WO2013056421A1 - 一种小区搜索方法及系统 - Google Patents

一种小区搜索方法及系统 Download PDF

Info

Publication number
WO2013056421A1
WO2013056421A1 PCT/CN2011/080930 CN2011080930W WO2013056421A1 WO 2013056421 A1 WO2013056421 A1 WO 2013056421A1 CN 2011080930 W CN2011080930 W CN 2011080930W WO 2013056421 A1 WO2013056421 A1 WO 2013056421A1
Authority
WO
WIPO (PCT)
Prior art keywords
correlation
sss
detection
frequency offset
pss
Prior art date
Application number
PCT/CN2011/080930
Other languages
English (en)
French (fr)
Inventor
易立强
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to PCT/CN2011/080930 priority Critical patent/WO2013056421A1/zh
Publication of WO2013056421A1 publication Critical patent/WO2013056421A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B1/00Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/69Spread spectrum techniques
    • H04B1/707Spread spectrum techniques using direct sequence modulation
    • H04B1/7073Synchronisation aspects
    • H04B1/7083Cell search, e.g. using a three-step approach
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04JMULTIPLEX COMMUNICATION
    • H04J11/00Orthogonal multiplex systems, e.g. using WALSH codes
    • H04J11/0069Cell search, i.e. determining cell identity [cell-ID]

