CN101056302B

CN101056302B - Channel and Carrier Frequency Offset Estimation Method Based on UKF in OFDM System

Info

Publication number: CN101056302B
Application number: CN200710041459A
Authority: CN
Inventors: 顾婷婷; 梁永明; 罗汉文; 李洪星
Original assignee: Shanghai Jiao Tong University; Sharp Corp
Current assignee: Shanghai Jiao Tong University; Sharp Corp
Priority date: 2007-05-31
Filing date: 2007-05-31
Publication date: 2010-05-19
Anticipated expiration: 2027-05-31
Also published as: CN101056302A

Abstract

A UKF-based channel and carrier frequency offset estimation method in an OFDM system is used in the technical field of wireless transmission. The channel estimation method in the present invention starts from the frequency domain, first uses the LS method to obtain the rough estimation value of the channel at the pilot frequency, and then uses the path acquisition method and the UKF method in the time domain to obtain the precise estimation value of the channel. The carrier frequency offset estimation method based on UKF in the OFDM system in the present invention estimates the carrier frequency offset in the time domain through the UKF filtering method, and then corrects the carrier frequency offset to eliminate ICI. The present invention also proposes a joint estimation method of channel and carrier frequency based on UKF for the OFDM system with inter-subcarrier interference and time-varying wireless channel, using a specific pilot structure and convolution matrix change form, which can effectively Eliminate ICI and improve the accuracy of channel estimation, and has good convergence and robustness.

Description

Channel and Carrier Frequency Offset Estimation Method Based on UKF in OFDM System

技术领域technical field

本发明涉及一种无线通信技术领域的信道和频偏估计的方法，具体是一种OFDM系统中基于UKF(无迹卡尔曼滤波器)的信道与载波频率偏移估计方法。The invention relates to a method for channel and frequency offset estimation in the technical field of wireless communication, in particular to a method for estimating channel and carrier frequency offset based on UKF (unscented Kalman filter) in an OFDM system.

背景技术Background technique

无线通信中信号在传播过程受到信道衰减、多径时延扩展和多普勒频率扩展等因素的影响，在接收端为了能够较好地恢复出发送信号通常采用相干解调，而相干解调需要信道参数信息，它通过信道估计来获得，因此信道估计器的性能直接影响系统性能。正交频分复用(OFDM)系统能够有效抵抗多径扩展造成的符号间干扰，使得在恶劣无线衰落信道下进行数据传输成为可能。在OFDM系统中，基于二维最小均方误差准则(MMSE)的信道估计方法是理论分析中性能最优的信道估计方法，但这类方法不仅复杂度极高，还需要知道部分或全部信道先验信息，因此工程应用价值不高。另外，基于最小平方准则(LS)的信道估计方法，虽然运算复杂度不高，但在信噪比下降时性能下降很快，不适合用于中低信噪比的场合。因此，寻找工程上适用的信道估计方法直接关系到接收机检测和译码性能的优劣，对提高宽带移动通信系统性能起着至关重要的作用。In wireless communication, the signal is affected by factors such as channel attenuation, multipath delay spread, and Doppler frequency spread during the propagation process. In order to better recover the sent signal at the receiving end, coherent demodulation is usually used, and coherent demodulation requires Channel parameter information, which is obtained through channel estimation, so the performance of the channel estimator directly affects the system performance. The Orthogonal Frequency Division Multiplexing (OFDM) system can effectively resist the inter-symbol interference caused by multipath spreading, making it possible to carry out data transmission under harsh wireless fading channels. In the OFDM system, the channel estimation method based on the two-dimensional minimum mean square error criterion (MMSE) is the channel estimation method with the best performance in theoretical analysis, but this kind of method is not only extremely complex, but also needs to know some or all of the channel Therefore, the engineering application value is not high. In addition, although the channel estimation method based on the least square criterion (LS) has low computational complexity, its performance drops rapidly when the signal-to-noise ratio drops, so it is not suitable for occasions with medium and low signal-to-noise ratios. Therefore, finding an engineering-applicable channel estimation method is directly related to the performance of receiver detection and decoding, and plays a vital role in improving the performance of broadband mobile communication systems.

OFDM建立在子载波间严格正交的基础上，任何传输过程中载波信号波形畸变均会影响子载波之间的正交性。数据符号周期的增加会使得OFDM系统比普通单载波系统对载波频率偏移和多普勒扩展更加敏感，从而使子载波之间干扰大大增加，引入子载波间干扰(ICI)，大大降低系统的性能。目前已经两类方法用来消除因载波频率偏移引起的ICI：一类是对频率偏移进行估计后进行校正；另一类是采取措施降低系统对频率偏移的敏感性。自消除(SC)方法中，在相邻子载波中加入冗余数据来降低系统对频率偏移的敏感性，达到消除ICI的目的；最大似然估计(MLE)方法对频率偏移进行估计然后进行校正。当频率偏移的值比较小时，SC方法具有较好性能，这是因为此时相邻载波间产生的ICI之间的差别很小，SC方法中相邻子载波间的ICI可有效抵消。当频率偏移值比较大时，SC方法中相邻子载波之间不能完全消除ICI，而此时MLE方法对频率偏移有比较精确的估计。SC和MLE方法都需要一倍的冗余带宽，带宽效率低。因此要寻找一种带宽效率高，同时对频率偏移有精确的估计的方法。实际无线通信系统中同时受信道的多径效应、时变性以及载波频率偏移的影响，因此还必须寻找一种能够同时解决信道估计和载波频率偏移估计两个问题的方法。OFDM is based on strict orthogonality between subcarriers, and any distortion of the carrier signal waveform during transmission will affect the orthogonality between subcarriers. The increase of the data symbol period will make the OFDM system more sensitive to carrier frequency offset and Doppler spread than the ordinary single-carrier system, thus greatly increasing the interference between sub-carriers, introducing inter-sub-carrier interference (ICI), and greatly reducing the system performance. At present, two types of methods have been used to eliminate the ICI caused by carrier frequency offset: one is to correct the frequency offset after estimation; the other is to take measures to reduce the system's sensitivity to frequency offset. In the self-cancellation (SC) method, redundant data is added to adjacent subcarriers to reduce the sensitivity of the system to frequency offset and achieve the purpose of eliminating ICI; the maximum likelihood estimation (MLE) method estimates the frequency offset and then Make corrections. When the value of the frequency offset is relatively small, the SC method has better performance, because at this time the difference between the ICIs generated between adjacent carriers is very small, and the ICI between adjacent subcarriers in the SC method can be effectively offset. When the frequency offset value is relatively large, the ICI cannot be completely eliminated between adjacent subcarriers in the SC method, while the MLE method can estimate the frequency offset more accurately at this time. Both SC and MLE methods require doubled redundant bandwidth, which is inefficient in bandwidth. Therefore, it is necessary to find a method with high bandwidth efficiency and accurate estimation of frequency offset. The actual wireless communication system is affected by channel multipath effect, time variation and carrier frequency offset at the same time, so it is necessary to find a method that can solve the two problems of channel estimation and carrier frequency offset estimation at the same time.

经对现有技术的文献检索发现，Dieter Schafhuber等人在2003年的IEEE国际无线通信的信号处理学术会议(IEEE Workshop on Signal ProcessingAdvances in Wireless Communication)上发表文章“无线MIMO-OFDM系统中双选择性衰落信道的自适应辨识和跟踪(Adaptive Identification and Trackingof Doubly Selective Fading Channels for Wireless MIMO-OFDM Systems)”，该文提出一种适用于多天线OFDM系统的基于最小均方自适应滤波器(LMS)的信道估计方法，在无需知道信道统计信息的基础上，利用导频符号和LMS滤波器，对时变信道进行跟踪和估计，但该方法有个明显的缺点就是估计精度不是很高，尤其是在中低信噪比的环境下，远远小于基于二维最小均方误差准则的信道估计方法。After searching the literature of the prior art, it was found that Dieter Schafhuber et al. published an article "Dual selectivity in wireless MIMO-OFDM system" in the 2003 IEEE International Wireless Communication Signal Processing Academic Conference (IEEE Workshop on Signal Processing Advances in Wireless Communication). Adaptive Identification and Tracking of Doubly Selective Fading Channels for Wireless MIMO-OFDM Systems” (Adaptive Identification and Tracking of Doubly Selective Fading Channels for Wireless MIMO-OFDM Systems), this paper proposes a least mean square adaptive filter (LMS) based The channel estimation method uses pilot symbols and LMS filters to track and estimate the time-varying channel without knowing the channel statistical information, but this method has an obvious shortcoming that the estimation accuracy is not very high, especially in In the environment of medium and low signal-to-noise ratio, it is far less than the channel estimation method based on the two-dimensional minimum mean square error criterion.

