WO2007079680A1 - Procédé de correction de décalage de fréquence pour système de communication à cellules duplex à répartition dans le temps à large bande et procédé de première recherche de cellule - Google Patents

Procédé de correction de décalage de fréquence pour système de communication à cellules duplex à répartition dans le temps à large bande et procédé de première recherche de cellule Download PDF

Info

Publication number
WO2007079680A1
WO2007079680A1 PCT/CN2007/000079 CN2007000079W WO2007079680A1 WO 2007079680 A1 WO2007079680 A1 WO 2007079680A1 CN 2007000079 W CN2007000079 W CN 2007000079W WO 2007079680 A1 WO2007079680 A1 WO 2007079680A1
Authority
WO
WIPO (PCT)
Prior art keywords
frequency offset
pilot
time domain
sequence
offset correction
Prior art date
Application number
PCT/CN2007/000079
Other languages
English (en)
French (fr)
Inventor
Shaohui Sun
Yang Yu
Yingmin Wang
Original Assignee
Shanghai Ultimate Power Communications Technology Co., Ltd.
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 Shanghai Ultimate Power Communications Technology Co., Ltd. filed Critical Shanghai Ultimate Power Communications Technology Co., Ltd.
Publication of WO2007079680A1 publication Critical patent/WO2007079680A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/26—Systems using multi-frequency codes
    • H04L27/2601—Multicarrier modulation systems
    • H04L27/2647—Arrangements specific to the receiver only
    • H04L27/2655—Synchronisation arrangements
    • H04L27/2657—Carrier synchronisation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/26—Systems using multi-frequency codes
    • H04L27/2601—Multicarrier modulation systems
    • H04L27/2647—Arrangements specific to the receiver only
    • H04L27/2655—Synchronisation arrangements
    • H04L27/2668—Details of algorithms
    • H04L27/2673—Details of algorithms characterised by synchronisation parameters
    • H04L27/2675—Pilot or known symbols
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/26—Systems using multi-frequency codes
    • H04L27/2601—Multicarrier modulation systems
    • H04L27/2647—Arrangements specific to the receiver only
    • H04L27/2655—Synchronisation arrangements
    • H04L27/2668—Details of algorithms
    • H04L27/2681—Details of algorithms characterised by constraints
    • H04L27/2684—Complexity
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L7/00—Arrangements for synchronising receiver with transmitter
    • H04L7/04—Speed or phase control by synchronisation signals
    • H04L7/10—Arrangements for initial synchronisation

