CN101577576A - TD-SCDMA terminal initial downlink synchronization method - Google Patents

Abstract

In the TD-SCDMA system, the direct correlation method of the downlink synchronization in the initial cell search stage after the terminal is turned on will be greatly affected by the large frequency offset. In order to overcome the influence of large frequency offset, the present invention proposes an initial downlink synchronization method for TD-SCDMA terminals on the basis of the direct correlation method. A certain initial frequency offset is added in the initial stage, which can not only effectively eliminate the impact of large frequency offset on Influenced by the data, the correct SYNC_DL code number and synchronization position can be detected, and the frequency offset can be estimated relatively accurately, so that the large frequency offset can be reduced to an appropriate range, so that the subsequent frequency adjustment algorithm can be carried out effectively, and the community Accurate downlink synchronization during initial search provides an effective solution.

Description

A kind of TD-SCDMA terminal initial downlink synchronization method

Technical field

The present invention relates to wireless communication technology field, relate in particular in a kind of TD SDMA (TD-SCDMA) mobile communication system user terminal initial cell search down link synchronisation method.

Background technology

In digital cellular mobile communication systems, after terminal equipment (UE, User Equipment) was started shooting in Radio Network System, what at first will carry out was the cell initial search stage.UE need obtain the timing position of base station down signal in the cell initial search stage, allows terminal equipment access network as early as possible, monitors broadcast channel, receives cell information, carries out follow-up communication process then.

Therefore down-going synchronous is the important step of cell initial search process, and in the TD-SCDMA system, this process is finished by searching downlink pilot frequency signal (SYNC_DL).With other signal down-bound pilot frequency signal is produced interference owing to having noise in the transmission course, influence the correct detection of down-bound pilot frequency signal, so selection is quick, accurate and the reliable down method for synchronous is most important for terminal equipment.The down link synchronisation method that can adopt has: a kind of is characteristics window, and another kind is directly related method.

For the frame structure form of mobile communication TD-SCDMA system as shown in Figure 1.Among the figure, each wireless sub-frame is by 7 common time slots (TS0～TS6) constitute with three special time slots.Wherein, common time slot is used for transmitting data; three special time slots are respectively DwPTS (descending pilot frequency time slot; be used for the transmission of the down-going synchronous information of system); UpPTS (uplink pilot time slot; be used for the uplink synchronous information transmission that the user inserts), GP (the translation and protection time slot is used to provide the time interval of descending sending time slots to up sending time slots conversion).The DwPTS time slot is made up of the GP of 32chip and the SYNC_DL of 64chip, and the GP time slot does not have transmitting power.

Synchronous finishing all will be carried out related operation in any code vision multiple address cellular mobile communication system, the TD-SCDMA system cell is just searched if adopt the method for feature window at present, it is the position that utilizes the architectural feature searching down-bound pilot frequency signal of DwTPS special time slot, on whole frame reception data, press the symbol rank and slide, press chip rated output feature window value in each position.With the power of 64chip in the middle of the feature window with its before and after each 32chip power and ratio obtain feature window position information, seek maximum feature window power ratio and and the threshold value thresholding relatively adjudicate, can accurately obtain the position of feature window one to five continuous subframes, after the position of determining the feature window, receive the 128chip data again at feature window position place relevant respectively with 32 SYNC_DL sign indicating numbers respectively, confirms the employed downlink frequency pilot code in this sub-district according to the related power value of maximum.Its amount of calculation is little, be convenient to Base-Band Processing and on dsp chip, realize, up high reject signal is had the good restraining effect, but the restriction of characteristics window makes that signal to noise ratio must be greater than 0dB, could obtain good correct detection probability, thereby have influence on the receiving sensitivity of DwPTS.

