fig. 6, after the interpolation process, one chip interval (the interval (i-2, i) on the horizontal axis of fig. 6) is divided into 16 sampling positions, where the signal values x (i-2), x (i-1), and x (i) at the positions i-2, i-1, and i are input signal values twice the chip rate, the interval between i-2 and i-1, and the interval between i-1 and i are divided into 8 equal intervals, and the signal values at the boundaries between these intervals are interpolation values.

As shown in fig. 7, an algorithm structure of the interpolation algorithm takes a section between i-2 and i-1 as an example, and a signal value y (u) at a certain interval boundary is:

y(u)＝(a·u/8+b)·u/8+c (0≤u/8≤1) (14)

wherein,

$a=\frac{1}{2}x\left(i\right)-\frac{1}{2}x(i-1)-\frac{1}{2}x(i-2)+\frac{1}{2}x(i-3)---\left(15\right)$

$b=-\frac{1}{2}x\left(i\right)+\frac{3}{2}x(i-1)-\frac{1}{2}x(i-2)-\frac{1}{2}x(i-3)---\left(16\right)$

c＝x(i-2) (17)

as can be seen from equation (14), u denotes the number of the interval boundary within the section (i-2, i-1), and for example, u-0 corresponds to the position i-2, u-1 corresponds to the position of the first interval boundary from the position i-2, u-2 corresponds to the position of the second interval boundary from the position i-2, … …, and so on, and u-8 corresponds to the position i-1. Since the received signal with twice chip input is a complex signal, the resampling unit should perform interpolation operation on the real part and the imaginary part of the complex signal, respectively.

It is worth pointing out that other interpolation algorithm configurations may also be employed, which are not related to the aforementioned basic idea of the present invention, and therefore the specific interpolation algorithm configuration should not be construed as limiting the spirit and scope of the present invention.

In step 52, as in the prior art, the resampling unit samples the 16 times chip rate signal obtained after the interpolation processing at three positions to obtain three signals, namely, the early-path signal, the middle-path signal and the late-path signal, assuming that the sampling position of the initial middle-path is located at the boundary position corresponding to the sequence number u in the interval (i-2, i-1), the early-path sampling position and the late-path sampling position are respectively located at the boundary positions corresponding to the sequence numbers u +4 and u-4 in the interval (i-2, i-1), and the three signal positions are sequentially different in time or phase by 1/4 chips.

In step 53, the three signals of the early path, the middle path and the late path obtained after the descrambling unit descrambles the signals of the middle path, the early path and the late path respectively are:

r_early(t)＝r_R，early(t)+jr_I，early(t)

r_ontime(t)＝r_R，ontime(t)+jr_I，ontime(t)

r_late(t)＝r_R，late(t)+jr_I，late(t)

for non-coherent despreading units c_n ^(I)(t) and c_n ^(Q)(t) for each of the aboveThe I (real) and Q (imaginary) signals of each signal are despread. For each signal, let z_n，l ^(I)(h，m)、z_n，l ^(Q)(h, m) represents the output of the matched filter at the mth symbol position in the mth slot of the lth path, then there is

$z_{n,l}^{\left(l\right)}(h,m)=\frac{1}{{\mathrm{Td}}^{\left(Q\right)}}\underset{h{T}_{\mathrm{slot}}+(m-1){\mathrm{Td}}^{\left(Q\right)}+{\mathrm{\&tau;}}_{n,l}}{\overset{h{T}_{\mathrm{slot}}+m{\mathrm{Td}}^{\left(Q\right)+{\mathrm{\&tau;}}_{n,l}}}{\&Integral;}}{r}_R\left(t\right)c_n^{\left(Q\right)}(t-{\mathrm{\&tau;}}_{n,l})\mathrm{dt}---\left(18\right)$

$Z_{n,l}^{\left(Q\right)}(h,m)=\frac{1}{{\mathrm{Td}}^{\left(Q\right)}}\underset{h{T}_{\mathrm{slot}}+(m-1){\mathrm{Td}}^{\left(Q\right)}+{\mathrm{\&tau;}}_{n,l}}{\overset{h{T}_{\mathrm{slot}}+m{\mathrm{Td}}^{\left(Q\right)}+{\mathrm{\&tau;}}_{n,l}}{\&Integral;}}{r}_I\left(t\right)c_n^{\left(Q\right)}(t-{\mathrm{\&tau;}}_{n,l})\mathrm{dt}---\left(19\right)$

Wherein T is_slotIs the slot width.

According to equations (18) and (19), the early, middle and late position energies of the mth symbol in the h slot of the ith path are obtained as follows:

$W_{n,l}(h,m)={\left|z_{n,l}^{\left(I\right)}(h,m)\right|}^2+{\left|z_{n,l}^{\left(Q\right)}(h,m)\right|}^2---\left(20\right)$

the mutual interference between two paths during tracking is directly related to the size of the interval between them, and when the mutual interference is greater than or equal to a certain value, the mutual interference is small, so that the sampling phases of the paths can be tracked independently, but when the mutual interference is less than the certain value, the mutual interference is large, so that the mutual interference between them needs to be eliminated before tracking.