Definitions

  • the present invention relates to the field of wireless communications, and in particular, to a cell search method and system. Background technique
  • the UE performs cell search immediately after being turned on, idle, and in an active state, and acquires detailed information of the cell and the neighboring cell, so as to further monitor paging on the network or initiate a call to establish a connection.
  • Physical layer process For the cell search process of the third generation mobile communication Long Term Evolution (LTE) system based on Orthogonal Frequency Division Multiplexing (OFDM) technology, the UE not only needs to acquire time and frequency synchronization with the serving cell, It is also necessary to detect the physical layer cell identity and cell/system feature information of the cell.
  • LTE Long Term Evolution
  • OFDM Orthogonal Frequency Division Multiplexing
  • the synchronization signal is used as the system characteristic information for implementing the cell search.
  • the hierarchical design is adopted, which is divided into a primary synchronization signal (PSS) and a secondary synchronization signal (SSS, Secondary Synchronization Signal).
  • PSS primary synchronization signal
  • SSS Secondary Synchronization Signal
  • the primary synchronization signal is mainly used to implement 5-leg timing, ID identification in a cell group
  • the secondary synchronization signal is mainly used to implement radio frame timing synchronization and cell ID identification.
  • the primary synchronization signal sequence and the secondary synchronization signal sequence are related to the cell identity (the primary synchronization signal sequence is only related to the intra-cell identity, and the secondary synchronization signal sequence is jointly determined by the cell identification group and the cell identifier), It is generated in a known manner and mapped to 31 subcarrier positions on both sides of the DC carrier. After symmetrically adding the reserved protection subcarriers, the OFDM time domain symbols are generated by OFDM modulation.
  • the spectrum resource after the primary synchronization signal and the secondary synchronization signal resource mapping is the minimum bandwidth that can be supported in the LTE system is 1.4 MHz, which ensures that the UE can realize fast and low complexity in different bandwidth configurations even if the UE does not know the bandwidth of the LTE system.
  • PSS and SSS are time-division multiplexed and transmitted twice in a 10 ms radio frame every 5 ms.
  • the PSS sent twice in one frame can be used for 5ms timing.
  • the SSS sent twice in a frame can be 10ms.
  • Figure 1A for the FDD frame structure type, the PSS is at 0. The first symbol of the last number of the time slot and the 10th slot, the SSS is located in the previous symbol of the PSS. As shown in FIG.
  • the PSS is located on the third symbol of subframe 1 and subframe 6; the SSS is located on the last symbol of subframe 0 and subframe 5, that is, the third of the SSS before the PSS. On the symbol.
  • Cell search is a very important process in the terminal system, and its performance directly affects the performance of the entire system.
  • the desired benchmark for cell search method design is fast and accurate acquisition, low overhead, simple signal processing, and support for both synchronous and asynchronous operations.
  • the existing LTE cell search uses CP correlation or time domain PSS detection for time synchronization, wherein the CP correlation method is simple to implement, but the timing accuracy is not high and the performance is poor at low SNR.
  • time domain PSS detection There are three methods for time domain PSS detection: The first one is the cross-correlation method, which relates the local sequence to the received data sliding, and its timing accuracy is high but the initial search is susceptible to frequency offset; the second is autocorrelation method, using sequence Repeatability, the two receiving sequences of the interval are slidingly correlated, the method can eliminate the influence of a certain frequency offset, but the timing deviation is large in the case of low SNR; the third method is the combination of the foregoing two methods, The autocorrelation finds the coarse timing position and the coarse frequency offset estimation.
  • the cross correlation is performed in the search window determined by the coarse timing to find the precise timing.
  • the coherent method in order to improve the detection performance, the coherent method is generally used, and the channel estimation is performed by using the PSS, and the channel estimation value of the SSS symbol is approximated, and the SSS detection is completed.
  • the Doppler shift is large. Underneath, the channel changes greatly, which will affect the SSS detection performance.
  • the channel estimation value of the weak cell includes the channel estimation value of the strong cell, and the SSS detection performance also decreases.
  • SSS coherent detection can obtain better performance than non-coherent detection, it has high complexity and large resource consumption.
  • the SSS non-coherent detection method without segmentation the timing and frequency offset are obviously affected.
  • LTE cell search is applied to multi-mode systems, such as GSM, WCDMA, TD-SCDMA, and the search window reserved for LTE by CDMA2000 is difficult to receive PSS and SSS in the same field at the same time.
  • the traditional coherent detection application will restricted. Summary of the invention
  • the main purpose of the embodiments of the present invention is to provide a cell search method and system, which can implement efficient and low complexity cell search based on system performance.
  • the present invention provides a cell search method, including:
  • RF radio frequency
  • the processed received signal is subjected to primary matching signal (PSS) detection based on composite matched filtering, and the carrier frequency offset estimation is performed by cyclic prefix (CP) correlation, and segmented non-coherent secondary synchronous signal (SSS) detection is performed.
  • PSS primary matching signal
  • CP cyclic prefix
  • SSS segmented non-coherent secondary synchronous signal
  • the PSS detection based on the composite matching filtering on the processed received signal is:
  • the filter is configured to obtain the intra-group ID of the cell.
  • the performing the composite matching filtering process is: performing segmentation matching filtering in the initial cell search phase, and performing non-segment symmetric matching filtering in the synchronization maintenance and neighbor cell search phases.
  • the method before the segment matching filtering is performed in the initial cell search phase, the method further includes: performing preset carrier frequency offset processing if the initial frequency offset exceeds a frequency offset range tolerated by the PSS detection;
  • the preset carrier frequency offset processing is: when the PSS does not detect the relevant correlation peak, the preset carrier frequency offset module is activated, and the frequency offset setting value is +10 KHz or - ⁇ or the RF voltage controlled oscillator is adjusted in turn. (VCO) Perform PSS detection again to obtain the initial carrier frequency offset estimate.
  • the carrier frequency offset estimation using the CP correlation is:
  • J i the correlation value corresponding to the CP boundary deviation of the i-th mode
  • the SSS time domain symbol on each antenna is obtained in the received signal after the PSS detection, and the SSS frequency domain symbol is obtained after time-frequency conversion, and the SSS frequency domain symbol and the local SSS sequence molecular frame are obtained.
  • Parallel correlation is performed with two phases of sub-frame 5 to obtain energy, and energy accumulation between segments and energy accumulation between antennas are performed, and SSS correlation energy obtained by accumulating energy between segments and accumulating energy between antennas is performed, and threshold calculation is performed;
  • the correlation values of the plurality of fields are non-coherently accumulated according to the sub-frame 0 and the sub-frame 5, the sub-frame 5 and the sub-frame 0, and the total energy is obtained, and the threshold calculation result and the total energy are peak-searched.
  • the method further includes:
  • the interference cancellation is not started, and the normal SSS detection process is performed; otherwise, the interference cancellation is started, and the interference estimation is performed according to the ID number in the set to be checked group.
  • the kth subcarrier value of the local SSS code and the kth subcarrier value of the local SSS code of the interfering cell L represents the length of each segment of the segment, N represents the number of samples of the time domain symbol, and d represents the number of samples of the detected cell and the interfering cell timing deviation ,
  • ⁇ N/i 6 indicating the received converted SSS frequency domain, indicating the interference factor of the mth segment for the i-th local code, indicating the correlation value of the m-th interference signal, indicating the m-th segment before the interference cancellation
  • the method further includes: performing cell type detection by using a CP correlation;
  • the relevant accumulated value in the CP correlation window is the threshold coefficient.
  • the present invention also provides a cell search system, including: a front end processing module, a PSS detection module, a frequency offset estimation module, and an SSS detection module;
  • a front-end processing module configured to perform RF front-end processing and digital front-end processing on the received signal to obtain a processed received signal
  • a PSS detection module configured to perform PSS detection based on composite matching filtering on the processed received signal
  • a frequency offset estimation module configured to perform carrier frequency offset estimation by using CP correlation
  • SSS detection module for segmented non-coherent SSS detection.
  • the cell search method and system provided by the embodiments of the present invention perform RF front-end processing and digital front-end processing on the received signal to obtain a processed received signal; perform PSS detection based on composite matched filtering on the processed received signal, and use CP correlation Carrier frequency offset estimation is performed, and segmented non-coherent SSS detection is performed, and efficient and low complexity cell search is realized on the basis of ensuring system performance.
  • 1 is a time domain structure diagram of a synchronization signal in the prior art
