WO2018000914A1 - 一种单载波信道估计方法 - Google Patents

一种单载波信道估计方法 Download PDF

Info

Publication number
WO2018000914A1
WO2018000914A1 PCT/CN2017/080944 CN2017080944W WO2018000914A1 WO 2018000914 A1 WO2018000914 A1 WO 2018000914A1 CN 2017080944 W CN2017080944 W CN 2017080944W WO 2018000914 A1 WO2018000914 A1 WO 2018000914A1
Authority
WO
WIPO (PCT)
Prior art keywords
channel
value
signal
channel estimation
frame signal
Prior art date
Application number
PCT/CN2017/080944
Other languages
English (en)
French (fr)
Inventor
张锦红
牛进
刘小同
王纯
卜世喜
Original Assignee
晶晨半导体(上海)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 晶晨半导体(上海)股份有限公司 filed Critical 晶晨半导体(上海)股份有限公司
Priority to US15/560,139 priority Critical patent/US10270623B1/en
Publication of WO2018000914A1 publication Critical patent/WO2018000914A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/0212—Channel estimation of impulse response
    • H04L25/0214—Channel estimation of impulse response of a single coefficient
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/309—Measuring or estimating channel quality parameters
    • H04B17/336—Signal-to-interference ratio [SIR] or carrier-to-interference ratio [CIR]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/022—Channel estimation of frequency response
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
    • H04L25/03006—Arrangements for removing intersymbol interference
    • H04L25/03159—Arrangements for removing intersymbol interference operating in the frequency domain
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
    • H04L25/03878—Line equalisers; line build-out devices
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L5/00—Arrangements affording multiple use of the transmission path
    • H04L5/003—Arrangements for allocating sub-channels of the transmission path
    • H04L5/0048—Allocation of pilot signals, i.e. of signals known to the receiver

