WO2012062098A1 - 一种时间提前量估计方法及系统 - Google Patents

一种时间提前量估计方法及系统 Download PDF

Info

Publication number
WO2012062098A1
WO2012062098A1 PCT/CN2011/074516 CN2011074516W WO2012062098A1 WO 2012062098 A1 WO2012062098 A1 WO 2012062098A1 CN 2011074516 W CN2011074516 W CN 2011074516W WO 2012062098 A1 WO2012062098 A1 WO 2012062098A1
Authority
WO
WIPO (PCT)
Prior art keywords
channel estimation
search
estimation result
energy
matrix
Prior art date
Application number
PCT/CN2011/074516
Other languages
English (en)
French (fr)
Inventor
陈昱良
侯晓辉
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
Publication of WO2012062098A1 publication Critical patent/WO2012062098A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L7/00Arrangements for synchronising receiver with transmitter
    • H04L7/0054Detection of the synchronisation error by features other than the received signal transition
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00Baseband systems
    • H04L25/02Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202Channel estimation
    • H04L25/024Channel estimation channel estimation algorithms
    • H04L25/025Channel estimation channel estimation algorithms using least-mean-square [LMS] method
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L7/00Arrangements for synchronising receiver with transmitter
    • H04L7/0004Initialisation of the receiver
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L7/00Arrangements for synchronising receiver with transmitter
    • H04L7/04Speed or phase control by synchronisation signals
    • H04L7/041Speed or phase control by synchronisation signals using special codes as synchronising signal