  • FIG. 2 is a schematic flowchart of implementing a cell search method according to an embodiment of the present invention
  • FIG. 3 is a schematic flow chart of performing RF front-end processing and digital front-end processing on a received signal according to an embodiment of the present invention
  • FIG. 4 is a schematic flow chart of performing PSS detection based on composite matching filtering on a processed received signal according to an embodiment of the present invention
  • FIG. 5 is a schematic diagram of a matched filter using a folded structure according to an embodiment of the present invention
  • FIG. 6 is a schematic diagram of performing carrier frequency offset estimation by using cyclic prefix correlation according to an embodiment of the present invention
  • FIG. 8 is a schematic flowchart of performing segmented non-coherent SSS detection according to an embodiment of the present invention
  • FIG. 9 is a schematic flowchart of interference cancellation according to an embodiment of the present invention.
  • FIG. 10 is a schematic diagram of a timing relationship between a cell to be detected and an interfering cell according to an embodiment of the present invention
  • FIG. 11 is a schematic flowchart of a method for simplifying threshold detection according to an embodiment of the present invention
  • FIG. 12 is a schematic diagram of a process flow of each step in a cell search method according to an embodiment of the present invention
  • FIG. 13 is a schematic structural diagram of a cell search system according to an embodiment of the present invention. detailed description
  • the basic idea of the embodiment of the present invention is: performing RF front-end processing and digital front-end processing on the received signal to obtain a processed received signal; performing PSS detection based on composite matched filtering on the processed received signal, and performing carrier frequency using CP correlation Partial estimation, and segmentation of non-coherent SSS detection.
  • FIG. 2 is a schematic flowchart of a cell search method according to an embodiment of the present invention. As shown in FIG. 2, the method includes the following steps:
  • Step 201 Perform RF front-end processing and digital front-end processing on the received signal to obtain a processed received signal.
  • the signal received by the antenna is processed by a radio frequency (RF, Radio Frequency) front end to perform digital front end processing; wherein the RF front end processing includes RF signal conditioning, filtering, down conversion, and analog to digital conversion processing.
  • the digital front-end processing includes RF defect reception IQ data compensation, AGC, downsampling filtering, inter-antenna weighting processing (inter-antenna energy balance), etc., to obtain the processed received signal; wherein the downsampling filtering process is to sample the frequency of 30.72 MHz.
  • the data is downsampled to 1.92MHz; the cell search part mainly includes PSS detection, carrier frequency offset detection and SSS detection, and the cell search part further detects the received signal processed by the digital front end. And processing.
  • Step 202 Perform PSS detection based on composite matching filtering on the processed received signal. Specifically, as shown in FIG. 4, when performing PSS detection on the processed time domain data, in the initial cell search phase, if the initial frequency offset exceeds The PSS detects the frequency offset range to be tolerated, and performs preset carrier frequency offset processing to resist the large frequency offset effect, and also obtains a large frequency offset range, and obtains an initial carrier frequency offset estimation; wherein, the preset carrier frequency offset processing The method is to start the preset carrier frequency offset module when the PSS does not detect the relevant correlation peak, and then use the frequency offset setting value +10KHz or - ⁇ or adjust the RF voltage controlled oscillator (VCO, Voltage Controlled Oscillator). , PSS detection is performed again, so that even if there is a frequency offset of an integer multiple of subcarriers, PSS detection can obtain better performance;
  • VCO Voltage Controlled Oscillator
  • the processed received signal and the ITS transform PSS time domain sequence are matched and filtered, wherein the matched filtering process is a composite matching filtering process, and in the initial cell search phase, segment matching filtering is performed.
  • the matched filtering process is a composite matching filtering process
  • segment matching filtering is performed in the initial cell search phase.
  • non-segment symmetric matching filtering is performed in the synchronous maintenance and neighbor cell search phase; after the matched filtering process is performed, the matched filtered energy and the accumulation of energy between the antennas are obtained to obtain the total energy:
  • Pu is the total energy of the "PSS" ra received signal sample offset matching filter
  • s (Z) is the time domain conversion of the Zadoff-Chu code used by the Mth PSS.
  • the length, ⁇ is the number of segments of the PSS time domain sequence, ⁇ represents the segment number, L is the length of each segment of the segment, / ⁇ is the number of accumulated antennas; in the initial cell search phase, because the UE is not connected to the base station
  • the segmentation matching filtering process is adopted.
  • the number of segments can be 2, and the frequency can be obtained within [-5, 5] ⁇ frequency offset.
  • non-segmented symmetric matching filtering is used, and the number of segments M is 1.
  • the coefficients of the matched filter have partial symmetry, non-fraction is adopted.
  • a folding structure can also be adopted. As shown in FIG.
  • the order of the matched filter can be 64 or 128, and the matched filter of order 64 needs to downsample to a frequency of 1.92 MHz to 960 KHz.
  • the filtering operation of the matched filter of order 64 will be saved by half; for the matched filter bank, the coefficients of the matched filter of Zadoff-Chu code numbers 29 and 34 are mutually common.
  • the yoke relationship, the filtering result when the code number is 34 can be obtained by the combination of the filtering result of the code number 29, so the final two sets of matched filters;
  • the UE Due to the inevitable existence of the UE in the actual network, the UE is subject to large signal interference when performing cell search, such as other UEs in the vicinity of the strong gain direction of the base station beamforming and the base station is forming. Etc., these may cause the correlation value of the strong interference position to mask the correlation value of the correct position, resulting in timing synchronization failure and ID misdetection in the cell group; therefore, in order to overcome the influence of large signal interference, the matched filtering is adopted in this embodiment. And obtaining energy and normalizing the received signal after the energy accumulation between the antennas;
  • the signal energy is estimated as follows:
  • formula (2) it is the doctorth sample offset signal symbol energy estimation value, the first sample offset of the P antenna receives the signal value, and N is the time domain symbol length, that is, the PSS time domain sequence.
  • the length of the ⁇ is the number of antennas accumulated by the UE.
  • the formula (2) can be converted into a recursive form to reduce the amount of calculation.
  • the calculation process of the half-frame filtered energy value in the correlation energy normalization process is as follows :
  • is the half-frame filtered energy value of the first "PSS" sample offset
  • the “the total energy of the nth receive signal sample offset matching filter of the PSS, W”) is the “sample offset signal signal energy estimation value,” is the energy discrimination threshold
  • half-frame non-coherent accumulation is needed.
  • the threshold calculation and the peak search discriminant are performed to obtain the peak value of the cell. From the position of the peak, the timing position of the cell PSS symbol can be obtained, and the matched intra-group ID of the cell can be obtained from the matched filter bank.
  • Step 203 Perform carrier frequency offset estimation by using CP correlation.
  • step 202 the frequency offset range is controlled, and the initial carrier frequency offset estimation is obtained, so that the frequency offset is within [-5, 5] KHz, and the secondary carrier frequency offset estimation is also needed, so that the frequency offset control is performed subsequently.
  • the carrier frequency offset estimation is completed by using a cyclic prefix (CP, Cyclic Prefix) correlation, as shown in FIG. 6, the carrier frequency offset estimation is performed.
  • CP Cyclic Prefix
  • J I . argmax
  • J f ⁇ (C ij )
  • ⁇ O 1, where 0 represents the extended CP mode, 1 represents the normal CP mode, and ⁇ is the correlation window correction value, which can still be included in the symbol boundary mainly for the deviation of the PSS detection timing
  • indicates that the first mode is relative
  • the correlation value of the CP boundary deviation ⁇ is the received signal of the first downlink symbol of the ⁇ th antenna, indicating the timing of the qth downlink symbol of the first mode, “is the correlation cumulative sequence number, L ; is the associated accumulated length; N is the time domain Symbol Number length, P is the number of antennas accumulated at the receiving end, ⁇ is the cumulative number of symbols, and 7 is the normalization factor of the peak comparison of the two CP modes, which is determined by the length relationship of ⁇ and ⁇ and the number of related symbols, indicating the corresponding first species
  • the correlation value of the mode relative to the CP boundary deviation ⁇ ; ⁇ / represents the obtained normalized frequency offset value, that is, the required carrier frequency offset estimation, referred to as the frequency offset
  • the range of the carrier frequency offset estimation based on the CP correlation is [-7.5, 7.5] KHz, which can well achieve the frequency offset of the remaining carrier obtained in step 202, and because of the large number of CPs, multiple symbols Accumulating correlation values to obtain better estimation performance;
  • the downlink service symbol is related to the CP, as shown in FIG. 7(A), under TDD-LTE, select half.
  • is the correlation window correction value, A e [-£, £ ⁇ ], Ar, is the relative CP. The specific location of the boundary deviation;
  • the CP type can also be detected by using the CP correlation.
  • the CP mode corresponding to the maximum peak is the detected CP mode, which belongs to the prior art and will not be described here.