发明内容Contents of the invention

本发明的目的在于克服现有技术中的不足，提供一种OFDM系统中基于UKF的信道与载波频率偏移估计方法，使其在不知道信道统计信息的情况下具有估计性能稳健、鲁棒性强、抗加性高斯白噪声的能力强、抗子载波间干扰的特点，且易于实现。The purpose of the present invention is to overcome the deficiencies in the prior art, to provide a method for estimating channel and carrier frequency offset based on UKF in an OFDM system, so that it has robust and robust estimation performance without knowing the channel statistical information Strong, strong anti-additive Gaussian white noise, anti-subcarrier interference characteristics, and easy to implement.

本发明是通过以下技术方案实现的，本发明是一种在OFDM系统中运用UKF处理导频信息得到信道与载波频率偏移估计值的方法，在已消除载波频偏的OFDM系统中，只考虑多径信道对系统的影响时，采用基于UKF的信道估计方法；在循环前缀的长度大于信道冲激响应的长度，且频域均衡器对信道进行了有效补偿的OFDM系统中，此时可以认为信道的影响主要来自噪声，而忽略信道多径效应的影响，采用基于UKF的载波频率偏移估计方法；在同时存在载波偏移和多径效应的OFDM系统中，采用基于UKF的信道与载波频率偏移联合估计方法。The present invention is achieved through the following technical solutions. The present invention is a method of using UKF to process pilot information in an OFDM system to obtain an estimated value of channel and carrier frequency offset. In an OFDM system where carrier frequency offset has been eliminated, only the When the influence of multipath channel on the system, the channel estimation method based on UKF is used; in the OFDM system where the length of the cyclic prefix is greater than the length of the channel impulse response, and the frequency domain equalizer has effectively compensated the channel, it can be considered at this time The influence of the channel mainly comes from the noise, while ignoring the influence of the multipath effect of the channel, the carrier frequency offset estimation method based on UKF is adopted; in the OFDM system with carrier offset and multipath effect at the same time, the channel and carrier frequency Offset Joint Estimation Method.

所述的基于UKF的信道估计方法，包括如下步骤：The described channel estimation method based on UKF comprises the steps:

步骤一：使用导频信息通过最小平方的方法获取频域上导频符号处信道状态信息的初始值；Step 1: using the pilot information to obtain the initial value of the channel state information at the pilot symbol in the frequency domain by the method of least squares;

步骤二：通过对所述初始值进行快速逆傅立叶变换，得到时间/时延域的未考虑噪声作用下的信道状态信息；Step 2: Obtain channel state information in the time/delay domain without considering the effect of noise by performing fast inverse Fourier transform on the initial value;

步骤三：针对所述未考虑噪声作用下的信道状态信息运用自适应路径捕获方法获得信道的有效路径的信道状态信息；Step 3: Obtain channel state information of the effective path of the channel by using an adaptive path acquisition method for the channel state information without considering the effect of noise;

步骤四：以捕获到的所述信道的有效路径的信道状态信息作为观测值，再利用UKF滤波器跟踪和估计出当前时刻的考虑噪声作用下的准确的信道状态信息；Step 4: Taking the captured channel state information of the effective path of the channel as the observed value, and then using the UKF filter to track and estimate the accurate channel state information under consideration of noise at the current moment;

步骤五：对所述的准确的信道状态信息进行快速傅立叶变换，得到导频符号的时间/频率域的信道传输函数；Step 5: performing fast Fourier transform on the accurate channel state information to obtain the channel transfer function of the time/frequency domain of the pilot symbols;

步骤六：对所述导频符号的时间/频率域的信道传输函数通过插值方法得到数据符号所对应的信道传输函数，把所述传输函数送给信号检测模块进行检测或译码。Step 6: Obtain the channel transfer function corresponding to the data symbol by interpolating the channel transfer function of the time/frequency domain of the pilot symbol, and send the transfer function to the signal detection module for detection or decoding.

所述的信道状态信息包括有效路径的数目和位置，以及所述数目和位置所对应的信道的幅度和相位。自适应路径捕获方法是通过采用根据排序得到的信道的路径和当前时刻的信噪比来实时确定有效路径的数目和位置。在同一信噪比下，采用相同的有效路径的数目和位置。The channel state information includes the number and position of effective paths, and the amplitude and phase of the channel corresponding to the number and position. The adaptive path acquisition method is to determine the number and position of effective paths in real time by using the paths of the channels obtained according to the ranking and the signal-to-noise ratio at the current moment. Under the same signal-to-noise ratio, the same number and position of effective paths are used.

所述的基于UKF的载波频率偏移估计方法，包括如下步骤：The described carrier frequency offset estimation method based on UKF comprises the steps:

步骤一：导频结构为在时间域插入一个导频符号用作等值导频，导频符号长为子载波数；Step 1: The pilot structure is to insert a pilot symbol in the time domain as an equivalent pilot, and the length of the pilot symbol is the number of subcarriers;

步骤二：把时域接收信号构造为发送信号与衰减系数相乘的形式；Step 2: Construct the received signal in the time domain into a form in which the transmitted signal is multiplied by the attenuation coefficient;

步骤三：在时域内，运用UKF滤波器对时域接收信号进行载波频率偏移估计；Step 3: In the time domain, use the UKF filter to estimate the carrier frequency offset of the received signal in the time domain;

步骤四：利用估计出的载波频移进行载波频率偏移校正，得到消除子载波干扰后的系统。Step 4: Carrier frequency offset correction is performed using the estimated carrier frequency shift to obtain a system after subcarrier interference is eliminated.

所述基于UKF的信道与载波频率偏移联合估计方法，包括如下步骤：The method for jointly estimating channel and carrier frequency offset based on UKF comprises the following steps:

步骤一；导频结构为在时间域每隔既定数目的OFDM符号插入一个导频符号用作等值导频，导频符号长为子载波数；Step 1: The pilot structure is to insert a pilot symbol every predetermined number of OFDM symbols in the time domain as an equivalent pilot, and the length of the pilot symbol is the number of subcarriers;

步骤二：将时间域接收信号构造为受载波频率偏移影响的发送信号与信道的卷积关系形式；Step 2: Constructing the received signal in the time domain as a convolution relationship between the transmitted signal and the channel affected by the carrier frequency offset;

步骤三：将时间域接收信号的所述卷积关系形式转换为矩阵关系形式；Step 3: converting the convolution relational form of the received signal in the time domain into a matrix relational form;

步骤四：在时间域内，运用UKF滤波器对所述矩阵关系形式进行信道估计和载波频率偏移估计；Step 4: In the time domain, use the UKF filter to perform channel estimation and carrier frequency offset estimation on the matrix relational form;

步骤五：利用估计出的载波频移进行载波频率偏移校正，得到消除子载波干扰后的系统。Step 5: Carrier frequency offset correction is performed using the estimated carrier frequency shift to obtain a system after subcarrier interference is eliminated.

步骤六：对所述信道估计的结果通过插值方法得到数据符号所对应的信道传输函数，把所述传输函数送给信号检测模块进行检测或译码。Step 6: Obtain the channel transfer function corresponding to the data symbol by interpolating the result of the channel estimation, and send the transfer function to the signal detection module for detection or decoding.

