CN102122997A - Method, device and terminal for detecting long term evolution (LTE) master synchronizing signal - Google Patents

Abstract

The present invention relates to a method, device and terminal for detecting an LTE primary synchronization signal, wherein the method includes: dividing the frequency offset range into a plurality of branches, each branch corresponding to a frequency offset value, and dividing each frequency offset value Add the CORDIC algorithm to the input time-domain data; separate the sampling points from the time-domain data with frequency offset added, and perform time-sliding correlation between the separated time-domain data and the three local feature sequences; Calculate the power value of the correlation result, and output the one with the larger power value of the separated sampling point to obtain the multi-channel frequency offset branch data; obtain the maximum value among the correlation peak values of the multi-channel frequency offset branch data, and the main synchronization signal corresponding to the maximum value is For the primary synchronization signal of the current cell, the frequency offset value corresponding to the maximum value is the initial frequency offset estimation value. The invention greatly improves the detection success rate of the position of the main synchronization signal, expands the detection range of the frequency deviation, reduces system resources, and is convenient to implement.

Description

Method, device and terminal for detecting LTE primary synchronization signal

technical field

The present invention relates to the field of communication technology, in particular to a method, device and terminal for detecting an LTE (Long Term Evolution, long-term evolution) primary synchronization signal.

Background technique

The LTE system is a standardized next-generation wireless communication technology that uses a Constant Envelope Zero Autocorrelation (CAZAC) sequence for a primary synchronization signal.

In the traditional cell primary synchronization signal search scheme, the local characteristic sequence is usually used to perform sliding correlation with the received signal, and then the correlation peak is detected to obtain the location of the primary synchronization signal.

This prior art has the following disadvantages: when the crystal oscillator deviation between transceivers is small, the constant envelope zero autocorrelation sequence has a very good correlation in the time domain, so that the correlation peak detection can be used to realize time synchronization; however when When the deviation of the crystal oscillator is large, the time-domain correlation of the constant-envelope zero autocorrelation sequence will deteriorate, which will lead to the deviation of the correlation peak detection, and the synchronization performance of the system and the detection performance of the main synchronization signal will be affected.

Contents of the invention

The main purpose of the present invention is to provide a method, device and terminal for detecting an LTE primary synchronization signal, which can improve the detection success rate of the primary synchronization signal under different initial frequency offsets.

In order to achieve the above object, the present invention proposes a method for detecting an LTE primary synchronization signal, including:

Divide the frequency offset range into multiple branches, each branch corresponds to a frequency offset value, and attach each frequency offset value to the input time domain data through the agreed algorithm;

Separating the sampling points of the time-domain data with the frequency offset value added, and performing time-domain sliding correlation with the time-domain data after sampling point separation and three kinds of local feature sequences;

Calculate the power value of the correlation result, and output the one with the larger power value of the separated sampling point to obtain multi-channel frequency offset branch data;

Obtain the maximum value among the correlation peak values of the multi-channel frequency offset branch data, and the local characteristic sequence corresponding to the maximum value is the primary synchronization signal of the current cell.

Preferably, the step of dividing the frequency offset range into multiple branches, each branch corresponding to a frequency offset value, and adding each frequency offset value to the input time domain data through an agreed algorithm includes:

Divide the frequency offset range into multiple branches according to the current environment;

determining a frequency offset value corresponding to each branch;

Calculate the angle value required by the agreed algorithm according to the frequency offset value corresponding to each branch and the sampling rate of the data;

Each frequency offset value is added to the input time domain data by using the agreed algorithm to obtain multiple time domain data.

Preferably, the agreed algorithm includes at least a CORDIC algorithm or a rotation search algorithm.

Preferably, said calculating the power value of the correlation result includes accumulating the power value of each antenna of the receiver.

Preferably, the sampling rate before separation of sampling points is an integer multiple of 960K, at least including one of 1.92M or 3.84M.

Preferably, the sampling point separation includes odd-even sampling point separation.

The present invention also proposes a device for detecting an LTE primary synchronization signal, including:

An operation module, configured to divide the frequency offset range into a plurality of branches, each branch corresponds to a frequency offset value, and attach each frequency offset value to the input time domain data through an agreed algorithm;

The matched filter module is used to separate the sampling points of the time-domain data with the frequency offset value added, and perform time-domain sliding correlation with the time-domain data after the sampling point separation and three kinds of local feature sequences;

The power value calculation selection module is used to calculate the power value of the relevant results, and output the one with the larger power value of the separated sampling point to obtain multi-channel frequency offset branch data;

The peak search module is used to obtain the maximum value among the correlation peak values of the multi-channel frequency offset branch data, and the local characteristic sequence corresponding to the maximum value is the primary synchronization signal of the current cell.