  • c O o and 3 ⁇ 4 ⁇ respectively represent the associated accumulated values in the extended CP and the regular CP correlation window, which are threshold coefficients, and the calculation method of the value of the CP mode is lower in computational complexity and complexity than the conventional calculation method, and The performance is superior.
  • the estimation performance is good.
  • the secondary carrier frequency offset estimation and the CP mode detection using the CP correlation are performed only in the initial cell search phase, and the synchronization is entered. In the maintenance and neighbor cell search phase, the secondary carrier frequency offset estimation function can be turned off because the carrier frequency offset is in the 3 track state.
  • Step 204 Perform segmental non-coherent SSS detection.
  • the SSS detection is performed by using the timing position of the PSS symbol of the cell obtained in step 202 and the intra-group ID of the cell after the further carrier frequency offset compensation, to obtain the cell ID and the radio frame boundary, in step 203.
  • the detection of the CP mode can also be placed in this step;
  • the following takes the method of not supporting the CP mode as an example.
  • the SSS time domain symbol on each antenna is obtained in the received signal after the PSS detection, and the SSS time domain symbol is subjected to time-frequency conversion.
  • the SSS frequency domain symbol is obtained, and then the SSS frequency domain symbol is segmentally correlated with the local SSS sequence molecular frame 0 and the subframe 5 phase to obtain energy, and the inter-segment energy accumulation and the inter-antenna energy accumulation are sequentially performed.
  • the correlation values of multiple fields are non-coherently accumulated according to the pairing manner of subframe 0 and subframe 5, subframe 5 and subframe 0, and the total energy is obtained:
  • s-. , s: ubframe5 ⁇ k denotes the i-th local code of sub-ton 0 and sub-ton 5, respectively.
  • the SSS sequence value of the subcarriers, R p , 2 fc) represents the kth subcarrier value on the second SSS frequency domain symbol received by the pth antenna, and represents the 2 ⁇ 1 SSS frequency domain symbol received by the pth antenna.
  • the kth subcarrier value, L represents the length of each segment of the segment, represents the segment number, M represents the number of segments, represents the number of antennas accumulated at the receiving end, ⁇ represents the number of accumulated PSS symbols, and ⁇ represents the number of associated ID groups.
  • the SSS correlation energy obtained by accumulating the energy between the segments and accumulating the energy between the antennas is performed, and the threshold is calculated, and the peak calculation is performed on the threshold calculation result and the total energy; here, for the mode supporting the CP mode detection, it is necessary to follow the two CPs.
  • the SSS time domain symbols obtained by the mode are respectively subjected to SSS detection;
  • the SSS sequence uses the M-sequence scrambling method.
  • the cross-correlation property of the SSS sequence is significantly worse than that of the M-sequence.
  • the ratio of the maximum correlation interference peak to the true correlation peak is obtained. 0.4839, therefore, in a multi-cell environment, especially a micro-cell, the cell detection with weak signal strength will be affected by the interference peak of the cell with higher signal strength; in the non-initial cell search phase, the cell with larger signal strength can be obtained.
  • the obtained method may be a serving cell or may be determined by measurement information, such as reference signal received power (RSRP, Reference Signal Receiving Power), signal to interference ratio (SINR), etc., so that the SSS is obtained in the SSS. Interference cancellation processing can be performed during detection;
  • RSRP reference signal received power
  • SINR signal to interference ratio
  • the specific interference cancellation method is shown in Figure 9.
  • the timing relationship between the cell to be detected and the interfering cell is shown in Figure 10.
  • S and (W respectively represent the kth subcarrier value and interference of the i-th local SSS code
  • the kth subcarrier value of the local SSS code of the cell L represents the length of each segment of the segment, N represents the number of samples of the time domain symbol, and d represents the number of samples of the detected cell and the interfering cell timing deviation.
  • a threshold detection simplification method is adopted. As shown in FIG. 11, in the SSS detection of the cell, the SSS time domain symbol is subjected to time-frequency conversion to obtain an SSS frequency domain symbol, if the cell If the timing (PSS detection is obtained) and the interfering cell timing does not meet the set timing deviation, the interfering cell is not activated, and the normal SSS detection procedure is performed. Otherwise, the interference cancellation function is activated; the interference estimation is performed according to the set ID number to be checked.
  • the ID group number to be checked is all ID group numbers; for non-coherent accumulation over multiple symbols, the last threshold detection threshold threshold corresponding ID group number is determined next time The ID group number to be detected, so that the number of ID group numbers to be detected can be reduced, and the amount of calculation can be reduced; if the interference is not activated, the threshold selection is relatively loose;
  • the method of segmented non-coherent SSS detection is less complex and consumes less resources than the coherent method, and can overcome the performance of the coherent detection method in the case of high-speed mobile and the same-frequency PSS group ID in the same cell. With the drop, the ability to meet the demand can be obtained. Moreover, SSS detection only needs to utilize SSS symbols, so it can be better applied to TDD-LTE, FDD-LTE and multi-mode systems.
  • the cell search includes two stages: an initial cell search and a non-initial cell search, and each step is invoked as shown in FIG. 12, and the CP mode detection is performed as an example in the SSS detection, and the initial cell search phase scheduling process is as shown in FIG.
  • the carrier frequency offset is estimated by first starting the PSS detection, and then performing the carrier correlation estimation for the maximum correlation peak of the PSS detection. After the partial estimation, the SSS detection is performed. Since the PSS detection may detect multiple thresholds of PSS related to the threshold, the SSS detection needs to be called multiple times, so the serial processing mode is adopted; for the non-initial cell search, the step scheduling is as shown in FIG. 12 (B).
  • the carrier frequency offset estimation does not start, and a similar pipeline mode, that is, parallel processing of PSS detection and SSS detection, and multiple symbol non-coherent accumulation for PSS detection and SSS detection may be used.
  • the peak search is started with each accumulation, and the high signal-to-noise ratio cell adopts an advanced output mechanism to shorten the detection time.
  • FIG. 13 is a schematic structural diagram of a cell search system according to an embodiment of the present invention.
  • the system includes: a front end processing module 131 and a PSS detection module. 132, a frequency offset estimation module 133, an SSS detection module 134;
  • the front end processing module 131 is configured to perform RF front end processing and digital front end processing on the received signal to obtain the processed received signal;
  • the PSS detection module 132 is configured to perform PSS detection based on composite matching filtering on the processed received signal
  • a frequency offset estimation module 133 configured to perform carrier frequency offset estimation by using CP correlation
  • the SSS detection module 134 is configured to perform segmented non-coherent SSS detection.
  • the PSS detection module 132 performs PSS detection based on the composite matching filtering on the processed received signal as follows: performing composite matching filtering processing on the processed received signal and the IFFT transformed PSS time domain sequence, and obtaining the matched filtered energy. Performing energy accumulation between the antennas and correlation energy normalization processing to obtain a half-frame filtered energy value; accumulating the plurality of half-frame filtered energy values, performing threshold calculation and peak search discrimination, and obtaining a peak of the cell, according to the position of the peak, The timing position of the cell PSS symbol is obtained, and the intra-group ID of the cell is obtained according to the matched filter.
  • the performing the composite matching filtering process is: performing segmentation matching filtering in the initial cell search phase, and performing non-segment symmetric matching filtering in the synchronization maintenance and neighbor cell search phases.
  • the preset carrier frequency offset processing is performed before the segmentation matching filtering; the preset carrier frequency offset processing is: when the PSS is not When the relevant peak value is detected, the preset carrier frequency offset module is started, and the PSR detection is performed again by using the frequency offset setting value +10 KHz or - ⁇ or by adjusting the RF voltage controlled oscillator (VCO) to obtain the initial carrier frequency offset. estimate.
  • the SSS detection module 134 performs segmented non-coherent SSS detection as follows: According to the timing position obtained by the PSS detection, the SSS time domain symbols on each antenna are acquired in the received signal after the PSS detection, and the SSS is obtained after time-frequency conversion. The frequency domain symbol, the SSS frequency domain symbol is segmentally correlated with the local SSS sequence molecular frame 0 and the subframe 5 phase to obtain energy, and the inter-segment energy accumulation and the inter-antenna energy accumulation are performed, and the inter-segment energy accumulation and the antenna are performed.
  • the SSS correlation energy obtained by accumulating the energy is used to perform threshold calculation; the correlation values of the plurality of fields are according to subframe 0 and subframe 5, and the subframe 5 and sub-frame 0 pairing mode for non-coherent accumulation, get the total energy, peak search for the threshold calculation result and total energy.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Databases & Information Systems (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本发明实施例公开一种小区搜索方法及系统,该方法包括:对接收的信号进行RF前端处理和数字前端处理,得到处理后的接收信号;对处理后的接收信号进行基于复合匹配滤波的PSS检测,利用CP相关进行载波频偏估计,并进行分段非相干的SSS检测。根据本发明实施例的技术方案,在保证系统性能的基础上实现高效低复杂度的小区搜索。