Definitions

  • the present invention relates to the field of communications technologies, and in particular, to a method and system for estimating a timing advance.
  • Timing Advance is one of the core technologies of the GSM system.
  • the estimation of the timing advance directly affects the implementation of several key GSM technologies such as channel estimation, demodulation, and antenna combining.
  • Chinese patent application CN200510087745.7 discloses a method for estimating the timing advance, but the TA estimation method mainly has the following technical problems: The method can cope with the noise-limited TA estimation problem, but in the interference When the time is limited, the estimation effect is greatly reduced, and it is difficult to obtain an accurate TA estimation value, which in turn affects the accuracy of the channel estimation result.
  • the present invention provides a method for estimating a timing advance, the method comprising: performing an adaptive energy window time advance (TA) search according to an initial channel estimation result of each receiving antenna, and obtaining a coarse received signal After searching for the TA value, performing least squares (LS) channel estimation based on the coarse search TA value, and calculating a noise variance matrix according to the LS channel estimation result; and obtaining a whitening matrix according to the noise variance matrix, The whitening matrix performs whitening correction on the initial channel estimation result, and performs adaptive energy again according to the corrected channel estimation result.
  • TA adaptive energy window time advance
  • LS least squares
  • the window TA searches for the exact threshold of the received signal.
  • the step of performing the adaptive energy window search according to the initial channel estimation result comprises: calculating the channel estimation of each of the receiving antennas according to the initial channel estimation results of the respective receiving antennas obtained through correlation
  • the energy is equal-gain combined with the energy estimated by each of the receiving antenna channels, and an adaptive energy window search is performed according to the energy of the combined channel to obtain a coarse search threshold of the received signal.
  • the step of calculating the noise variance matrix according to the LS channel estimation result comprises: respectively obtaining channel estimates of the respective antennas of the received signals according to the channel estimation; and training sequences for the received signals
  • the partial (TSC) signal is reconstructed to calculate the noise and interference mixed signal of the main set and the mixed noise and interference mixed signal, respectively; and calculate the noise variance matrix.
  • the step of whitening the initial channel estimation result by using the whitening matrix comprises: performing whitening correction on the initial channel estimation result of each receiving antenna by using the whitening matrix, and obtaining the corrected Channel estimation result.
  • the step of performing the adaptive energy window search again according to the corrected channel estimation result includes: separately calculating the corrected channel estimation energy of each receiving antenna, and estimating the energy of each of the receiving antenna channels The equal gain combining is performed, and the adaptive energy window search is performed again according to the energy of the combined channel, and the accurate threshold of the received signal is obtained.
  • the present invention further provides a timing advance estimation system, the system comprising: a coarse search module, configured to: perform an adaptive energy window timing advance according to an initial channel estimation result of each receiving antenna ( ⁇ ) search, get the rough search threshold of the received signal; a channel estimation module, configured to: perform channel estimation on the basis of the coarse search TA value; and a noise variance matrix estimation module, configured to: calculate a noise variance matrix according to the channel estimation result;
  • a matrix decomposition module configured to: obtain a whitening matrix according to the noise variance matrix; a whitening correction module, configured to: perform whitening correction on the initial channel estimation result by using the whitening matrix; and a fine search module, which is set to : Perform adaptive energy window TA search again according to the corrected channel estimation result, and finally obtain an accurate TA value of the received signal.
  • the coarse search module includes: a first energy combining unit, configured to: calculate, according to initial channel estimation results of the receiving antennas obtained through correlation, energy of each of the receiving antenna channels And performing equal gain combining on the energy estimated by each of the receiving antenna channels; and a first TA searching unit configured to: perform an adaptive energy window TA search according to the energy of the combined channel to obtain a coarse search TA value of the received signal .
  • the channel estimation module is configured to: for receiving signals of two or more receiving antennas, obtain channel estimates of respective antennas of the received signals according to channel estimation; and the noise variance matrix estimation module is set Calculating a noise variance matrix according to the channel estimation result as follows: reconstructing, according to the channel estimation result, a signal of a training sequence portion of the received signal, respectively calculating a noise and interference mixed signal of each antenna; The module is arranged to obtain a whitening matrix according to the noise variance matrix as follows: By using the noise variance matrix, a whitening matrix is obtained.
  • the whitening correction module is configured to perform whitening correction on the initial channel estimation result by using the whitening matrix as follows: using the whitening matrix to whiten initial channel estimation results of each receiving antenna Correction, the corrected channel estimation result is obtained.
  • the fine search module includes: a second energy combining unit, configured to: separately calculate the corrected channel of each receiving antenna Estimating energy, and performing equal gain combining on the energy estimated by each of the receiving antenna channels; and a second TA searching unit configured to: perform an adaptive energy window TA search again according to the energy of the combined channel to obtain an accurate received signal TA value.
  • the TA estimation scheme proposed by the invention is mainly for NB pulse (Gaussian filtering minimum frequency shift keying
  • GMSK Gallium Filtered Minimum Shift Keying
  • 8PSK Phase Shift Keying
  • FIG. 1 is a schematic diagram of a TA estimation method according to an embodiment of the present invention.
  • the present invention provides a TA estimation method, which corrects the initial channel estimation of the antenna through the whitening matrix, thereby suppressing the interference component, and then using the correction.
  • the subsequent channel estimation performs the TA estimation of the adaptive energy window, thus optimizing the performance of the TA estimation, thereby improving the accuracy of the channel estimation.
