WO2004066533A1

WO2004066533A1 - Method and device for searching and tracking multi-path

Info

Publication number: WO2004066533A1
Application number: PCT/CN2003/000062
Authority: WO
Inventors: Jingbo Liu; Li Fang
Original assignee: Linkair Communications, Inc.
Priority date: 2003-01-23
Filing date: 2003-01-23
Publication date: 2004-08-05
Also published as: WO2004066533A8; CN1640039A

Abstract

The present invention provides a method and device for searching and tracking multi-path, by using matching filter for demodulating spread spectrum received signals; when local codes in matching filter align with a certain path signals, correlative peak value is outputted from matching filter; power of different delay signal is gained after the signals of the two branches I, Q are squared and added, wherein the maximum is selected as strongest path signal, and the strongest path signal is subtracted from the output of matching filter, then the maximum power is selected from the rest signals as another path signal ; the position information is sent to RAKE receiving unit. It provides for searching the strongest power signal in the downlink synchronous channel signals.

Description

Method and device for multipath search and tracking

The present invention belongs to the technical field of wireless communications, and particularly relates to the implementation of multipath search in a downlink synchronization subsystem of a CDMA system, and specifically relates to a method and device for multipath search and tracking. Background technique

In a CDMA communication system, multi-path search and tracking is to search and select a suitable path among the received signals that reach the mobile station through multiple paths after the mobile station obtains frame synchronization, code synchronization, and carrier synchronization, and determines the target cell. Provided to RAKE receiver.

In the mobile communication environment, signals sent by the transmitter reach the receiver via different paths. The signal of each path undergoes different random amplitude attenuation and phase rotation, as well as different time delays. After these signals reach the receiver, they are superimposed on each other, causing the fading of the synthesized signal. The existence of signals on these paths is also random, that is, they exist at certain moments, disappear at some moments, some moments are stronger than other path signals, and some times may have stronger path signals.

In a CDMA communication system, a RAKE receiver including multiple branches can be used to combat fading. The RAKE receiver uses multiple paths with stronger signals detected by a search unit and assigns them to each branch. Each branch is responsible for synchronizing with a path signal and demodulating the signal in it, and then combining the path information according to certain rules.

The specific tasks of the multipath search unit include:

Search for the position (timing) of the strongest path signal;

Accurately track the signal timing of each path and report to the RAKE receiver;

Check the quality of each path signal;

Search for new and more suitable paths.

Summary of the Invention

An object of the present invention is to provide a method and a device for multipath search and tracking, which are used to find a signal with the strongest energy among downlink synchronization channel signals. The technical solution of the present invention is:

A method for multipath search and tracking, including the following steps:

Use matched filter to despread the received signal;

When the local codeword in the matched filter is aligned with a certain path signal of the received signal, the matched filter outputs a correlation peak;

The energy of the different delay signals is obtained by adding the square of the signals of the two branches I and Q, and the maximum value is selected as the strongest path signal. The strongest path signal is subtracted from the output of the matched filter, and then the remaining signal is subtracted. Select the maximum energy as the other path signal;

The position information of the two paths is sent to the RAKE receiving unit.

The use of a matched filter to despread a received signal refers to: before the start of D-SYNCH (downlink synchronization channel) 4 /; the matched filter despreads the received signal.

In the method of the present invention, the steps may further include:

Before the start of D-SYNCH (downlink synchronization channel), the N matched filter despreads the received signal, where N is the sampling frequency;

Save the data in the range of RANGE near the D-SYNCH synchronization position and accumulate the signals of consecutive N_FRAMES-Acc frames; the initial synchronization position comes from the frame boundary detection unit, and the subsequent position can be the first one found in the previous multipath search径位置； Path position;

Find the maximum value in the accumulated RANGE long signal, which can represent a path; record the maximum value and the position of the maximum value

The average V of the currently selected maximum value and the maximum value of the previous AVERAGE_COUNTER frame. In comparison, if V ^ Kx V ₀ , the RAKE receiver uses the path position of the previous frame and does not update; otherwise it uses P;