Description

一种小区搜索方法及系统 技术领域 本发明实施例涉及无线通信领域, 特别涉及一种小区搜索方法及系统。 背景技术
在无线通信系统中, UE在刚开机时、 空闲时以及处于激活状态时, 都 要进行小区搜索, 获取小区及相邻小区的详细信息, 以便进一步监听网络 上的寻呼或发起呼叫建立连接等物理层过程。 对于基于正交频分复用 ( OFDM, Orthogonal Frequency Division Multiplexing )技术的第三代移动 通信长期演进(LTE, Long Term Evolution ) 系统的小区搜索过程, UE不 仅要与服务小区取得时间和频率同步, 还必须检测小区的物理层小区标识 和小区 /系统特征信息等。
同步信号作为用于实施小区搜索的可利用的系统特征信息, 在 LTE 系 统中,采用分层设计,分为主同步信号( PSS, Primary Synchronization Signal ) 和辅同步信号 (SSS, Secondary Synchronization Signal )。 通常地, 主同步 信号主要用来实现 5腿定时、 小区组内 ID识别等, 辅同步信号主要用来实 现无线帧定时同步以及小区 ID识别等。
由 3GPP协议可知, 由于主同步信号序列、辅同步信号序列是与小区标 识有关 (主同步信号序列仅与小区组内标识有关, 辅同步信号序列生成由 小区标识组与小区标识共同确定), 由公知方式生成, 并映射至直流载波两 侧各 31个子载波位置, 对称增加预留保护子载波后, 经 OFDM调制生成 OFDM时域符号。 主同步信号、 辅同步信号资源映射后的频谱资源为 LTE 系统中可支持的最小带宽 1.4MHz,保证了即便事先 UE不知道 LTE系统的 带宽, 也能保证在不同带宽配置下实现快速、 低复杂度、 低开销的小区搜 索。
图 1是现有技术中同步信号的时域结构图, 如图 1所示, PSS和 SSS 采用时分复用, 在一个 10ms无线帧内发送两次, 每 5ms—次。 对于特定小 区,一帧内两次发送的 PSS相同,可以做 5ms定时;一帧内两次发送的 SSS 不同, 可以实现 10ms定时; 如图 1A所示, 对于 FDD帧结构类型, PSS 位于第 0号和第 10号时隙的倒数第一个符号, SSS位于 PSS的前一个符号。 如图 1B所示, 对于 TDD帧结构类型, PSS位于子帧 1和子帧 6的第 3个 符号上; SSS位于子帧 0和子帧 5的最后一个符号上, 即 SSS在 PSS前的 第 3个符号上。
小区搜索作为终端系统中非常重要过程, 其性能直接影响到整个系统 的性能。 小区搜索方法设计的期望基准是快速准确获取、 低开销、 简单的 信号处理过程, 以及支持同步和异步操作。
现有的 LTE小区搜索对于时间同步采用 CP相关或时域 PSS检测, 其 中 CP相关方法的实现简单,但是定时精度不高且在低信噪比情况下性能不 佳。 时域 PSS检测大体有三种方法: 第一种是互相关法, 将本地序列与接 收数据滑动相关, 其定时精度较高但初始搜索易受频偏影响; 第二种是自 相关法, 利用序列的重复性, 将间隔的两接收序列进行滑动相关, 该方法 能够消除一定频偏的影响, 但在低信噪比情况下定时偏差较大; 第三种方 法是前面两种方法的结合, 通过自相关找出粗定时位置和粗频偏估计, 频 偏补偿后, 再在粗定时确定的搜索窗口内进行互相关, 找出精准定时, 这 种方法结合了前面两种方法的优点, 但是需要消耗额外的搜索时间, 而且 运用于 FDD-LTE系统中, 其优势难以发挥。
SSS检测中, 为了提升检测性能一般采用相干法, 利用 PSS进行信道 估计,近似为 SSS符号的信道估计值,完成 SSS检测。然而在 TDD模式下, 时域中 PSS和 SSS之间间隔了 2个符号, 特别地, 在多普勒频移较大情况 下, 信道变化很大, 这样将影响 SSS检测性能。 同时在多小区情况下, 对 于相同 PSS组内 ID的小区检测,弱小区的信道估计值包含了强小区的信道 估计值, SSS检测性能也将下降。 从复杂度上来说, SSS相干检测虽然能获 得相比非相干检测较好的性能, 但其复杂度高, 资源消耗大。 但是, 如果 采用不分段处理的 SSS非相干检测法, 对定时和频偏影响明显。
另夕卜, LTE小区搜索运用于多模系统,如 GSM、 WCDMA、 TD-SCDMA, CDMA2000预留给 LTE的搜索窗口出现难以同时接收到同一半帧内的 PSS 和 SSS, 传统的相干检测运用将受到限制。 发明内容
有鉴于此, 本发明实施例的主要目的在于提供一种小区搜索方法及系 统, 在保证系统性能的基础上实现高效低复杂度的小区搜索。
为达到上述目的, 本发明实施例的技术方案是这样实现的:
本发明提供一种小区搜索方法, 包括:
对接收的信号进行射频 (RF )前端处理和数字前端处理, 得到处理后 的接收信号;
对处理后的接收信号进行基于复合匹配滤波的主同步信号( PSS )检测, 利用循环前缀(CP )相关进行载波频偏估计, 并进行分段非相干的辅同步 信号 (SSS )检测。
上述方法中, 所述对处理后的接收信号进行基于复合匹配滤波的 PSS 检测为:
对处理后的接收信号与 IFFT变换的 PSS时域序列进行复合匹配滤波处 理, 并求取匹配滤波后的能量, 进行天线间能量的累加和相关能量归一处 理, 得到半帧滤波能量值;
将多个半帧滤波能量值进行累加, 进行门限计算和峰值搜索判别, 得 到小区的峰值, 根据峰值的位置, 得到小区 PSS符号的定时位置, 根据匹 配滤波器, 得到小区的组内 ID。
上述方法中, 所述进行复合匹配滤波处理为: 在初始小区搜索阶段, 进行分段匹配滤波, 在同步维护和邻小区搜索阶段, 进行非分段对称匹配 滤波。
上述方法中, 在初始小区搜索阶段, 所述进行分段匹配滤波之前, 该 方法还包括: 如果初始频偏超过 PSS检测所容忍的频偏范围, 进行预设载 波频偏处理;
所述预设载波频偏处理为: 当 PSS未检测到符合要求的相关峰值时, 启动预设载波频偏模块, 依次利用频偏设置值 +10KHz或 -ΙΟΚΗζ或通过调 整 RF的电压控制振荡器( VCO )再次进行 PSS检测, 获取初始载波频偏 估计。
上述方法中, 所述利用 CP相关进行载波频偏估计为:
所述载波频偏估计为 Δ/ =— arg(CiAr ) , 其中 , 和 根据
7, 确 定 , |C。J 和 I 或 | 根 据 Ci
Figure imgf000005_0001
q(n + T.q + Δτ + Λ 确定; 其中, ζ·=0或 1, 其中 0代表扩展 CP模式, 1代表常规 CP模式, Δτ为 相关窗口修正值, Δ^ [-£·,£·], 表示第 种模式相对 CP边界偏差 的 相关值, rM(M)为第 ρ根天线第 个下行符号的接收信号, 表示第 种模式 第 个下行符号的定时, "为相关累加序号, L;为相关累加长度; N为时域 符号长度, P为接收端累加的天线数目, β为累加的符号数, 为两种 CP 模式峰值比较的归一化因子, 由! 1和 的长度关系以及相关符号数决定, Ji ]表示对应第 i种模式相对 CP边界偏差 的相关值, Δ/表示求得的归一 化频率偏移值, arg(*)e [-π, π), 4=32 , 1^=9。 上述方法中, 所述进行分段非相干的 SSS检测为:
根据 PSS检测得到的定时位置, 在经过 PSS检测后的接收信号中获取 各天线上的 SSS时域符号, 经过时频转换后得到 SSS频域符号, 将 SSS频 域符号与本地 SSS序列分子帧 0和子帧 5两种相位进行分段相关, 求取能 量, 并进行段间能量累加和天线间能量累加, 对段间能量累加和天线间能 量累加得到的 SSS相关能量, 进行门限计算;
将多个半帧的相关值按照子帧 0与子帧 5、子帧 5与子帧 0两种配对方 式进行非相干累加, 得到总能量, 对门限计算结果和总能量进行峰值搜索。
上述方法中, 该方法还包括:
当小区定时和干扰小区定时不满足设定的定时偏差时, 不启动干扰消 除, 按照正常的 SSS检测流程进行; 否则启动干扰消除, 干扰估计按设定 待检组内 ID号进行。
上述方法中, 所述干扰消除为: 利 用 - 和
Figure imgf000006_0001
m = R(W进行干扰消除; 其中, S k、、 分别表示第 i组本 k=mL
地 SSS码第 k个子载波值和干扰小区的本地 SSS码第 k个子载波值, L表 示分段的每段长度, N表示时域符号的样点数, d表示检测小区和干扰小区 定时偏差样点数, | < N/i6 , 表示接收的转换后的 SSS 频域, 表示 第 m段针对于第 i组本地码的干扰因子, 表示第 m段干扰信号的相关值, 表示干扰消除前第 m段第 i组本地码的相关值, ^表示干扰消除后第 m 段第 i组本地码的相关值。
上述方法中,
该方法还包括: 利用 CP相关进行小区 CP类型的检测;
所述利用 CP相关进行小区 CP类型的检测为: 基于相关窗的 CP模式 进行检测, 选择相同的相关累加长度, /4 = L2 = 9 , CP 模式通过
Γ 2 2
CP模式 = 0 > - c &To 判别, 其中 <¾Δ7和<¾,^分别表示扩展 CP和常
[l 其他 规 CP相关窗内相关累加值, 为门限系数。
本发明还提供一种小区搜索系统, 包括: 前端处理模块、 PSS检测模 块、 频偏估计模块、 SSS检测模块; 其中,
前端处理模块, 用于对接收的信号进行 RF前端处理和数字前端处理, 得到处理后的接收信号;
PSS检测模块,用于对处理后的接收信号进行基于复合匹配滤波的 PSS 检测;
频偏估计模块, 用于利用 CP相关进行载波频偏估计;
SSS检测模块, 用于进行分段非相干的 SSS检测。