In step 54, it is determined whether or not the interval between the two paths is smaller than the set value. Let d (h) be the interval between two paths in the h-th slot, which can be calculated by the following formula:

where path1 and path2 are the path delays (in chips, and path2 > path1) for paths 1 and 2, respectively, in the first slot (h ═ 0), au_i(h) The unit of the change in the sampling phase of the ith slot (h > 0) from the h-1 th slot is 1/16 chips.

If d (h) is greater than or equal to the set value, it indicates that mutual interference is small, and therefore the routine proceeds to step 56, where the sampling phases of paths 1 and 2 are tracked independently. If d (h) is smaller than the set value, it indicates that the mutual interference is large, and therefore, the process proceeds to step 55, where the early, middle and late position energies of the mth symbol in the h-th time slots of the 1 st and 2 nd paths are corrected to eliminate the mutual interference therebetween. The setting value is preferably 1.5 times the chip interval.

In step (b)Step 55, correcting the position energy of the early, middle and late paths of the mth symbol in the h slot of the 1 st and 2 nd paths, assuming W_{n，l，early}(h，m)、W_{n，l，ontime}(h, m) and W_n，l，late(h, m) represents the energy of the m-th symbol of the early, middle and late path positions of the ith path (l ═ 1 or 2) of the nth user in the h-th slot, and the correction algorithm is as follows:

W′_{n，1，early}(h，m)＝W_{n，1，early}(h，m)-W_{n，2，ontime}(h，m)×a[64-d(h)-4] (23-1)

W′_{n，1，ontime}(h，m)＝W_{n，1，ontime}(h，m)-W_{n，2，ontime}(h，m)×a[64-d(h)] (23-2)

W′_n，1，late(h，m)＝W_n，1，late(h，m)-W_{n，2，ontime}(h，m)×a[64-d(h)+4] (23-3)

W′_{n，2，early}(h，m)＝W_{n，2，early}(h，m)-W_{n，1，ontime}(h，m)×a[64+d(h)-4] (23-4)

W′_{n，2，ontime}(h，m)＝W_{n，2，ontime}(h，m)-W_{n，1，ontime}(h，m)×a[64+d(h)] (23-5)

W′_n，2，late(h，m)＝W_n，2，late(h，m)-W_{n，1，ontime}(h，m)×a[64+d(h)+4] (23-6)

wherein the coefficient a [ n ] is a weight factor of the symbol energy of the middle path position of one path at the positions of the early path, the middle path and the late path of the other path, and represents the normalized distribution form of the energy of the path. The weighting factor may be determined as follows: regarding the middle position of the path as the center, consider the relative values of the energy values at several equally spaced sampling positions within a certain range on the left and right sides, in this embodiment, the range is 2 chips on the left and right, the number of sampling positions is 64, and the interval between two adjacent sampling positions is 1/16 chips. Specifically, for each sampling position, a path of chip-level data is separated, each path of chip-level data is descrambled and is despread by a DPCCH spreading code, so that symbol energy and time slot energy are calculated, then the time slot energy is subjected to multi-slot smoothing filtering, and finally the group of time slot energy is subjected to normalization processing to obtain a weighting factor a [ n ] of the signal energy of the other path at each equal-interval sampling position.

It is worth pointing out that the product of the symbol energy of the middle position of one path and the weighting factors at the early, middle and late positions of another path is regarded as the symbol energy of the path at the early, middle and late positions of the other path, but this approximate form is not unique, and other approximate ways capable of accurately representing the symbol energy of the path at the early, middle and late positions of the other path are well known to those skilled in the art, and therefore will not be described herein again.

Next, at step 56, the intermediate symbol energy W 'of the mth symbol of the h slot corrected at step 55 is subjected to the infinite impulse response filter according to the following equation'_{n，l，ontime}(h, m) and the difference between the symbol energies of the early and late roads Δ W_n，l(h, m) (i.e.. DELTA.W)_n，l(h，m)＝W′_{n，l，early}(h，m)-W′_n，l，late(h, m)) are subjected to IIR filtering processing to obtain reference values and error values:

y(n)＝x(n)+0.99y(n-1) (24)

where x (n) is the input value and y (n) and y (n-1) are the current and previous filtered output values, respectively. It is to be noted that the form and parameters of the infinite impulse response filter are not limited to only one defined by equation (24), and other forms may be adopted, as will be apparent to those skilled in the art, and therefore, will not be further developed herein.

Subsequently, in step 57, the error value in step 56 is compared with the reference value by the comparator, and the adjustment direction of the sampling phase of the same symbol in the next slot is determined according to the following rule:

if R | ≦ V_TIf so, the middle path sampling position is the actual path position, and therefore, no adjustment is made;

if R > V_TThen it indicates that the mid-way sample position lags the actual path position, thus indicating that the three sample positions of the same symbol of the next slot are shifted forward in time by 1/16 chips, i.e., so that u in equation (14) is equal to u + 1;

if R < -V_TIt indicates that the mid-way sample position is ahead of the actual path position, thus indicating that three of the same symbol in the next slot have the sample position shifted backward in time by 1/16 chips, i.e., let u be u-1 in equation (14).

