CN101540623B

CN101540623B - Method for initial cell search in TD-SCDMA system and device therefor

Info

Publication number: CN101540623B
Application number: CN2008100244138A
Authority: CN
Inventors: 朱志明; 吴子熙
Original assignee: AST TECHNOLOGY (SUZHOU) Co Ltd
Current assignee: MediaTek Inc
Priority date: 2008-03-21
Filing date: 2008-03-21
Publication date: 2012-12-19
Anticipated expiration: 2028-03-21
Also published as: CN101540623A

Abstract

The invention relates to a method and device for initially searching a cell in a TD-SCDMA system, and belongs to the technical field of electric communication. The method adopts a unique cell initial search device, and calculates the SYNC-DL code and the position of the maximum value in the power spectrum through the steps of receiving subframe signal data, obtaining power spectrum, accumulating power spectrum, and finding the maximum value. By converting the power value into a floating-point number representation, even through the accumulation of many subframes, the power spectrum will not have the risk of overflow, which can improve the signal-to-noise ratio of the power spectrum, thereby improving the performance of the initial cell search. Moreover, the capacity requirement of the storage unit does not increase with the increase of the number of accumulated subframes, and the requirement for the storage unit can be reduced while improving performance. Therefore, the present invention has substantive technical characteristics and remarkable technical progress, and its application prospect is very broad.

Description

A kind of method and device thereof that is used for TD-SCDMA system cell initial search

Technical field

The present invention designs a kind of method and device thereof of cell initial search, particularly relates to the method and the device thereof of a kind of TD-SCDMA of being used for system cell initial search, belongs to technical field of telecommunications.

Background technology

In the TD-SCDMA system; The purpose of cell initial search is to use the synchronous of family equipment receiver acquisition and base station; Confirm original position, SYNC-DL (down-going synchronous) sign indicating number, basic Midamble sign indicating number, the TTI head of radio frames; And then detect receiver through single sub-district or multi-plot joint and correctly solve system message (this system message is sent through the PCCPCH channel), finally select persistent district.

In the frame structure of TD-SCDMA system, one has 32 SYNC-DL (down-going synchronous) sign indicating number, representes to 31 with ID 0 respectively.A SYNC-DL sign indicating number is used and only used in each base station.The SYNC-DL sign indicating number can be sent by bigger power in the base station in the DwPCH channel.The original position of radio frames is confirmed in the position that the subscriber equipment receiver need detect the SYNC-DL sign indicating number.Simultaneously, also need detect the ID of SYNC-DL sign indicating number.For subscriber equipment, down-going synchronous is that the completion sub-district is resident, and then accomplishes the service access prerequisite, and its performance will directly influence the speed and the overall performance of subscriber equipment access network.

Summary of the invention

The object of the invention is exactly the problems referred to above that exist in the prior art in order to solve, and the method and the device thereof of a kind of TD-SCDMA of being used for system cell initial search is provided.

The object of the invention is realized through following technical scheme:

A kind of method that is used for TD-SCDMA system cell initial search, it may further comprise the steps:

Step 1: receive the subframe signal data since any time;

Step 2: for each the subframe signal data that receives, do the displacement related operation with 32 SYNC-DL sign indicating numbers respectively, obtain the power spectrum of a sub-frame length;

Step 3: will belong to the power spectrum of first subframe, and directly deposit in the memory cell, the power spectrum of each subframe deposits memory cell in after all adding up with the power spectrum that deposits memory cell before in afterwards;

Step 4: after the number of sub frames that adds up reaches threshold value, search for the power spectral value in all memory cell successively, find out maximum wherein, write down peaked position;

Step 5: pass through the peaked position of being write down, extrapolate SYNC-DL sign indicating number and the position of this maximum in power spectrum under this maximum.

Above-mentioned a kind of method that is used for TD-SCDMA system cell initial search; Wherein: the result of calculation of power spectrum transfers floating number to by integer and representes in the said step 2; Floating number is represented an integer through the long mantissa of N Bit with the index that M Bit grows, and promptly the signal of each subframe all obtains the power spectrum of representing with floating number that the length of 32 SYNC-DL sign indicating numbers is a sub-frame respectively.