With respect to the limitation of characteristics window, directly related method has the advantage of himself.Tradition realizes the directly related method of down-going synchronous, at first, be locked on the working frequency points, relevant then is to serve as to slide and carry out in the basis with each chip (chip) even with the magnitude of a part chip on whole frame data, do relevant with each group in 32 groups of SYNC sign indicating numbers the signal that UE receives, every group of SYNC_DL sign indicating number all obtains 6400 correlated results, in 32 groups of results, seek a maximal correlation peak value, and correlation peak and predefined threshold value compared, if just write down its position and the ID of corresponding SYNC_DL greater than predefined threshold value.Adopt directly related method, receiving sensitivity relative characteristic window method can be brought up to 4-6dB.

In the TD-SCDMA system, base station and terminal all send and receive with the carrier frequency of nominal.Because the restriction of device level always has certain deviation between actual transmission and receive frequency and the nominal frequency.At the base station end, owing to the restriction of temperature, volume, power consumption, cost etc. is less, the frequency accuracy of oscillator can satisfy the requirement of standard.The carrier frequency error of the base station of TD-SCDMA system requirements is less than ± 0.05ppm.And, being subjected to the restriction of temperature, volume, power consumption, cost etc. bigger at user side, the frequency accuracy of selected oscillator does not satisfy standard-required usually.The carrier frequency error of the user terminal of TD-SCDMA system requirements is less than ± 0.1ppm.Like this, the known down-bound pilot frequency signal that user terminal just need utilize the base station to send is carried out Nonlinear Transformation in Frequency Offset Estimation, and utilizes the Nonlinear Transformation in Frequency Offset Estimation value to carry out automatic frequency control (AFC), makes the carrier frequency error of user terminal satisfy system requirements.

Terminal received signal expression formula is as follows:

$r_k=e^{j\·2\mathrm{\π}\·\mathrm{\Δf}\·k\·T}\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}{h}_l\·s_{k-l}+n_k$

Wherein, s represents the transmission signal of transmitting terminal, and h represents channel impulse response, and n represents white Gaussian noise, multiply by e ^{J2 π Δ fkT}Representative is through being superimposed upon the frequency deviation on the signal after the Channel Transmission.

The calculating formula of directly related method is as follows:

$h_k=\underset{i=0}{\overset{N-1}{\mathrm{\Σ}}}{r}_{k+i}\·s_i^{*}$

$=\underset{i=0}{\overset{N-1}{\mathrm{\Σ}}}(e^{j\·2\mathrm{\π}\·\mathrm{\Δf}\·(k+i)\·T}\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}{h}_l\·s_{k+i-l}+n_{k+i})\·s_i^{*}$

$=h_k\·\underset{i=0}{\overset{N-1}{\mathrm{\Σ}}}{e}^{j\·2\mathrm{\π}\·\mathrm{\Δf}\·(k+i)\·T}\·s_i\·s_i^{*}+\underset{l\≠k}{\mathrm{\Σ}}{h}_l\·\underset{i=0}{\overset{N-1}{\mathrm{\Σ}}}{e}^{j\·2\mathrm{\π}\·\mathrm{\Δf}\·(k+i)\·T}\·s_{k-l+i}\·s_i^{*}+\underset{i=0}{\overset{N-1}{\mathrm{\Σ}}}{n}_{k+i}\·s_i^{*}$

First of following formula is desired value, and the phase deviation after the merging is essentially

Second belongs between the footpath and disturbs, and the 3rd be Gaussian noise, and it is big that we expect that first absolute value can be enough, to resist The noise.

As can be seen, for the TD-SCDMA system, correlation length N is a certain value, and related power not only depends on the autocorrelation of SYNC_DL sign indicating number, also depends on frequency deviation Δ f.The correlation size is as shown in the table during corresponding different frequency deviation:

The SYNC_DL sign indicating number autocorrelation value of the different frequency deviations of table 1

N\Δf(KHz)	1	5	10	15	20
						64	63.737	57.622	40.748	19.211	5.3841e- 015

As can be seen from Table 1, directly related method is based on SYNC_DL sign indicating number and the size that receives data correlation value, yet the size of correlation not only depends on the autocorrelation of SYNC_DL sign indicating number, also with initial frequency deviation much relations is arranged.When frequency deviation is 20kHz, autocorrelation value even become 0.Therefore directly related method exist under the situation of big frequency deviation unavailable.So the autocorrelation severe exacerbation of SYNC_DL sign indicating number when initial frequency deviation is big, thereby had a strong impact on the validity that detects.