Definitions

  • the present invention relates to an LTE-SCDMA (Time Division Synchronous Code Division Multiple Access) evolution scheme broadband time division duplex cellular system, and more particularly to a frequency offset correction method for a broadband time division duplex cellular system and application method thereof User terminal
  • LTE-SCDMA Time Division Synchronous Code Division Multiple Access
  • TD-SCDMA is the only three international standards for third-generation mobile communication systems that adopt Time Division Duplex (TDD) mode to support uplink and downlink asymmetric service transmission, and has greater flexibility in spectrum utilization.
  • the system combines advanced technologies in wireless communications such as smart antennas, uplink synchronization, joint detection and software radio to provide high performance and spectrum utilization.
  • TD-SCDMA time division duplex cellular system
  • Broadband time division duplex cellular systems also present some problems while providing higher speed and greater capacity.
  • the base station and the UE both transmit and receive at a nominal carrier frequency. Since the crystal oscillator as a local oscillator in the UE has limited precision, it will cause crystal drift. As a result of the crystal drift, after a period of time, the frequency of the UE deviates from the system, causing great damage to the performance of the UE. Therefore, when performing initial cell search, the UE needs to estimate the frequency offset by using the signal sent by the base station, and complete the frequency offset correction accordingly.
  • the transmission of the data portion may adopt an OFDM (Orthogonal Frequency Division Multiplexing) method.
  • OFDM puts forward higher requirements for frequency offset correction.
  • the UE needs to achieve a frequency offset less than ⁇ (kHz) during the initial cell search to ensure the performance of the OFDM system.
  • the second is to use two or more adjacent OFDM pilot symbols in the time slot TS0 of the broadcast channel.
  • the frequency offset information is obtained by differential correlation between the previous symbol and the latter symbol. , to perform frequency offset correction. Also, due to the limitation of the OFDM pilot symbol length, the accuracy of the frequency offset correction is limited to only a few KHz.
  • the present invention is to solve the problem that the prior art frequency offset correction method cannot meet the accuracy requirements required for a broadband time division duplex cellular system.
  • the frequency offset correction method of the broadband time division duplex cellular system of the present invention comprises the steps of: receiving a time domain sequence of two columns of pilot symbols having a predetermined interval in a time slot for transmitting a broadcast channel; and using two received pilot time domains The sequence performs a conjugate correlation operation with a pilot time domain sequence of the locally stored corresponding pilot symbols;
  • the frequency offset correction is performed based on the obtained calculation result.
  • the conjugate correlation operation between the two received pilot time domain sequences and the locally stored pilot time domain sequence is specifically: combining two received pilot time domain sequences with locally stored corresponding pilot symbols The pilot time domain sequences are conjugate multiplied separately, and the two conjugate multiplied results are correlated.
  • the method further comprises: after receiving the frequency offset correction, receiving two columns of pilot symbols and a time domain sequence of data symbols adjacent to the respective column pilot symbols as two receiving sequences;
  • the combination of the two comprising pilot symbols and adjacent data symbols has the same predetermined interval.
  • the time domain sequence of receiving two columns of pilot symbols in a time slot for transmitting a broadcast channel is specific
  • the method is: sampling according to the maximum bandwidth supported by the user terminal and the cell access system, and obtaining a time domain sequence of two columns of pilot symbols for transmitting the time slot of the broadcast channel.
  • the method further includes: determining a bandwidth of the cell access system according to the time domain signal sampling sequence of the pilot symbol and the bandwidth matching sequence set; and the pilot symbol including the cell in the bandwidth matching sequence set corresponds to different A pilot sample sequence of bandwidth.
  • the determining the bandwidth of the cell access system is specifically:
  • the method further includes: performing frequency offset estimation and correction by calculating a phase offset between the received downlink synchronization codes.
  • the two columns of pilot symbols are respectively located on respective subcarriers of the frequency domain of two OFDM symbols.
  • the two columns of pilot symbols are respectively located on subcarriers with a predetermined carrier spacing in two frequency bands of OFDM symbols, and the difference between the transmit power and the average transmit power of the data symbols is determined by a predetermined carrier interval; In the OFDM symbol in which it is located, if its subcarrier does not transmit a pilot symbol, the subcarrier does not transmit any other symbols.
  • the pilot symbols are omnidirectionally transmitted by using a common pilot manner.
  • the time slot for transmitting the broadcast channel is a TS0 time slot.
  • the present invention provides a user terminal, including a time domain sampling unit, a storage unit, a frequency offset estimation sequence unit, and a frequency offset correction unit, where:
  • the time domain sampling unit is configured to sample a signal in a time slot in which the broadcast channel is transmitted to generate a time domain sequence
  • the storage unit is configured to store information, including a pilot time domain sequence of pilot symbols
  • the frequency offset estimation sequence unit is configured to generate two received frequency offset estimation sequences with predetermined intervals according to the time domain sequence output by the time domain sampling unit, and each of the received frequency offset estimation sequences includes two columns of pilots for transmitting the broadcast channel. a time domain sequence of one of the symbols; generating two local frequency offset estimation sequences respectively corresponding to the received frequency offset estimation sequence, and each local frequency offset estimation sequence includes a corresponding one in the storage unit a pilot time domain sequence of pilot symbols;
  • the frequency offset correction unit performs frequency offset correction on the conjugate correlation operation result of the two received frequency offset estimation sequences generated by the frequency offset estimation sequence unit and the corresponding local frequency offset estimation sequence.
  • the frequency offset estimation sequence unit generates two received frequency offset estimation sequences according to the time domain sequence output by the time domain sampling unit, specifically: using a time domain sequence of two columns of pilot symbols output by the time domain sampling unit as the receiving frequency. Partial estimation sequence;
  • the frequency offset estimation sequence unit generates two local frequency offset estimation sequences, specifically: using the pilot time domain sequence corresponding to the two columns of pilot symbols stored in the storage unit as two corresponding local frequency offset estimation sequences.
  • the frequency offset estimation sequence unit generates two received frequency offset estimation sequences according to the time domain sequence output by the time domain sampling unit, specifically: two columns of pilot symbols output by the time domain sampling unit and data adjacent thereto a time domain sequence of symbols as a sequence of received frequency offset estimates;
  • the frequency offset estimation sequence unit generates two local frequency offset estimation sequences, which are: performing hard decision on the time domain sequence of the data symbol and combining with the pilot time domain sequence corresponding to the adjacent pilot symbols in the storage module respectively. Two local frequency offset estimation sequences.
  • the user terminal further includes a bandwidth control unit, configured to determine a bandwidth supported by the cell access system and the user terminal, as a sampling frequency of the time domain sampling unit.
  • a bandwidth control unit configured to determine a bandwidth supported by the cell access system and the user terminal, as a sampling frequency of the time domain sampling unit.
  • the information stored by the storage unit further includes a bandwidth matching sequence set, where the pilot symbol, the bandwidth, and a pilot sampling sequence corresponding to each pilot symbol and each bandwidth are included;
  • the user terminal further includes a system bandwidth unit, configured to determine a bandwidth of the cell access system according to a time domain sequence of the pilot symbols output by the time domain sampling unit and a correlation peak of the pilot sample sequence in the bandwidth matching sequence set, and output to the Bandwidth control unit.
  • a system bandwidth unit configured to determine a bandwidth of the cell access system according to a time domain sequence of the pilot symbols output by the time domain sampling unit and a correlation peak of the pilot sample sequence in the bandwidth matching sequence set, and output to the Bandwidth control unit.
  • the time slot for transmitting the broadcast channel is a TS0 time slot.
  • the invention provides a cell initial search method for a broadband time division duplex cellular system, which comprises the following steps:
  • the user terminal performs coarse frequency offset correction
  • Fine-precision offset correction according to a received reception time domain sequence of two columns of pilot symbols having a predetermined interval, and a conjugate correlation result of two pilot time-domain sequences stored on the pilot symbol stored by the user terminal;
  • the user terminal reads the cell broadcast information.
  • the method further comprises: after performing fine frequency offset correction:
  • the performing the coarse frequency offset correction by the user terminal is specifically: performing frequency offset estimation and correction by calculating a phase offset between the received downlink synchronization codes.
  • the present invention performs frequency offset correction by using a time domain sequence of two columns of pilot symbols and a conjugate correlation value of a time domain sequence of corresponding pilot symbols stored on a user terminal, since two columns of pilot sequences exist in TS0. a certain interval, when the phase difference existing between the two columns of correlation signals can be used for the frequency offset estimation by correlation, the frequency offset value corresponding to each phase angle difference unit is higher with the increase of the interval; The accuracy of the frequency offset estimation is improved by the predetermined interval of the two columns of pilot symbols, which can meet the requirements of the broadband time division duplex cellular system, and is simple to implement;
  • the present invention performs the conjugate correlation operation together with the time domain sequence of the adjacent data symbols, and the length of the sequence participating in the correlation operation becomes longer, the accuracy of the calculation is improved, and the frequency offset is further realized. Correction.
  • FIG. 1 is a schematic diagram of a frame structure of a TD-SCMDA and an evolved system thereof;
  • FIG. 2 is a diagram showing an example of setting of a TS0 time slot pilot symbol in the present invention
  • FIG. 3 is a flowchart of a frequency offset correction method according to the present invention.
  • FIG. 5 is a diagram showing an example of selecting a data symbol for frequency offset correction in the present invention.
  • FIG. 6 is a schematic structural diagram of a user terminal to which the frequency offset correction method of the present invention is applied;
  • FIG. 7 is a flowchart of a cell initial search method applying the frequency offset correction method according to the present invention.
  • each wireless sub The frame consists of 7 time slots TS0, TS1 to TS6 and three special time slots.
  • the time slots TS0 to TS6 are used to transmit data, and the three special time slots are DwPTS (downlink pilot channel), UpPTS (uplink pilot channel), and GP (conversion guard slot), where DwPTS is used to transmit the initial cell.
  • the downlink pilot is searched, the UpPTS is used to send a random access signal, and the GP is a guard interval for the downlink time slot to switch to the uplink time slot, and its length determines the maximum value of the cell coverage radius.
  • the cell initial search of the UE refers to a process in which the UE searches from the power-on to the login to the appropriate cell. After the UE logs in to the cell, it can obtain more information about the cell and information of the neighboring cell, and listen to the paging or initiate the call of the wireless network.
  • the initial search of the cell includes synchronization and frequency offset correction.
  • the synchronization mainly uses the signal transmitted by the DwPTS time slot base station to complete the frequency offset correction, and the UE can obtain the cell SYNC-DL transmitted by the DwPTS time slot during the synchronization process. Therefore, before performing the frequency offset correction, the UE can obtain the pilot symbols used by the cell according to the SYNC-DL.
  • the subcarrier of the OFDM symbol is 15 ⁇ . Therefore, when the length of the CP (Cyclic Prefix) is relatively short, it can be included in one downlink time slot. 9 OFDM symbols; if the CP length is long, 8 OFDM symbols can be placed.
  • the OFDM pilot symbols are typically placed in time slot TS0.
  • the TS0 slot in the present invention has two columns of pilot symbols, and there is a certain interval between the two columns of pilot symbols. This interval can improve the accuracy of fine synchronization and facilitate the UE to perform channel estimation and decoding.
  • One possible setup is shown in Figure 2.
  • the 0.60 ms (millisecond) TS0 slot includes 9 OFDM symbols, with the second and eighth columns being pilot symbols.
  • the subcarriers in which the pilot symbols are located in the same OFDM symbol may have a certain carrier spacing, or the pilot symbols may be placed on each subcarrier without the subcarrier spacing. If there is a subcarrier spacing, no symbols are placed in the OFDM symbol including the pilot symbols on the subcarriers as the interval, that is, the subcarriers in which the pilot symbols are not placed in the same OFDM symbol transmitting the pilot symbols do not transmit signals.
  • the OFDM symbol in which the pilot symbol is located is referred to as a pilot OFDM symbol. In the pilot OFDM symbol, if its subcarrier does not transmit a pilot symbol, the subcarrier does not transmit any other symbol.
  • the transmit power of each subcarrier of the pilot OFDM symbol can be adjusted according to the placement interval of the pilot symbols on the subcarriers. For example, if one pilot symbol is placed every two subcarriers, then the transmit power of each subcarrier of the pilot OFDM symbol can be higher than when all pilot carriers are placed on all subcarriers.