A kind of device that is used for TD-SCDMA system cell initial search, it includes:

SYNC-DL sign indicating number generating unit is used to generate SYNC-DL) sign indicating number;

The displacement correlation unit is used for first sampled value with the reception signal of each subframe, and the length that takes away the beginning is the reception signal of 64chips, carries out related calculation with 32 SYNC-DL sign indicating numbers respectively and calculates its performance number; With second sampled value of the reception signal of each subframe, the length that takes away the beginning is the reception signal of 64chips afterwards, carry out related calculation with 32 SYNC-DL sign indicating numbers respectively and calculate its performance number, by that analogy, up to the power spectrum that calculates whole subframe;

Integer is to the floating number converting unit, is used for the form that the performance number of representing with integer form with the output of displacement correlation unit converts floating-point into and representes;

The floating number addition unit is used to accomplish the addition of 2 floating numbers;

Memory cell is used to store the performance number after adding up;

The floating number comparing unit is used for the size of two floating numbers of comparison;

Maximum and position storage unit thereof are used to store the maximum that the floating number comparing unit is exported, and store its corresponding position in power spectrum of this maximum simultaneously.

The outstanding substantive distinguishing features and the obvious improvement of technical scheme of the present invention is mainly reflected in: represent through converting performance number into floating number; Even through adding up of a lot of subframes; Can there be the risk of overflowing in power spectrum yet; Can improve the signal to noise ratio of power spectrum, and then improve the performance of cell initial search.And in conjunction with the characteristics of floating number addition, after two floating number additions with the long exponential representation of long mantissa of N Bit and M Bit, the long index of mantissa that still available N Bit is long and M Bit is represented.The Bit bit wide (N+M) of back floating number does not increase owing to add up, and the capacity requirement of memory cell does not increase with the increase of the number of sub frames that adds up like this, when improving performance, can reduce the demand to memory cell.Moreover, adopt floating number, can accomplish big several filtrations automatically to decimal, help improving the performance of cell initial search.Therefore, the present invention has substantive technical characterstic and significant technological progress, and its application prospect is boundless.

Description of drawings

The object of the invention, advantage and characteristics will illustrate through the non-limitative illustration of following preferred embodiment and explain.These embodiment only are the prominent examples of using technical scheme of the present invention, and all technical schemes of taking to be equal to replacement or equivalent transformation and forming all drop within the scope of requirement protection of the present invention.In the middle of these accompanying drawings,

Fig. 1 is the subframe structure sketch map of TD-SCDMA system;

Fig. 2 is the running sketch map that is used for the device of TD-SCDMA system cell initial search;

Fig. 3 is the work sketch map of displacement correlation unit;

Fig. 4 is the work sketch map of floating number comparing unit;

Fig. 5 is a maximum value search process sketch map.

The implication of each Reference numeral is following among the figure:

1 SYNC-DL sign indicating number generating unit, 2 displacement correlation units

3 integers are to floating number converting unit 4 floating number addition unit

5 memory cell, 6 floating number comparing units

7 maximums and position storage unit 8 data shift register sequences thereof

9 yards register series 10 power computation module

11 related operation modules

Embodiment

In order to set forth the present invention better, combine Fig. 1～5 for example to enumerate down earlier:

As shown in Figure 1; The 5ms subframe of each TD-SCDMA is divided into seven common time slots, and (TS0～TS6) and three special time slots, three special time slots comprise: a descending pilot frequency time slot (DwPTS), a uplink pilot time slot (UpPTS) and a protection period (GP).The subframe of each 5ms is made up of 6400 chips (chip).In the TD-SCDMA system, Node B all can send specific SYNC-DL (down-going synchronous) sign indicating number with bigger power in the descending pilot frequency time slot (DwPTS) of its main carrier, and different SYNC-DL sign indicating numbers is used in the sub-district of position near being positioned at.Further, the length of SYNC-DL sign indicating number is 64chips, and subscriber equipment can be accomplished the down-going synchronous with Node B through search SYNC-DL sign indicating number.Simultaneously as shown in Figure 1, before obtaining down-going synchronous, subscriber equipment requires to circulate in 5ms with the counter of certain over-sampling rate (C times of chip speed) according to timing accuracy.

In conjunction with the represented device that is used for TD-SCDMA system cell initial search of Fig. 2, this device includes SYNC-DL sign indicating number generating unit 1, displacement correlation unit 2, integer to floating number converting unit 3, floating number addition unit 4, memory cell 5, floating number comparing unit 6, maximum and position storage unit 7 thereof.