Summary of the invention

In order to overcome of the influence of big frequency deviation to the cell initial search down-going synchronous, the present invention proposes a solution, promptly be a kind of TD-SCDMA terminal initial downlink synchronization method, at first,

The described terminal in start back is carried out Cell searching with initial carrier frequency, and described terminal receives and calculate the related power peak value and the position of N subframe of number of sub frames, and this step comprises content:

Obtain n sub-frame data, n=1,2,3 ..., N, described sub-frame data length is N _Frame

To descending synchronous code length N of the many receptions of described n sub-frame data _SyncdlData, obtain receiving sequence r _i, i=1,2, Λ Λ, N _Frame+ N _Syncdl-1

The current carrier frequency people of described n sub-frame data for adding extra frequency deviation-K+n* (2K+1)/N, is obtained

r _i′＝r _i×e ^{j·2π[-K+n*(2K+1)/N]·iT}，i＝1，2，ΛΛ，N _frame+N _syncdl-1

Here the artificial scope that increases described extra frequency deviation of K representative is by the big frequency deviation value decision of estimating to eliminate

Here r _iReception data after the described extra frequency deviation of the artificial increase of ' representative;

Slide by chip (chip), with every synchronous code length N _SyncdlThe reception data calculate a correlation with 32 groups of basic SYNC_DL yardages respectively, obtain

$R_{k,i}=\underset{l=1}{\overset{64}{\mathrm{\Σ}}}\left\{\right[\mathrm{real}\left(r_{l+i}\right)+\mathrm{imag}\left(r_{l+i}\right)*j]*[\mathrm{real}\left({\mathrm{SYNC}\_\mathrm{DL}}_{k,l}\right)-\mathrm{imag}\left({\mathrm{SYNC}\_\mathrm{DL}}_{k,l}\right)*j\left]\right\}$

P _k，i＝{real[R _k，i]} ²+{imag[R _k，i]} ²

R _{K, i}The correlation that expression calculates, P _{K, i}Be every section related power value that receives data (64chip) and 32 groups of basic codes,

Here, i represent i chip (i ∈ [and 1, N _Frame]), k represents k SYNC_DL sign indicating number (k ∈ [1,32]), and l represents to calculate the cyclic variable of correlation, is incremented to 64 from 1, and real represents to get the real operation, and imag represents to get the imaginary part operation of plural number;

Described related power value is carried out normalized, and the 64chip that calculating begins from every bit receives the average power of data:

$P_{k,i}^{\′}=1/64*\underset{l=0}{\overset{63}{\mathrm{\Σ}}}[\mathrm{real}{\left(r_{l+i}\right)}^2+\mathrm{imag}{\left(r_{l+i}\right)}^2]$

With the related power value of described every bit divided by the corresponding average power of this point as the normalization related power that compares:

P″ _k，i＝P _k，i/P′ _k，i

k∈[1，32]，i∈[1，N _frame]

Search maximal correlation performance number is at N _FrameIndividual correlation P " _{K, i}The middle related power value of selecting maximum is as correlation peak, and the position of definite described correlation peak and its corresponding SYNC_DL sign indicating number:

[pos _k, peak _k]=max (P " _{K, i}), k=1,2 ... 32; I=1,2 ... N _Frame, wherein, max represents to look for peaked operation, the maximum peak that finds _kExpression, it is at N _FrameIn position pos _kExpression, pos _k∈ [1, N _Frame].

Repeating said steps 1.1 is total to N time to step 1.6 process, keeps [the pos of N subframe _k, peak _k], with [pos _{N, k}, peak _{N, k}], n ∈ [1～N], k ∈ [1～32] represents.