Preferably, the computing module includes:

A frequency offset branch division unit, configured to divide the frequency offset range into multiple branches according to the current environment;

a frequency offset value determining unit, configured to determine a frequency offset value corresponding to each branch;

An angle value calculation unit, configured to calculate the angle value required by the agreed algorithm according to the frequency offset value corresponding to each branch and the sampling rate of the data;

An adding unit, configured to use the agreed algorithm to add each frequency offset value to the input time domain data to obtain multiple channels of time domain data.

Preferably, the sampling rate before separation of sampling points is an integer multiple of 960K, at least including one of 1.92M or 3.84M.

Preferably, the sampling point separation includes at least odd-even sampling point separation.

The present invention also proposes a terminal for detecting an LTE primary synchronization signal, and the terminal includes the above-mentioned device.

A method, device, and terminal for detecting an LTE primary synchronization signal proposed by the present invention greatly improve the success rate of detecting the position of the primary synchronization signal under different initial frequency offsets, expand the range of frequency offset detection, and at the same time In the IC (Integrated Circuit, integrated circuit) hardware design, the system resources are greatly reduced, and the implementation is very convenient. Specifically, compared with the prior art, it has the following beneficial effects:

1. Instead of multiplying the local eigensequence and frequency-phase offset binding, the input time-domain data is phase-rotated, so that configurable frequency offset can be realized, and the hardware flexibility is strong;

2. The received signal is not slidingly correlated with the frequency offset branch sequence, but the phase-rotated received signal is slidingly correlated with the local main synchronization sequence, which can realize multiplexing of matched filters and save resources;

3. Under the oversampled data, select the sampling point, which greatly improves the accuracy of the peak detection of the main synchronization sequence, and does not increase the storage RAM;

4. Use CORDIC or rotation search algorithm to offset the frequency of the received signal, with low resource consumption and strong configurability.

Description of drawings

Fig. 1 is a schematic diagram of a frame format of TDD LTE in the prior art;

Fig. 2 is a schematic flow chart of an embodiment of a method for detecting an LTE primary synchronization signal of the present invention;

Fig. 3 is the angle initial value generation schematic diagram of CORDIC iteration in the foregoing embodiment;

FIG. 4 is a schematic structural diagram of a matched filter in the foregoing embodiment;

FIG. 5 is a schematic diagram of the principle of master synchronization signal detection in the foregoing embodiment;

Fig. 6 is a schematic flowchart of dividing the frequency offset range into multiple branches in the above-mentioned embodiment, each branch corresponds to a frequency offset value, and attaching each frequency offset value to the input time domain data through the agreed algorithm ;

7 is a schematic structural diagram of an embodiment of a device for detecting an LTE primary synchronization signal according to the present invention;

8 is a schematic structural diagram of an arithmetic module in an embodiment of an apparatus for detecting an LTE primary synchronization signal according to the present invention;

FIG. 9 is a schematic structural diagram of an embodiment of a terminal for detecting an LTE primary synchronization signal according to the present invention.

In order to make the technical solution of the present invention clearer and clearer, it will be further described below in conjunction with the accompanying drawings.

Detailed ways

The core of the embodiment of the present invention is to avoid the influence of a large frequency offset on the synchronization accuracy and the detection performance of the main synchronization signal without prior information, to add the frequency offset to the input data, and to add the additional frequency offset The input signal and the local feature sequence are slidingly correlated to realize the calculation of the position of the primary synchronization signal and the initial estimation of the frequency offset.

Taking the TDD LTE system as an example, the technical solution of the present invention will be described in detail in combination with the accompanying drawings and implementation modes.

As shown in Figure 1, it is the frame format of TDD LTE, and the first and sixth subframes are used for special subframes. The sending period of the main synchronization signal is 5ms, and the contents sent in the two half-frames before and after are the same. According to the existing standard protocol, PSCH adopts Zadoff-Chu (ZC) sequence, and its expression is:

${\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; u</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}{\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; j</span>}\frac{\mathrm{<span\; class="notranslate">\; \&pi;un</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 63</span>}}}& \mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0,1</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 30</span>}\\ {\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; j</span>}\frac{\mathrm{<span\; class="notranslate">\; \&pi;u</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 63</span>}}}& \mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 31,32</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 61</span>}\end{array}\right.$

Among them, u contains the information of the ID within the group, and the ZC sequence has a very good correlation in both the time domain and the frequency domain. The embodiment of the present invention utilizes the time-domain correlation of the ZC sequence.