  • the transmit power of each subcarrier is 3 dB (decibel) higher; if one pilot symbol is placed every 4 subcarriers, the transmit power of each subcarrier can be as high as 6 dB, and so on.
  • the subcarrier spacing of the pilot symbols can be adjusted according to the size of the system bandwidth.
  • the pilot symbols adopt a common pilot manner of omnidirectional transmission.
  • FIG. 3 is a flow chart showing the frequency offset correction method of the present invention.
  • step S310 pilot sequences of respective pilot symbols corresponding to respective bandwidths are pre-stored in the UE.
  • the bandwidth can be above 20 MHz and can support different bandwidth operations, such as 1.25MHz, 2.5MHz, 5MHz, 10MHz and 20MHz.
  • bandwidth of the cell access system is different, the pilot time domain sequence transmitted by the cell base station is different for the same pilot symbol.
  • the sample reception is always performed according to the minimum requirement of the access system bandwidth.
  • the access system transmits a pilot sequence corresponding to a bandwidth of 5 MHz in a pilot symbol of TS0, and performs an IFFT (Inverse Fast Fourier Transform) transform on the OFDM pilot symbol sequence, by using a frequency domain.
  • the signal is converted into a time domain signal.
  • the UE samples the time domain signal, and the minimum bandwidth required by the access system is 1.25 MHz, and the sampling frequency of the UE adopts the minimum bandwidth requirement of the system is 1.25 MHz, and the pilot sampling with the same sampling number as the minimum system bandwidth is obtained by sampling. sequence.
  • the bandwidth of the system is different, and the pilot sequence is different.
  • the pilot sample sequence sampled according to the minimum system bandwidth is different.
  • other systems larger than the minimum system bandwidth requirement may have a bandwidth corresponding to a pilot sample sequence, from which a set of bandwidth matching sequences may be generated.
  • step S320 the bandwidth of the access system is determined according to the time domain signal sampling sequence of the pilot symbols and the bandwidth matching sequence set.
  • the UE samples the time domain signal of the pilot symbol with the minimum bandwidth required by the access system,
  • the time-domain sequence obtained by the sampling is correlated with all the pilot sample sequences of the pilot symbols used by the cell in the bandwidth matching sequence set.
  • the bandwidth corresponding to the pilot sample sequence with the highest correlation value is the cell access system. Bandwidth.
  • the bandwidth of the access system is determined in steps S310 and S320 in order to enable the UE to operate at the highest possible bandwidth to speed up the frequency offset correction. These two steps can also be omitted.
  • step S330 a time domain sequence of two columns of pilot symbols is received in the TS0 slot.
  • the pilot symbol can be received by using the maximum bandwidth supported by both the UE and the cell access system.
  • the bandwidth access capability of the UE is greater than or equal to the bandwidth of the current cell access system, the UE may sample according to the bandwidth of the cell access system, and receive a time domain sequence of pilot symbols; when the UE has the bandwidth access capability
  • the bandwidth of the current cell access system is smaller than that of the current cell access system, the UE may sample according to its maximum bandwidth and receive a time domain sequence of pilot symbols.
  • step S340 the two received pilot time domain sequences are conjugate-correlated with the locally stored pilot time domain sequence of the corresponding pilot symbols.
  • Equation (3) L is between the two pilot time domain sequences according to the sampling rate of the UE The number of sampling intervals.
  • step S350 frequency offset correction is performed based on the result of the conjugate correlation operation.
  • step S340 the frequency offset ⁇ / can be obtained:
  • the UE can be frequency offset corrected.
  • a relatively accurate frequency offset correction can be obtained by the correlation operation of two columns of pilot symbols having a predetermined interval in TS0.
  • the obtained frequency offset estimation is more accurate.
  • the signal-to-noise ratio of the correlation output is related to the sequence length N of the correlation operation. The longer N is, the larger the signal-to-noise ratio of the correlation output is.
  • the time domain sequence length of the pilot symbols is also determined. As the frequency offset becomes smaller during the frequency offset correction process, the data symbols in the vicinity of the pilot symbols can be relatively accurately estimated. Therefore, it is possible to introduce a time domain sequence of the data symbols to perform correlation operations to increase the sequence length. More accurate frequency offset estimation.
  • step S360 a time domain sequence of two columns of pilot symbols and data symbols respectively adjacent to each column of pilot symbols is received as two receiving sequences, that is, a time domain sequence of the received pilot symbols and adjacent data symbols thereof.
  • Common composition (and ( ⁇ ) is, a time domain sequence of the received pilot symbols and adjacent data symbols thereof.
  • step S370 the time domain sequence of the data symbols is hard-decised and combined with the corresponding locally stored pilot time domain sequences to form two local sequences.
  • Figure 5 shows a possible data symbol selection.
  • the received first column pilot symbol and the time domain sequence of the left data symbol may be combined to form a receiving sequence r/( ), which is twice the length; likewise, the received second
  • the column pilot symbol and the time domain sequence of the data symbol to its right together form another reception sequence r 2 '(/t) having a length that is twice (after a hard decision on the sampled value of the received data symbol,
  • the generated local sequences (k) and S (k) are also twice as large as (k) and ⁇ (k), respectively.
  • the additional data symbols in Figure 5 are selected at both ends of the pilot symbols such that the two combinations of pilot symbols and adjacent data symbols used to perform the frequency offset estimation still have a predetermined interval. This ensures that the sampling interval L between the two received sequences is constant for better frequency offset correction.
  • step S380 further frequency offset correction is performed according to the conjugate correlation result of the two received sequences and the corresponding local sequence.
  • the conjugate correlation operation and the frequency offset estimation method of the two received sequences and the local sequence are the same as those in steps S340 and S350, except that the sequence of the operation is changed and will not be repeated here.
  • the above-described frequency offset correction method of the present invention can be used in combination with the frequency offset correction method in the prior art.
  • the coarse frequency offset correction is performed by the method in the prior art, that is, the frequency offset estimation and correction are performed by calculating the phase offset of the received downlink synchronization code, so that the frequency offset error is reduced to several KHz, and then the present invention is used.
  • the above method of the invention performs fine frequency offset correction, so that the speed and accuracy of the frequency offset correction can be ensured at the same time.
  • FIG. 6 is a schematic structural diagram of a user terminal to which the frequency offset correction method of the present invention is applied.
  • the storage unit 620 is connected to the frequency offset estimation series unit 630 and the system bandwidth unit 660, respectively; the frequency offset estimation series unit 630 is connected to the frequency offset correction unit 640; and the bandwidth control unit 650 is connected to the system bandwidth unit 660.
  • the storage unit 620 stores information required for the user terminal to perform frequency offset correction, such as a pilot time domain sequence of pilot symbols, a bandwidth matching sequence set for determining system bandwidth, and the like.
  • the time domain sampling unit 610 samples the downlink radio frequency signal of the TS0 time slot according to the bandwidth used by the user terminal to generate a time domain sequence, which includes a time domain sequence of pilot symbols and data symbols in the TS0 time slot.
  • the frequency offset estimation sequence unit 630 generates the four sequences required for performing the frequency offset correction based on the time domain sequence output by the time domain sampling unit 610 and the pilot time domain sequence stored in the storage unit 620.
  • 2 The received frequency offset estimation sequence is generated according to the output of the time domain sampling unit 610, and the two received frequency offset estimation sequences have predetermined intervals, each of which includes a time domain sequence of one of the two columns of pilot symbols of the TS0 slot.
  • the other two sequences are local frequency offset estimation sequences respectively corresponding to the two received frequency offset estimation sequences, and each local frequency offset estimation sequence includes pilots stored in the storage unit 620 for pilot symbols in the corresponding received frequency offset estimation sequence.
  • Time domain sequence When the time-frequency sequence of the data symbols is included in the received frequency offset estimation sequence, the local frequency offset estimation sequence further includes a time domain sequence generated according to the hard decision result of the time domain sample values of the corresponding data symbols.
  • the output sequence of the two columns of pilot symbols in the TS0 slot in the time domain sampling unit 610 may be used as two received frequency offset estimation sequences; accordingly, the pilot time domain sequence of the corresponding pilot symbols stored in the storage unit 620 is used as Two local frequency offset estimation sequences.
  • the frequency offset estimation sequence unit 630 outputs the generated four sequences for frequency offset correction to the frequency offset correction unit 640.
  • the frequency offset estimation sequence unit 630 may also output different received frequency offset estimation sequences and corresponding local frequency offset estimation sequences to the frequency offset correction unit 640 in order to perform frequency offset correction with different precision.
  • the frequency offset correction unit 640 performs a conjugate correlation operation on the two received frequency offset estimation sequences generated by the frequency offset estimation sequence unit 630 and the corresponding local frequency offset estimation sequence, and performs frequency offset correction according to the operation result.
  • the bandwidth control unit 650 determines the bandwidth used by the user terminal according to the bandwidth supported by the cell access system and the bandwidth supported by the user terminal, and uses the determined bandwidth as the sampling frequency of the time domain sampling unit 610.
  • System bandwidth unit 660 is used to determine the bandwidth used by the cell access system.
  • the system bandwidth unit 660 correlates the time domain sequence of the pilot symbols output by the time domain sampling unit 610 with the respective pilot samples of the corresponding pilot symbols in the bandwidth matching sequence set in the storage unit 620, and has a correlation maximum value.
  • the bandwidth corresponding to the pilot sampling sequence is the bandwidth of the cell access system.
  • the system bandwidth unit 660 outputs the determined cell access system bandwidth to the bandwidth control unit 650. It is used to select the bandwidth used by the user terminal.
  • FIG. 7 is a flow chart showing a cell initial search method applying the frequency offset correction method of the present invention.
  • the UE performs cell coarse synchronization.
  • DwPTS can be used for cell synchronization search and coarse frequency offset correction.
  • the DwPTS slot consists of two parts, one part being an idle period during which the base station does not transmit any signal; the other part is a SYNC-DL code, which is a finite length pseudo-random sequence.
  • the wideband time division duplex cellular system has a set of SYNC-DL code sequences, and each cell is assigned a SYNC-DL code as the ID number of the cell.
  • the SYNC-DL code transmitted by the DwPTS slot is a single-carrier signal.
  • the bandwidth of the transmitted signal can be set according to the minimum bandwidth of the broadband time division duplex cellular system. For example, the minimum bandwidth is 1.25MHz or 1.6MHz, and the SYNC-DL code is transmitted.
  • the single carrier bandwidth is less than or equal to this bandwidth.
  • the SYNC_DL pseudo-random code of the DwPTS There are guard intervals on both sides of the SYNC_DL pseudo-random code of the DwPTS. Since the base station does not signal at the guard interval, the UE receives little RF power at the guard interval, and the SYNC-DL block base station transmits at full power. From the received power spectrum of the UE, the SYNC-DL code segment is peak compared to the received power of the guard interval between the two sides. When the sum of the power received by the guard interval between the two sides is divided by the sum of the power of the SYNC-DL segment, the ratio is small.
  • step S720 the UE receives the signal of the approximate location of the SYNC_DL code segment, and uses the correlation algorithm to confirm the used downlink synchronization sequence.
  • SYNC-DL uses a different pseudo-random code sequence. In this way, the cell is further distinguished by confirming the SYNC-DL code.
  • step S730 the UE performs fine synchronization of the cell.
  • the UE performs the correlation algorithm by using the confirmed downlink synchronization pilot to confirm the relevant peak position, thereby implementing downlink fine synchronization of the UE.
  • the UE performs coarse frequency offset correction.
  • the coarse frequency offset correction can use the prior art frequency offset adjustment method to estimate and correct the frequency offset of the phase offset at both ends of the received SYNC-DL code by using the SYNC-DL code of the DwPTS slot.
  • step S750 the UE performs fine frequency offset correction.
  • Fine frequency offset correction can adopt the steps in the present invention The method in S320 to step S380 or some of the steps thereof are not repeated here.
  • the UE reads the broadcast information of the cell to complete the process of initial cell search.
  • the new frequency offset correction method and the cell initial search method proposed by the invention can realize fast cell initial search.
  • the pilot symbols and data symbols of the TS0 slot can be used for frequency offset estimation and correction.
  • This method can improve the accuracy of UE frequency offset estimation and reduce the complexity of implementation. It provides an effective solution for downlink synchronization and frequency offset correction for broadband time division duplex cellular systems.