The operation of this device is roughly following:

At first, receive the data of sufficient length since any time.The implication of sufficient length is to utilize these data can accomplish cell initial search.Specifically, data length representes that with subframe the length of a sub-frame is 6400 chip (chip) in the TD-SCDMA system, therefore receives the signal that the K sub-frame constitutes, the add up number of sub frames of K for setting.

Then, for the signal of each subframe that receives, accomplish with 32 SYNC-DL sign indicating numbers with displacement correlation unit 2 and to divide other displacement relevant, and obtain the power spectrum of a sub-frame length.The performance number of displacement correlation unit 2 outputs of this moment is represented with integer.

Then, use integer the form that power spectrum converts floating-point into to be represented to floating number converting unit 3.Specifically, represent an integer through N Bit long mantissa and the long index of M Bit.Subsequently, the power spectrum of representing with floating number of first subframe is directly deposited in memory cell 5, the memory cell 5 of restoring after the power spectrum of each subframe all needs to add up with the power spectrum that deposits memory cell 5 before in afterwards.After the number of sub frames that adds up reaches threshold value, search for the power spectral value in all memory cell 5 successively, the maximum of using floating number comparing unit 6 to find out wherein, and deposit maximum and position thereof in maximum and position storage unit 7 thereof.

Again further, the realization of displacement correlation unit 2 is as shown in Figure 3.The I of input, the Q circuit-switched data, adopting the bit width is L bit, data shift register sequence 8 all is 64 with the length of sign indicating number register series 9.Related operation module 11 completion reception data are relevant with descending synchronous code.Value with reference to the related definition SYNC-DL sign indicating number of agreement is :+1 ,-1, and+j ,-j.Related operation is as shown in the table:

The input data	Enter code word	Output
			I+jQ	00(+1)	I+jQ
I+jQ	01(-1)	-I-jQ
			I+jQ	10(+j)	Q-jI
I+jQ	11(-j)	-Q+jI

Having 64 related operation modules 11 in the whole displacement correlation unit 2, is exactly the correlated results of whole sequence with adding up of these 64 correlated results.If the correlated results after adding up is C _i+ jC _q, C _iAnd C _qBe respectively the signed integer that L+6bit representes.

Simultaneously, the power computation module 10 among Fig. 3 is accomplished the calculating of power, and the correlated results of establishing input power computing module 10 is: C _i+ jC _q, the output valve of power computation module 10 is:

C_{Pow} = Max (| C_{i} |, | C_{q} |) + \frac{1}{2} Min (| C_{i} |, | C_{q} |),

Wherein: | C _i| with | C _q| be respectively the L+5bit signless integer, C _PowBe the L+6bit signless integer.

Integer is accomplished the conversion of L+6bit signless integer to floating number (N bit mantissa, M bit index) to following its of the course of work of floating number converting unit 3.Mantissa is normalized to MSB and equals 1, and the scope of mantissa is 0 to (2^N-1).Index is that signed number and scope do--2 ^(M-1)To 2 ^(M-1)-1.If (I, width Q) are L bit to the reception data of input cell search first step, and scope is-2 ^(L-1)～.2 ^(L-1)-1。Process length is that the relevant later range of results of 64chip is-2 ^(L+5)～2 ^(L+5)-1, so the absolute value of correlated results is 2 to the maximum ^(L+5)If correlated results | I|=|Q|=2 ^(L+5), its performance number is 2 ^(L+5)+ 2 ^(L+4)

In conjunction with above-mentioned expression formula, choose suitable index figure place, can be expressed as P*2 ^QWherein, P is the Nbit number, and Q is the Mbit number, even then can guarantee F sub-frame (F＜=(2 of adding up continuously ^(M-1)-1-Q)) power spectrum, still can not exceed maximum expression scope.Moreover, if the power spectrum of first subframe, then it is directly deposited in the relevant position of memory cell 5, otherwise just with the performance number addition that adds up of reading from memory cell 5 relevant positions after the restore relevant position of memory cell 5.