As shown in Figure 2, an embodiment of the present invention proposes a method for detecting an LTE primary synchronization signal, including:

Step S101, dividing the frequency offset range into multiple branches, each branch corresponding to a frequency offset value, and adding each frequency offset value to the input time domain data through an agreed algorithm;

The agreed algorithm can be the CORDIC algorithm or the rotation search algorithm, etc. Compared with the rotation search algorithm, the CORDIC algorithm has better operation accuracy, so this embodiment prefers the CORDIC algorithm, and in the following implementation modes, the CORDIC algorithm is taken as an example for illustration .

In this embodiment, division of frequency offset range (hereinafter referred to as frequency offset) branches is determined according to the current environment. Since the subcarrier spacing of the LTE system is 15KHz, the branches can be divided into five branches: -15KHz～-9KHz, -9KHz～-3KHz, -3KHz～+3KHz, +3KHz～+9KHz and +9KHz～+12KHz. The frequency offset values Δf corresponding to the five branches are -12KHz, -6KHz, 0KHz, +6KHz, +12KHz. Then, according to the frequency offset value of each branch and the sampling rate of the data, the angle value of each CORDIC iteration is calculated.

The specific angle initial value generation of CORDIC iteration is shown in Figure 3.

In this embodiment, the CORDIC algorithm is used to add five kinds of frequency offset values to the input time domain data. In this way, the accuracy of the initial frequency offset estimation is ±3KHz. After the iteration of CORDIC, output 5 channels of time domain data.

Wherein, in order to reduce the complexity and amount of calculation, the input time-domain data may adopt a sampling rate of 1.92M, so that the number of sampling points for each symbol is 128.

Since the angle value of each frequency branch of the CORDIC iteration in this embodiment is configurable, multiple CORDIC iterations can be performed to realize further estimation of the frequency offset. For example, the branches are divided into -2.5KHz～-1.5KHz, -1.5KHz～-0.5KHz, -0.5KHz～+0.5KHz, +5KHz～+1.5KHz and +1.5KHz～+2.5KHz, so that each branch corresponds to The frequency offset value Δf is -2KHz, -1KHz, 0KHz, +1KHz, +2KHz. At this time, the initial frequency offset estimation accuracy is ±0.5KHz.

Step S102, performing sampling point separation on the time-domain data to which the frequency offset value has been added, and performing time-domain sliding correlation on the time-domain data after sampling point separation and three kinds of local characteristic sequences;

Among them, the sampling rate before the separation of sampling points is an integer multiple of 960K, and the separation of sampling points can be separated by odd-even point separation or quarter-point separation, etc. The specific separation used corresponds to the aforementioned adoption rate. M, it is base-even separation, if the adoption rate is 3.84M, then it is quarter-point separation, this embodiment uses parity separation as an example for illustration.

After CORDIC iteration, 5 channels of 1.92M time-domain data with different frequency offsets are separated for parity, which is equivalent to a 2-fold downsampling process. In this way, 10 channels of time domain data (2 types of downsampling points * 5 types of frequency offset) will be obtained. After down-sampling, use three local feature sequences (corresponding to three kinds of intra-group IDs) to perform sliding correlation with 10 channels of data. Since the data rate of the time-domain data is 960K after downsampling, the number of sampling points per symbol is 64 at this time, and the number of sampling points per 5 ms is 4800. Therefore, the process of sliding correlation can be realized with a 64-order matched filter in the time domain.

The expression of the matched filter in the time domain is:

$\mathrm{<span\; class="notranslate">\; Y</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; 63</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; P</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>$

where X(k) is the input time-domain data, and P ^* (nk) is the conjugate of the local feature sequence. The structure of the specific matched filter is shown in Fig. 4 .

Among them, the generation method of the local characteristic sequence is as follows: in the LTE system, there are three local primary synchronization characteristic sequences in the frequency domain, and after the three 62-point primary synchronization characteristic sequences in the frequency domain are filled with 0, a 64-point IFFT operation is performed , the primary synchronization sequences P(K) in the three time domains can be obtained, and the primary synchronization sequences P(K) in the three time domains are the three local signature sequences described in this embodiment.

Thus, three local feature sequences*10 IQs=30 branch data can be obtained in total, and each branch data is 4800 pieces.