以下对本发明作进一步详细说明：The present invention is described in further detail below:

1、基于UKF的信道估计方法，具体实现如下：1. The channel estimation method based on UKF, the specific implementation is as follows:

(1)使用导频信息通过最小平方的方法获取频域上导频符号处信道状态信息的初始值(1) Use the pilot information to obtain the initial value of the channel state information at the pilot symbol in the frequency domain by the method of least squares

假设采用N_T根发射天线，N_R根接收天线，n个OFDM符号以及K个子载波的MIMO-OFDM系统。

为发送信号向量，a⁽ⁱ⁾[n，k]表示天线i在n时刻，第k个子载波上的发送信号。对a[n，k]作IFFT并加上长为L_CP的循环前缀(CP)得到第n个OFDM符号

每个OFDM符号持续时间为N＝K+L_cp。全部基带发送信号为

It is assumed that a MIMO-OFDM system with _NT transmitting antennas, _NR receiving antennas, n OFDM symbols and K subcarriers is adopted.

is the transmitted signal vector, a ⁽ⁱ⁾ [n, k] represents the transmitted signal of antenna i on the kth subcarrier at time n. Perform IFFT on a[n, k] and add a cyclic prefix (CP) of length L _CP to get the nth OFDM symbol

Each OFDM symbol has a duration of N=K+L _cp . All baseband signals are

${s the s}_{n no} [[m m]] = = \{\begin{matrix} \frac{11}{\sqrt{{KN KN}_{T T}}} {Σ Σ}_{k k = = 00}^{K K - - 11} a a [[n no,, k k]] {e e}^{j j 22 πmk πmk / / K K} & m m = = - - {L L}_{CP CP},, . . . . . .,, K K - - 11 \\ 00 & else else \end{matrix} - - - - - - ((11))$

信道的冲激响应可以表示为The impulse response of the channel can be expressed as

$h h ((t t,, τ τ)) = = \underset{k k}{Σ Σ} {γ γ}_{k k} ((t t)) c c ((τ τ - - {τ τ}_{k k})) - - - - - - ((22))$

在t时刻的频率响应为The frequency response at time t is

$H h ((t t,, f f)) = = {&Integral; &Integral;}_{- - \infty \infty}^{+ + \infty \infty} h h ((t t,, τ τ)) {e e}^{j j 22 πfτ πfτ} dτ dτ = = C C ((f f)) \underset{k k}{Σ Σ} {γ γ}_{k k} {e e}^{j j 22 πf πf {τ τ}_{k k}} - - - - - - ((33))$

$C C ((f f)) = = {&Integral; &Integral;}_{- - \infty \infty}^{+ + \infty \infty} c c ((τ τ)) {e e}^{- - j j 22 πfτ πfτ} dτ dτ - - - - - - ((44))$

接收天线接收到的信号由衰落MIMO N_R×N_T的信道矩阵H[m，l]，传输信号S[m]和噪声η[m]形成。η[m]为分布为N(0，σ_η ²)高斯白噪声The signal received by the receiving antenna is formed by the fading MIMO N _R × _NT channel matrix H[m, l], the transmission signal S[m] and the noise η[m]. η[m] is Gaussian white noise with distribution N(0, σ _η ² )

$r r [[m m]] = = {Σ Σ}_{l l = = 00}^{L L - - 11} H h [[m m,, l l]] S S [[m m - - l l]] + + {σ σ}_{η η} [[m m]] - - - - - - ((55))$

接收信号r[m]去循环前缀做FFT得到The received signal r[m] is obtained by removing the cyclic prefix and doing FFT

$X x [[n no,, k k]] = = \frac{11}{\sqrt{K K}} {Σ Σ}_{m m = = 00}^{K K - - 11} r r [[nN n + + m m]] {e e}^{- - j j 22 km km / / K K} - - - - - - ((66))$

如果Nf_Doppler＜＜1，H[m，l]在一个OFDM符号的变化可以忽略，本发明中的MIMO-OFDM系统的输入/输出关系可以表示为If Nf _Doppler <<1, the change of H[m, l] in one OFDM symbol can be ignored, and the input/output relationship of the MIMO-OFDM system in the present invention can be expressed as

$X x [[n no,, k k]] = = \overset{^^}{H h} [[n no,, k k]] a a [[n no,, k k]] + + {\overset{^^}{σ σ}}_{η η} [[n no,, k k]] - - - - - - ((77))$

本发明的导频信息为结构采用块状的导频符号，每个导频符号包括所有的子载波上的时频块，导频的帧结构采用在一帧数据前面加入若干个导频符号的结构，在多天线系统中不同发射天线在同一时刻的导频是互相正交的。为了方便地处理信号，在一根天线上的同一个OFDM符号的不同子载波上的导频信号是相同的，不同发送天线发送的导频互相正交。P为所有天线上的导频序列所组成的导频矩阵，这里P_n是一根天线上的一串导频序列，P_n是一串周期为N_T的循环序列，

The pilot information of the present invention adopts block-shaped pilot symbols in structure, and each pilot symbol includes time-frequency blocks on all subcarriers, and the frame structure of the pilot adopts the method of adding several pilot symbols in front of a frame of data In a multi-antenna system, the pilots of different transmitting antennas at the same time are mutually orthogonal. In order to process signals conveniently, the pilot signals on different subcarriers of the same OFDM symbol on one antenna are the same, and the pilot signals sent by different transmitting antennas are orthogonal to each other. P is the pilot matrix composed of pilot sequences on all antennas, Here P _n is a series of pilot sequences on an antenna, P _n is a series of cyclic sequences with a period of _NT ,

在有N_T根发射天线的MIMO-OFDM系统中，发送信号和接收信号的关系表示为In a MIMO-OFDM system with _NT transmitting antennas, the relationship between the transmitted signal and the received signal is expressed as

$X x [[n no,, k k]] = = \sqrt{\frac{{W W}_{T T}}{{N N}_{T T}}} T T [[n no,, k k]] H h [[n no,, k k]] + + {σ σ}_{η η} [[n no,, k k]] - - - - - - ((88))$

T[n，k]是考虑发送能量的a[n，k]，W_T为发射信号T[n，k]的能量。采用频域LS信道估计，即不考虑噪声的影响得到T[n, k] is a[n, k] considering the transmitted energy, and W _T is the energy of the transmitted signal T[n, k]. Using frequency-domain LS channel estimation, that is, without considering the influence of noise, we get

$\overset{^^}{H h} [[n no,, k k]] = = \sqrt{\frac{{N N}_{T T}}{{W W}_{T T}}} {T T}^{- - 11} [[n no,, k k]] X x [[n no,, k k]] - - - - - - ((99))$

导频处的信道估计式Channel Estimation Equation at Pilot

$\overset{^^}{H h} [[n no,, k k]] = = \sqrt{\frac{{N N}_{T T}}{{W W}_{T T}}} {P P}^{H h} X x [[n no,, k k]] - - - - - - ((1010))$

信道传输方程表示为The channel transmission equation is expressed as

$\overset{^^}{H h} [[n no,, k k]] = = H h [[n no,, k k]] + + ψ ψ [[n no,, k k]] - - - - - - ((1111))$

因此通过对

做IFFT得到信道的时域状态信息

(11)式转化为Therefore through the

Do IFFT to get the time domain state information of the channel

(11) is transformed into

${\overset{~ ~}{h h}}_{l l} [[n no]] = = {h h}_{l l} [[n no]] + + {z z}_{l l} [[n no]] - - - - - - ((1212))$

其中z_l[n]为满足N(0，σ_z ²)分布的零均值，高斯向量。Where z _l [n] is a zero-mean and Gaussian vector satisfying N(0, σ _z ² ) distribution.

(2)通过对所述初始值进行快速逆傅立叶变换，得到时间/时延域的未考虑噪声作用下的信道状态信息(2) By performing fast inverse Fourier transform on the initial value, the channel state information in the time/delay domain without considering the effect of noise is obtained

通过快速逆傅立叶变换(IFFT)得到时间/时延域的未考虑噪声作用下的信道状态信息。即接收机对LS方法估计出来的信道估计初值做快速逆傅立叶变换，得到时间/时延域的信道估计值。The channel state information in the time/delay domain without considering the effect of noise is obtained by fast inverse Fourier transform (IFFT). That is, the receiver performs fast inverse Fourier transform on the initial channel estimation value estimated by the LS method to obtain the channel estimation value in the time/delay domain.

(3)针对所述未考虑噪声作用下的信道状态信息运用自适应路径捕获方法获得信道的有效路径的信道状态信息(3) Using an adaptive path acquisition method for the channel state information without considering the effect of noise to obtain the channel state information of the effective path of the channel

本发明中采用了自适应门限捕获方法，该方法根据排序的信道抽头和当前时刻的信噪比来实时确定重要抽头的数目L和位置，使L的取值具有一定的自适应性，这种方法要优于传统的确定路径捕获法和传统的门限捕获法。In the present invention, an adaptive threshold acquisition method is adopted, which determines the number L and position of important taps in real time according to the sorted channel taps and the signal-to-noise ratio at the current moment, so that the value of L has a certain degree of self-adaptability. The method is superior to the traditional definite path capture method and the traditional threshold capture method.