Secondly, described terminal is utilized described correlation peak and position, determines descending pilot frequency synchronous code (SYNC_DL) and sync bit, and this step comprises content:

The peak value and the position of all corresponding N the subframe of each SYNC_DL sign indicating number (the k value is identical), adjacent position (pos _{N, k}Difference is in ± 2chip) peak value think same position, with the related power peak value addition of same position, arithmetic expression is as follows,

${\mathrm{tap}}_{k,i}=\underset{n=1}{\overset{N}{\mathrm{\Σ}}}|{\mathrm{pos}}_{n,k}-i|\≤2{\mathrm{peak}}_{n,k}:0,$ i＝1～N _frame

This formulate is at sequence tap _{K, i}I ∈ [1, N _Frame] in, if position i ± 2chip has a peak value with interior, then this peak value is superimposed upon tap _{K, i}On, if position i ± 2chip do not have peak value, then a tap with interior _{K, i}Value is zero,

Try to achieve tap _{K, i}Middle largest peaks and position [pos ' _k, peak ' _k], so repeatedly, obtain corresponding 32 the SYNC_DL sign indicating numbers in 32 groups of peak values and position;

Search for corresponding k value maximum in 32 groups of peak values $(k_{\max }=\underset{k}{\arg }\max \left({\mathrm{peak}}_k^{\′}\right))$ Be the sign indicating number number of SYNC_DL sign indicating number, and its corresponding position

Be the original position of descending DwPCH, wherein,

Each value of expression traversal k;

Performance number with maximum

Be divided by with the mean value of all the other 31 groups of performance numbers, as follows,

${\mathrm{peak}}_{k_{\max }}^{\′\′}=31\·{\mathrm{peak}}_{k_{\max }}^{\′}/\underset{\begin{array}{c}k=1\\ k\≠k_{\max }\end{array}}{\overset{32}{\mathrm{\Σ}}}{\mathrm{peak}}_k^{\′}$

Will With predefined thresholding P _hCompare,, think that then SYNC_DL sign indicating number number and sync bit seek correctly, otherwise just need carry out the initial downlink synchronizing process again if more than or equal to this thresholding.

Once more, described terminal is utilized described descending pilot frequency synchronous code (SYNC_DL) and sync bit, determines to finish the carrier frequency of described terminal behind the Cell searching and the adjustment amount of the frequency deviation between the described initial carrier frequency, according to the downlink frequency pilot code sequence number k that obtains _MaxAnd sync bit

Its corresponding N subframe related power value

In, find out peak pos _{N, k}With sync bit

Between deviation be no more than the subframe numbers n of value correspondence of the maximum of L _f, L is a preset value here, the described frequency deviation that draws described terminal correspondence is-(K+n _f(2K+1)/N), use this value big frequency deviation can be reduced in the acceptable scope, be convenient to follow-up careful frequency deviation adjustment process.

The present invention is on the basis of directly related method, a kind of TD-SCDMA terminal initial downlink synchronization method has been proposed, can not only effectively eliminate of the influence of big frequency deviation to data, detect correct SYNC_DL sign indicating number number and sync bit, can also estimate this frequency deviation relatively accurately, big frequency deviation is reduced in the suitable scope,, provides effective solution for carrying out down-going synchronous in the cell initial search process exactly so that follow-up frequency adjustment algorithm is effectively carried out.

Description of drawings

Fig. 1 is TD-SCDMA system frame structure figure

Fig. 2 is a method flow diagram of the present invention

Embodiment

For proving absolutely the technical scheme of method of the present invention, be illustrated with a real example below in conjunction with accompanying drawing.At first calculate the related power peak value and the position of N subframe, process is as follows:

The first step if artificially increase frequency deviation by the AFC method, then at the n subframe that will receive, is adjusted extra frequency deviation-K+n* (the 2K+1)/N of this subframe correspondence; Otherwise directly entered for second step.