  • the above TA estimation method specifically uses the following technical solutions: Performing an adaptive energy window TA search according to the initial channel estimation result, performing Least Square (LS) channel estimation based on the obtained rough search TA value, and calculating a noise variance matrix according to the LS channel estimation result;
  • the whitening matrix is obtained according to the noise variance matrix, and the initial channel estimation result is whitened and corrected by the whitening matrix, and the adaptive energy window TA search is performed again according to the corrected channel estimation result.
  • LS Least Square
  • the step of performing the adaptive energy window TA search according to the initial channel estimation result includes: obtaining an initial channel estimation of each receiving antenna by correlation; and separately calculating energy of each receiving antenna channel estimation, and respectively, for each receiving antenna channel
  • the estimated energy is equal gain combining
  • the adaptive energy window TA search is performed according to the energy of the combined channel to obtain a coarse search TA value of the received signal.
  • the whitening matrix is obtained by performing cholesky decomposition on the noise variance matrix.
  • the method further includes: correcting the initial channel estimation result by using the whitening matrix w; separately calculating the channel estimation energy of the two receiving antennas for whitening correction, and performing equal gain combining on the energy of the two receiving antenna channels.
  • the adaptive energy window TA search is performed again according to the energy of the combined channel, and the accurate TA value of the received signal is obtained.
  • the foregoing solution may include the following steps:
  • Equal gain combining is performed on the energy estimated by the two channels, and the energy of the combined channel is
  • TA5 On the basis of obtaining TA1, perform LS channel estimation, obtain channel estimation/sum of the received signal, reconstruct the signal of the training sequence (TSC) part of the received signal, and calculate the correlation matrix of noise and interference. . Perform cholesky decomposition (a matrix mathematical decomposition method) to obtain a whitening matrix w;
  • This application example provides a method for estimating the TA of the GSM system, which is applicable to a system with N times of N antennas.
  • This application example illustrates the method by taking a double double antenna as an example.
  • Step 2 performing equal gain combining on the energy of the two channels, and the energy of the combined channel is:
  • Step 3 Perform the first adaptive energy window TA search to obtain a coarse search TA value, which is denoted as TA1.
  • W refers to the conjugate conversion.
  • E means "f, (Step 6, by Cholesky decomposition theory, which can be expressed as
  • the angular matrix, R's formula can be defined by Cholesky:
  • the embodiment of the present invention further provides a time advance estimation system, which mainly includes the following functional modules: a coarse search module, which is configured to: perform adaptive energy window time advancement according to initial channel estimation results of each receiving antenna. A quantity (TA) search, obtaining a coarse search TA value of the received signal;
  • a coarse search module which is configured to: perform adaptive energy window time advancement according to initial channel estimation results of each receiving antenna.
  • a quantity (TA) search obtaining a coarse search TA value of the received signal;
  • An LS channel estimation module configured to: perform LS channel estimation on the basis of the coarse search TA value; and a noise variance matrix estimation module, configured to: calculate a noise side according to the LS channel estimation result Difference matrix
  • a matrix decomposition module configured to: obtain a whitening matrix according to the noise variance matrix; a whitening correction module, configured to: perform whitening correction on the initial channel estimation result by using the whitening matrix; and a fine search module, which is set to : Perform adaptive energy window TA search again according to the corrected channel estimation result, and finally obtain an accurate TA value of the received signal.
  • the coarse search module includes: a first energy combining unit, configured to: calculate, according to initial channel estimation results of the receiving antennas obtained through correlation, energy of each of the receiving antenna channels, and The energy of each receiving antenna channel is estimated to be equal gain combining; and the first TA searching unit is configured to: perform an adaptive energy window TA search according to the energy of the combined channel to obtain a coarse search TA value of the received signal.
  • a first energy combining unit configured to: calculate, according to initial channel estimation results of the receiving antennas obtained through correlation, energy of each of the receiving antenna channels, and The energy of each receiving antenna channel is estimated to be equal gain combining
  • the first TA searching unit is configured to: perform an adaptive energy window TA search according to the energy of the combined channel to obtain a coarse search TA value of the received signal.
  • the noise variance matrix estimation module is configured to: reconstruct, according to the LS channel estimation result, a signal of a training sequence portion of the received signal, and separately calculate a noise and a mixed interference signal of the main set and the diversity :
  • u M ⁇ k) y M ⁇ k) _ h M * Tsc
  • the noise variance matrix Qu is calculated according to the following formula.
  • u M H ⁇ k is the conjugate conversion of u M ⁇ k)
  • E is the expected value of the inner product
  • the matrix decomposition module is set to: by performing cholesky sub-RRT on the acoustic variance matrix, R is a complex number
  • the whitening correction module is configured to: perform whitening correction on the initial channel estimation result “M” of each receiving antenna by using the whitening matrix W, and obtain a corrected channel estimation result wM
  • the fine search module includes: a second energy combining unit, configured to: separately calculate channel corrected energy corrected by each receiving antenna, and perform equal gain combining on energy estimated by each receiving antenna channel;
  • the second TA search unit is configured to: perform an adaptive energy window TA search again according to the energy of the combined channel to obtain an accurate TA value of the received signal.
  • the TA estimation scheme proposed by the present invention is mainly directed to TA of NB pulse (GMSK, 8PSK) Search for improvements that improve the performance of TA estimates in situations where interference is limited.
  • GMSK, 8PSK NB pulse
  • the solution of the invention can complete high-precision TA estimation with low complexity for voice or data services in static or multi-path environments.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Power Engineering (AREA)
  • Radio Transmission System (AREA)