Landscapes

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

Description

宽带时分双工蜂窝系统的频偏校正方法及小区初搜方法 技术领域
本发明涉及 TD-SCDMA ( Time Division Synchronous Code Division Multiple Access, 时分同步码分多址）的演进方案宽带时分双工蜂窝系统， 尤 其涉及宽带时分双工蜂窝系统的频率偏移校正方法及应用该方法的用户终 端
背景技术
TD-SCDMA是第三代移动通信系统的三种大国际标准中唯一采用时分 双工 （TDD， Time Division Duplex)方式， 支持上下行非对称业务传输， 在 频谱利用上具有较大的灵活性。 该系统综合采用了智能天线、 上行同步、 联 合检测和软件无线电等无线通信中的先进技术， 使系统具有较高的性能和频 谱利用率。
随着社会的发展以及技术的进步， 人们对移动通信的要求不断提高， 希 望系统能够提供大容量、 高速率、 低时延的数据传输服务。 为了满足这种日 益增长的需求， TD-SCDMA系统同样需要不断演进和提高性能。 TD-SCDMA 的演进系统为了得到高速率大容量的服务， 需要占用更宽的带宽， 因而将其 称为宽带时分双工蜂窝系统。 宽带时分双工蜂窝系统在提供更高速率更大容 量的同时也带来了一些问题。
在 TD-SCDMA系统中， 基站和 UE (User Equipment, 用户终端）均以 标称的载波频率进行发送和接收。 由于 UE中作为本振源的晶振精度有限， 会造成晶振漂移。晶振漂移的结果是过了一段时间后， UE的频率与系统出现 偏差，对 UE的性能造成较大的损害。因而 UE在进行小区初始搜索时， 需要 利用基站发送的信号对其频率偏移进行估计， 并据以完成频偏校正。
在宽带时分双工蜂窝系统中， 数据部分的传输可以采用 OFDM (Orthogonal Frequency Division Multiplexing, 正交频分复用）方式。 OFDM 对频偏校正提出了更高的要求， UE在小区初搜过程中需要达到频率偏移小于 ΙΚΗζ (千赫）， 以保证 OFDM系统的性能。
现有技术中， 宽带时分双工蜂窝系统中常用的频偏校正方法有两种： 其一是利用 DwPTS时隙中的 SYNC-DL (下行同步码）码， 通过对接收 SYNC-DL码两端的相位偏移进行频偏估计和校正。 由于 SYNC-DL码长度的 限制， 频偏校正的精度有限， 通过 SYNC-DL码进行的频偏校正精度只可能 达到几个 KHz;
其二是采用发送广播信道的时隙 TS0中两个以上相邻的 OFDM导频符 号， 根据 OFDM导频符号在时域的重复特性， 通过前一个符号和后一个符号 进行差分相关获得频偏信息， 以进行频偏校正。 同样， 由于 OFDM导频符号 长度的限制， 频偏校正的精度有限， 只能达到几个 KHz。
发明内容
本发明要解决的是现有技术中频偏校正方法不能达到宽带时分双工蜂窝 系统所需的精度要求。
本发明所述宽带时分双工蜂窝系统的频偏校正方法包括以下步骤： 在发送广播信道的时隙接收具有预定间隔的两列导频符号的时域序列； 将两个接收的导频时域序列与本地存储的对应导频符号的导频时域序列 进行共轭相关运算；
根据所得的运算结果进行频偏校正。
优选地， 所述将两个接收的导频时域序列与本地存储的导频时域序列进 行共轭相关运算具体为： 将两个接收的导频时域序列与本地存储的对应导频 符号的导频时域序列分别共轭相乘， 将两个共轭相乘结果进行相关运算。
优选地， 所述方法在进行频偏校正之后还包括- 接收两列导频符号及与分别与各列导频符号相邻的数据符号的时域序列 作为两个接收序列；
将数据符号的时域序列进行硬判决后与对应于其相邻导频符号的本地存 储的导频时域序列分别组合构成两个本地序列；
根据两个接收序列与对应的本地序列的共轭相关运算结果进行进一步频 偏校正。
优选地， 所述两个包括导频符号和相邻的数据符号的组合具有同样的预 定间隔。
优选地， 所述在发送广播信道的时隙接收两列导频符号的时域序列具体 为： 根据用户终端和小区接入系统都支持的最大带宽进行采样， 获得发送广 播信道的时隙两列导频符号的时域序列。
优选地， 所述方法还包括： 根据导频符号的时域信号采样序列和带宽匹 配序列集合来确定小区接入系统的带宽； 所述带宽匹配序列集合中包括该小 区的导频符号对应于不同带宽的导频采样序列。
优选地， 所述确定小区接入系统的带宽具体为：
以小区接入系统的最小带宽对导频符号的时域信号进行采样；
将时域信号的采样序列与带宽匹配序列集合中该导频符号的各个导频采 样序列进行相关运算；
查找相关值最大的采样序列， 其对应的带宽即为小区接入系统的带宽。 优选地， 所述方法在确定小区接入系统的带宽之前还包括： 通过计算接 收的下行同步码两端的相位偏移进行频偏估计和校正。
可选地，所述两列导频符号分别位于两个 OFDM符号频域的各个子载波 上。
优选地，所述两列导频符号分别位于两个 OFDM符号频域具有预定载波 间隔的子载波上， 其发射功率高于数据符号平均发射功率的差值由预定载波 间隔确定；在导频符号所在的 OFDM符号中，如果其子载波不发送导频符号， 则该子载波也不发送其他任何符号。
优选地， 所述导频符号采用公共导频方式进行全向发送。
优选地， 所述发送广播信道的时隙为 TS0时隙。
本发明提供了一种用户终端， 包括时域采样单元、 存储单元、 频偏估计 序列单元和频偏校正单元， 其中：
时域采样单元用来对发送广播信道的时隙中的信号进行采样以生成时域 序列；
存储单元用来存储信息， 其中包括导频符号的导频时域序列；
频偏估计序列单元用来根据时域采样单元输出的时域序列生成两个具有 预定间隔的接收频偏估计序列， 每个接收频偏估计序列中各包括发送广播信 道的时隙两列导频符号中一个的时域序列； 生成与接收频偏估计序列分别对 应的两个本地频偏估计序列， 每个本地频偏估计序列中包括存储单元中对应 导频符号的导频时域序列；
频偏校正单元根据频偏估计序列单元生成的两个接收频偏估计序列与对 应的本地频偏估计序列的共轭相关运算结果进行频偏校正。
优选地， 所述频偏估计序列单元根据时域采样单元输出的时域序列生成 两个接收频偏估计序列具体为： 将时域采样单元输出的两列导频符号的时域 序列作为接收频偏估计序列；
所述频偏估计序列单元生成两个本地频偏估计序列具体为： 将存储单元 中存储的对应于两列导频符号的导频时域序列作为两个对应的本地频偏估计 序列。
优选地， 所述频偏估计序列单元根据时域采样单元输出的时域序列生成 两个接收频偏估计序列具体为： 将时域采样单元输出的两列导频符号及与其 分别相邻的数据符号的时域序列作为接收频偏估计序列；
所述频偏估计序列单元生成两个本地频偏估计序列具体为： 将数据符号 的时域序列进行硬判决后与存储模块中对应于其相邻导频符号的导频时域序 列分别组合生成两个本地频偏估计序列。
优选地， 所述用户终端还包括带宽控制单元， 用来确定小区接入系统和 所述用户终端均支持的带宽， 作为时域采样单元的采样频率。
优选地， 所述存储单元所存储的信息还包括带宽匹配序列集合， 其中包 括导频符号、 带宽及对应于每个导频符号和每个带宽的导频采样序列；
所述用户终端还包括系统带宽单元， 用来根据时域采样单元输出的导频 符号的时域序列与带宽匹配序列集合中导频采样序列的相关峰值确定小区接 入系统的带宽， 并输出至带宽控制单元。
优选地， 所述发送广播信道的时隙为 TS0时隙。
本发明提供了一种宽带时分双工蜂窝系统的小区初搜方法， 包括以下步 骤：
用户终端进行小区同步；
用户终端进行粗频偏校正；
根据接收的具有预定间隔的两列导频符号的接收时域序列、 用户终端存 储的对 导频符号的两个导频时域序列的共轭相关结果进行精频偏校正； 用户终端读取小区广播信息。
优选地， 所述方法在进行精频偏校正后还包括：
接收两列导频符号及与该导频符号相邻的数据符号的时域序列作为两个 接收序列；
将数据符号的时域序列进行硬判决后与对应于其相邻导频符号的本地存 储的导频时域序列分别组合构成两个本地序列；
根据两个接收序列与对应的本地序列的共轭相关运算结果进行进一步频 偏校正。
优选地， 所述用户终端进行粗频偏校正具体为： 通过计算接收的下行同 步码两端的相位偏移进行频偏估计和校正。
本发明利用两列导频符号的时域序列， 与在用户终端上存储的对应导频 符号的时域序列的共轭相关值来进行频偏校正， 由于 TS0中两列导频序列之 间存在一定的间隔， 当通过相关计算两列相关信号之间存在的相位差可以用 于频偏估计时， 每个相位角度差单元对应的频偏值随着间隔的增加， 其精度 越高； 本发明通过两列导频符号的预定间隔提高了频偏估计的精度， 能够满 足宽带时分双工蜂窝系统的要求， 并且实现简单；