本发明实施例提供的小区搜索方法及系统,对接收的信号进行 RF前端 处理和数字前端处理, 得到处理后的接收信号; 对处理后的接收信号进行 基于复合匹配滤波的 PSS检测,利用 CP相关进行载波频偏估计,并进行分 段非相干的 SSS检测, 在保证系统性能的基础上实现高效低复杂度的小区 搜索。 附图说明
图 1是现有技术中同步信号的时域结构图;
图 2是本发明实施例实现小区搜索方法的流程示意图;
图 3是本发明实施例对接收的信号进行 RF前端处理和数字前端处理的 流程示意图;
图 4是本发明实施例对处理后的接收信号进行基于复合匹配滤波的 PSS检测的流程示意图;
图 5是本发明实施例采用折叠结构的匹配滤波器的示意图; 图 6是本发明实施例利用循环前缀相关完成载波频偏估计的示意图; 图 8是本发明实施例进行分段非相干的 SSS检测的流程示意图; 图 9是本发明实施例干扰消除的流程示意图;
图 10是本发明实施例待检测小区和干扰小区的定时关系的示意图; 图 11是本发明实施例门限检测简化方法的流程示意图;
图 12是本发明实施例小区搜索方法中各个步驟调度流程示意图; 图 13是本发明实施例实现小区搜索系统的结构示意图。 具体实施方式
本发明实施例的基本思想是:对接收的信号进行 RF前端处理和数字前 端处理, 得到处理后的接收信号; 对处理后的接收信号进行基于复合匹配 滤波的 PSS检测, 利用 CP相关进行载波频偏估计, 并进行分段非相干的 SSS检测。
下面通过附图及具体实施例对本发明实施例再做进一步的详细说明。 本发明实施例提供一种小区搜索方法, 图 2是本发明实施例实现小区 搜索方法的流程示意图, 如图 2所示, 该方法包括以下步驟:
步驟 201 , 对接收的信号进行 RF前端处理和数字前端处理, 得到处理 后的接收信号;
具体的,如图 3所示,天线接收的信号通过射频(RF, Radio Frequency ) 前端处理后, 进行数字前端处理; 其中, RF前端处理包括 RF信号调理、 滤波、 下变频以及模数转换等处理; 数字前端处理包括射频缺陷接收 IQ数 据补偿、 AGC、 下采样滤波、 天线间加权处理(天线间能量平衡)等处理, 得到处理后的接收信号;其中下采样滤波处理是将采样频率 30.72MHz的数 据降采样到 1.92MHz; 小区搜索部分主要包括 PSS检测、 载波频偏检测和 SSS检测, 小区搜索部分对数字前端处理后的接收信号进行进一步的检测 和处理。
步驟 202, 对处理后的接收信号进行基于复合匹配滤波的 PSS检测; 具体的, 如图 4所示, 对处理后的时域数据进行 PSS检测时, 在初始 小区搜索阶段, 如果初始频偏超过 PSS检测所容忍的频偏范围, 则进行预 设载波频偏处理, 用于抵制大频偏影响, 同时也能获得大频偏范围, 获得 初始载波频偏估计; 其中, 预设载波频偏处理的方法为当 PSS未检测到符 合要求的相关峰值时, 启动预设载波频偏模块, 依次利用频偏设置值 +10KHz 或 -ΙΟΚΗζ 或通过调整 RF 的电压控制振荡器 (VCO , Voltage Controlled Oscillator ), 再次进行 PSS检测, 这样即使存在整数倍子载波的 频偏, PSS检测也能获得较好的性能;
然后, 根据下述公式组, 将处理后的接收信号与 IFFT变换的 PSS时域 序列进行匹配滤波处理, 其中, 匹配滤波处理为复合匹配滤波处理, 在初 始小区搜索阶段, 进行分段匹配滤波, 在同步维护和邻小区搜索阶段, 进 行非分段对称匹配滤波; 进行匹配滤波处理后, 求取匹配滤波后的能量, 以及天线间能量的累加, 得到总能量:
Figure imgf000009_0001
公式组( 1 ) 中, Pu 为第 "个 PSS第 ra个接收信号样点偏移匹配滤波的 总能量, s„ (Z)为第 M个 PSS采用的 Zadoff-Chu码进行时域转换后的第 个系 数, "可以为 25、 29、 34, (·)*表示共轭操作, +ζ)为第 ρ根天线第„+ /个样 点偏移量接收信号值, Ν为 PSS时域序列的长度, Μ为 PSS时域序列的分 段个数, ^表示分段序号, L为分段的每段长度, /^为1¾累加的天线数目; 在初始小区搜索阶段, 由于 UE未与基站取得频率同步, 为了增大 PSS检 测对抗频偏的能力, 采用分段匹配滤波处理, 本实施例中, 分段个数 Μ可 以为 2, 在 [-5,5]ΚΗζ频偏内能够获得较好的 PSS检测性能; 在初始小区搜 索阶段之后的同步维护和邻小区搜索阶段, UE已经与基站取得基本的频率 同步, 频偏控制在较小范围内, 为了不影响搜索性能, 采用非分段对称匹 配滤波处理, 分段个数 M为 1; 此外, 由于匹配滤波器的系数具有部分对称 性质, 采用非分段对称匹配滤波方法时, 还可以采用折叠结构, 如图 5 所 示, 匹配滤波器的阶数可采用 64或 128, 其中阶数 64的匹配滤波器需要将 采用频率 1.92MHz降采样到 960KHz, 和阶数 128的匹配滤波器相比, 阶 数 64 的匹配滤波器的滤波运算量将节省一半; 对于匹配滤波器组, 由于 Zadoff-Chu码号 29和 34的匹配滤波器的系数互为共轭关系, 码号为 34时 的滤波结果可以由码号为 29的滤波结果通过组合转化得到, 所以最终为两 组匹配滤波器;
由于在实际网络中不可避免的存在 UE在进行小区搜索时受到大信号 的干扰, 如旁边有其他 UE在进行通话和数据业务, 或 UE处于基站波束赋 形的强增益方向上且基站正在赋形等, 这些将可能造成强干扰位置的相关 值掩盖了正确位置的相关值, 导致定时同步失败及小区组内 ID误检; 因此 为了克服大信号干扰的影响, 本实施例中采用对经过匹配滤波、 求取能量 和天线间能量累加处理后的接收信号进行相关能量归一处理;
信号能量的估计如下述公式:
Figure imgf000010_0001
公式(2 ) 中, 为第„个样点偏移量信号符号能量估计值, 、为 第 P根天线第 个样点偏移量接收信号值, N为时域符号长度, 即 PSS时 域序列的长度, /^为 UE累加的天线数目。 实际计算中, 可以将公式(2 ) 转换成递推形式, 减少运算量。 相关能量归一处理中的半帧滤波能量值的 计算过程如下述公式:
[Pu (n) 其他
其中^ )为第 "个 PSS第"个样点偏移量的半帧滤波能量值, 为 第"个 PSS第 n个接收信号样点偏移量匹配滤波的总能量, W")为第"个 样点偏移量信号符号能量估计值, 《为能量判别门限;
接着, 需要进行半帧非相干累加, 为了获得较好的 PSS检测性能, 将 多个半帧滤波能量值进行累加 (")=∑^ ), 其中 β为累加的半帧中 PSS 符号数, )为第 "个 PSS第"个样点偏移量的半帧滤波能量值, ^(")为 第"个 PSS第"个样点偏移量的 β个半帧滤波累加总能量值;
最后, 进行门限计算和峰值搜索判别, 得到小区的峰值, 由峰值的位 置, 可得到小区 PSS符号的定时位置, 由所在的匹配滤波器组, 可得到小 区的组内 ID。
步驟 203, 利用 CP相关进行载波频偏估计;
具体的, 在步驟 202 中, 进行频偏范围的控制, 得到初始载波频偏估 计, 使得频偏在 [-5,5]KHz内, 还需要进行二次载波频偏估计, 使频偏控制 在后续处理需要的正常工作范围内, 如 [-l,l]KHz内; 本实施例中, 利用循 环前缀(CP , Cyclic Prefix )相关完成载波频偏估计, 如图 6所示, 载波 频偏估计的计算过程如下:
CI =∑∑∑ rVA (n + TI Q +Ατ)- rpq (n + TI Q +Ατ + Ν), ATG [-ε,ε]
JI . =argmax|c0AJ2, J f =^(Ci j)
其中, ^O, 1, 其中 0代表扩展 CP模式, 1代表常规 CP模式, Δτ为 相关窗口修正值, 主要为了 PSS检测定时存在偏差情况下, 仍 能包括到符号边界, ς 表示第 种模式相对 CP 边界偏差 ΔΓ的相关值, 为第 ρ根天线第 个下行符号的接收信号, 表示第 种模式第 q 个下行符号的定时, "为相关累加序号, L;为相关累加长度; N为时域符 号长度, P为接收端累加的天线数目, β为累加的符号数, 7为两种 CP模 式峰值比较的归一化因子, 由 ^和^的长度关系以及相关符号数决定, 表示对应第 种模式相对 CP边界偏差 Δτ;的相关值, Δ /表示求得的归一化 频率偏移值, 即所需的载波频偏估计, 简称频偏, 公式中的 argC)e [- , π), 一般地, ^ =32 , ½=9; 这里, 通过 J; ^计算还能得到所需的 值。 实际计 算过程中, 的计算可以转换成递推形式完成, 减小运算量。
理论上,基于 CP相关的载波频偏估计得到的范围为 [-7.5, 7.5] KHz, 能 够很好的实现与步驟 202得到的剩余载波频偏衔接,而且由于 CP的数目较 多 , 多个符号间进行相关值进行累加能够获得较好的估计性能;
例如, 对于小区对于下行业务符号不发送的情况, 为不影响载波频偏 估计的性能,优选的,下行业务符号和 CP相关,如图 7( A )所示,在 TDD-LTE 下, 选择半帧第一时隙和第二时隙的前 3个符号, 在 FDD-LTE下, 选择第 一时隙的后 3个符号和第二时隙前 4个符号; 相关值分别按照偶数半帧和 奇数半帧进行累加,若 C:和 分别表示第 i种模式相对 CP边界偏差 Δτ的 偶数半帧相关累加值和奇数半帧相关累加值, 这里利用下述公式计算 .:
Figure imgf000012_0001
其中 7为两种 CP模式峰值比较的归一化因子, 由 ^和^的长度关系以 及相关符号数决定, Δτ为相关窗口修正值, A e [-£,£·], Ar,为相对 CP边界 偏差的具体位置;
此外, 利用 CP相关也能进行小区 CP类型的检测, 上述的最大峰值对 应的 CP模式即是检测的 CP模式, 属于现有技术, 这里不再赘述; 另一种 错开两种 CP模式的重叠部分,基于相关窗的 CP模式检测方法,如图 7(B ) 所示, 这种方法中, 选择相同的相关累加长度, ^ = 2=9 , CP模式通过如 下公式判别: CP模式 =