Abstract

本发明公开了一种时间提前量估计方法,所述方法包括:根据各接收天线的初始信道估计结果进行自适应能量窗时间提前量(TA)搜索,得到接收信号的粗搜索TA值后,在所述粗搜索TA值的基础上进行最小二乘(LS)信道估计,并根据LS信道估计结果计算噪声方差矩阵;以及根据所述噪声方差矩阵得到白化矩阵,利用所述白化矩阵对所述初始信道估计结果进行白化校正,根据校正后的信道估计结果再次进行自适应能量窗TA搜索,最终得到接收信号的准确TA值。本发明还公开了一种时间提前量估计系统。本发明能够显著提高干扰受限模型下TA估计的准确性。

Description

一种时间提前量估计方法及系统
技术领域 本发明涉及通信技术领域, 更具体地, 涉及一种时间提前量估计方法及 系统。
背景技术 在全球移动通讯系统 ( Global System for Mobile Communications, GSM ) 中, 时间提前量( Timing Advance , TA )估计是 GSM系统的核心技术之一。 对时间提前量的估计直接影响着信道估计,解调, 以及天线合并等多项 GSM 关键技术的实现。 目前已有的相关技术中, 中国专利申请 CN200510087745.7公开了一种 时间提前量估计方法, 但该 TA估计方法主要存在如下技术问题: 该方法能 够应对噪声受限的 TA估计问题, 但在干扰受限时估计效果却大大降低, 很 难得到准确的 TA估计值, 进而还影响了信道估计结果的准确性。
发明内容 本发明解决的技术问题是提供一种时间提前量估计方法及系统, 能够显 著提高干扰受限模型下 TA估计的准确性。 为解决上述技术问题, 本发明提供了一种时间提前量估计方法, 所述方 法包括: 根据各接收天线的初始信道估计结果进行自适应能量窗时间提前量 ( TA )搜索, 得到接收信号的粗搜索 TA值后, 在所述粗搜索 TA值的基础 上进行最小二乘(LS )信道估计, 并根据 LS信道估计结果计算噪声方差矩 阵; 以及 才艮据所述噪声方差矩阵得到白化矩阵, 利用所述白化矩阵对所述初始信 道估计结果进行白化校正, 根据校正后的信道估计结果再次进行自适应能量 窗 TA搜索, 最终得到接收信号的准确 ΤΑ值。 本发明的方法中, 根据初始信道估计结果进行自适应能量窗 ΤΑ搜索的 步骤包括: 才艮据通过相关得到的所述各接收天线的初始信道估计结果, 分别计算所 述各接收天线信道估计的能量, 并对所述各接收天线信道估计的能量进行等 增益合并, 根据合并后信道的能量进行自适应能量窗 ΤΑ搜索, 得到接收信 号的粗搜索 ΤΑ值。 本发明的方法中, 对于 2根接收天线的接收信号, 根据 LS信道估计结 果计算噪声方差矩阵的步骤包括: 根据信道估计分别得到接收信号的各个天线的信道估计; 对所述接收信号的训练序列部分(TSC ) 的信号进行重构, 分别计算主 集的噪声和干扰混合信号和分集的噪声和干扰混合信号; 以及 计算得到所述噪声方差矩阵。 本发明的方法中, 利用所述白化矩阵对所述初始信道估计结果进行白化 校正的步骤包括: 利用所述白化矩阵, 对所述各接收天线的初始信道估计结果进行白化校 正, 得到校正后的信道估计结果。 本发明的方法中,根据校正后的信道估计结果再次进行自适应能量窗 ΤΑ 搜索的步骤包括: 分别计算所述各接收天线校正后的信道估计能量, 并对所述各接收天线 信道估计的能量进行等增益合并, 根据合并后信道的能量再次进行自适应能 量窗 ΤΑ搜索, 得到接收信号的准确 ΤΑ值。 为解决上述技术问题, 本发明还提供了一种时间提前量估计系统, 所述 系统包括: 粗搜索模块, 其设置为: 根据各接收天线的初始信道估计结果进行自适 应能量窗时间提前量(ΤΑ )搜索, 得到接收信号的粗搜索 ΤΑ值; 信道估计模块, 其设置为: 在所述粗搜索 TA值的基础上进行信道估计; 噪声方差矩阵估计模块, 其设置为: 根据信道估计结果计算噪声方差矩 阵;
矩阵分解模块, 其设置为: 根据所述噪声方差矩阵得到白化矩阵; 白化校正模块, 其设置为: 利用所述白化矩阵对所述初始信道估计结果 进行白化校正; 以及 精搜索模块, 其设置为: 根据校正后的信道估计结果再次进行自适应能 量窗 TA搜索, 最终得到接收信号的准确 TA值。 本发明的系统中, 所述粗搜索模块包括: 第一能量合并单元, 其设置为: 根据通过相关得到的所述各接收天线的 初始信道估计结果, 分别计算所述各接收天线信道估计的能量, 并对所述各 接收天线信道估计的能量进行等增益合并; 以及 第一 TA搜索单元, 其设置为: 根据合并后信道的能量进行自适应能量 窗 TA搜索 , 得到接收信号的粗搜索 TA值。 本发明的系统中, 所述信道估计模块是设置为: 对于两根或者两根以上 接收天线的接收信号,根据信道估计分别得到接收信号各个天线的信道估计; 所述噪声方差矩阵估计模块是设置为按如下方式根据信道估计结果计算 噪声方差矩阵: 根据所述信道估计结果, 对所述接收信号的训练序列部分的 信号进行重构, 分别计算各个天线的噪声和干扰混合信号; 所述矩阵分解模块是设置为按如下方式根据所述噪声方差矩阵得到白化 矩阵: 通过利用所述噪声方差矩阵, 进而得到白化矩阵。 本发明的系统中, 所述白化校正模块是设置为按如下方式利用所述白化 矩阵对所述初始信道估计结果进行白化校正: 利用所述白化矩阵, 对各接收 天线的初始信道估计结果进行白化校正, 得到校正后的信道估计结果。 本发明的系统中, 所述精搜索模块包括: 第二能量合并单元, 其设置为: 分别计算所述各接收天线校正后的信道 估计能量, 并对所述各接收天线信道估计的能量进行等增益合并; 以及 第二 TA搜索单元, 其设置为: 根据合并后信道的能量再次进行自适应 能量窗 TA搜索, 得到接收信号的准确 TA值。