自适应门限捕获法：Adaptive threshold capture method:

L为重要抽头的数目L is the number of important taps

L_m是所能接受的最大复杂度的多径数，一般取循环前缀数L _m is the number of multipaths with the maximum complexity that can be accepted, and generally takes the number of cyclic prefixes

为快速逆傅立叶变换(IFFT)，得到时间/时延域的信道估计值

For the fast inverse Fourier transform (IFFT), the channel estimation value in the time/delay domain is obtained

为

的降序排列

for

in descending order of

L_i为满足下式的多径抽头数L _i is the number of multipath taps satisfying the following formula

${| | \overset{^^}{h h} [[l l]] | |}^{22} > > η η \cdot &Center Dot; \frac{11}{α α} {Σ Σ}_{l l = = 00}^{α α - - 11} {| | {\overset{^^}{h h}}_{sort sort} [[l l]] | |}^{22} - - - - - - ((1313))$

L为L_i和L_m的最小值L is the minimum of L _i and L _m

η是一个门限值，η的取值决定于SNR，

η is a threshold value, and the value of η depends on the SNR,

(4)以捕获到的所述信道的有效路径的信道状态信息作为观测值，再利用UKF滤波器跟踪和估计出当前时刻的考虑噪声作用下的准确的信道状态信息(4) Take the captured channel state information of the effective path of the channel as an observation value, and then use the UKF filter to track and estimate the accurate channel state information under the noise effect at the current moment

以当前时刻未考虑噪声作用下的信道状态信息作为观测值，利用UKF跟踪和估计当前时刻的考虑噪声作用下的准确的信道状态信息，在迭代过程中，状态噪声和测量噪声不断更新。Taking the channel state information without considering the noise at the current moment as the observed value, the UKF is used to track and estimate the accurate channel state information at the current moment under the effect of noise. During the iterative process, the state noise and measurement noise are constantly updated.

状态方程：Equation of state:

x_n+1＝F(x_n)+v_n (14)x _n+1 ＝F(x _n )+v _n (14)

测量方程：Measurement equation:

y_n＝H(x_n)+n_n (15)y _n =H(x _n )+n _n (15)

其中 $x_{n} = [\begin{matrix} h_{0} [n] \\ . \\ . \\ . \\ h_{L - 1} [n] \\ vec {T} \end{matrix}],$ $G (x_{n}) = [\begin{matrix} T h_{0} [n] \\ . \\ . \\ . \\ T h_{L - 1} [n] \\ vec {T} \end{matrix}]$ $H (x_{n}) = [\begin{matrix} h_{0} [n] \\ . \\ . \\ . \\ h_{L - 1} [n] \\ h_{L - 1} [n] \end{matrix}]$ in $x_{no} = [\begin{matrix} h_{0} [no] \\ . \\ . \\ . \\ h_{L - 1} [no] \\ vec {T} \end{matrix}],$ $G (x_{no}) = [\begin{matrix} T h_{0} [no] \\ . \\ . \\ . \\ T h_{L - 1} [no] \\ vec {T} \end{matrix}]$ $h (x_{no}) = [\begin{matrix} h_{0} [no] \\ . \\ . \\ . \\ h_{L - 1} [no] \\ h_{L - 1} [no] \end{matrix}]$

v_n，n_n分别为状态噪声向量和测量噪声向量。v _n , n _n are state noise vector and measurement noise vector respectively.

LM×1维测量向量

M＝N_T×N_R，N_T为发射天线数，N_R为接收天线数。LM×1-dimensional measurement vector

M=N _T ×N _R , where _NT is the number of transmitting antennas, and _NR is the number of receiving antennas.

(5)对所述的准确的信道状态信息进行快速傅立叶变换，得到导频符号的时间/频率域的信道传输函数(5) Carry out fast Fourier transform to described accurate channel state information, obtain the channel transfer function of the time/frequency domain of pilot symbol

以上步骤求得时间/时延域的信道估计的准确值，但是信号检测需要的是信道的时间/频率域的状态信息，因此，对时间/时延域的信道估计值做快速傅立叶变换(FFT)变换得到时间/频率域信道估计值。The above steps obtain the accurate value of the channel estimation in the time/delay domain, but what signal detection needs is the state information of the time/frequency domain of the channel, therefore, do fast Fourier transform (FFT) to the channel estimation value in the time/delay domain ) transformation to obtain the time/frequency domain channel estimation value.

(6)对所述导频符号的时间/频率域的信道传输函数通过插值方法得到数据符号所对应的信道传输函数，把所述传输函数送给信号检测模块进行检测或译码(6) Obtain the channel transfer function corresponding to the data symbol by interpolation to the channel transfer function of the time/frequency domain of the pilot symbols, and send the transfer function to the signal detection module for detection or decoding

2、基于UKF的载波频率偏移估计方法，具体实现如下：2. The carrier frequency offset estimation method based on UKF, the specific implementation is as follows:

(1)导频结构为在时间域插入一个导频符号用作等值导频，导频符号长为子载波数(1) The pilot structure is to insert a pilot symbol in the time domain as an equivalent pilot, and the length of the pilot symbol is the number of subcarriers

考虑接收到的信号中引入了频率偏移的情况。频率偏移可以看作是信道中引入的一个乘性因子，假设采用2发2收的MIMO-OFDM系统，Consider the case where a frequency offset is introduced into the received signal. The frequency offset can be regarded as a multiplicative factor introduced in the channel. Assuming a 2-transmit and 2-receive MIMO-OFDM system,

${r r}_{n no}^{11} = = {t t}_{n no}^{11} \times \times {e e}^{j j \frac{22 π π}{N N} n no {ϵ ϵ}_{n no}^{1111}} + + {t t}_{n no}^{22} \times \times {e e}^{j j \frac{22 π π}{N N} n no {ϵ ϵ}_{n no}^{21 twenty one}} + + {w w}_{n no}^{11} - - - - - - ((1616))$

${r r}_{n no}^{22} = = {t t}_{n no}^{11} \times \times {e e}^{j j \frac{22 π π}{N N} n no {ϵ ϵ}^{21 twenty one}} + + {t t}_{n no}^{22} \times \times {e e}^{j j \frac{22 π π}{N N} n no {ϵ ϵ}^{22 twenty two}} + + {w w}_{n no}^{22} - - - - - - ((1717))$

t_n ⁱ为n时刻第i根发射天线的发送信号，r_n ^j为n时刻第j根接收天线的接收信号，ε_n ^ij为发射天线i到接收天线j传输路径上的归一化的频率频移值，w_n ^j为传输路径上引入的加性高斯白噪声，N是子载波数，L_CP是循环前缀数。t _n ⁱ is the sending signal of the i-th transmitting antenna at time n, r _n ^j is the receiving signal of the j-th receiving antenna at n time, ε _n ^ij is the normalized frequency on the transmission path from transmitting antenna i to receiving antenna j Frequency shift value, w _n ^j is the additive Gaussian white noise introduced on the transmission path, N is the number of subcarriers, and L _CP is the number of cyclic prefixes.

在时间域插入一个OFDM符号用作等值导频。An OFDM symbol is inserted in the time domain as an equivalent pilot.

$p_{n}^{i} = 1,$ k＝1，2...N+L_CP-1，i＝1，2 $p_{no}^{i} = 1,$ k=1, 2...N+L _CP -1, i=1, 2

其中，N是子载波数，L_CP是循环前缀数。Wherein, N is the number of subcarriers, and L _CP is the number of cyclic prefixes.

(2)把时域接收信号构造为发送信号与衰减系数相乘的形式导频处的频率偏移模型为：(2) Construct the received signal in the time domain as the form of multiplying the transmitted signal and the attenuation coefficient. The frequency offset model at the pilot is:

${r r}_{n no}^{11} = = {p p}_{n no}^{11} \times \times {e e}^{j j \frac{22 π π}{N N} n no {ϵ ϵ}_{n no}^{1111}} + + {p p}_{n no}^{22} \times \times {e e}^{j j \frac{22 π π}{N N} n no {ϵ ϵ}_{n no}^{21 twenty one}} + + {w w}_{n no}^{11} - - - - - - ((1818))$

${r r}_{n no}^{22} = = {p p}_{n no}^{11} \times \times {e e}^{j j \frac{22 π π}{N N} n no {ϵ ϵ}^{21 twenty one}} + + {p p}_{n no}^{22} \times \times {e e}^{j j \frac{22 π π}{N N} n no {ϵ ϵ}^{22 twenty two}} + + {w w}_{n no}^{22} - - - - - - ((1919))$

(3)在时域内，运用UKF滤波器对时域接收信号进行载波频率偏移估计UKF估计载波频率偏移：(3) In the time domain, use the UKF filter to estimate the carrier frequency offset of the received signal in the time domain UKF estimates the carrier frequency offset:

状态方程：Equation of state:

x_n+1＝F(x_n)+v_n x _n+1 ＝F(x _n )+v _n

测量方程：Measurement equation:

y_n＝H(x_n)+n_n y _n ＝H(x _n )+n _n

其中

F(x_n)＝x_n，

为观测向量，v_n，n_n分别为状态噪声与测量噪声向量。in

F(x _n )=x _n ,

is the observation vector, v _n , n _n are the state noise and measurement noise vectors respectively.