Second goes on foot, and receives the data of n subframe continuously, and length is N _Frame, get N _Frame=6400.Consider that when cell initial search the synchronizing signal position of DwPTS do not determine, may appear at the end of receiving sequence and produce to detect and omit, thereby receive a descending synchronous code length N more _Syncdl(N _Syncdl=64) data obtain receiving sequence:

r _i，i＝1，2，ΛΛ，N _frame+N _syncdl-1

In the 3rd step,, then forwarded for the 4th step to if artificially increase frequency deviation by the AFC method; Otherwise artificially increase frequency deviation by numeric field, on the whole sub-frame data that receives, be superimposed with extra frequency deviation-K+n* (the 2K+1)/N of this subframe correspondence:

r′ _i＝r _i×e ^{j·2π·[-K+n*(2K+1)/N]·iT}，i＝1，2，ΛΛ，N _frame+N _syncdl-1

In the 4th step, slide by chip (chip), every synchronous code length N _SyncdlThe reception data calculate a correlation with 32 groups of basic SYNC_DL yardages respectively:

$R_{k,i}=\underset{l=1}{\overset{64}{\mathrm{\Σ}}}\left\{\right[\mathrm{real}\left(r_{l+i}\right)+\mathrm{imag}\left(r_{l+i}\right)*j]*[\mathrm{real}\left({\mathrm{SYNC}\_\mathrm{DL}}_{k,l}\right)-\mathrm{imag}\left({\mathrm{SYNC}\_\mathrm{DL}}_{k,l}\right)*j\left]\right\}$

P _k，i＝{real[R _k，i]} ²+{imag[R _k，i]} ²

P _{K, i}Be every section related power value that receives data (64chip) and 32 groups of basic codes.

Wherein, i represents i chip (i ∈ [1,6400]), and k represents k SYNC_DL sign indicating number (k ∈ [1,32])

In the 5th step, carry out power normalization and handle.The 64chip that calculating begins from every bit receives the average power of data:

$P_{k,i}^{\′}=1/64*\underset{l=0}{\overset{63}{\mathrm{\Σ}}}[\mathrm{real}{\left(r_{l+i}\right)}^2+\mathrm{imag}{\left(r_{l+i}\right)}^2]$

With the related power of every bit divided by the corresponding average power of this point as the normalization related power that compares:

P″ _k，i＝P _k，i/P′ _k，i

k∈[1，32]，i∈[1，6400]

The 6th step, search maximal correlation performance number.At 6400 correlation P " _{K, i}The middle correlation peak of selecting maximum related power value as this yard, and the position of definite relevant peaks and its corresponding SYNC_DL sign indicating number:

[pos _k，peak _k]＝max(P″ _k，i)，k＝1，2，......32；i＝1，2，......6400

Repeat above-mentioned five step of the first step to the process N time altogether, keep [the pos of N subframe _k, peak _k], with [pos _{N, k}, peak _{N, k}], n ∈ [1～N], k ∈ [1～32] represents again.

Secondly, determine descending SYNC_DL sign indicating number and sync bit according to related power peak value and position, process is as follows:

1) peak value and the position of all corresponding N the subframe of each SYNC_DL sign indicating number (the k value is identical), adjacent position (pos _{N, k}Difference is in ± 2chip) peak value think same position, with the related power peak value addition of same position, arithmetic expression is as follows,

${\mathrm{tap}}_{k,i}=\underset{n=1}{\overset{N}{\mathrm{\Σ}}}|{\mathrm{pos}}_{n,k}-i|\≤2{\mathrm{peak}}_{n,k}:0,$ i＝1～6400

Try to achieve tap _{K, i}Middle largest peaks and position [pos ' _k, peak ' _k], so repeatedly, obtain corresponding 32 the SYNC_DL sign indicating numbers in 32 groups of peak values and position.

2) maximum corresponding k value in 32 groups of peak values of search $(k_{\max }=\underset{k}{\arg }\max \left({\mathrm{peak}}_k^{\′}\right))$ Be the sign indicating number number of SYNC_DL sign indicating number, and its corresponding position Be the original position of descending DwPCH.