本发明提出的 TA估计方案主要针对 NB脉冲 (高斯滤波最小频移键控
( Gaussian Filtered Minimum Shift Keying, GMSK )以及 8相移键控 ( 8 Phase Shift Keying, 8PSK ) 的 TA搜索进行改进, 能够改善干扰受限情况下的 TA 估计的性能。 釆用本发明方案, 能够显著提高干扰受限模型下 TA估计的准 确性, 从而提高信道估计准确性, 显著提升解调性能。 本发明方案对静态或 者多径环境的语音或者数据业务都能以较低的复杂度完成高精度的 TA估 计。
附图概述 此处所说明的附图用来提供对本发明的进一步理解, 构成本申请的一部 分, 本发明的示意性实施例及其说明用于解释本发明, 并不构成对本发明的 不当限定。 在附图中: 图 1为本发明实施例的 TA估计方法的示意图。
本发明的较佳实施方式 为使本发明的目的、 技术方案和优点更加清楚明白, 下文中将结合附图 对本发明的实施例进行详细说明。 需要说明的是, 在不冲突的情况下, 本申 请中的实施例及实施例中的特征可以相互任意组合。 针对现有 TA估计方法在干扰受限模型下准确性的问题, 本发明提供一 种 TA估计方法, 该方法通过白化矩阵对天线的初始信道估计进行校正, 从 而对干扰成分进行抑制, 再利用校正后的信道估计进行自适应能量窗的 TA 估计, 因此, 优化了 TA估计的性能, 进而提高了信道估计的准确性。 上述 TA估计方法具体釆用如下技术方案: 根据初始信道估计结果进行自适应能量窗 TA搜索 ,在得到的粗搜索 TA 值的基础上进行最小二乘(Least Square, LS )信道估计, 并根据 LS信道估 计结果计算噪声方差矩阵; 以及 才艮据噪声方差矩阵得到白化矩阵, 利用所述白化矩阵对所述初始信道估 计结果进行白化校正,根据校正后的信道估计结果重新进行自适应能量窗 TA 搜索。 上述方法中, 根据初始信道估计结果进行自适应能量窗 TA搜索的步骤 包括: 通过相关得到各接收天线的初始信道估计; 以及 分别计算各接收天线信道估计的能量, 并对所述各接收天线信道估计的 能量进行等增益合并, 根据合并后信道的能量进行自适应能量窗 TA搜索, 得到接收信号的粗搜索 TA值。 其中, 对于 2根接收天线的接收信号, 根据 LS信道估计结果计算噪声 方差矩阵的步骤包括: 根据 LS信道估计得到接收信号的主、分集信道估计 //M(/) , hs (l) , l=0~L; 对接收信号的训练序列 (Training Sequence, TSC )部分信号进行重构, 计算主、 分集的噪声和干扰混合信号: uM
Figure imgf000007_0001
i SC
us (k) = ys (k) -hs * Tsc , k=L〜N,其中 Μ (A)和 ^ (A)对应接收信号 (w)中的 训练序列部分数据; 以及 按照下式计算得到所述噪声方差矩阵 :
Figure imgf000007_0002
其中, )是指 ¾ W的共轭转置; E表示 )和 内积的期望值。 上述方法中, 通过对所述噪声方差矩阵进行 cholesky分解, 得到所述白 化矩阵。 上述方法 , 按照以下方式得到所述白化矩阵 W:
Figure imgf000008_0001
其中, ¾„= RRT , R为下三角矩阵, gn 0
R = 其中 g„=V^, 为实 数; , 为实数, 是 g21的共轭。
Figure imgf000008_0002
上述方法还包括, 利用所述白化矩阵 w, 校正初始信道估计结果; 再次分别计算两根接收天线白化校正后的信道估计能量, 并对所述两根 接收天线信道估计的能量进行等增益合并, 根据合并后信道的能量再次进行 自适应能量窗 TA搜索, 得到接收信号的准确 TA值。 具体地, 上述方案可以包括如下步骤:
1)通过相关的方法得到 2根天线的初始信道估计 和/ ^ , 其中 表示主 集的初始信道估计, h .表示分集的初始信道估计;
2)计算 2个信道估计的能量 £Μ = | |和 =
3)对 2 个信道估计的能量进行等增益合并, 合并后信道的能量为
F = FM +卞 L FS ',
4)进行第 1次自适应能量窗搜索, 得到接收信号的粗搜索 TA值, 记为 TA1 ;
5)在得到 TA1的基础上,进行 LS信道估计,得到接收信号的信道估计/ 和 , 对接收信号的训练序列 (Training Sequence , TSC )部分的信号进行重 构, 计算出噪声和干扰的相关矩阵 。 对 进行 cholesky分解(一种矩阵 数学分解方法) , 获得白化矩阵 w;
6)用白化矩阵 w , 对初始信道估计进行干 4尤抑制合并 (Interference Rejection Combining , IRC )校正, 得到校正后的信道估计 hw' M和 hw' S
7)重新计算 2个信道的能量 M = \hwJ和 £ws = Ks '
8)对 2个信道的能量进行等增益合并 ,合并后信道的能量为 E = EwM + EwS , 9)进行第 2次自适应能量窗搜索, 得到接收信号的准确 TA值。
为了便于阐述本发明, 以下将结合附图及应用示例对本发明技术方案的 实施作进一步详细描述。 应用示例一 本应用示例提供一种 GSM系统 TA估计的方法, 适用于 N倍釆样 N根 天线的系统, 本应用示例以双倍釆样双天线为例对该方法进行说明。 如图 1所示, 本实施例方法主要包括如下步骤: 步骤一,对两天线接收信号 y(n)= yM \n 的训练序列部分数据匹配滤波, 得到 2根天线各自的初始信道估计 •M ' , 1=0~N0。 其中, NO为初始信道 估计的长度。 步骤二, 对 2个信道的能量进行等增益合并, 合并后信道的能量为:
'M + , /=0~N0。 本发明中, 计算 的目的在于, 通过合并利用两天线带来的增益, 使 粗搜索的 TA也更加准确。 步骤三, 进行第 1次自适应能量窗 TA搜索, 得到粗搜索 TA值, 记为 TA1。 步骤四, 在得到 TA1的基础上, 进行 LS信道估计, 得到较准确的主、 分集信道估计 (/)和 hs (/) , /=0~L; 步骤五,利用 TSC进行信号重构,计算主、分集的噪声和干扰混合信号:
- * Tsc ,
us (k) = ys (k) - hs * Tsc , k=L〜N , 其中 , Λ 对应接收信号 中的 训练序列部分数据, 进而得到噪声方差矩阵
Figure imgf000010_0001
其中, W是指 的共轭转制。 E表示" f,( 步骤六, 由 Cholesky分解理论, 可表示为
角矩阵, R的计算公式可由 Cholesky定义可得:
8u 0
R.. ,其中 = ,为实数; = ,为复数; g22 = lQ2l-g g: 为实数, 是 g21的共轭 最后才艮据 R求得白化矩阵 w = gll 0
R-
Figure imgf000010_0002
步骤七,用白化矩阵 w,对初始信道估计 •M 1=0~N0进行白化校正: 得到校正后的信道估计:
Figure imgf000010_0003
Figure imgf000010_0004
1=0- No 0 步骤八, 合并白化后的信道能量为 (/) = k wM + 'wS, , /=0〜N0 步骤九, 进行第 2次自适应能量窗 TA估计, 得到准确的 TA值, 记为
TA。
此外, 本发明实施例中还提供了一种时间提前量估计系统, 该系统主要 包括以下功能模块: 粗搜索模块, 其设置为: 根据各接收天线的初始信道估计结果进行自适 应能量窗时间提前量(TA)搜索, 得到接收信号的粗搜索 TA值;
LS信道估计模块, 其设置为: 在所述粗搜索 TA值的基础上进行 LS信 道估计; 噪声方差矩阵估计模块, 其设置为: 根据 LS信道估计结果计算噪声方 差矩阵;
矩阵分解模块, 其设置为: 根据所述噪声方差矩阵得到白化矩阵; 白化校正模块, 其设置为: 利用所述白化矩阵对所述初始信道估计结果 进行白化校正; 以及 精搜索模块, 其设置为: 根据校正后的信道估计结果再次进行自适应能 量窗 TA搜索, 最终得到接收信号的准确 TA值。 其中, 所述粗搜索模块包括: 第一能量合并单元, 其设置为: 根据通过相关得到的所述各接收天线的 初始信道估计结果, 分别计算所述各接收天线信道估计的能量, 并对所述各 接收天线信道估计的能量进行等增益合并; 以及 第一 TA搜索单元, 其设置为: 根据合并后信道的能量进行自适应能量 窗 TA搜索 , 得到接收信号的粗搜索 TA值。 其中, 所述 LS信道估计模块是设置为: 对于 2根接收天线的接收信号, 根据 LS 信道估计分别得到接收信号的主集信道估计// M(/)和分集信道估计 (I) , 其中 /=0~ ; 所述噪声方差矩阵估计模块是设置为: 根据所述 LS信道估计结果, 对 所述接收信号的训练序列部分的信号进行重构, 分别计算主集、 分集的噪声 和干扰混合信号: uM {k) = yM {k) _ hM * Tsc , us (k) = ys (k) - hs * Tsc , k=L~N, 其 中 _y。( , _y。W对应接收信号 _y。 照下式计算 得到所述噪声方差矩阵 Qu 其中
Figure imgf000011_0001
uM H {k)为 uM {k)的共轭转制, E为内积的期望值; 所述矩阵分解模块是设置为: 通过对 声方差矩阵 进行 cholesky 分 RRT, R 为 为复数;
Figure imgf000011_0002
所述白化校正模块是设置为: 利用所述白化矩阵 W, 按照以下方式对所 述各接收天线的初始信道估计结果 "M 进行白化校正, 得到校正后的信道 估 计 结口 果 wM
wS
Figure imgf000012_0001
Figure imgf000012_0002
其中, 所述精搜索模块包括: 第二能量合并单元, 其设置为: 分别计算所述各接收天线校正后的信道 估计能量, 并对所述各接收天线信道估计的能量进行等增益合并; 以及 第二 TA搜索单元, 其设置为: 根据合并后信道的能量再次进行自适应 能量窗 TA搜索, 得到接收信号的准确 TA值。
以上仅为本发明的优选实施案例而已, 并不用于限制本发明, 本发明还 可有其他多种实施例, 在不背离本发明精神及其实质的情况下, 熟悉本领域 的技术人员可根据本发明做出各种相应的改变和变形, 但这些相应的改变和 变形都应属于本发明所附的权利要求的保护范围。 显然, 本领域的技术人员应该明白, 上述的本发明的各模块或各步骤可 以用通用的计算装置来实现, 它们可以集中在单个的计算装置上, 或者分布 在多个计算装置所组成的网络上, 可选地, 它们可以用计算装置可执行的程 序代码来实现, 从而, 可以将它们存储在存储装置中由计算装置来执行, 并 且在某些情况下, 可以以不同于此处的顺序执行所示出或描述的步骤, 或者 将它们分别制作成各个集成电路模块, 或者将它们中的多个模块或步骤制作 成单个集成电路模块来实现。 这样, 本发明不限制于任何特定的硬件和软件 结合。
工业实用性 本发明提出的 TA估计方案主要针对 NB脉冲 (GMSK、 8PSK ) 的 TA 搜索进行改进, 能够改善干扰受限情况下的 TA估计的性能。 釆用本发明方 案, 能够显著提高干扰受限模型下 TA估计的准确性, 从而提高信道估计准 确性, 显著提升解调性能。 本发明方案对静态或者多径环境的语音或者数据 业务都能以较低的复杂度完成高精度的 TA估计。