Figure imgf000013_0001
其中 cO o和 ¾ΔΓ分别表示扩展 CP和常规 CP相关窗内相关累加值, 为门限系数, 这种 CP模式的 值的计算方法比传统的计算方法运算量更 低, 复杂度更低, 且性能较优; 此外, 本实施例中由于符号和 CP较多, 因 而具有良好的估计性能; 这里, 利用 CP相关进行二次载波频偏估计以及 CP模式检测仅在初始小区搜索阶段进行,进入同步维护和邻小区搜索阶段, 由于载波频偏出于 3艮踪状态, 二次载波频偏估计功能可以关闭。
步驟 204, 进行分段非相干的 SSS检测;
具体的, 如图 8所示, SSS检测是在进一步载波频偏补偿后, 利用步 驟 202中得到的小区 PSS符号的定时位置和小区的组内 ID, 得到小区 ID 和无线帧边界, 步驟 203中的 CP模式的检测也可以放在本步驟中;
下面以不支持 CP模式检测的方式为例,依据经过 PSS检测得到的定时 位置, 在经过 PSS检测后的接收信号中获取各天线上的 SSS时域符号, 所 述 SSS时域符号经过时频转换后, 得到 SSS频域符号, 然后将 SSS频域符 号与本地 SSS序列分子帧 0和子帧 5两种相位进行分段相关, 求取能量, 并依次进行段间能量累加和天线间能量累加, 将多个半帧的相关值按照子 帧 0与子帧 5、子帧 5与子帧 0两种配对方式进行非相干累加,得到总能量:
L(2- 」
p 〉: 〉:
L(2- p ( 〉:」〉:
Figure imgf000013_0002
= 0,1,' " , 7VG— 1
其中 s—。 、 s:ubframe5{k)分别表示子顿 0和子顿 5的第 i组本地码第 k 个子载波的 SSS序列值, Rp,2 fc)表示第 p根天线接收的第 2 个 SSS频域符 号上第 k个子载波值, 表示第 p根天线接收的第 2^1个 SSS频域符 号上第 k个子载波值, L表示分段的每段长度, 表示分段序号, M表示 分个数, 表示接收端累加的天线数目, β表示累加的 PSS符号数, ^表 示相关的 ID组数;最后对段间能量累加和天线间能量累加得到的 SSS相关 能量, 进行门限计算, 对该门限计算结果和总能量进行峰值搜索; 这里, 对于支持 CP模式检测的方式, 需要对按照两种 CP模式获取的 SSS时域符 号, 分别进行 SSS检测;
3GPP协议中, SSS序列采用的是 M序列加扰方式, SSS序列的互相关 特性明显差于 M序列的性能, 相同组内 ID的条件下, 互相关最大相关干 扰峰与真实相关峰值之比达到 0.4839, 因此在多小区环境下, 特别是微蜂 窝 , 信号强度较弱的小区检测将受到信号强度较大的小区的干扰峰值影响; 在非初始小区搜索阶段, 可以获得信号强度较大的小区, 获取的方法可以 是服务小区或由测量信息, 如参考信号接收功率 (RSRP, Reference Signal Receiving Power )、 信号与干扰力口噪声比 ( SINR, Signal to Interference plus Noise Ratio )等判别得到, 这样在 SSS检测中可进行干扰消除处理;
具体的干扰消除方法如图 9所示, 待检测小区和干扰小区的定时关系 如图 10所示, 干扰消除的具体过程如下: = ∑ {S' ik)} - S! (k) - e
Figure imgf000014_0001
公式中 S 、 (W分别表示第 i组本地 SSS码第 k个子载波值和干扰 小区的本地 SSS码第 k个子载波值, L表示分段的每段长度, N表示时域 符号的样点数, d表示检测小区和干扰小区定时偏差样点数, 一般地,
| | < N/i6 , 表示接收的转换后的 SSS频域, 表示第 m段针对于第 i 组本地码的干扰因子, 表示第 m段干扰信号的相关值, C表示干扰消除 前第 m段第 i组本地码的相关值, 表示干扰消除后第 m段第 i组本地码 的相关值;
为减少干扰消除的运算量, 本实施例中, 采用一种门限检测简化方法, 如图 11所示, 小区的 SSS检测中, SSS时域符号经过时频转换后得到 SSS 频域符号, 若小区定时 (PSS检测获得)和干扰小区定时不满足设定的定 时偏差, 则不启动干扰小区, 按照正常的 SSS检测流程进行, 否则启动干 扰消除功能; 干扰估计按设定待检 ID组号进行,对于初始的第一个 SSS频 域符号, 待检的 ID组号为所有的 ID组号; 对于经过多个符号进行非相干 累加的,由上一次门限检测过门限峰值对应 ID组号确定下一次待检测的 ID 组号, 这样可以缩小待检测的 ID组号的数目, 减小运算量; 不启动干扰消 该门限选择相对较宽松;
这里, 采用分段非相干 SSS检测的方法, 相比于相干方法, 复杂度更 低, 资源消耗更小,还能够克服高速移动和同频相同小区 PSS组内 ID情况 下的相干检测方法性能的下降问题, 能够获得满足需求的性能, 而且, SSS 检测仅需要利用 SSS 符号, 因而能够更好的应用于 TDD-LTE、 FDD-LTE 和多模系统中。
本发明实施例中, 小区搜索包含初始小区搜索和非初始小区搜索两个 阶段,各个步驟调用如图 12所示, 以在 SSS检测时进行 CP模式检测为例, 初始小区搜索阶段调度流程如图 12 ( A )所示, 按照首先启动 PSS检测, 接着对 PSS检测符合要求的最大相关峰值进行载波频偏估计, 经历载波频 偏估计后, 进行 SSS检测, 由于 PSS检测可能检测到多个过门限 PSS相关 峰值, 需要多次调用 SSS检测, 因而采用串行处理模式; 对于非初始小区 搜索, 步驟调度如图 12 ( B ), 由于已经完成了基本的小区频率同步, 载波 频偏估计不启动, 可以采用类似流水方式, 即 PSS检测和 SSS检测并行处 理, 而且对于 PSS检测和 SSS检测采用多个符号非相干累加, 还可以采用 每次累加便启动峰值搜索, 高信噪比小区采用提前输出机制, 缩短其检测 时间。
为实现上述方法, 本发明实施例还提供一种小区搜索系统, 图 13是本 发明实施例实现小区搜索系统的结构示意图, 如图 13所示, 该系统包括: 前端处理模块 131、 PSS检测模块 132、 频偏估计模块 133、 SSS检测模块 134; 其中,
前端处理模块 131 , 用于对接收的信号进行 RF前端处理和数字前端处 理, 得到处理后的接收信号;
PSS检测模块 132,用于对处理后的接收信号进行基于复合匹配滤波的 PSS检测;
频偏估计模块 133 , 用于利用 CP相关进行载波频偏估计;
SSS检测模块 134, 用于进行分段非相干的 SSS检测。
所述 PSS检测模块 132对处理后的接收信号进行基于复合匹配滤波的 PSS检测为:对处理后的接收信号与 IFFT变换的 PSS时域序列进行复合匹 配滤波处理, 并求取匹配滤波后的能量, 进行天线间能量的累加和相关能 量归一处理, 得到半帧滤波能量值; 将多个半帧滤波能量值进行累加, 进 行门限计算和峰值搜索判别, 得到小区的峰值, 根据峰值的位置, 得到小 区 PSS符号的定时位置, 根据匹配滤波器, 得到小区的组内 ID。
其中, 所述进行复合匹配滤波处理为: 在初始小区搜索阶段, 进行分 段匹配滤波, 在同步维护和邻小区搜索阶段, 进行非分段对称匹配滤波。 在初始小区搜索阶段, 所述进行分段匹配滤波之前, 如果初始频偏超 过 PSS检测所容忍的频偏范围, 进行预设载波频偏处理; 所述预设载波频 偏处理为: 当 PSS未检测到符合要求的相关峰值时, 启动预设载波频偏模 块,依次利用频偏设置值 +10KHz或 -ΙΟΚΗζ或通过调整 RF的电压控制振荡 器( VCO )再次进行 PSS检测, 获取初始载波频偏估计。
所述频偏估计模块 133利用 CP相关进行载波频偏估计为:利用循环前 缀(CP)相关完成载波频偏估计, 所述载波频偏估计为 4"=larg(CiA7), 其中, 和 .根据 J =argmax|c。,Ar|2, ·| |2|确定, |c。,Ar|和 /或 根据
Q-l P-l h-l
CI =∑∑∑ rVA (n + TI Q + AT) · rp q (n + TI Q + Δ + N)确定;
=0 p=0 n=0
其中, =0或 1, 其中 0代表扩展 CP模式, 1代表常规 CP模式, Δτ为 相关窗口修正值, Δ^[-£·,£·], 表示第 种模式相对 CP边界偏差 的 相关值, rM(M)为第;根天线第 个下行符号的接收信号, 表示第 种模式 第 个下行符号的定时, "为相关累加序号, L;为相关累加长度; N为时域 符号长度, P为接收端累加的天线数目, β为累加的符号数, 7为两种 CP 模式峰值比较的归一化因子, 由! ^口 的长度关系以及相关符号数决定, Ji j表示对应第 i种模式相对 CP边界偏差 Δτ;的相关值, Δ/表示求得的归一 化频率偏移值, arg(*)e [-π, π), L, =32 , ½=9;
所述 SSS检测模块 134进行分段非相干的 SSS检测为: 根据 PSS检测 得到的定时位置, 在经过 PSS检测后的接收信号中获取各天线上的 SSS时 域符号, 经过时频转换后得到 SSS频域符号, 将 SSS频域符号与本地 SSS 序列分子帧 0和子帧 5两种相位进行分段相关, 求取能量, 并进行段间能 量累加和天线间能量累加, 对段间能量累加和天线间能量累加得到的 SSS 相关能量, 进行门限计算; 将多个半帧的相关值按照子帧 0与子帧 5、 子帧 5与子帧 0两种配对方式进行非相干累加,得到总能量,对门限计算结果和 总能量进行峰值搜索。
以上所述, 仅为本发明实施例的较佳实施例而已, 并非用于限定本发 明实施例的保护范围, 凡在本发明实施例的精神和原则之内所作的任何修 改、 等同替换和改进等, 均应包含在本发明实施例的保护范围之内。