可以看出测量方程为非线性方程，需要采用非线性滤波器来进行处理。采用UKF滤波器处理(18)，(19)式。It can be seen that the measurement equation is a nonlinear equation, which needs to be processed by a nonlinear filter. Use UKF filter to process (18), (19) formula.

(4)利用估计出的载波频移进行载波频率偏移校正，得到消除子载波干扰后的系统(4) Use the estimated carrier frequency shift to correct the carrier frequency offset, and obtain the system after subcarrier interference is eliminated

3、基于UKF的信道与载波频率偏移联合估计方法，具体实现如下：3. The joint estimation method of channel and carrier frequency offset based on UKF, the specific implementation is as follows:

(1)导频结构为在时间域每隔既定数目的OFDM符号插入一个导频符号用作等值导频，导频符号长为子载波数；(1) The pilot structure is to insert a pilot symbol every predetermined number of OFDM symbols in the time domain as an equivalent pilot, and the length of the pilot symbol is the number of subcarriers;

假设采用N_T根发射天线，N_R根接收天线，n个OFDM符号以及N个子载波的MIMO-OFDM系统。aⁱ(n，k)表示天线i在n时刻，第k个子载波上的发送信号。对aⁱ(n，k)作IFFT并加上长为L_CP的循环前缀(CP)得到第个OFDM符号s_n ⁱ(m)，每个OFDM符号持续时间为N＝K+L_cp。It is assumed that a MIMO-OFDM system with _NT transmitting antennas, _NR receiving antennas, n OFDM symbols and N subcarriers is adopted. a ⁱ (n, k) represents the transmission signal of antenna i on the kth subcarrier at time n. Perform IFFT on a ⁱ (n, k) and add a cyclic prefix (CP) whose length is L _CP to obtain the th OFDM symbol s _n ⁱ (m), and the duration of each OFDM symbol is N=K+L _cp .

信道模型为：The channel model is:

${s the s}_{ϵn ϵn}^{ij ij} ((m m)) = = {s the s}_{n no}^{i i} ((m m)) \times \times {e e}^{\frac{j j 22 πm πm {ϵ ϵ}^{ij ij}}{N N}} - - - - - - ((2020))$

其中

为为发射天线i到接收天线j传输路径上传送第n个OFDM符号时的归一化的频率频移值，f_d ^ij是频率偏移，f_s＝1/T_s是数据符号s_n ⁱ(m)的传输速率，Δf＝f_s/N是子载波间隔。in

is the normalized frequency shift value when the nth OFDM symbol is transmitted on the transmission path from the transmitting antenna i to the receiving antenna j, f _d ^ij is the frequency offset, and f _s =1/T _s is the data symbol s _n ⁱ (m) transmission rate, Δf=f _s /N is the subcarrier spacing.

导频的帧结构采用在一帧数据前面加入一个导频符号的结构。The frame structure of the pilot adopts a structure in which a pilot symbol is added in front of a frame of data.

(2)将时间域接收信号构造为受载波频率偏移影响的发送信号与信道的卷积关系形式(2) The received signal in the time domain is constructed as the convolutional relationship between the transmitted signal and the channel affected by the carrier frequency offset

考虑系统所经过的信道为慢衰落信道，在传输一个OFDM符号的时间内，信道信息不变。Considering that the channel passed by the system is a slow fading channel, the channel information remains unchanged during the time of transmitting one OFDM symbol.

如果Nf_Doppler＜＜1，信道在一个OFDM符号的变化可以忽略，本发明的MIMO-OFDM系统的输入输出关系可表示为If Nf _Doppler << 1, the change of the channel in one OFDM symbol can be ignored, and the input-output relationship of the MIMO-OFDM system of the present invention can be expressed as

${r r}_{n no}^{j j} = = {Σ Σ}_{i i = = 11}^{{N N}_{T T}} {h h}_{n no}^{ij ij} * * {s the s}_{ϵn ϵn}^{ij ij} + + {η η}_{n no}^{j j} - - - - - - ((21 twenty one))$

其中r_n ^j＝[r_n ^j(1)，...，r_n ^j(N)]^T为接收天线j上的接收信号向量，h_n ^ij＝[h_n ^ij(1)，...，h_n ^ij(L)]^T为发射天线i到接收天线j传输路径的信道冲激响应向量，s_εn ^ij＝[s_εn ^ij(1)，...，s_εn ^ij(N)]^T为存在载波频偏的发送信号向量，η_n ^j＝[η_n ^j(1)，...，η_n ^j(N)]^T为加性高斯白噪声列矢量，L为信道冲激响应的最大长度。Where r _n ^j =[r _n ^j (1),...,r _n ^j (N)] ^T is the received signal vector on receiving antenna j, h _n ^ij =[h _n ^ij (1),... , h _n ^ij (L)] ^T is the channel impulse response vector of the transmission path from transmitting antenna i to receiving antenna j, s _εn ^ij =[s _εn ^ij (1),...,s _εn ^ij (N)] ^T is the transmitted signal vector with carrier frequency offset, η _n ^j =[η _n ^j (1), ..., η _n ^j (N)] ^T is the additive Gaussian white noise column vector, L is the channel impulse response The maximum length.

(3)将时间域接收信号的所述卷积关系形式转换为矩阵关系形式(3) Convert the convolution relational form of the time domain received signal into a matrix relational form

卷积形式可以表示为矩阵相乘的形式有The convolution form can be expressed as a matrix multiplication form with

${r r}_{n no}^{j j} = = {Σ Σ}_{i i = = 11}^{{N N}_{T T}} {S S}_{ϵn ϵn}^{ij ij} {h h}_{n no}^{ij ij} + + {η η}_{n no}^{j j} - - - - - - ((22 twenty two))$

经过无线信道，去循环前缀的接收信号表示为：Through the wireless channel, the received signal with the cyclic prefix removed is expressed as:

${r r}_{n no}^{11} = = [[{C C}_{ϵ ϵ}^{1111} \cdot \cdot {p p}_{n no}^{11}]] * * {h h}_{n no}^{1111} + + [[{C C}_{ϵ ϵ}^{21 twenty one} \cdot \cdot {p p}_{n no}^{22}]] * * {h h}_{n no}^{21 twenty one} + + {η η}_{n no}^{11} - - - - - - ((23 twenty three))$

${r r}_{n no}^{22} = = [[{C C}_{ϵ ϵ}^{1212} \cdot &Center Dot; {p p}_{n no}^{11}]] * * {h h}_{n no}^{1212} + + [[{C C}_{ϵ ϵ}^{22 twenty two} \cdot &Center Dot; {p p}_{n no}^{22}]] * * {h h}_{n no}^{22 twenty two} + + {η η}_{n no}^{22} - - - - - - ((24 twenty four))$

其中p_n ⁱ为导频列向量where p _n ⁱ is the pilot column vector

${r r}_{n no}^{11} = = [[{P P}_{ϵn ϵn}^{1111}]] * * {h h}_{n no}^{1111} + + [[{p p}_{ϵn ϵn}^{21 twenty one}]] * * {h h}_{n no}^{21 twenty one} + + {η η}_{n no}^{11} - - - - - - ((2525))$

${r r}_{n no}^{22} = = [[{P P}_{ϵn ϵn}^{1212}]] * * {h h}_{n no}^{1212} + + [[{p p}_{ϵn ϵn}^{22 twenty two}]] * * {h h}_{n no}^{22 twenty two} + + {η η}_{n no}^{22} - - - - - - ((2626))$