Claims

权 利 要 求 书
1、 一种时间提前量估计方法, 所述方法包括: 根据各接收天线的初始信道估计结果进行自适应能量窗时间提前量 ( TA )搜索, 得到接收信号的粗搜索 TA值后, 在所述粗搜索 TA值的基础 上进行最小二乘 ( LS )信道估计, 并根据 LS信道估计结果计算噪声方差矩 阵; 以及 才艮据所述噪声方差矩阵得到白化矩阵, 利用所述白化矩阵对所述初始信 道估计结果进行白化校正, 根据校正后的信道估计结果再次进行自适应能量 窗 TA搜索 , 最终得到接收信号的准确 TA值。
2、如权利要求 1所述的方法, 其中,根据初始信道估计结果进行自适应 能量窗 TA搜索的步骤包括: 才艮据通过相关得到的所述各接收天线的初始信道估计结果, 分别计算所 述各接收天线信道估计的能量, 并对所述各接收天线信道估计的能量进行等 增益合并, 根据合并后信道的能量进行自适应能量窗 TA搜索, 得到接收信 号的粗搜索 TA值。
3、 如权利要求 1或 2所述的方法, 其中, 对于 2根接收天线的接收信号, 根据 LS信道估计结果计算噪声方差矩 阵的步骤包括: 根据信道估计分别得到接收信号的各个天线的信道估计; 对所述接收信号的训练序列部分(TSC ) 的信号进行重构, 分别计算主 集的噪声和干扰混合信号和分集的噪声和干扰混合信号; 以及 计算得到所述噪声方差矩阵。
4、如权利要求 3所述的方法, 其中, 利用所述白化矩阵对所述初始信道 估计结果进行白化校正的步骤包括: 利用所述白化矩阵, 对所述各接收天线的初始信道估计结果进行白化校 正, 得到校正后的信道估计结果。
5、如权利要求 4所述的方法, 其中,根据校正后的信道估计结果再次进 行自适应能量窗 TA搜索的步骤包括: 分别计算所述各接收天线校正后的信道估计能量, 并对所述各接收天线 信道估计的能量进行等增益合并, 根据合并后信道的能量再次进行自适应能 量窗 TA搜索, 得到接收信号的准确 TA值。
6、 一种时间提前量估计系统, 所述系统包括: 粗搜索模块, 其设置为: 根据各接收天线的初始信道估计结果进行自适 应能量窗时间提前量(TA )搜索, 得到接收信号的粗搜索 TA值; 信道估计模块, 其设置为: 在所述粗搜索 TA值的基础上进行信道估计; 噪声方差矩阵估计模块, 其设置为: 根据信道估计结果计算噪声方差矩 阵;
矩阵分解模块, 其设置为: 根据所述噪声方差矩阵得到白化矩阵; 白化校正模块, 其设置为: 利用所述白化矩阵对所述初始信道估计结果 进行白化校正; 以及 精搜索模块, 其设置为: 根据校正后的信道估计结果再次进行自适应能 量窗 TA搜索, 最终得到接收信号的准确 TA值。
7、 如权利要求 6所述的系统, 其中, 所述粗搜索模块包括: 第一能量合并单元, 其设置为: 根据通过相关得到的所述各接收天线的 初始信道估计结果, 分别计算所述各接收天线信道估计的能量, 并对所述各 接收天线信道估计的能量进行等增益合并; 以及 第一 TA搜索单元, 其设置为: 根据合并后信道的能量进行自适应能量 窗 TA搜索, 得到接收信号的粗搜索 TA值。
8、 如权利要求 6或 7所述的系统, 其中, 所述信道估计模块是设置为: 对于两根或者两根以上接收天线的接收信 号, 根据信道估计分别得到接收信号各个天线的信道估计; 所述噪声方差矩阵估计模块是设置为按如下方式根据信道估计结果计算 噪声方差矩阵: 根据所述信道估计结果, 对所述接收信号的训练序列部分的 信号进行重构, 分别计算各个天线的噪声和干扰混合信号; 所述矩阵分解模块是设置为按如下方式根据所述噪声方差矩阵得到白化 矩阵: 通过利用所述噪声方差矩阵, 进而得到白化矩阵。
9、 如权利要求 8所述的系统, 其中, 所述白化校正模块是设置为按如下方式利用所述白化矩阵对所述初始信 道估计结果进行白化校正: 利用所述白化矩阵, 对各接收天线的初始信道估 计结果进行白化校正, 得到校正后的信道估计结果。
10、 如权利要求 9所述的系统, 其中, 所述精搜索模块包括: 第二能量合并单元, 其设置为: 分别计算所述各接收天线校正后的信道 估计能量, 并对所述各接收天线信道估计的能量进行等增益合并; 以及 第二 TA搜索单元, 其设置为: 根据合并后信道的能量再次进行自适应 能量窗 TA搜索, 得到接收信号的准确 TA值。
PCT/CN2011/074516 2010-11-09 2011-05-23 一种时间提前量估计方法及系统 WO2012062098A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201010537075.5A CN102469046B (zh) 2010-11-09 2010-11-09 一种时间提前量估计方法及系统
CN201010537075.5 2010-11-09