进一步地， 本发明将导频符号与相邻数据符号的时域序列一同进行共轭 相关运算， 随着参与相关运算的序列长度变长， 计算的准确度得到提高， 进 而更快地实现频偏校正。
附图说明
图 1为 TD-SCMDA及其演进系统的帧结构示意图；
图 2为本发明中 TS0时隙导频符号设置的示例图；
图 3为本发明所述频偏校正方法的流程图；
图 4为构成带宽匹配序列集合的示例图；
图 5为本发明中选择数据符号进行频偏校正的示例图；
图 6为应用本发明所述频偏校正方法的用户终端的结构示意图； 图 7为应用本发明所述频偏校正方法的小区初搜方法的流程图。
具体实施方式
TD-SCMDA及其演进系统的帧结构形式如图 1所示。 图中， 每个无线子 帧由 7个时隙 TS0、 TS1至 TS6和三个特殊时隙构成。其中，时隙 TS0至 TS6 用来传送数据， 三个特殊时隙分别为 DwPTS (下行导频信道)、 UpPTS (上 行导频信道）和 GP (转换保护时隙），其中 DwPTS用于发送小区初搜的下行 导频， UpPTS用于发送随机接入信号， GP是下行时隙向上行时隙转换的保 护间隔， 它的长度决定了小区覆盖半径的最大值。
UE的小区初搜是指 UE从开机搜索到登录合适小区的过程。 UE只有在 登录小区后， 才能获取本小区更多信息和邻近小区的信息， 以及监听无线网 络的寻呼或发起呼叫。小区初搜包括同步和频偏校正， 同步主要利用 DwPTS 时隙基站发射的信号在频偏校正之前完成， UE 在同步过程中能够获得 DwPTS 时隙发射的小区 SYNC-DL。 因而， 在进行频偏校正前， UE根据 SYNC-DL可以得到小区使用的导频符号。
对 TS0至 TS6中采用 OFDM调制方式的每个下行时隙， 其 OFDM符号 的子载波为 15ΚΉζ， 因此， 当 CP (Cyclic Prefix, 循环前缀） 的长度比较短 时， 在一个下行时隙中可以包含 9个 OFDM符号； 如果 CP长度比较长， 可 以放置 8个 OFDM符号。
OFDM导频符号通常设置在时隙 TS0中。本发明中的 TS0时隙具有两列 导频符号， 这两列导频符号之间有一定间隔。 这个间隔可以提高精同步的精 度， 并且有利于 UE进行信道估计和解码。 一种可能的设置方式如图 2所示， 0.675ms (毫秒） 的 TS0时隙包括 9个 OFDM符号， 其中第 2列和第 8列为 导频符号。
同一 OFDM符号中导频符号所在的子载波之间可以具有一定的载波间 隔， 也可以不存在子载波间隔而将导频符号放置在每个子载波上。 如果存在 子载波间隔时，作为间隔的子载波上在包括导频符号的 OFDM符号中不放置 任何符号， 即发射导频符号的同一 OFDM符号中未放置导频符号的子载波不 发送信号。导频符号所在的 OFDM符号简称导频 OFDM符号，在导频 OFDM 符号中，如果其子载波不发送导频符号， 则该子载波也不发送其他任何符号。 这样， 根据导频符号在子载波上的放置间隔， 导频 OFDM符号的每个子载波 的发射功率可以做调整。 例如， 每两个子载波放置一个导频符号， 则导频 OFDM符号的每个子载波发射功率可以比所有子载波全部放置导频符号时每 个子载波的发射功率高 3dB (分贝）； 如果每 4个子载波放置一个导频符号， 则每个子载波的发射功率可以高 6dB， 以此类推。 通过提高放置导频符号的 子载波的发射功率， 使得 UE能够更好地接收导频符号， 进行频偏校正和信 道估计。 导频符号的子载波间隔可以根据系统带宽的大小进行调整。
为了让小区的所有 UE都可以利用 TS0时隙的导频符号进行频偏校正， 导频符号采用全向发送的公共导频方式。
图 3所示为本发明所述频偏校正方法的流程图。 在步骤 S310， 在 UE中 预先存储各个导频符号对应于各个带宽的导频采样序列。
在宽带时分双工蜂窝系统中， 带宽可以到 20MHz以上，可以同时支持不 同带宽工作， 如 1.25MHz、 2.5MHz, 5MHz、 10MHz和 20MHz等。 当小区 接入系统的带宽不同时， 对相同的导频符号， 小区基站发射的导频时域序列 是不同的。
当 UE接收 TS0时隙的导频符号时， 总是按接入系统带宽的最小要求来 进行采样接收。为了能够通过导频符号的时域采样序列得知接入系统的带宽， 需要在 UE中存储对应于各个导频符号和各个带宽的时域采样序列， 以进行 带宽匹配， 在本文中称之为带宽匹配序列集合。 由于导频符号与小区 SYNC-DL的关联关系， 也可以根据 SYNC-DL的 ID (标识）号来生成时域 采样序列与带宽的对应关系。
例如， 在图 4中， 接入系统在 TS0的导频符号中发射对应为 5MHz带宽 的导频序列， OFDM导频符号序列进行 IFFT (Inverse Fast Fourier Transform, 快速傅立叶逆变换）变换， 由频域信号转换成时域信号。 UE对该时域信号进 行采样， 设接入系统要求的最小带宽是 1.25MHz, 则 UE的釆样频率采用系 统最小的带宽要求 1.25MHz, 通过采样得到与最小系统带宽相同抽样数目的 导频采样序列。 系统的带宽不同， 发射导频序列就不同， 按照最小系统带宽 抽样的导频采样序列也就不同。 同理， 其他大于最小系统带宽要求的系统， 其带宽可以与一个导频采样序列相对应， 据此可以生成带宽匹配序列集合。
回到图 3， 在步骤 S320, 按照导频符号的时域信号采样序列和带宽匹配 序列集合来确定接入系统的带宽。
通常， UE以接入系统要求的最小带宽对导频符号的时域信号进行采样， 再将采样所得的时域序列与带宽匹配序列集合中所在小区使用的导频符号的 所有导频采样序列进行相关运算， 显然相关值最高的导频采样序列所对应的 带宽即为小区接入系统的带宽。
可见， 当 UE工作于小区接入系统支持的不同带宽时， 所接收的导频符 号时域序列具有不同的序列长度。 在本发明中， 导频符号的时域序列越长， 可以达到越精确的频偏校正。 因而在步骤 S310和步骤 S320中确定接入系统 的带宽是为了使 UE工作于尽可能高的带宽， 以加快频偏校正的速度。 这两 步也可以省略。
在步骤 S330， 在 TS0时隙接收两列导频符号的时域序列。
如果执行本步骤前 UE已经得到接入系统的带宽，则可以采用 UE和小区 接入系统都支持的最大带宽来接收导频符号。 当 UE具有的带宽接入能力大 于或等于目前小区接入系统的带宽时， UE可以按该小区接入系统的带宽进行 采样， 接收导频符号的时域序列； 当 UE具有的带宽接入能力小于目前小区 接入系统的带宽时， UE可以按照其最大带宽进行采样，接收导频符号的时域 序列。
在步骤 S340， 将两个接收的导频时域序列与本地存储的对应导频符号的 导频时域序列进行共轭相关运算。
假设接收到第一列和第二列导频符号时域序列的值分别为 (k)和 r2 (k)， 其中t = l,...,N， N为序列 7 ( )和^( )的长度。
由于导频符号已知， 与该接收时域序列对应的本地存储的导频时域序列 为 )和 ^(：)， 其中 A = 1,...,N， 序列 和 W的长度同样为 N。
将接收的时域序列 与本地的导频时域序列 W共扼相乘，得到 ： gl{k) = A - e^ k^ , k = l,..,N ( 1 )； 将接收的时域序列 r2( 与本地的导频时域序列 共扼相乘， 得到 g2(k) = A - eJi2^f^] , k = i, ..,N (2)； 在公式 （1 )和 （2) 中， Δ/为频偏； A为解调后信号的幅度； 7为釆样 周期。 将 和 g2W进行相关， 得到相关结果 z :
N—\
Z =∑g (k) - g2 (k) = N - A - β 2π^ ) (3)； 在公式（3)， L为按照 UE的采样速率在两个导频时域序列之间具有的抽 样间隔数。
在步骤 S350, 根据共轭相关运算的结果进行频偏校正。
根据步骤 S340中的公式 （3)， 可以得到频偏 Δ/ :
Af (4)；
Figure imgf000011_0001
根据频偏 Δ/， 即可对 UE进行频偏校正。
可见， 通过 TS0中具有预定间隔的两列导频符号的相关运算可以获得比 较精确的频偏校正。 同时， 两个导频时域序列之间的抽样间隔 L越大， 与频 偏 Δ/对应的的相位值也越大， 因而得到的频偏估计就越准确。