Calculate the other sample energy of the corresponding path and subtract it from the accumulated corresponding signal of A ¾ "length;

Search for the maximum value V ₂ and position P ₂ in the RANAGE window;

in case

Then P ₂ represents the second path, otherwise it is regarded as noise; W

P _{2 is} sent to the RAKE receiver; '

If the number of consecutive LOSS_COUNTER times W, notify the synchronous capture to work again;

Compare P ₂ , and use the signal that arrives first as the first path, and the signal that arrives later as the second path, that is: P mirKP P ₂ ), P ₂ = raax (P _l5 P ₂ ).

The selection of the RANGE should consider the clock error and the multipath search error.

The selection of the N_FRAMES_Ac should consider the clock error.

The selection of the N-FRAMES- Acc to consider the clock error includes:

When the initial clock error is 3 ppm, 3 frames can be accumulated; when the clock error correction is less than 0.3 ppm, 10 frames can be accumulated.

The present invention also provides a multi-path search and tracking device, which includes: a matched filter, a square addition module, a storage module, a path search module, and a synchronization capture module;

The matched filter despreads a received signal;

Inputting the despread signal into the squared addition module;

Inputting the output signal of the square addition module into the storage module;

An output signal of the storage module is input to the path search module and the synchronization capture module, and the synchronization capture module is coupled to the path search module;

The output signal of the path search module is sent to a RAKE receiving unit.

The matched filter includes: the sampling rate is: N _c ; the matched filter rate is: N / _c . The output signal of the path search module includes: a position of a first path and a position of a second path.

An effect of the present invention is that, by providing a method and a device for multipath search and tracking, it is possible to find a signal with the strongest energy among downlink synchronization channel signals.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a subframe structure diagram of a downlink synchronization channel;

Figure 2a is the waveform diagram of the output signal of the matched filter under the condition of no noise and multipath; Figure 2b is when another signal goes through a different path with a time delay of 1 chip, Wave chart of output signal with filter;

Figure 2c is a waveform diagram of the output signal of the matched filter when the two signals are present at the same time and after the output signals of the code matched filter are superimposed;

Figure 3 is a waveform data diagram;

4 is a structural block diagram of a device according to the present invention;

FIG. 5 is a flowchart of a method according to the present invention.

detailed description

The present invention provides a method and a device for multipath search and tracking. The following takes the accompanying drawings as an example, and the implementation of multipath search in the downlink synchronization subsystem of the TD-LAS system designed by the physical layer specification of the TD-LAS system is taken as an example. The specific embodiments of the present invention will be described.

In the TD-LAS system, the structure of the downlink synchronization channel is shown in Figure 1. The downlink synchronization physical channels of all cells use the same LS spreading code (see Table 1), and different modulation symbol code groups are used to distinguish the cells. The downlink synchronization sub-frame has a total of 8 time slots, and the length of each time slot is shown in Table 2. Each time slot transmits a downlink synchronization pulse with a length of 72 Chips, including 24 chips of C code and 24 chips of S code. 24chips protection is reserved in the middle. At the beginning of each time slot, the downlink synchronization pulse is transmitted. The LS code used for the downlink synchronization channel is complex digital, and the corresponding spreading and despreading is similar to QPSK, as shown in Table 4. 8 modulation symbols can be transmitted in the 8 time slots of the downlink synchronization sub-frame, and a total of 8 bipolar mutually orthogonal code sequences can be obtained, such as each row or column of the 8 X 8Wal sh matrix in the TD-LAS system These code sequences are called LA polar sequences (see Table 3). 8 orthogonal LA polar sequences can support 8-cell / sector networking; to support a larger network size, consider combining with other cell identification methods such as continuous pilot cell identification, or you can Different cell / sector clusters use different LA guard intervals to expand the signal set of D-SYNPCH. Table 1. Spreading code (LS code) of the downlink synchronization channel (j = V ^ I)

W where: A = (+ j + + +-), B = (-}-+ +-), C = (-7 _— +), D = (-}-+ +-)

Table 2. Subframe slot lengths for downlink synchronization channels

Table 3. LA Polarity Codes for Downlink Synchronization Channels

Table 4. LA Polarity Codes for Downlink Synchronization Channels

Multi-path search and tracking is performed after the mobile station obtains frame synchronization, code synchronization, and carrier synchronization, and determines the target cell, then selects an appropriate path among the received signals that reach the mobile station through multiple paths, and provides it to the RAKE receiver.