In this embodiment, the local feature sequence is not multiplied by the frequency offset, but the input time domain data is multiplied by the frequency offset value, so that three matched filters (corresponding to three main Synchronization signal) can realize the time-domain correlation of various frequency offsets. In IC design, great resources are saved.

Step S103, calculate the power value of the correlation result, and output the one with the larger power value of the separated sampling point to obtain multi-channel frequency offset branch data;

Perform power calculation on the time-domain correlation results output by each matched filter. When the receiver uses dual antennas for reception, it is necessary to accumulate the power values of the dual antennas. Since the time-domain data is down-sampled in the CORDIC iteration, it is necessary to select the one with the larger power value to output at the odd point and the even point of the down-sampling. The selection of odd and even sample points can improve the success rate of master synchronization signal detection. Therefore, the 30-way branch data in the above correlation result becomes 15-way branch data, corresponding to 3 local feature sequences*5 kinds of frequency offsets.

The selection of the sampling point in this step can improve the precision of the main synchronization signal while realizing the oversampling rate without increasing the RAM.

In order to improve the success rate of detection, it is possible to accumulate and smooth the correlation results of multiple half-frames, that is, to accumulate the correlation values of the corresponding positions with a period of 4800 points.

Step S104, obtaining the maximum value among the correlation peak values of the multi-channel frequency offset branch data, and the primary synchronization signal corresponding to the maximum value is the primary synchronization signal of the current cell.

Taking the maximum value among multiple frequency offset branches, and using the local characteristic sequence corresponding to the maximum value as the primary synchronization signal of the current cell, and meanwhile, the frequency offset corresponding to the frequency offset branch is the initially estimated frequency offset value.

The principle of master synchronization signal detection in this embodiment is shown in FIG. 5 .

As shown in Figure 6, step S101 includes:

Step S1011, divide the frequency offset range into multiple branches according to the current environment;

Step S1012, determining the frequency offset value corresponding to each branch;

Step S1013, calculate the angle value required by the agreed algorithm according to the frequency offset value corresponding to each branch and the sampling rate of the data;

For the CORDIC algorithm, according to the frequency offset value corresponding to each branch and the sampling rate of the data, the angle value of each iteration of CORDIC can be calculated.

Step S1014, using the agreed algorithm to add each frequency offset value to the input time domain data to obtain multiple channels of time domain data.

As shown in FIG. 7 , an embodiment of the present invention proposes a device for detecting an LTE primary synchronization signal, including: an operation module 701, a matched filter module 702, a power value calculation and selection module 703, and a peak search module 704, wherein:

An operation module 701, configured to divide the frequency offset range into multiple branches, each branch corresponds to a frequency offset value, and attach each frequency offset value to the input time domain data through an agreed algorithm;

The matched filtering module 702 is used to separate the sampling points of the time domain data with the frequency offset value added, and perform time domain sliding correlation with the time domain data after the sampling point separation and three kinds of local feature sequences;

The power value calculation and selection module 703 is used to calculate the power value of the correlation result, and output the larger power value of the separated sampling point to obtain multi-channel frequency offset branch data;

The peak search module 704 is configured to obtain the maximum value among the correlation peak values of the multi-channel frequency offset branch data, and the local characteristic sequence corresponding to the maximum value is the primary synchronization signal of the current cell.

In this embodiment, the agreed algorithm can be the CORDIC algorithm or the rotation search algorithm, etc. Compared with the rotation search algorithm, the CORDIC algorithm has better operation accuracy, so the CORDIC algorithm is preferred in this embodiment, and in the following embodiments, the CORDIC algorithm is specifically used as example to illustrate.

The basic principles of this embodiment are as follows:

The operation module 701 divides the branches of the frequency offset range (hereinafter referred to as frequency offset) according to the current environment. Since the subcarrier spacing of the LTE system is 15KHz, the branches can be divided into five branches: -15KHz～-9KHz, -9KHz～-3KHz, -3KHz～+3KHz, +3KHz～+9KHz and +9KHz～+12KHz. The frequency offset values Δf corresponding to the five branches are -12KHz, -6KHz, 0KHz, +6KHz, +12KHz. Then, according to the frequency offset value of each branch and the sampling rate of the data, the angle value of each CORDIC iteration is calculated.

The specific angle initial value generation of CORDIC iteration is shown in Figure 3.

In this embodiment, the CORDIC algorithm is used to add five kinds of frequency offset values to the input time domain data. In this way, the accuracy of the initial frequency offset estimation is ±3KHz. After the iteration of CORDIC, output 5 channels of time domain data.