Publications (1)

Publication Number Publication Date
WO2012062098A1 true WO2012062098A1 (zh) 2012-05-18

Family

ID=46050363

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/074516 WO2012062098A1 (zh) 2010-11-09 2011-05-23 一种时间提前量估计方法及系统

Country Status (2)

Country Link
CN (1) CN102469046B (zh)
WO (1) WO2012062098A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106998557A (zh) * 2016-01-26 2017-08-01 中兴通讯股份有限公司 一种ta值的估计方法和装置
CN108462989A (zh) * 2017-02-21 2018-08-28 中兴通讯股份有限公司 时间提前量的处理方法和装置
CN111149394B (zh) * 2017-12-06 2022-01-25 上海朗帛通信技术有限公司 一种用于无线通信的通信节点中的方法和装置

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1913421A (zh) * 2005-08-08 2007-02-14 中兴通讯股份有限公司 一种适用于gsm/edge系统的时间提前量估计方法
CN101056286A (zh) * 2006-02-16 2007-10-17 西门子公司 改善突发干扰变化期间无线电网中的信道估计的方法

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1913421A (zh) * 2005-08-08 2007-02-14 中兴通讯股份有限公司 一种适用于gsm/edge系统的时间提前量估计方法
CN101056286A (zh) * 2006-02-16 2007-10-17 西门子公司 改善突发干扰变化期间无线电网中的信道估计的方法

Also Published As

Publication number Publication date
CN102469046B (zh) 2015-07-22
CN102469046A (zh) 2012-05-23

Similar Documents

Publication Publication Date Title
CN107426120B (zh) 一种基于最小均方误差的水声ofdm-mfsk信道均衡方法
CN103178947B (zh) 基于时域相关与编码辅助的载波同步方法
TW201004234A (en) Interference cancellation under non-stationary conditions
US20100034110A1 (en) Iterative reception method and iterative receiver
CN103929394A (zh) 基于迭代算法的高精度频偏估计方法
CN102244624A (zh) 基于正交匹配追踪的稀疏信道估计方法
CN101494874B (zh) Toa估计方法及装置
CN101022433A (zh) 高速数字接收机并行自适应盲均衡方法
CN103888145A (zh) 一种重构信号的方法
US8923447B2 (en) Channel estimation method and system based on time-domain training sequence
CN102307054B (zh) 一种直接序列扩频信号的捕获方法
CN103475602A (zh) 一种考虑同频干扰的mimo-ofdm信道估计方法
WO2012062098A1 (zh) 一种时间提前量估计方法及系统
WO2010057382A1 (zh) 一种用于无线通信系统的频偏估计方法及装置
CN102761502B (zh) 一种ofdm系统信道估计方法
CN105406956A (zh) 基于对称帧的载波同步方法
CN101789918B (zh) 并行信道均衡方法
CN100479444C (zh) 电力线通信系统中块状方式的信道估计方法
CN107623647A (zh) 一种基于离散导频辅助的载波同步方法
WO2012095052A2 (zh) 频偏估计和信道估计的方法、装置及系统
CN109639616B (zh) 联合使用同步序列和ofdm循环前缀的下行时频同步方法
CN104168240B (zh) 一种oqpsk突发信号的多路并行同步解调方法
CN104301282A (zh) 一种超高速移动ofdm系统的ici自适应抑制方法
CN101170366B (zh) 基于最小均方差线性块均衡实现信号联合检测的方法
CN110943949A (zh) 一种基于sc-fde传输体制的宽带角跟踪方法

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: 11839212

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: 11839212

Country of ref document: EP

Kind code of ref document: A1