其中 $p_{ϵn}^{ij} = C_{ϵ}^{ij} p_{n}^{i}$ in $p_{ϵn}^{ij} = C_{ϵ}^{ij} p_{no}^{i}$

将卷积形式表示成矩阵形式，有Expressing the convolution form as a matrix form, we have

${r r}_{n no}^{11} = = {P P}_{ϵn ϵn}^{1111} {h h}_{n no}^{1111} + + {P P}_{ϵn ϵn}^{21 twenty one} {h h}_{n no}^{21 twenty one} + + {η η}_{n no}^{11} - - - - - - ((2727))$

${r r}_{n no}^{22} = = {P P}_{ϵn ϵn}^{1212} {h h}_{n no}^{1212} + + {P P}_{ϵn ϵn}^{22 twenty two} {h h}_{n no}^{22 twenty two} + + {η η}_{n no}^{22} - - - - - - ((2828))$

(4)在时间域内，运用UKF滤波器对所述矩阵关系形式进行信道估计和载波频率偏移估计(4) In the time domain, use the UKF filter to perform channel estimation and carrier frequency offset estimation on the matrix relational form

首先用UKF来处理式(27)First use UKF to deal with equation (27)

x_n+1＝F(x_n)+v_n (29)x _n+1 ＝F(x _n )+v _n (29)

y_n＝H(x_n)+n_n (30)y _n =H(x _n )+n _n (30)

其中F(x_n)＝x_n，

为观测向量，v_n，n_n分别为状态噪声与测量噪声向量。in F(x _n )=x _n ,

运用UKF滤波器到信道的时域估计值

经FFT变换得到频域估计值

用于信道检测。Applying the UKF filter to the time-domain estimate of the channel

The estimated value in the frequency domain is obtained by FFT transformation

Used for channel detection.

$\overset{^^}{H h} ((n no,, k k)) = = \frac{11}{{N N}_{T T}} {Σ Σ}_{l l = = 00}^{L L - - 11} {\overset{^^}{h h}}_{00} ((l l)) {e e}^{- - j j 22 kl kl / / N N} - - - - - - ((3131))$

(5)利用估计出的载波频移进行载波频率偏移校正，得到消除子载波干扰后的系统(5) Use the estimated carrier frequency shift to correct the carrier frequency offset, and obtain the system after subcarrier interference is eliminated

(6)对所述信道估计的结果通过插值方法得到数据符号所对应的信道传输函数，把所述传输函数送给信号检测模块进行检测或译码。(6) Obtain the channel transfer function corresponding to the data symbol through the interpolation method from the channel estimation result, and send the transfer function to the signal detection module for detection or decoding.

本发明的优点在于：基于UKF的信道估计方法中采用了自适应路径捕获算法大大降低了计算复杂度并提高了估计精度；用UKF对信道估计值进行滤波处理，提高估计精度。The invention has the advantages that: the UKF-based channel estimation method adopts the self-adaptive path acquisition algorithm, which greatly reduces the computational complexity and improves the estimation precision; the UKF is used to filter the channel estimation value to improve the estimation precision.

基于UKF的载波频率偏移估计的方法来消除ICI，与传统的SC干扰自消除方法相比，在处理频率偏移较大的情况下，有更好的消除ICI的性能，而且带宽利用率高。本发明中基于UKF的载波频率偏移估计的方法来消除ICI的方法与ML方法的复杂度相似，但是ML方法需要重复发送码元，需要一倍的冗余，同SC方法具有一样的带宽效率，所以基于UKF的方法比ML方法的带宽利用率高。The carrier frequency offset estimation method based on UKF is used to eliminate ICI. Compared with the traditional SC interference self-cancellation method, it has better ICI elimination performance and high bandwidth utilization in the case of large frequency offset. . In the present invention, the method based on carrier frequency offset estimation of UKF to eliminate ICI is similar to the complexity of the ML method, but the ML method needs to repeatedly send symbols, requires twice the redundancy, and has the same bandwidth efficiency as the SC method , so UKF-based methods have higher bandwidth utilization than ML methods.

基于UKF的信道与载波频率偏移联合估计方法可以同时估计出信道信息与子载波间频率偏移。由于UKF比EKF更适合处理非线性问题，所以，这种基于UKF的联合估计方法能够更好地消除载波频偏对信道估计的影响，具有令人满意的性能，。The joint estimation method of channel and carrier frequency offset based on UKF can estimate the channel information and the frequency offset between subcarriers at the same time. Since UKF is more suitable for dealing with nonlinear problems than EKF, this joint estimation method based on UKF can better eliminate the influence of carrier frequency offset on channel estimation, and has satisfactory performance.

附图说明Description of drawings

图1是OFDM系统中基于UKF的信道与载波频率偏移估计方法原理框图Figure 1 is a block diagram of the channel and carrier frequency offset estimation method based on UKF in the OFDM system

图2是MIMO-OFDM系统中信道估计导频结构图Figure 2 is a structure diagram of the channel estimation pilot in the MIMO-OFDM system

图3是MIMO-OFDM系统中基于UKF的信道估计方法框图Figure 3 is a block diagram of a UKF-based channel estimation method in a MIMO-OFDM system

图4是信道估计中自适应路径捕获方法与确定路径数信道估计的MSE性能比较图Figure 4 is a comparison of the MSE performance between the adaptive path acquisition method and the determined path number channel estimation in channel estimation

图5是信道估计中基于UKF、LS和LMS的信道估计BER性能比较图Figure 5 is a comparison of channel estimation BER performance based on UKF, LS and LMS in channel estimation

图6是MIMO-OFDM系统中基于UKF的载波频率偏移估计原理图Figure 6 is a schematic diagram of carrier frequency offset estimation based on UKF in MIMO-OFDM system

图7是MIMO-OFDM系统存在频率偏移时的BER信能图Figure 7 is the BER signal energy diagram when there is a frequency offset in the MIMO-OFDM system

图8是MIMO-OFDM系统中信道与载波频率偏移联合估计导频结构图Figure 8 is a structure diagram of pilots for joint estimation of channel and carrier frequency offset in MIMO-OFDM system

图9是MIMO-OFDM系统中基于UKF的信道与载波频率偏移联合估计原理图Figure 9 is a schematic diagram of joint estimation of channel and carrier frequency offset based on UKF in MIMO-OFDM system

图10是基于UKF的信道与载波频率偏移联合估计的BER性能图Figure 10 is a BER performance diagram based on the joint estimation of channel and carrier frequency offset based on UKF

具体实施方式Detailed ways

下面结合附图对本发明的实施例作详细说明：本实施例在以本发明技术方案为前提下进行实施，给出了详细的实施方式和具体的操作过程，但本发明的保护范围不限于下述的实施例。The embodiments of the present invention are described in detail below in conjunction with the accompanying drawings: this embodiment is implemented on the premise of the technical solution of the present invention, and detailed implementation methods and specific operating procedures are provided, but the protection scope of the present invention is not limited to the following the described embodiment.

本发明其原理框图如图1所示，在已消除载波频偏的OFDM系统中，只考虑多径信道对系统的影响时，采用基于UKF的信道估计方法；在循环前缀的长度大于信道冲激响应的长度，且频域均衡器对信道进行了有效补偿的OFDM系统中，此时可以认为信道的影响主要来自噪声，而忽略信道多径效应的影响，采用基于UKF的载波频率偏移估计方法；在同时存在载波偏移和多径效应的OFDM系统中，Its functional block diagram of the present invention is shown in Figure 1, in the OFDM system that has eliminated carrier frequency offset, when only considering the impact of multipath channel on the system, adopt the channel estimation method based on UKF; In the OFDM system where the frequency domain equalizer has effectively compensated the channel, it can be considered that the influence of the channel mainly comes from noise, and the influence of channel multipath effect is ignored, and the carrier frequency offset estimation method based on UKF is adopted ; In an OFDM system with both carrier offset and multipath effects,

采用基于UKF的信道与载波频率偏移联合估计方法。The joint estimation method of channel and carrier frequency offset based on UKF is adopted.

下面给出一个具体的OFDM参数配置，来阐述本发明的实现步骤。需要说明的是，下例中的参数并不影响本发明的一般性。A specific OFDM parameter configuration is given below to illustrate the implementation steps of the present invention. It should be noted that the parameters in the following examples do not affect the generality of the present invention.