Wherein, in order to reduce the complexity and amount of calculation, the input time-domain data may adopt a sampling rate of 1.92M, so that the number of sampling points for each symbol is 128. Sampling points can be separated by odd-even point separation or quarter-point separation.

Since the angle value of each frequency branch input to the operation module 701 in this embodiment is configurable, the operation module can be scheduled multiple times to realize further estimation of the frequency offset. For example, the branch is divided into -2.5KHz～-1.5KHz, -1.5KHz～-0.5KHz, -0.5KHz～+0.5KHz, +5KHz～+1.5KHz and +1.5KHz～+2.5KHz, so that the frequency corresponding to each branch The offset value Δf is -2KHz, -1KHz, 0KHz, +1KHz, +2KHz. At this time, the initial frequency offset estimation accuracy is ±0.5KHz.

The matched filter module 702 separates parity points from the 5 channels of 1.92M time-domain data with different frequency offsets output by the operation module 701, which is equivalent to a 2-times downsampling process. In this way, 10 channels of time domain data (2 types of downsampling points * 5 types of frequency offset) will be obtained. After down-sampling, use three local feature sequences (corresponding to three kinds of intra-group IDs) to perform sliding correlation with 10 channels of data. Since the data rate of the time-domain data is 960K after downsampling, the number of sampling points per symbol is 64 at this time, and the number of sampling points per 5 ms is 4800. Therefore, the process of sliding correlation can be realized with a 64-order matched filter in the time domain.

The expression of the matched filter in the time domain is:

$\mathrm{<span\; class="notranslate">\; Y</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; 63</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; P</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>$

where X(k) is the input time-domain data, and P ^* (nk) is the conjugate of the local feature sequence. The structure of the specific matched filter is shown in Fig. 4 .

Among them, the generation method of the local characteristic sequence is as follows: in the LTE system, there are three local primary synchronization characteristic sequences in the frequency domain, and after the three 62-point primary synchronization characteristic sequences in the frequency domain are filled with 0, a 64-point IFFT operation is performed , the primary synchronization sequences P(K) in the three time domains can be obtained, and the primary synchronization sequences P(K) in the three time domains are the three local signature sequences described in this embodiment.

Thus, three local sequences*10 IQ=30 branch data can be obtained in total, and each branch data is 4800 pieces.

In this embodiment, the local feature sequence is not multiplied by the frequency offset, but the input time domain data is multiplied by the frequency offset value, so that three matched filters (corresponding to three main Synchronization signal) can realize the time-domain correlation of various frequency offsets. In IC design, great resources are saved.

The power value calculation and selection module 703 performs power calculation on the time-domain correlation results output by each matched filter. When the receiver uses dual antennas for reception, it needs to accumulate the power values of the dual antennas. Since the time-domain data is down-sampled in the CORDIC iteration, it is necessary to select the one with the larger power value to output at the odd point and the even point of the down-sampling. The selection of odd and even sample points can improve the success rate of master synchronization signal detection. Therefore, the 30-way branch data in the above correlation result becomes 15-way branch data, corresponding to 3 local feature sequences*5 kinds of frequency offsets.

The selection of the sampling point in this step can improve the precision of the main synchronization signal while realizing the oversampling rate without increasing the RAM.

In order to improve the success rate of detection, it is possible to accumulate and smooth the correlation results of multiple half-frames, that is, to accumulate the correlation values of the corresponding positions with a period of 4800 points.

The peak search module 704 takes the maximum value among multiple frequency offset branches, and uses the local characteristic sequence corresponding to the maximum value as the primary synchronization signal of the current cell, and at the same time, the frequency offset corresponding to the frequency offset branch is the initially estimated frequency offset value.

The principle of master synchronization signal detection in this embodiment is shown in FIG. 5 .

As shown in FIG. 8, the operation module 701 includes: a frequency offset branch division unit 7011, a frequency offset value determination unit 7012, an angle value calculation unit 7013, and an additional unit 7014, wherein:

A frequency offset branch division unit 7011, configured to divide the frequency offset range into multiple branches according to the current environment;

A frequency offset value determination unit 7012, configured to determine a frequency offset value corresponding to each branch;

The angle value calculation unit 7013 is used to calculate the angle value required by the agreed algorithm according to the frequency offset value corresponding to each branch and the sampling rate of the data;

The appending unit 7014 is configured to append each frequency offset value to the input time domain data by using an agreed algorithm to obtain multiple channels of time domain data.