另外， 如果相关输出的信噪比越高， 得到的频偏估计也越准确。 相关输 出的信噪比与进行相关运算的序列长度 N有关系， N越长， 相关输出的信噪 比越大。 当小区接入系统和 UE的带宽确定时， 导频符号的时域序列长度也 是确定的。 而随着频偏校正过程中频率偏移的变小， 可以比较准确地估计出 导频符号附近的数据符号， 因此， 可以考虑引入数据符号的时域序列进行相 关运算， 以通过增加序列长度得到更精确的频偏估计。
在步骤 S360， 接收两列导频符号及分别与各列导频符号相邻的数据符号 的时域序列作为两个接收序列， 即将接收的导频符号和其相邻的数据符号的 时域序列共同构成 ( 和 (^)。
在步骤 S370, 将数据符号的时域序列进行硬判决后与对应的本地存储的 导频时域序列分别组合构成两个本地序列。
将接收的数据符号的采样值进行硬判决， 按照数据符号与相邻导频符号 的时域顺序， 将硬判决结果和本地储存的相邻导频符号对应的导频时域序列 共同构成本地序列， 则可生成两个与接收队列相对应的本地序列。
根据图 2中 TS0时隙的导频符号设置， 图 5给出了一种可能的数据符号 选择方式。 可以将接收的第一列导频符号和其左边的数据符号的时域序列共 同组成一个接收序列 r/( )， 其长度是 的两倍； 同样， 可以将接收的第二 列导频符号和其右边的数据符号的时域序列共同组成另一个接收序列 r2'(/t)， 其长度是 ( 的两倍。 在对接收的数据符号的采样值进行硬判决后， 生成的 本地序列 (k)和 S (k)也同样分别是 (k)和 ^ (k)的两倍。
图 5中增加数据符号选在导频符号的两端， 使得由导频符号和相邻数据 符号组成的两个用来进行频偏估计的组合之间仍然具有预定间隔。 这样可以 保证两个接收序列之间的抽样间隔 L不变， 以实现更好的频偏校正效果。
在时域序列中增加了数据符号后，由于用来进行频偏估计的数据量增大， 就可以得到更为准确的频偏估计， 加快频偏校正的速度。
请再参阅图 3， 在步骤 S380, 根据两个接收序列与对应的本地序列的共 轭相关结果进行进一步频偏校正。 两个接收序列与本地序列的共轭相关运算 和频偏估算方法与步骤 S340及步骤 S350中相同， 只是进行运算的序列发生 了变化， 此处不再重复。
为了加快频偏校正的速度， 本发明的上述频偏校正方法可以与现有技术 中的频偏校正方法结合使用。 先采用现有技术中的方法进行粗频偏校正， '即 通过计算接收的下行同步码两端的相位偏移进行频偏估计和校正， 使频偏误 差缩小到几个 KHz之内， 再采用本发明的上述方法进行精频偏校正， 从而可 以同时保证频偏校正的速度和精度。
图 6所示为应用本发明所述频偏校正方法的用户终端的结构示意图， 时 域采样单元 610分别与频偏估计系列单元 630、频偏校正单元 640、带宽控制 单元 650和系统带宽单元 660连接； 存储单元 620分别与频偏估计系列单元 630和系统带宽单元 660连接； 频偏估计系列单元 630连接至频偏校正单元 640； 带宽控制单元 650连接至系统带宽单元 660。
存储单元 620中存储着用户终端进行频偏校正所需的信息， 例如导频符 号的导频时域序列， 以及用于确定系统带宽的带宽匹配序列集合等等。
时域釆样单元 610对 TS0时隙的下行射频信号根据用户终端所使用的带 宽进行采样， 生成时域序列， 其中包括 TS0时隙中导频符号和数据符号的时 域序列。
频偏估计序列单元 630根据时域采样单元 610输出的时域序列和存储单 元 620中存储的导频时域序列来生成进行频偏校正所需的 4个序列。 其中 2 个接收频偏估计序列根据时域采样单元 610的输出生成， 两个接收频偏估计 序列之间具有预定间隔，各包括 TS0时隙两列导频符号中的一个的时域序列。 另 2个序列为与 2个接收频偏估计序列分别对应的本地频偏估计序列， 每个 本地频偏估计序列中包括对应接收频偏估计序列中的导频符号在存储单元 620存储的导频时域序列。 当接收频偏估计序列中包括数据符号的时域序列 时， 本地频偏估计序列中还包括根据对应数据符号的时域采样值的硬判决结 果生成的时域序列。
可以将 TS0时隙中两列导频符号在时域采样单元 610的输出序列作为两 个接收频偏估计序列； 相应地， 将存储单元 620中存储的对应导频符号的导 频时域序列作为两个本地频偏估计序列。
还可以将时域采样单元 610输出的两列导频符号及与两列导频符号分别 相邻的数据符号的时域序列用作两个接收频偏估计序列； 相应地， 将数据符 号的时域序列进行硬判决后， 与存储模块 620中对应的导频时域序列分别组 合成两个本地频偏估计序列。
频偏估计序列单元 630将生成的用于频偏校正的 4个序列输出至频偏校 正单元 640。 频偏估计序列单元 630还可以按照顺序输出不同的接收频偏估 计序列及对应的本地频偏估计序列至频偏校正单元 640， 以进行具有不同精 度的频偏校正。
频偏校正单元 640将频偏估计序列单元 630生成的两个接收频偏估计序 列与对应的本地频偏估计序列进行共轭相关运算， 根据运算结果进行频偏校 正。
带宽控制单元 650根据小区接入系统支持的带宽以及用户终端支持的带 宽确定该用户终端使用的带宽， 并将确定的带宽用作时域采样单元 610的采 样频率。
系统带宽单元 660用来确定小区接入系统所使用的带宽。 系统带宽单元 660将时域采样单元 610输出的导频符号的时域序列， 与存储单元 620中带 宽匹配序列集合中对应导频符号的各个导频釆样序列进行相关运算， 具有相 关最大值的导频采样序列对应的带宽即为小区接入系统的带宽。
系统带宽单元 660将确定的小区接入系统带宽输出至带宽控制单元 650， 供其选择用户终端所使用的带宽。
图 7所示为应用本发明所述频偏校正方法的小区初搜方法的流程图。 在 步骤 S710， UE进行小区粗同步。
TD-SCDMA帧结构中 DwPTS可以用于小区同步搜索和粗频偏校正。 DwPTS时隙由两部分组成， 一部分是空闲时段， 在该时段内基站不发送任何 的信号； 另一部分是 SYNC-DL码， 该码为一有限长度的伪随机序列。 宽带 时分双工蜂窝系统有一个 SYNC-DL 码序列集合， 每个小区被分配一个 SYNC-DL码来作为小区的 ID号。 DwPTS时隙发送的 SYNC-DL码是一个单 载波的信号， 发送信号的带宽可以根据宽带时分双工蜂窝系统的最小带宽进 行设置，如最小带宽为 1.25MHz或 1.6MHz，发送 SYNC-DL码的单载波带宽 小于或等于该带宽。
在 DwPTS的 SYNC_DL伪随机码的两侧都有保护间隔，由于基站在保护 间隔没有信号发送， UE在保护间隔接收到的射频功率很小，而 SYNC一 DL码 段基站则以全功率发射。 从 UE的接收功率谱来看， 与两边保护间隔的接收 功率相比 SYNC— DL码段为峰值。 当用两边保护间隔接收的功率之和除以 SYNC—DL段功率之和时， 其比值很小。 由于在时间轴上存在这么一个特殊 的功率 "特征窗"， 在遍历整个接收数据时， 比值最小的位置即是 DwPTS的 位置， 由此判断出 SYNC— DL的大致位置， 从而实现 UE下行的粗同步。
在步骤 S720， UE接收 SYNC— DL码段大致位置的信号， 利用相关算法 确认所使用的下行同步序列。 对于不同的小区， SYNC— DL采用不同的伪随 机码序列。 这样， 通过确认 SYNC— DL码进而区分小区。
在步骤 S730， UE进行小区的精同步。 UE利用确认的下行同步导频进行 相关算法， 确认相关峰值位置， 从而实现 UE的下行精同步。
步骤 S710至步骤 S730的小区同步过程本发明采用与现有技术相同的方 法， 不再赘述。
在步骤 S740， UE进行粗频偏校正。 粗频偏校正可以采用现有技术中的 频偏调整方法， 利用 DwPTS时隙的 SYNC-DL码， 通过对接收 SYNC-DL码 两端的相位偏移进行频偏估计和校正。
在步骤 S750， UE进行精频偏校正。 精频偏校正可以采用本发明中步骤 S320至步骤 S380中的方法或其中的部分步骤， 此处不再重复。
在步骤 S760, 精频偏校正完成后， UE读取小区的广播信息， 完成小区 初搜的过程。
本发明提出的新的频偏校正方法和小区初搜方法可以实现快速的小区初 搜。 当 UE进行小区初搜时， 可以利用 TS0时隙的导频符号和数据符号进行 频偏估计和校正。 这种方法能够提高 UE进行频偏估计的精度， 同时降低了 实现的复杂度， 为宽带时分双工蜂窝系统实现下行同步和频率偏移校正提供 了有效的解决方案。
以上所述的本发明实施方式， 并不构成对本发明保护范围的限定。 任何 在本发明的精神和原则之内所作的任何修改、 等同替换和改进等（例如， 采 用 TS0以外的其他时隙作为发送广播信道的时隙），均应包含在本发明的权利 要求保护范围之内。