In a mobile communication environment, a transmitter sends signals to a receiver via different paths. The signal of each path undergoes different random amplitude attenuation and phase rotation, and different time delays. After these signals reach the receiver, they are superimposed on each other, causing the fading of the synthesized signal. The existence of signals on these paths is also random. They exist at certain moments, disappear at some moments, some moments are stronger than other path signals, and some moments may have stronger path signals.

Specifically, the tasks of the multipath search unit include: 1) Search for the position (timing) of the strongest path signal;

2) accurately track the signal timing of each path and report to the RAKE receiver;

3) Detect the quality of each path signal;

4) Search for new and more suitable paths;

In the TD-LAS system, the downlink multipath search uses a matched filter with a rate of 4 to perform matching filtering on the downlink synchronization channel signal, and after averaging over multiple frames, find the strongest signal in it.

In the present invention, the received signal has a rate of 4 ¾ and is first despread by a code matched filter. Due to the zero correlation window of the codeword, the waveform of the output signal of the matched filter in the absence of noise and multipath is shown in Figure 2a.

When another signal goes through a different path, the time delay is 1 chip, and the output waveform after the code matching filter is shown in Figure 2b. As a result, when two signals exist at the same time, after the code matched filter, the output signals are superimposed, and the waveform is shown in Fig. 2c.

Assuming that the signal of the first path is stronger than the signal of the second path, the position of the maximum value in the output waveform of the matched filter is the position of the first path signal. The maximum value of the transmitter output can determine the synchronization position of the first path signal, or the time delay, and allocate it to a RAKE branch. In order to determine the time delay of the second path, the component of the first path signal at the output of the matched filter is subtracted from the entire output to form a waveform as shown in FIG. 2b. After searching for the maximum value, the second path is detected Sync position.

Because only the maximum output value can be obtained when detecting the first path signal, the output amplitude at other sample moments needs to be calculated. For example, in Figure 3, after adding the baseband shaping filter, the output of the matched filter is ideally normalized to the maximum value. As long as you know the maximum energy and use the waveform data in Figure 3, you can calculate the actual energy value at other sample moments.

E „, is the maximum energy value, is the energy value of / time. Is the maximum position, the waveform at time S (i) i. As shown in FIG. 4 and FIG. 5, specific embodiments of the present invention are as follows:

The search unit starts to work after acquiring the rough position of the downlink synchronization channel through synchronization acquisition. The 4-rate matched filter despreads the received signal. When the local codeword in the matched filter is aligned with a path signal of the received signal, the matched filter Sensor output related peaks. The energy of the different delay signals is obtained by adding the square of the signals of the two branches I and Q. The maximum value is selected as the strongest path signal, which is subtracted from the output of the matched filter, and the maximum energy is selected from the remaining signals As the second path signal, the position information of the two paths is sent to the RAKE receiving unit. The algorithm implementation steps are as follows:

When the downlink synchronization initial acquisition algorithm is completed, the initial position of the D-SYNCH channel is sent, and the multipath search unit starts to work:

1) before D-SYNCH starts, the matched filter of 4 despreads the received signal;

2) Save the data in the vicinity of the D-SYNCH synchronization position, and

The signals of the N-FRAMES-Acc frame are accumulated. The initial synchronization position is from the frame boundary detection unit, and the subsequent position may be the first path position obtained by the previous multipath search.

Note: The selection should consider clock error and multipath search error.