本发明采用2发2收的MIMO-OFDM系统，即N_T＝2，N_R＝2.采用QPSK的调制方式，空时编码方案采样正交空时块码(O-STBC)，码率为1/2，空时解码采样最大似然(ML)译码方案，载频4GHz，带宽6M，子载波个数为64，考虑到本发明的信道估计方法比较适合于频率选择性慢衰落信道，因此信道模型采用ITU提出的多径瑞利衰落信道，信道多径数目为5，时延为[0 260 520 780 1040]ns，功率为[-1.78 0 -7.47 -10 -12.62]dB，运动速度3km/h。The present invention adopts the MIMO-OFDM system of 2 transmissions and 2 receptions, that is, _NT = 2, _NR = 2. The modulation mode of QPSK is adopted, and the space-time coding scheme samples the Orthogonal Space-Time Block Code (O-STBC), and the code rate is 1/2, space-time decoding sampling maximum likelihood (ML) decoding scheme, carrier frequency 4GHz, bandwidth 6M, the number of subcarriers is 64, considering that the channel estimation method of the present invention is more suitable for frequency selective slow fading channel, Therefore, the channel model adopts the multipath Rayleigh fading channel proposed by ITU, the number of channel multipath is 5, the time delay is [0 260 520 780 1040]ns, the power is [-1.78 0 -7.47 -10 -12.62]dB, the movement speed 3km/h.

基于UKF的信道估计方法具体实现如下：The specific implementation of the channel estimation method based on UKF is as follows:

(1)如图2所示，每隔六个OFDM符号插入两个导频符号，共发送600个OFDM符号，即插入200个导频符号。导频采样正交化设计，不同发送天线发送的导频互相正交，发送天线1到发送天线2的导频分别为P₁＝[1，1]，P₂＝[1，-1]。(1) As shown in FIG. 2, two pilot symbols are inserted every six OFDM symbols, and a total of 600 OFDM symbols are transmitted, that is, 200 pilot symbols are inserted. Orthogonal design of pilot sampling, the pilots transmitted by different transmitting antennas are orthogonal to each other, the pilots of transmitting antenna 1 to transmitting antenna 2 are respectively P ₁ =[1,1], P ₂ =[1,-1].

(2)如图3框图所示，在频域利用最小平方法(LS)求出每个导频处的信道的初始估计值

这实际上是一种不考虑噪声影响的理想化的信道估计方法。(2) As shown in the block diagram of Figure 3, use the least square method (LS) in the frequency domain to find the initial estimated value of the channel at each pilot

This is actually an idealized channel estimation method that does not consider the influence of noise.

(3)如图3所示，每个接收机对各自的LS方法估计出来的信道估计初值做快速逆傅立叶变换(IFFT)，得到时间/时延域的信道估计值

(3) As shown in Figure 3, each receiver performs fast inverse Fourier transform (IFFT) on the initial channel estimation value estimated by its own LS method to obtain the channel estimation value in the time/delay domain

(4)在自适应路径捕获算法中，η是一个门限值，η的取值决定于SNR，

观察可得在信噪比SNR值为10dB时取值范围为0.1-0.4。(4) In the adaptive path acquisition algorithm, η is a threshold value, and the value of η depends on the SNR,

It can be observed that when the signal-to-noise ratio SNR value is 10dB, the value range is 0.1-0.4.

(5)对所述的准确的信道状态信息进行快速傅立叶变换，得到导频符号的时间/频率域的信道传输函数；(5) performing fast Fourier transform on the accurate channel state information to obtain the channel transfer function of the time/frequency domain of the pilot symbols;

(6)对所述导频符号的时间/频率域的信道传输函数通过高斯插值方法得到数据符号所对应的信道传输函数，把所述传输函数送给信号检测模块进行检测或译码。(6) Obtain the channel transfer function corresponding to the data symbol by Gaussian interpolation method for the channel transfer function of the time/frequency domain of the pilot symbol, and send the transfer function to the signal detection module for detection or decoding.

如图4所示，在系统的MSE(均方误差)性能比较中，采用自适应路径捕获算法的UKF信道估计方法比确定路径数为16的UKF信道估计方法和确定路径数为8的UKF信道估计方法分别取得3dB和6dB的性噪比增益。As shown in Figure 4, in the system MSE (mean square error) performance comparison, the UKF channel estimation method using the adaptive path acquisition algorithm is better than the UKF channel estimation method with the number of determined paths being 16 and the UKF channel estimation method with the number of determined paths being 8. Estimation methods achieve 3dB and 6dB SNR gains, respectively.

如图5所示，在系统的MSE(均方误差)性能比较中，基于UKF的信道估计方法比LMS信道估计方法以及LS信道估计分别取得5dB和10dB的性噪比增益。As shown in Figure 5, in the MSE (mean square error) performance comparison of the system, the UKF-based channel estimation method achieves 5dB and 10dB SNR gains compared to the LMS channel estimation method and the LS channel estimation method, respectively.

基于UKF的载波频率估计方法具体实现如下：The specific implementation of the carrier frequency estimation method based on UKF is as follows:

(1)如图6所示，发明中采用2发2收的MIMO-OFDM系统。当OFDM系统存在频率偏移时，频率偏移可以看作是信道中引入的一个乘性因子。(1) As shown in FIG. 6 , the invention adopts a MIMO-OFDM system with 2 transmissions and 2 receptions. When there is a frequency offset in the OFDM system, the frequency offset can be regarded as a multiplicative factor introduced in the channel.

其导频结构为在时间域插入一个导频符号用作等值导频，导频符号长为子载波数，在具体实施方案中即插入一个长为64的值均为1的导频符号；Its pilot structure is to insert a pilot symbol in the time domain as an equivalent pilot, and the length of the pilot symbol is the number of subcarriers. In a specific implementation, a pilot symbol with a length of 64 and a value of 1 is inserted;

2)在时域内载波频率偏移的影响可以表示为

的形式，其中ε为归一化的频率偏移，在具体实施方案中的2发2收系统中存在4个载波频率偏移即ε¹¹发送天线1到接收天线1的载波频率偏移，ε¹²发送天线1到接收天线2的载波频率偏移，ε²¹发送天线2到接收天线1的载波频率偏移，ε²²发送天线2到接收天线2的载波频率偏移，把时域接收信号构造为发送信号与衰减系数相乘的形式；2) The impact of carrier frequency offset in the time domain can be expressed as

In the form of , where ε is the normalized frequency offset, there are 4 carrier frequency offsets in the 2 transmit 2 receive system in the specific implementation, that is, ε ¹¹ carrier frequency offset from transmitting antenna 1 to receiving antenna 1, ε ¹² Carrier frequency offset from transmitting antenna 1 to receiving antenna 2, ε ²¹ Carrier frequency offset from transmitting antenna 2 to receiving antenna 1, ε ²² Carrier frequency offset from transmitting antenna 2 to receiving antenna 2, construct the time domain received signal is the form of multiplying the transmitted signal by the attenuation coefficient;

(3)在时域内，运用UKF滤波器对时域接收信号进行载波频率偏移估计可以看出测量方程为非线性方程，需要采用非线性滤波器来进行处理。采用UKF滤波器处理来估计载波频率偏移。因为载波频率偏移为实数，所以对每一步更新得到频率偏移的估计值x_n取实部，直到时域导频信号估计完毕得到频率偏移ε¹¹，ε²¹，ε¹²，ε²²的估计值，再用频率偏移估计值对接收到的信号进行频率偏移校正；(3) In the time domain, using the UKF filter to estimate the carrier frequency offset of the received signal in the time domain, it can be seen that the measurement equation is a nonlinear equation, which needs to be processed by a nonlinear filter. The carrier frequency offset is estimated using UKF filter processing. Because the carrier frequency offset is a real number, the estimated value x _n of the frequency offset is updated for each step to take the real part until the time domain pilot signal is estimated to obtain the frequency offset ε ¹¹ , ε ²¹ , ε ¹² , ε ²² estimated value, and then use the frequency offset estimated value to perform frequency offset correction on the received signal;

(4)利用估计出的载波频移进行载波频率偏移校正，得到消除子载波干扰后的系统。(4) Carrier frequency offset correction is performed by using the estimated carrier frequency shift to obtain a system after subcarrier interference is eliminated.

仿真结果表明，载波频率偏移估计值ε¹¹，ε¹²，ε²¹和ε²²迅速地收敛到其真实值附近。Simulation results show that the estimated values of carrier frequency offset ε ¹¹ , ε ¹² , ε ²¹ and ε ²² quickly converge to their true values.