Claims

权利要求书
1、 一种小区搜索方法, 其特征在于, 该方法包括:
对接收的信号进行射频 (RF )前端处理和数字前端处理, 得到处理后 的接收信号;
对处理后的接收信号进行基于复合匹配滤波的主同步信号( PSS )检测, 利用循环前缀(CP )相关进行载波频偏估计, 并进行分段非相干的辅同步 信号 (SSS )检测。
2、 根据权利要求 1所述的方法, 其特征在于, 所述对处理后的接收信 号进行基于复合匹配滤波的 PSS检测为:
对处理后的接收信号与 IFFT变换的 PSS时域序列进行复合匹配滤波处 理, 并求取匹配滤波后的能量, 进行天线间能量的累加和相关能量归一处 理, 得到半帧滤波能量值;
将多个半帧滤波能量值进行累加, 进行门限计算和峰值搜索判别, 得 到小区的峰值, 根据峰值的位置, 得到小区 PSS符号的定时位置, 根据匹 配滤波器, 得到小区的组内 ID。
3、 根据权利要求 2所述的方法, 其特征在于, 所述进行复合匹配滤波 处理为: 在初始小区搜索阶段, 进行分段匹配滤波, 在同步维护和邻小区 搜索阶段, 进行非分段对称匹配滤波。
4、 根据权利要求 3所述的方法, 其特征在于, 在初始小区搜索阶段, 所述进行分段匹配滤波之前, 该方法还包括: 如果初始频偏超过 PSS检测 所容忍的频偏范围, 进行预设载波频偏处理;
所述预设载波频偏处理为: 当 PSS未检测到符合要求的相关峰值时, 启动预设载波频偏模块, 依次利用频偏设置值 +10KHz或 -ΙΟΚΗζ或通过调 整 RF的电压控制振荡器( VCO )再次进行 PSS检测, 获取初始载波频偏 估计。
5、 根据权利要求 1所述的方法, 其特征在于, 所述利用 CP相关进行 载波频偏估计为:
所述载波频偏估计为 Δ/ =— arg(CiAr ) , 其中 , 和 ΛΤ]根据 Ji . 确 定 , |C。J 和 I 或 |C1Ar| 根 据 I q (n + TI Q + Δ + N)确定;
Figure imgf000020_0001
其中, =0或 1, 其中 0代表扩展 CP模式, 1代表常规 CP模式, Δτ为 相关窗口修正值, Δ^[-£·,£·], CiA7表示第 种模式相对 CP边界偏差 的 相关值, rM(M)为第;根天线第 个下行符号的接收信号, 表示第 种模式 第 个下行符号的定时, "为相关累加序号, L;为相关累加长度; N为时域 符号长度, P为接收端累加的天线数目, β为累加的符号数, 7为两种 CP 模式峰值比较的归一化因子, 由! 1和 L2的长度关系以及相关符号数决定, Ji j表示对应第 i种模式相对 CP边界偏差 Δτ;的相关值, Δ/表示求得的归一 化频率偏移值, arg(*)e [-π, π), L, =32 , ½=9。
6、 根据权利要求 1所述的方法, 其特征在于, 所述进行分段非相干的 SSS检测为:
根据 PSS检测得到的定时位置, 在经过 PSS检测后的接收信号中获取 各天线上的 SSS时域符号, 经过时频转换后得到 SSS频域符号, 将 SSS频 域符号与本地 SSS序列分子帧 0和子帧 5两种相位进行分段相关, 求取能 量, 并进行段间能量累加和天线间能量累加, 对段间能量累加和天线间能 量累加得到的 SSS相关能量, 进行门限计算;
将多个半帧的相关值按照子帧 0与子帧 5、子帧 5与子帧 0两种配对方 式进行非相干累加, 得到总能量, 对门限计算结果和总能量进行峰值搜索。
7、 根据权利要求 6所述的方法, 其特征在于, 该方法还包括: 当小区定时和干扰小区定时不满足设定的定时偏差时, 不启动干扰消 除, 按照正常的 SSS检测流程进行; 否则启动干扰消除, 干扰估计按设定 待检组内 ID号进行。
8、 根据权利要求 7所述的方法, 其特征在于, 所述干扰消除为: 利 用 和
Figure imgf000021_0001
X {^(wi . RW进行干扰消除; 其中, ^、 (W分别表示第 i组本 k=mL
地 SSS码第 k个子载波值和干扰小区的本地 SSS码第 k个子载波值, L表 示分段的每段长度, N表示时域符号的样点数, d表示检测小区和干扰小区 定时偏差样点数, | < N/i6 , 表示接收的转换后的 SSS 频域, 表示 第 m段针对于第 i组本地码的干扰因子, 表示第 m段干扰信号的相关值, C表示干扰消除前第 m段第 i组本地码的相关值, ^表示干扰消除后第 m 段第 i组本地码的相关值。
9、 根据权利要求 5或 6所述的方法, 其特征在于,
该方法还包括: 利用 CP相关进行小区 CP类型的检测;
所述利用 CP相关进行小区 CP类型的检测为: 基于相关窗的 CP模式 进行检测, 选择相同的相关累加长度, Zi = L2 = 9 , CP 模式通过
CP模式 = 0
Figure imgf000021_0002
> ¾ 判别, 其中 c0 和 <¾,ΔΓ分别表示扩展 CP和常
[l 其他 规 CP相关窗内相关累加值, 为门限系数。
10、 一种小区搜索系统, 其特征在于, 该系统包括: 前端处理模块、 PSS检测模块、 频偏估计模块、 SSS检测模块; 其中,
前端处理模块, 用于对接收的信号进行 RF前端处理和数字前端处理, 得到处理后的接收信号;
PSS检测模块,用于对处理后的接收信号进行基于复合匹配滤波的 PSS 检测;
频偏估计模块, 用于利用 CP相关进行载波频偏估计; SSS检测模块, 用于进行分段非相干的 SSS检测。
PCT/CN2011/080930 2011-10-18 2011-10-18 一种小区搜索方法及系统 WO2013056421A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2011/080930 WO2013056421A1 (zh) 2011-10-18 2011-10-18 一种小区搜索方法及系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2011/080930 WO2013056421A1 (zh) 2011-10-18 2011-10-18 一种小区搜索方法及系统

Publications (1)

Publication Number Publication Date
WO2013056421A1 true WO2013056421A1 (zh) 2013-04-25

Family

ID=48140303

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/080930 WO2013056421A1 (zh) 2011-10-18 2011-10-18 一种小区搜索方法及系统

Country Status (1)

Country Link
WO (1) WO2013056421A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20150072983A (ko) * 2013-12-20 2015-06-30 주식회사 쏠리드 Lte 프레임 동기 검출 방법 및 장치, 이를 적용한 중계 장치
CN110463088A (zh) * 2017-05-04 2019-11-15 华为技术有限公司 同步信号的发送方法、同步信号的接收方法及相关设备

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101552762A (zh) * 2009-02-25 2009-10-07 北京天碁科技有限公司 循环前缀长度类型及辅同步信号检测的方法及装置
CN101689930A (zh) * 2007-07-06 2010-03-31 Lg电子株式会社 在无线通信系统中执行小区搜索的方法
CN102158250A (zh) * 2010-02-11 2011-08-17 富士通株式会社 初始小区搜索方法及装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101689930A (zh) * 2007-07-06 2010-03-31 Lg电子株式会社 在无线通信系统中执行小区搜索的方法
CN101552762A (zh) * 2009-02-25 2009-10-07 北京天碁科技有限公司 循环前缀长度类型及辅同步信号检测的方法及装置
CN102158250A (zh) * 2010-02-11 2011-08-17 富士通株式会社 初始小区搜索方法及装置

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20150072983A (ko) * 2013-12-20 2015-06-30 주식회사 쏠리드 Lte 프레임 동기 검출 방법 및 장치, 이를 적용한 중계 장치
KR101626457B1 (ko) * 2013-12-20 2016-06-01 주식회사 쏠리드 Lte 프레임 동기 검출 방법 및 장치, 이를 적용한 중계 장치
US10206192B2 (en) 2013-12-20 2019-02-12 Solid, Inc. Method and device for detecting LTE frame synchronization, and relay device applying the same
CN110463088A (zh) * 2017-05-04 2019-11-15 华为技术有限公司 同步信号的发送方法、同步信号的接收方法及相关设备
US10652843B2 (en) 2017-05-04 2020-05-12 Huawei Technologies Co., Ltd. Techniques for synchronization in wireless communications
CN110463088B (zh) * 2017-05-04 2020-07-24 华为技术有限公司 同步信号的发送方法、同步信号的接收方法及相关设备
US11330539B2 (en) 2017-05-04 2022-05-10 Huawei Technologies Co., Ltd. Techniques for synchronization in wireless communications
US11991650B2 (en) 2017-05-04 2024-05-21 Huawei Technologies Co., Ltd. Techniques for synchronization in wireless communications

Similar Documents

Publication Publication Date Title
EP2437453B1 (en) Secondary synchronization signal detection method and device
EP2497223B1 (en) Frequency and time synchronization methods and apparatus
CA2570748C (en) Signal acquisition in a wireless communication system
EP2987370B1 (en) Secondary cell synchronization for carrier aggregation
CN102695263B (zh) 一种鲁棒的lte系统下行定时同步方法
JP4339362B2 (ja) 直交周波数分割多元接続システムにおけるセルを検出する方法及び装置
CN101719890B (zh) 一种应用于长期演进系统的小区搜索方法和装置
CN107241794B (zh) 一种针对tdd-ofdm下行链路的快速同步跟踪方法及装置
CN102694763A (zh) 一种用于td-lte系统整数倍频偏估计的方法
WO2007140670A1 (fr) procédé pour réaliser une synchronisation dans un système de multiplexage par répartition orthogonale de la fréquence à entrées multiples, sorties multiples
WO2008048060A1 (en) Tdm based cell search method for ofdm system
JP2011514721A (ja) 周波数誤差推定
TW200913522A (en) Preamble sequence detection and integral carrier frequency offset estimation method for OFDM/OFDMA wireless communication system
WO2008155238A2 (en) Simultaneous cell group and cyclic prefix detection method, apparatus and system
CN108282434B (zh) 一种lte下行主同步信号的检测方法
WO2014063275A1 (zh) 一种远距离同频干扰源的确定方法以及定位方法
CN112714448B (zh) 动态频谱共享中的主同步信号检测方法
CN101001234B (zh) 正交频分复用系统发送信号的方法及小区搜索的方法
WO2017140590A1 (en) Nb-iot receiver operating at minimum sampling rate
CN105532051B (zh) 用于载波聚合的辅小区同步
CN112738000B (zh) 一种pss分块互相关检测方法
WO2013056421A1 (zh) 一种小区搜索方法及系统
CN104980184A (zh) 用于td-scdma系统的小区搜索方法和设备
WO2012171407A1 (zh) 一种确定时间同步位置的方法及设备
CN103532896B (zh) 可变带宽系统的定时估计方法和装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11874220

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 11874220

Country of ref document: EP

Kind code of ref document: A1