Claims

权利 要 求 书
1.一种宽带时分双工蜂窝系统的频偏校正方法， 其特征在于， 包括以下 步骤：
在发送广播信道的时隙接收具有预定间隔的两列导频符号的时域序列； 将两个接收的导频时域序列与本地存储的对应导频符号的导频时域序列 进行共轭相关运算；
根据所得的运算结果进行频偏校正。
2.如权利要求 1所述宽带时分双工蜂窝系统的频偏校正方法， 其特征在 于， 所述将两个接收的导频时域序列与本地存储的导频时域序列进行共轭相 关运算具体为： 将两个接收的导频时域序列与本地存储的对应导频符号的导 频时域序列分别共轭相乘， 将两个共轭相乘结果进行相关运算。
3.如权利要求 2所述宽带时分双工蜂窝系统的频偏校正方法， 其特征在 于， 所述方法在进行频偏校正之后还包括：
接收两列导频符号及与分别与各列导频符号相邻的数据符号的时域序列 作为两个接收序列；
将数据符号的时域序列进行硬判决后与对应于其相邻导频符号的本地存 储的导频时域序列分别组合构成两个本地序列；
根据两个接收序列与对应的本地序列的共轭相关运算结果进行进一步频 偏校正。
4. 如权利要求 3所述宽带时分双工蜂窝系统的频偏校正方法， 其特征在 于： 所述两个包括导频符号和相邻的数据符号的组合具有同样的预定间隔。
5.如权利要求 1至 4任意一项所述宽带时分双工蜂窝系统的频偏校正方 法， 其特征在于， 所述在发送广播信道的时隙接收两列导频符号的时域序列 具体为： 根据用户终端和小区接入系统都支持的最大带宽进行采样， 获得发 送广播信道的时隙两列导频符号的时域序列。
6.如权利要求 5所述宽带时分双工蜂窝系统的频偏校正方法， 其特征在 于， 所述方法还包括： 根据导频符号的时域信号采样序列和带宽匹配序列集 合来确定小区接入系统的带宽； 所述带宽匹配序列集合中包括该小区的导频 符号对应于不同带宽的导频采样序列。
7.如权利要求 6所述宽带时分双工蜂窝系统的频偏校正方法， 其特征在 于， 所述确定小区接入系统的带宽具体为：
以小区接入系统的最小带宽对导频符号的时域信号进行采样；
将时域信号的采样序列与带宽匹配序列集合中该导频符号的各个导频采 样序列进行相关运算；
查找相关值最大的采样序列， 其对应的带宽即为小区接入系统的带宽。
8.如权利要求 6所述宽带时分双工蜂窝系统的频偏校正方法， 其特征在 于， 所述方法在确定小区接入系统的带宽之前还包括： 通过计算接收的下行 同步码两端的相位偏移进行频偏估计和校正。
9.如权利要求 1所述宽带时分双工蜂窝系统的频偏校正方法， 其特征在 于： 所述两列导频符号分别位于两个 OFDM符号频域的各个子载波上。
10. 如权利要求 1 所述宽带时分双工蜂窝系统的频偏校正方法， 其特征 在于：所述两列导频符号分别位于两个 OFDM符号频域具有预定载波间隔的 子载波上， 其发射功率高于数据符号平均发射功率的差值由预定载波间隔确 定； 在导频符号所在的 OFDM符号中， 如果其子载波不发送导频符号， 则该 子载波也不发送其他任何符号。
11. 如权利要求 1 所述宽带时分双工蜂窝系统的频偏校正方法， 其特征 在于： 所述导频符号采用公共导频方式进行全向发送。
12.如权利要求 1 所述宽带时分双工蜂窝系统的频偏校正方法， 其特征 在于： 所述发送广播信道的时隙为 TS0时隙。
13. 一种用户终端， 其特征在于， 包括时域采样单元、 存储单元、 频偏 估计序列单元和频偏校正单元， 其中：
时域采样单元用来对发送广播信道的时隙中的信号进行采样以生成时域 序列；
存储单元用来存储信息， 其中包括导频符号的导频时域序列； 频偏估计序列单元用来根据时域采样单元输出的时域序列生成两个具有 预定间隔的接收频偏估计序列， 每个接收频偏估计序列中各包括发送广播信 道的时隙两列导频符号中一个的时域序列； 生成与接收频偏估计序列分别对 应的两个本地频偏估计序列， 每个本地频偏估计序列中包括存储单元中对应 导频符号的导频时域序列；
频偏校正单元根据频偏估计序列单元生成的两个接收频偏估计序列与对 应的本地频偏估计序列的共轭相关运算结果进行频偏校正。
14. 如权利要求 13所述的用户终端， 其特征在于： 所述频偏估计序列单 元根据时域采样单元输出的时域序列生成两个接收频偏估计序列具体为： 将 时域采样单元输出的两列导频符号的时域序列作为接收频偏估计序列；
所述频偏估计序列单元生成两个本地频偏估计序列具体为： 将存储单元 中存储的对应于两列导频符号的导频时域序列作为两个对应的本地频偏估计 序列。
15.如权利要求 13所述的用户终端， 其特征在于： 所述频偏估计序列单 元根据时域采样单元输出的时域序列生成两个接收频偏估计序列具体为： 将 时域采样单元输出的两列导频符号及与其分别相邻的数据符号的时域序列作 为接收频偏估计序列；
所述频偏估计序列单元生成两个本地频偏估计序列具体为： 将数据符号 的时域序列进行硬判决后与存储模块中对应于其相邻导频符号的导频时域序 列分别组合生成两个本地频偏估计序列。
16. 如权利要求 13至 15任意一项所述的用户终端， 其特征在于： 所述 用户终端还包括带宽控制单元， 用来确定小区接入系统和所述用户终端均支 持的带宽， 作为时域采样单元的采样频率。
17. 如权利要求 16所述的用户终端， 其特征在于： 所述存储单元所存储 的信息还包括带宽匹配序列集合， 其中包括导频符号、 带宽及对应于每个导 频符号和每个带宽的导频采样序列；
所述用户终端还包括系统带宽单元， 用来根据时域采样单元输出的导频 符号的时域序列与带宽匹配序列集合中导频采样序列的相关峰值确定小区接 入系统的带宽， 并输出至带宽控制单元。
18.如权利要求 13所述的用户终端， 其特征在于： 所述发送广播信道的 时隙为 TS0时隙。
19. 一种宽带时分双工蜂窝系统的小区初搜方法， 其特征在于， 包括以 下步骤：
用户终端进行小区同步；
用户终端进行粗频偏校正；
根据接收的具有预定间隔的两列导频符号的接收时域序列、 用户终端存 储的对应导频符号的两个导频时域序列的共轭相关结果进行精频偏校正； 用户终端读取小区广播信息。
20.如权利要求 19所述宽带时分双工蜂窝系统的小区初搜方法， 其特征 在于， 所述方法在进行精频偏校正后还包括：
接收两列导频符号及与该导频符号相邻的数据符号的时域序列作为两个 接收序列；
将数据符号的时域序列进行硬判决后与对应于其相邻导频符号的本地存 储的导频时域序列分别组合构成两个本地序列；
根据两个接收序列与对应的本地序列的共轭相关运算结果进行进一步频 偏校正。
21. 如权利要求 19或 20所述宽带时分双工蜂窝系统的小区初搜方法， 其特征在于， 所述用户终端进行粗频偏校正具体为： 通过计算接收的下行同 步码两端的相位偏移进行频偏估计和校正。
PCT/CN2007/000079 2006-01-12 2007-01-09 Procédé de correction de décalage de fréquence pour système de communication à cellules duplex à répartition dans le temps à large bande et procédé de première recherche de cellule WO2007079680A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200610001298.3 2006-01-12
CN2006100012983A CN101001231B (zh) 2006-01-12 2006-01-12 宽带时分双工蜂窝系统的频偏校正方法及小区初搜方法