Adding up of power spectrum realized through floating number addition unit 4.Data format after the process integer is changed to floating number is Nbit mantissa and Mbit index.The realization of the performance number addition of add up performance number and the present frame of reading from memory cell 5 is following: comparison index at first all normalizes to same index (big index) with mantissa part then.This means the bit number that the mantissa part of the number that index is less will move to right certain.The bit number that moves to right is the difference of two number indexes.Afterwards with 2 mantissa's additions.If the N+1bit of addition result is 0 then the low Nbit of addition result just is the mantissa of output, the index of output is an index bigger in 2.If the N+1bit of addition result is 1, then get the mantissa of high Nbit for output, the index of output is index+1 bigger in 2.The result of output is at the appropriate address of write storage unit 5.

After the number of sub frames that adds up reaches threshold value, search for the power spectral value in all memory cell 5 successively, find out maximum wherein, and write down peaked position.Search for maximum through floating number comparing unit 6.

The course of work of floating number comparing unit 6 is as shown in Figure 4: A wherein and B are 2 floating numbers of input floating number comparing unit 6.The bit width of A and B is N+M (Nbit mantissa, Mbit index).The long exponential part of Mbit that at first compares A and B is if the index of A, is then represented A greater than the index of B greater than B, if the index of A equals the index of B then continues the comparison mantissa part, if the index of A, representes then that A is less than B less than the index of B.Equal at the index of A under the situation of index of B, the relatively mantissa of A and the mantissa of B are if the mantissa of A is greater than the mantissa of B at this moment; Then represent A greater than B, if this moment A mantissa equal the mantissa of B, then A equals B in expression; If the mantissa of A representes then that less than the mantissa of B A is less than B at this moment.Under the situation of A more than or equal to B, floating number comparing unit 6 is output as 1, otherwise is output as 0.

Peaked search procedure is as shown in Figure 5: at first, from memory cell 5, read first power spectral value, read the maximum (value in maximum and the position storage unit 7 thereof always is initialized to 0) of storage in maximum and the position storage unit 7 thereof simultaneously.Here represent the power spectral value that from memory cell 5, reads with C, represent the maximum of storage in maximum and the position storage unit 7 thereof with D.With C and D input floating number comparing unit 6 size of these two numbers relatively, if be output as 1 expression C greater than D, C is updated to maximum and position storage unit 7 thereof with its position in memory cell 5, otherwise then is left intact.Afterwards, from memory cell 5, read second power spectral value and continue the corresponding process of repetition, all value in running through memory cell 5.At last, the output of the content in maximum and the position storage unit 7 thereof as last output result, can be accomplished.

Can find out that through above-mentioned text description the present invention has following beneficial effect:

Represent through converting performance number into floating number that (1) even through the adding up of a lot of subframes, can there be the risk of overflowing in power spectrum, and can improve the signal to noise ratio of power spectrum yet, and then improve the performance of cell initial search.

(2) because the floating number addition, after two floating number additions with the long exponential representation of long mantissa of N Bit and M Bit, the index of mantissa that still available N Bit is long and M Bit length is represented.The Bit bit wide (N+M) of back floating number does not increase owing to add up, and the capacity requirement of memory cell does not increase with the increase of the number of sub frames that adds up like this, when improving performance, can reduce the demand to memory cell.

(3) adopt floating number, can accomplish big several filtrations automatically, help improving the performance of cell initial search decimal.

Claims

1. a kind of method that is used for cell initial search in the TD-SCDMA system is characterized in that comprising the following steps:

Step 1: Start receiving subframe signal data from any moment;

Step 2: For each received subframe signal data, respectively use 32 SYNC-DL codes to perform shift correlation operations to obtain a power spectrum of a subframe length;

Step 3: directly store the power spectrum belonging to the first subframe in the storage unit, and then store the power spectrum of each subframe in the storage unit after being accumulated with the power spectrum previously stored in the storage unit;

Step 4: After the accumulated number of subframes reaches the threshold value, search the power spectrum values in all storage units in turn, find out the maximum value, and record the position of the maximum value;

Step 5: Calculate the SYNC-DL code to which the maximum value belongs and the position of the maximum value in the power spectrum through the position of the recorded maximum value.

2. a kind of method that is used for sub-district search at first in the TD-SCDMA system according to claim 1 is characterized in that: the calculation result of power spectrum in the described step 2 is converted into floating-point number by integer and represents, and floating-point number passes through The mantissa of N Bit length and the exponent of M Bit length represent an integer, that is, the signal of each subframe obtains the power spectrum represented by a floating point number of 32 SYNC-DL codes whose length is one subframe.