3) with the performance number of maximum

Be divided by with the mean value of all the other 31 groups of performance numbers, as follows,

${\mathrm{peak}}_{k_{\max }}^{\′\′}=31\·{\mathrm{peak}}_{k_{\max }}^{\′}/\underset{\begin{array}{c}k=1\\ k\≠k_{\max }\end{array}}{\overset{32}{\mathrm{\Σ}}}{\mathrm{peak}}_k^{\′}$

Will

With predefined thresholding P _h(can be taken as 2dB) compares, if more than or equal to this thresholding, thinks that then SYNC_DL sign indicating number number and sync bit seek correctly, otherwise just need carry out the initial downlink synchronizing process again.

At last, determine the initial frequency deviation adjustment amount according to SYNC_DL sign indicating number number and sync bit information:

According to the downlink frequency pilot code sequence number k that obtains _MaxAnd sync bit

Its corresponding N subframe related power value

In, find out peak pos _{N, k}With sync bit

Between deviation be no more than the subframe numbers n of value correspondence of the maximum of L (L can be set at 7) herein _f, the roughly frequency deviation that draws the terminal correspondence at last is :-(K+n _f(2K+1)/N), use this value big frequency deviation can be reduced in the acceptable scope, be convenient to follow-up careful frequency deviation adjustment process.

Artificial maximum frequency deviation KHz and the number of sub frames N that increases depends on initial frequency deviation and the frequency interval (generally getting 3kHz) that real system is possible.

The method of using the present invention to propose can be in the performance that improves on the basis that does not increase memory cost under big frequency deviation, eliminate of the influence of big frequency deviation to correlated performance, add the decision method of multiframe stack, can reduce the probability of erroneous judgement more, adjust the frequency deviation of terminal, obtain higher performance, obtain down-going synchronous fast and accurately.

Claims (2)