As shown in FIG. 9 , an embodiment of the present invention proposes a terminal for detecting an LTE primary synchronization signal, and the terminal includes the above-mentioned apparatus 901 .

The method, device and terminal for detecting the LTE primary synchronization signal in the embodiment of the present invention, compared with the prior art, does not multiply the local signature sequence and the frequency phase offset binding, but performs phase rotation on the input time domain data, thus Configurable frequency offset can be realized, and the hardware is flexible; the received signal is not slidingly correlated with the frequency offset branch sequence, but the phase-rotated received signal is slidingly correlated with the local main synchronization sequence, which can realize The multiplexing of matched filters saves resources; under the oversampled data, the sampling point selection greatly improves the accuracy of the peak detection of the main synchronization sequence, and does not increase the storage RAM; the received signal is processed by a conventional algorithm such as the CORDIC algorithm. frequency offset, low resource consumption, and strong configurability.

The above is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or process transformation made by using the description of the present invention and the contents of the accompanying drawings, or directly or indirectly used in other related technical fields , are all included in the scope of patent protection of the present invention in the same way.

Claims (11)

1. A method for Long Term Evolution LTE primary synchronization signal detection, characterized in that, comprising: Divide the frequency offset range into multiple branches, each branch corresponds to a frequency offset value, and attach each frequency offset value to the input time domain data through the agreed algorithm; Separating the sampling points of the time-domain data with the frequency offset value added, and performing time-domain sliding correlation with the time-domain data after sampling point separation and three kinds of local feature sequences; Calculate the power value of the correlation result, and output the one with the larger power value of the separated sampling point to obtain multi-channel frequency offset branch data; Obtain the maximum value among the correlation peak values of the multi-channel frequency offset branch data, and the local characteristic sequence corresponding to the maximum value is the primary synchronization signal of the current cell. 2. The method according to claim 1, wherein the frequency offset range is divided into a plurality of branches, each branch corresponds to a frequency offset value, and each frequency offset value is added by an agreed algorithm The steps on the input time domain data include: Divide the frequency offset range into multiple branches according to the current environment; determining a frequency offset value corresponding to each branch; Calculate the angle value required by the agreed algorithm according to the frequency offset value corresponding to each branch and the sampling rate of the data; Each frequency offset value is added to the input time domain data by using the agreed algorithm to obtain multiple time domain data. 3. The method according to claim 1, wherein the agreed algorithm at least comprises a CORDIC algorithm or a rotation search algorithm. 4. The method according to claim 1, wherein said calculating the power value of the correlation result comprises accumulating the power value of each antenna of the receiver. 5. The method according to claim 1, wherein the sampling rate before the separation of the sampling points is an integer multiple of 960K, at least including one of 1.92M or 3.84M. 6. The method according to any one of claims 1-5, wherein the sampling point separation comprises odd-even sample point separation. 7. A device for LTE primary synchronization signal detection, characterized in that, comprising: An operation module, configured to divide the frequency offset range into a plurality of branches, each branch corresponds to a frequency offset value, and attach each frequency offset value to the input time domain data through an agreed algorithm; The matched filter module is used to separate the sampling points of the time-domain data with the frequency offset value added, and perform time-domain sliding correlation with the time-domain data after the sampling point separation and three kinds of local feature sequences; The power value calculation selection module is used to calculate the power value of the relevant results, and output the one with the larger power value of the separated sampling point to obtain multi-channel frequency offset branch data; The peak search module is used to obtain the maximum value among the correlation peak values of the multi-channel frequency offset branch data, and the local characteristic sequence corresponding to the maximum value is the primary synchronization signal of the current cell. 8. The device according to claim 7, wherein the computing module comprises: A frequency offset branch division unit, configured to divide the frequency offset range into multiple branches according to the current environment; a frequency offset value determining unit, configured to determine a frequency offset value corresponding to each branch; An angle value calculation unit, configured to calculate the angle value required by the agreed algorithm according to the frequency offset value corresponding to each branch and the sampling rate of the data; An adding unit, configured to use the agreed algorithm to add each frequency offset value to the input time domain data to obtain multiple channels of time domain data. 9 . The device according to claim 8 , wherein the sampling rate before separation of sampling points is an integer multiple of 960K, at least including one of 1.92M or 3.84M. 10. The device according to claim 7, 8 or 9, wherein the separation of sampling points at least includes separation of even and odd samples. 11. A terminal for detecting an LTE primary synchronization signal, characterized in that the terminal comprises the device according to any one of claims 7-10.

Images