Where R is the ratio of the error value to the reference value, V_TIs a preset threshold value greater than 0.

The following describes a comparison of simulation results of the multipath tracking method of the present invention with the multipath tracking method of the prior art with the aid of fig. 8a and 8 b. The simulation environment parameters are as follows:

channel: white noise + rayleigh fading

Fading (fading) characteristics: 2 paths, the relative power of each path is 0dB and 0dB respectively, and the path delay is 260ns and 520ns respectively

Speed of the mobile station: 120 km/h

Background white noise power (dBm): 8X 10^-9

Channel attenuation (dB): 100-116dB

Tracking threshold value: 0.15, 0.20, 0.25, 0.3, 0.35, 0.4

Fig. 8a is a three-dimensional schematic diagram of a simulation result, in which the vertical axis represents the bit error rate, the horizontal axis represents the channel attenuation (in decibels (db)) and the threshold value, respectively, "+" represents the simulation result of the multipath tracking method according to the present invention.

Fig. 8b is a two-dimensional schematic diagram of the simulation result, where the threshold value is 0.2, the vertical axis in the diagram represents the bit error rate, and the horizontal axis represents the channel attenuation (unit is decibel (db)), "" represents the simulation result of the multipath tracking method in the prior art, "o" represents the simulation result in the ideal sampling phase, and "+" represents the simulation result of the multipath tracking method in the present invention.

The simulation results of fig. 8a and 8b show that the bit error rate of the system can be obviously reduced by using the multipath tracking method of the present invention, and particularly, the performance of the system is greatly improved when the tracking threshold is small and the channel attenuation is low. In addition, as can be seen from fig. 8b, the system error rate obtained by the multipath tracking method of the present invention is very close to the system error rate at the ideal sampling phase, so that it can be inferred that when the interval between two paths is less than one chip, the multipath tracking method of the present invention will effectively overcome the disadvantage of path loss in the multipath tracking method of the prior art, and ensure the performance of the system.

Claims (6)

1. An uplink multi-path tracking method for a wideband code division multiple access system, which adjusts the sampling phase of a tracked path according to the symbol energy of the early, middle and late positions of the path, characterized in that if the interval between one tracked path and other paths is less than a set value, the value corrected according to the following steps is taken as the symbol energy of the early, middle and late positions of the tracked path:

(1) determining the energy of the other paths at the early, mid and late positions of the tracked path; and

(2) and (3) respectively subtracting the energy at the positions of the early road, the middle road and the late road determined in the step (1) from the symbol energy of the positions of the early road, the middle road and the late road of the tracked path.

2. The uplink multipath tracking method for a wideband code division multiple access system as claimed in claim 1 wherein said step (1) determines the energy of each of said other paths at the early, mid and late positions of said tracked path in the following manner:

(1a) determining the symbol energy of the path position in the other path and the weight factors of the path position at the early path, the middle path and the late path of the tracked path; and

(1b) multiplying the symbol energy of the mid-way position of the other path by its weighting factors at the early, mid and late way positions of the tracked path, respectively.

3. The method of claim 2 wherein the weighting factors are values of normalized distribution function of symbol energy of the middle path position of the other path over time at early, middle and late path positions of the tracked path.

4. The uplink multipath tracking method for a wideband code division multiple access system as claimed in claim 3, wherein the weighting factor is determined in the following manner:

(a) determining the time slot energy on a plurality of equally spaced sampling positions in a certain range at the left side and the right side of the other path by taking the middle path position of the other path as a center;

(b) performing multi-time slot smoothing filtering processing on the time slot energy at the plurality of equal-interval sampling positions; and

(c) and normalizing the time slot energy after the smoothing filtering processing on the plurality of equal-interval sampling positions to obtain a weight factor of the symbol energy of the other path on each equal-interval sampling position.

5. The uplink multipath tracking method for W-CDMA system as claimed in any one of claims 1 to 4, wherein the sampling position of the tracked path is adjusted according to the symbol energy of the early, middle and late positions of the tracked path in the following way:

(3) respectively carrying out infinite impulse response filtering on the symbol energy of the middle path position of the tracked path and the difference between the symbol energy of the early path position and the symbol energy of the late path position;

(4) and (3) dividing the difference between the early road position symbol energy and the late road position symbol energy processed in the step (3) with the middle road position symbol energy, if the absolute value of the ratio is less than or equal to a preset threshold, not adjusting the sampling position of the tracked path, if the ratio is greater than the preset threshold, moving the sampling position of the tracked path forward in time, and if the ratio is less than the preset threshold, moving the sampling position of the tracked path backward in time.

6. The uplink multipath tracking method for a wideband code division multiple access system as claimed in claim 5, wherein the set value of the path interval is 1.5 times the chip interval.

Images