1. A TD-SCDMA terminal initial downlink synchronization method, characterized in that, comprising: Step 1: After starting up, the terminal performs cell search with the initial carrier frequency, and the terminal receives and calculates the relevant power peaks and positions of N subframes, the step 1 includes: 1.1 Obtain the nth subframe data, n=1, 2, 3, ..., N, the length of the subframe data is N _frame ; 1.2 Receive one more data of downlink synchronization code length N _syncdl for the nth subframe data, and obtain the received sequence r _i , i=1, 2, ΛΛ, N _frame +N _syncdl -1; 1.3 Artificially add an additional frequency offset -K+n*(2K+1)/N to the current carrier frequency of the nth subframe data to obtain ${{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}<span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; [</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; K</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; *</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; K</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; \&CenterDot;</span>\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&Lambda;\&Lambda;</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; frame</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; syncdl</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>$ Here K represents the range of artificially increasing the additional frequency offset, which is determined by the large frequency offset value that is expected to be eliminated. Here r _i ' represents the received data after artificially increasing the additional frequency offset; 1.4 Slide by chip (chip), calculate the correlation value once with the received data of each synchronization code length N _syncdl and 32 groups of basic SYNC_DL codes, and obtain $R_{k,i}=\underset{l=1}{\overset{64}{\mathrm{\&Sigma;}}}\left\{\right[\mathrm{real}\left(r_{l+i}\right)+\mathrm{imag}\left(r_{l+i}\right)*j]*[\mathrm{real}(\mathrm{SYNC}\_{\mathrm{DL}}_{k,l})-\mathrm{imag}(\mathrm{SYNC}\_{\mathrm{DL}}_{k,l})*j\left]\right\},$ R _k,i table $P_{k,i}={\left\{\mathrm{real}\right[R_{k,i}\left]\right\}}^2+{\left\{\mathrm{imag}\right[R_{k,i}\left]\right\}}^2$ shows the calculated correlation value, P _{k, i} is the correlation power value of each received data (64chip) and 32 groups of basic codes, Here, i represents the i-th chip (i∈[1, N _frame ]), k represents the k-th SYNC_DL code (k∈[1, 32]), and l represents the loop variable for calculating the correlation value, which increases from 1 to 64, real represents the operation of taking the real part of a complex number, and imag represents the operation of taking the imaginary part of a complex number; 1.5 Normalize the relevant power value, and calculate the average power of the 64chip received data from each point: ${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; 64</span>}<span\; class="notranslate">\; *</span>\underset{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; 63</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; real</span>}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; imag</span>}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; ,</span>$ Divide the correlation power value of each point by the average power corresponding to the point as the normalized correlation power for comparison, and obtain P″ _k,i =P _k,i /P′ _k,i , k ∈ [1, 32], i ∈ [1, N _frame ]; 1.6 Search for the maximum correlation power value, select the maximum correlation power value as the correlation peak among N _frame correlation values P″ _{k, i} , and determine the position of the correlation peak and its corresponding SYNC_DL code: [pos _k , peak _k ]=max(P″ _{k, i} ), k=1, 2, ... 32; i = 1, 2, ... N _frame , Among them, max represents the operation of finding the maximum value, the maximum value found is represented by peak _k , and its position in N _frame is represented by pos _k , pos _k ∈ [1, N _frame ]; 1.7 Repeat steps 1.1 to 1.6 for a total of N times, retain [pos _k , peak _k ] of N subframes, use [pos _{n, k} , peak _{n, k} ], n∈[1～N], k∈ [1～32] to represent; Step 2: The terminal determines the downlink pilot synchronization code (SYNC_DL) and the synchronization position by using the correlation peak value and the position, and the step 2 includes: 2.1 Each SYNC_DL code (with the same k value) corresponds to the peak value and position of N subframes, and the peak value of adjacent positions (pos _{n, k} difference within ±2chip) is considered to be the same position, and the correlation power peak value of the same position is compared to add, get, ${\mathrm{<span\; class="notranslate">\; tap</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; pos</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; ?</span>{\mathrm{<span\; class="notranslate">\; peak</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; :</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ~</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; frame</span>}}<span\; class="notranslate">\; ,</span>$ This formula means that in the sequence tap _{k, i} i∈[1, N _frame ], if there is a peak within ±2chip of position i, then the peak will be superimposed on tap _{k, i} , if there is no peak within ±2chip of position i peak value, then tap _{k, i} take the value of zero. Obtain tap _{k according to the formula, the largest peak value and position [pos'k, peak'k] in i , and} so on, obtain 32 groups of peak values and positions corresponding to 32 SYNC_DL codes; 2.2 Search for the largest corresponding k value among the 32 groups of peaks, $k_{\max }=\underset{k}{\arg }\max \left({\mathrm{peak}}_k^{\&prime;}\right)$ It is the code number of the SYNC_DL code, and its corresponding position That is, the starting position of the downlink DwPCH, where, Indicates traversing each value of k; 2.3 Set the maximum power value

Divided with the average value of the remaining 31 groups of power values, as follows, ${\mathrm{<span\; class="notranslate">\; peak</span>}}_{{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}}^{<span\; class="notranslate">\; \&prime;</span><span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 31</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; peak</span>}}_{{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; /</span>\underset{\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; \&NotEqual;</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}}{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}}{\overset{\mathrm{<span\; class="notranslate">\; 32</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; peak</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; ,</span>$ Will

Compared with the preset threshold _Ph , if it is greater than or equal to the threshold, it is considered that the SYNC_DL code number and synchronization position are found correctly, otherwise step 1 is performed again. 2. The method for initial downlink synchronization of a TD-SCDMA terminal according to claim 1, wherein said terminal utilizes said downlink pilot synchronization code (SYNC_DL) and a synchronization position to determine that said terminal has completed cell search The adjustment amount of the frequency offset between the carrier frequency and the initial carrier frequency, according to the obtained downlink pilot code sequence number k _max and synchronization position

Correlation power values in its corresponding N subframes

, find out the peak position pos _{n, k} and the synchronous position

The subframe number n _f corresponding to the maximum value whose deviation does not exceed L, where L is a preset value, it is obtained that the frequency offset corresponding to the terminal is -(-K+n _f (2K+1 )/N).

Images