Publications (1)

Publication Number Publication Date
WO2007079680A1 true WO2007079680A1 (fr) 2007-07-19

Family

ID=38255986

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2007/000079 WO2007079680A1 (fr) 2006-01-12 2007-01-09 Procédé de correction de décalage de fréquence pour système de communication à cellules duplex à répartition dans le temps à large bande et procédé de première recherche de cellule

Country Status (2)

Country Link
CN (1) CN101001231B (zh)
WO (1) WO2007079680A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113194051A (zh) * 2021-03-17 2021-07-30 深圳市力合微电子股份有限公司 电力双模通信中无线通信频偏的估计方法

Families Citing this family (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101394221B (zh) * 2007-09-17 2012-09-05 大唐移动通信设备有限公司 测量td-scdma小区时的同步方法、装置和ue
CN101425839B (zh) * 2007-10-31 2011-09-14 电信科学技术研究院 一种确定数据发送偏移量的方法、系统和装置
CN101557240B (zh) * 2008-04-11 2012-09-19 电信科学技术研究院 一种移动通信系统中频偏估计的方法及装置
CN101404517B (zh) * 2008-11-06 2012-11-21 华为技术有限公司 信号频偏纠正的方法、装置和系统
CN102045815B (zh) * 2009-10-10 2013-01-30 中国科学院计算技术研究所 用于lte系统的小区搜索装置和方法
CN102201824B (zh) * 2010-03-23 2013-12-04 上海华虹集成电路有限责任公司 一种联合数字下变频和频偏校正电路的实现方法
WO2011120226A1 (zh) 2010-03-31 2011-10-06 海能达通信股份有限公司 信号带宽自适应识别方法和系统
CN101848176B (zh) * 2010-03-31 2013-01-16 海能达通信股份有限公司 信号带宽自适应识别方法和系统
EP2728923B1 (en) 2011-06-28 2016-06-01 ZTE Corporation Measurement method and apparatus
CN102497240B (zh) * 2011-12-20 2014-03-26 北京泰美世纪科技有限公司 一种数字广播系统的采样同步装置和采样同步方法
CN103905361A (zh) * 2012-12-25 2014-07-02 普天信息技术研究院有限公司 一种ofdm系统中的采样频率同步方法
CN103269322B (zh) * 2013-05-08 2016-09-07 京信通信系统（广州）有限公司 一种确定频偏值的方法和装置
CN104243372B (zh) * 2013-06-07 2019-03-12 中兴通讯股份有限公司 频偏估计的方法和装置
CN105337909B (zh) * 2014-08-07 2020-02-14 中兴通讯股份有限公司 一种频偏估计的方法和装置
CN104579513B (zh) * 2015-01-29 2017-01-18 武汉虹旭信息技术有限责任公司 一种终端频偏校正方法
CN107453853B (zh) * 2016-05-31 2020-10-16 华为技术有限公司 一种导频传输的方法及设备
CN108075872B (zh) * 2016-11-17 2021-03-16 上海高清数字科技产业有限公司 基于导频辅助ofdm系统
CN111585924B (zh) * 2020-03-25 2021-07-16 北京理工大学 频偏校正方法及接收器

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1130967A (zh) * 1994-07-25 1996-09-11 摩托罗拉公司 数字接收机的频率偏移跟踪性能最大化的设备和方法
CN1345140A (zh) * 2000-09-20 2002-04-17 日本电气株式会社 校正频率偏移的方法和装置及存储其控制程序的媒体
WO2005062564A1 (en) * 2003-12-05 2005-07-07 Advanced Micro Devices, Inc. Residual frequency error estimation in an ofdm receiver

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1622653A (zh) * 2003-11-28 2005-06-01 皇家飞利浦电子股份有限公司 一种用于对td－scdma系统下行链路进行频率估测的装置和方法
CN1719815B (zh) * 2004-07-07 2011-02-02 华为技术有限公司 频偏估计和纠正方法及其装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1130967A (zh) * 1994-07-25 1996-09-11 摩托罗拉公司 数字接收机的频率偏移跟踪性能最大化的设备和方法
CN1345140A (zh) * 2000-09-20 2002-04-17 日本电气株式会社 校正频率偏移的方法和装置及存储其控制程序的媒体
WO2005062564A1 (en) * 2003-12-05 2005-07-07 Advanced Micro Devices, Inc. Residual frequency error estimation in an ofdm receiver

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113194051A (zh) * 2021-03-17 2021-07-30 深圳市力合微电子股份有限公司 电力双模通信中无线通信频偏的估计方法
CN113194051B (zh) * 2021-03-17 2022-06-10 深圳市力合微电子股份有限公司 电力双模通信中无线通信频偏的估计方法

Also Published As

Publication number Publication date
CN101001231A (zh) 2007-07-18
CN101001231B (zh) 2011-11-30

Similar Documents

Publication Publication Date Title
WO2007079680A1 (fr) Procédé de correction de décalage de fréquence pour système de communication à cellules duplex à répartition dans le temps à large bande et procédé de première recherche de cellule
US9954715B2 (en) Preambles in OFDMA system
US9232486B2 (en) Method and apparatus for transmitting synchronization signals in an OFDM based cellular communications system
US7580400B2 (en) Apparatus and method for generating preamble signal for cell identification in an orthogonal frequency division multiplexing system
US8537931B2 (en) Methods and apparatus for synchronization and detection in wireless communication systems
CN101005307B (zh) 宽带时分双工蜂窝系统的同步方法及小区初搜方法及终端
EP2353263B1 (en) Codes and preambles for single carrier and ofdm transmissions
CN101682605B (zh) 同时检测小区组和循环前缀的方法、装置和系统
WO2008048060A1 (en) Tdm based cell search method for ofdm system
EP2286562A1 (en) Methods and arrangements in a wireless communication system for producing signal structure with cyclic prefix
US7801088B2 (en) Synchronous orthogonal frequency division multiplexing receiving apparatus in multi-cell environment and method thereof
WO2008067721A1 (fr) Procédé de processus de synchronisation, station de base, dispositif utilisateur et système de communication
CN113315610A (zh) 一种无线通信方法及装置
Sharma et al. Fast cell synchronization for beyond 3G OFDMA based system
WO2008063031A1 (en) Method for allocating code to cells and planning cell in ofdm cellular system

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application
NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 07702011

Country of ref document: EP

Kind code of ref document: A1