如图7所示，在系统的BER(误比特率)性能比较中，可以看出OFDM系统的误码率受到频率偏移的影响很大，随着频率偏移的增加，OFDM系统的误码率将大大的增加，尤其当|ε|＞0.3时，信噪比的改善无助于误码率的降低。As shown in Figure 7, in the performance comparison of the BER (Bit Error Rate) of the system, it can be seen that the bit error rate of the OFDM system is greatly affected by the frequency offset. With the increase of the frequency offset, the bit error rate of the OFDM system The rate will increase greatly, especially when |ε|>0.3, the improvement of signal-to-noise ratio will not help the reduction of bit error rate.

如图8所示，采用基于UKF的载波频率偏移估计方法的系统性能逼近无载波频率偏移的系统。As shown in Figure 8, the system performance of the carrier frequency offset estimation method based on UKF is close to that of the system without carrier frequency offset.

基于UKF的信道与载波频率联合估计方法具体实现如下：The specific implementation of the joint estimation method of channel and carrier frequency based on UKF is as follows:

(1)图9给出该方法的系统模型。导频结构为在时间域每隔7个OFDM符号插入一个导频符号用作等值导频，导频符号长为子载波数；(1) Figure 9 shows the system model of the method. The pilot structure is to insert a pilot symbol every 7 OFDM symbols in the time domain as an equivalent pilot, and the length of the pilot symbol is the number of subcarriers;

(2)将时间域接收信号构造为受载波频率偏移影响的发送信号与信道的卷积关系形式；(2) Constructing the received signal in the time domain as a convolutional relationship between the transmitted signal and the channel affected by the carrier frequency offset;

(3)将时间域接收信号的所述卷积关系形式转换为矩阵关系形式；(3) converting the convolution relational form of the received signal in the time domain into a matrix relational form;

(4)在时间域内，运用UKF滤波器对所述矩阵关系形式进行信道估计和载波频率偏移估计，因为载波频率偏移为实数，所以对每一步更新得到频率偏移的估计值ε_n ^ij取实部，直到时域导频信号估计完毕得到频率偏移。经过基于UKF滤波得到频率偏移的估计值后，用这一估计值对接收到的信号进行频率偏移校正；得到信道的时域估计值经FFT变换得到频域估计值用于信道检测；(4) In the time domain, use the UKF filter to perform channel estimation and carrier frequency offset estimation on the matrix relational form. Since the carrier frequency offset is a real number, the estimated value ε _n ^ij of the frequency offset is obtained for each step update Take the real part until the time-domain pilot signal is estimated to obtain the frequency offset. After obtaining the estimated value of the frequency offset based on UKF filtering, use this estimated value to correct the frequency offset of the received signal; obtain the time-domain estimated value of the channel The estimated value in the frequency domain is obtained by FFT transformation for channel detection;

(5)利用估计出的载波频移进行载波频率偏移校正，得到消除子载波干扰后的系统。(5) Carrier frequency offset correction is performed using the estimated carrier frequency shift to obtain a system after subcarrier interference is eliminated.

本发明提出的方法能够很好地跟踪载波频率偏移的变化，载波频率偏移估计值迅速地收敛到真实值附近，准确快速地估计出载波频移。基于UKF的信道与载波频率偏移联合估计方法的系统BER性能如图10所示。The method proposed by the invention can well track the change of the carrier frequency offset, the estimated value of the carrier frequency offset quickly converges to the vicinity of the real value, and the carrier frequency offset can be accurately and quickly estimated. The system BER performance of the joint estimation method of channel and carrier frequency offset based on UKF is shown in Fig. 10 .

由图可见，基于UKF的信道与载波频率偏移联合估计方法能够很好地估计出载波频移和信道信息，消除ICI和信道多径效应，使系统的性能得到了很大的改善。通过仿真性能比较，在BER为10^-3的时候，本发明中的UKF方法可以比EKF方法取得0.3dB的信噪比增益，可以体现出在处理非线性问题时，UKF的性能要优于EKF。It can be seen from the figure that the joint estimation method of channel and carrier frequency offset based on UKF can estimate carrier frequency offset and channel information well, eliminate ICI and channel multipath effects, and greatly improve the performance of the system. Through the simulation performance comparison, when the BER is 10 ^-3 , the UKF method in the present invention can obtain 0.3dB SNR gain than the EKF method, which can reflect that the performance of the UKF is better than that of the EKF when dealing with nonlinear problems .

Claims

In the ofdm system based on channel and the carrier frequency deviation estimating method of UKF, it is characterized in that, in the ofdm system of eliminating carrier wave frequency deviation, when only considering multipath channel, adopt channel estimation methods based on UKF to the influencing of system; In the length of Cyclic Prefix length greater than channel impulse response, and frequency-domain equalizer has carried out in the ofdm system of effective compensation to channel, this moment, the influence of channel came self noise, ignored the influence of channel multipath effect, adopted the carrier frequency deviation estimating method based on UKF; Exist at the same time in the ofdm system of carrier shift and multipath effect, adopt channel and carrier frequency shift combined estimation method based on UKF;

Described channel estimation methods based on UKF comprises the steps:

1.1): use pilot frequency information to obtain the initial value of frequency pilot sign place channel condition information on the frequency domain by the method for least square;

1.2): by described initial value is carried out fast adverse Fourier transform, obtain the channel condition information under the noise contributions do not considered of time-domain;

1.3): do not consider that at described the utilization of channel condition information under noise contributions Adaptive Path catching method obtains the channel condition information of the active path of channel;

1.4):, utilize the UKF filter tracks again and estimate channel condition information accurately under the consideration noise contributions of current time as measured value with the channel condition information of the active path of the described channel that captures;

1.5): described channel condition information is accurately carried out fast fourier transform, obtain the channel function of the frequency domain of frequency pilot sign;

1.6): the channel function to the frequency domain of described frequency pilot sign obtains the pairing channel function of data symbol by interpolation method, gives signal detection module the pairing channel function of described data symbol and detects or decipher;

Described carrier frequency deviation estimating method based on UKF comprises the steps:

2.1): pilot configuration is to insert a frequency pilot sign as equivalent pilot tone in time-domain, and frequency pilot sign is long to be sub-carrier number;

2.2): the time domain received signal is configured to send the form that signal and attenuation coefficient multiply each other;

2.3): in time domain, utilization UKF filter carries out Carrier frequency offset estimation to the time domain received signal;

2.4): utilize the carrier frequency shift that estimates to carry out carrier frequency shift and proofread and correct the system after the subcarrier interference that is eliminated;

Described channel and carrier frequency shift combined estimation method based on UKF comprise the steps:

3.1): pilot configuration is to insert a frequency pilot sign as equivalent pilot tone in time-domain every the OFDM of prearranged number symbol, and frequency pilot sign is long to be sub-carrier number;

3.2): the convolution relation form that the time-domain received signal is configured to be subjected to the transmission signal and the channel of carrier frequency bias effect;

3.3): with the described convolution relation formal transformation of time-domain received signal is the matrix relationship form;

3.4): in time-domain, utilization UKF filter carries out channel estimating and Carrier frequency offset estimation to described matrix relationship form;

3.5): utilize the carrier frequency shift that estimates to carry out carrier frequency shift and proofread and correct the system after the subcarrier interference that is eliminated;

3.6): the result to described channel estimating obtains the pairing channel function of described data symbol by interpolation method, gives signal detection module the pairing channel function of described data symbol and detects or decipher.
2. in the ofdm system according to claim 1 based on channel and the carrier frequency deviation estimating method of UKF, it is characterized in that: 1.1) described in channel condition information comprise the number and the position of active path and the amplitude and the phase place of the pairing channel of described number and position.
3. based on channel and the carrier frequency deviation estimating method of UKF, it is characterized in that: in the ofdm system according to claim 1 1.3) described in the Adaptive Path catching method be number and the position that the noise of path by adopting the channel that obtains according to ordering and current time is recently determined active path in real time.
4. based on channel and the carrier frequency deviation estimating method of UKF, it is characterized in that: in the ofdm system according to claim 3 1.3) in, under same signal to noise ratio, adopt the number and the position of identical active path.
5. in the ofdm system according to claim 1 based on channel and the carrier frequency deviation estimating method of UKF, it is characterized in that: 1.4) in, utilizing the UKF filtering method to follow the tracks of and estimating the current time process of channel condition information accurately, bring in constant renewal in state-noise in each iteration and measure noise.
6. based on channel and the carrier frequency deviation estimating method of UKF, it is characterized in that: in the ofdm system according to claim 1 2.3) in, described Carrier frequency offset estimation adopts nonlinear filter to realize by the frequency shift (FS) model in conjunction with the pilot tone place.