^ — ^ KS ^ cc should also be selected in consideration of clock errors. When the initial clock error is 3 ppm, 3 frames can be accumulated; when the clock error correction is less than 0.3 ppm, 10 frames can be accumulated.

3) Find the maximum value in the accumulated long signal, which represents a path. Record the maximum and maximum position P].

4) The currently selected maximum value is compared with the average value of the maximum values of the previous A VERAGE_COUNTER frames. If <Γχ V ₀ , the RAKE receiver uses the path position of the previous frame without updating, otherwise it uses P,

5) Calculate other sample energy corresponding to the Λ path, and subtract from the accumulated corresponding signals of ^ W length;

6) Search for the maximum value and position P ₂ 'in the RANAGE window,

7) If VV ^ TH—Pa tl ^, it represents the second path, otherwise it is regarded as noise. Send ^, to the RAKE receiver.

8) If LOSS_COUNTER times consecutively ₀ , notify the synchronous capture to work again.

9) Comparative, P _2, the first to arrive as the first signal path, a second path after arrival. That is P min min CP ₇ , P, P ₂ = x (P _1: P.

The I / O and related parameters of the present invention are: Table 5 shows the parameters of multipath search and tracking, and Table 6 shows the input and output parameters.

Table 5. Parameters for multipath search and tracking

Table 6. Inputs and outputs for multipath search and tracking

The effect of the present invention is that, by providing a method and a device for multipath search and tracking, it is possible to find a signal with the strongest energy among downlink synchronization channel signals.

The above specific implementations are only used to illustrate the present invention, but not intended to limit the present invention.

Claims

Rights request

1. A method for multipath search and tracking, characterized by comprising the following steps:

Use matched filter to despread the received signal;

The two path position information is sent to the RAKE receiving unit.

2. The method according to claim 1, wherein the despreading the received signal by using a matched filter means: before the start of D-SYNCH (downlink synchronization channel), the matched signal is received by the N _c matched filter. Despread.

3. The method according to claim 1, further comprising: before the start of the D-SYNCH (downlink synchronization channel), the Nf _c matched filter despreads the received signal; and the D-SYNCH is near the synchronization position. Data in the RANGE range are saved and

The signals of the N_FRAMES_Acc frame are accumulated; the initial synchronization position is from the frame boundary detection unit, and the subsequent position may be the first-path position obtained by the previous multipath search;

The currently selected maximum value is compared with the maximum average value V _{Q of the} previous AVERAGE-COUNTER frame. If V ^ Kx V ^ RAKE is set, the receiver uses the path position of the previous frame and does not update; otherwise, P _{1 is used; the} corresponding Pi path is calculated. Other sample energies are subtracted from the corresponding signals of the accumulated JUW length;

Search for the maximum value ^ and position P ₂ in the RANAGE window;

If / V ^ TH—Pa th ", P ₂ represents the second path, otherwise it is regarded as noise; P _{2 is} sent to the RAKE receiver;

If consecutive LOSS — COUNTER times <V ₀ , notify the synchronous capture to work again;

Compare P ₂ , and use the signal that arrives first as the first path, and the signal that arrives later as the second path, that is, P mii PP ₂ ), and P ₂ = max (P _1- P ₂ ).

4. The method according to claim 3, wherein the selection of the RANGE takes into account clock errors and multipath search errors.

5. The method according to claim 3, wherein the N_FRAMES-Acc is selected in consideration of a clock error.

6. The method according to claim 5, wherein the selection of the N-FRAMES_Acc to consider a clock error comprises:

7. A multipath search and tracking device, comprising: a matched filter, a squared addition module, a storage module, a path search module, and a synchronization capture module;

The matched filter despreads a received signal;

Inputting the despread signal into the squared addition module;

The output signal of the path search module is sent to a RAKE receiving unit.

8. The apparatus according to claim 7, wherein the matched filter comprises: a sampling rate is: N; the matched filter rate is: N / _c .

9. The apparatus according to claim 7, wherein the output signal of the path search module comprises: a position of a first path and a position of a second path.