Definitions

  • the present invention relates to the field of channel coding and modulation technologies, and in particular, to a single carrier channel estimation method.
  • the channel is estimated by passing the input signal through a correlator or by frequency domain division to obtain an initial channel estimate; then the initial channel estimation for multiple frames is performed with a fixed coefficient filter. Filtering to obtain channel estimation; for reconstructed channel, channel estimation is obtained by alpha (alpha) filtering.
  • the initial channel estimation performance obtained by this channel estimation method is not good; and because of the fixed coefficient filtering, the best channel estimation cannot be obtained, because the characteristics of the channel are unknown, the bandwidth of the filter is determined by Doppler, and more Puller is unknown, resulting in an estimated best channel performance.
  • the present invention aims to provide a novel channel estimation method, which obtains a better past initial channel estimation by channel reconstruction; and obtains an initial initial channel estimation by correlator or frequency domain division; The initial channel estimate is adaptively filtered to obtain the best channel estimate.
  • a single carrier channel estimation method comprising:
  • the current frame signal and the channel where the current frame signal is located are subjected to carrier equalization by using a single carrier equalizer.
  • the initial channel estimation value is a frequency domain response of the channel
  • the time domain correlation algorithm is performed on the current frame signal, the sequence code, and the signal to noise ratio estimation value. Or frequency domain division is calculated.
  • the time domain correlation algorithm comprises a linear correlation algorithm or a cyclic correlation algorithm.
  • the transmission sequence decision value includes a sequence code and a transmission sequence estimation value.
  • the reference channel is a frequency domain response of a channel
  • the reference channel is filtered by a filter to obtain the filtered A channel value, and the bandwidth of the filter is equal to the length of the sequence code.
  • the reference channel is filtered by a Fourier transform to obtain the filtered channel value.
  • the adaptive filtering includes:
  • a channel estimate value for generating the next frame signal is calculated using the latest initial channel estimate and the most recently filtered channel value.
  • the current frame signal is a time domain signal
  • the time domain signal is a baseband signal converted into an analog-to-digital converted signal by down-conversion, filtering, and sampling rate conversion, and the rate of the baseband signal is a single symbol rate.
  • each frame signal includes a sequence code and single carrier data connected at the end of the sequence code, and the sequence codes of each frame signal are identical to form a cyclic correlation.
  • the present invention provides a single-carrier channel estimation method, which uses channel estimation and input current frame signal to obtain a sequence code and a signal-to-noise ratio of a current frame signal by equalization, and obtains a reference.
  • Channel H_ref and filtered channel value H_better H_better may also be referred to as a preferred channel estimate, which will be H_better is called the preferred channel estimation value;
  • the initial channel estimation H_init is obtained by the correlator; then the initial channel estimation H_init and the filtered channel value H_better are used as reference values for the adaptive channel estimation and update coefficients. , get the channel estimate for the next frame.
  • the technical solution of the present invention realizes that the channel estimation can be calculated for any channel environment (whether it is stationary or a channel environment moving at different speeds), thereby improving the performance of the receiver.
  • 1 is a frame structure of a DTMB system of the present invention
  • FIG. 3 is a schematic diagram of adaptive filtering of the present invention.
  • the single carrier channel estimation method of the present invention performs carrier equalization on the input current frame signal eqin and the channel H_flt where eqin is located, obtains pn code of eqin, transmits sequence decision value slicer_out, and signal to noise ratio estimated value snr_esti; then according to eqin, pn And snr_esti, the initial channel estimate H_init is calculated; according to eqin, slicer_out and snr_esti, Calculating the reference channel H_ref; filtering the reference channel H_ref to obtain the filtered channel value H_better; and using the reference channel H_ref as a reference value, performing adaptive channel estimation on the initial channel estimation value H_init and the filtered channel value H_better, updating The coefficient gives the channel estimate of the next frame signal.
  • each frame signal (hereinafter referred to as "Symbol") of the present embodiment includes a frame header and a frame body, and the frame header is a sequence code.
  • the frame body is a data block (DATA)
  • the connection mode between the respective frame signals is: the PN connection of the DATA of the previous frame signal and the signal of the subsequent frame, for example, the DATA of the previous frame signal Symbol 0 in FIG. 0 PN 1 1 is connected to a rear signal Symbol, Symbol M-1 is connected to the DATA M-1 Symbol M of PN M, and so on, not repeated here.
  • the sequence code PN of this embodiment has three possibilities, namely PN420, PN595 and PN945, and the corresponding PNs have lengths of 420, 595 and 945 respectively; the PN of each Symbol may be constant or may vary, this implementation The example only describes the same situation.
  • the DATA of this embodiment has two possibilities, corresponding to a single carrier and a multiple carrier, respectively, and has a length of 3780. If DATA is a single carrier, the PNs in each Symbol are identical, and they constitute a cyclic correlation.
  • the lengths of all Fourier transforms (FFT) and inverse Fourier transform (IFFT) operations involved in this paper are both lengths of PN + 3780.
  • Figure 2 is a single carrier channel estimate divided into four parts. Respectively: 1 acquisition of initial channel estimate (H_init); 2 acquisition of reference channel (H_ref); 3 preferred channel Estimation of the estimated value (H_better); 4 adaptive filtering.
  • the single carrier equalizer in FIG. 2 can select any method to complete the equalization by using the known current frame signal (eqin) and the channel estimation (H_flt) of eqin, and obtain the decision value of the transmission sequence (slicer_out, the decision value includes The sequence code (pn) and the estimate of the transmitted sequence) and the signal-to-noise ratio (SNR) estimate (snr_esti) of the current frame signal eqin.
  • eqin is a time domain signal, which is a baseband signal converted to a signal after analog-to-digital conversion (ADC) after down-conversion, filtering, and sampling rate conversion, and the rate of the signal is a single symbol rate in the DTMB system.
  • ADC analog-to-digital conversion
  • Medium is 7.56MHz.
  • H_init is an input to the channel estimate. It is the frequency domain response of the channel. It can be obtained by time domain correlation (including linear correlation or cyclic correlation) or by frequency domain division.
  • the length of H_init is ftrlen.
  • loop correlation Take loop correlation as an example.
  • mod(a,b) is a for modulo b.
  • H1 is a time domain channel response obtained by cyclic correlation
  • Hcir is padded with zeros on the basis of h1 to obtain the frequency domain channel response H_init of fftlen.
  • H_len is the length of the loop correlation.
  • the length is 165,595,434 respectively.
  • corrlen is 255,595,512 respectively.
  • pnlen is the length of the PN, which is 420,595,945 respectively.
  • Gdlen is the length of the guard interval in pn.
  • pnAMP is the amplitude of PN and is a known value.
  • the inputs eqin, pn and snr_esti are known, and channel estimation can be done by frequency domain division. Specifically: fill pn to the length of fften, and then do fft, you can get X.
  • Y is the frequency domain signal after the input signal eqin is transformed by fft, and its length For fftlen.
  • noisePow is the estimated noise power
  • Cof is used to correct the energy loss of the entire division so that the resulting H_init is an unbiased MMSE estimate.
  • H_init is the estimated frequency domain channel response, and its length is fften.
  • This module is implemented after the equalizer. Its input is eqin, the decision value slicer_out of the equalizer output, and the signal-to-noise ratio estimate snr_esti.
  • the acquisition method of H_ref is very similar to the frequency domain division described above, except that pn is replaced by slicer_out.
  • the decision value slicer_out contains an estimate of the pn and the transmitted sequence. The specific formula is as follows:
  • H_better is another input of the adaptive filter. By filtering H_ref, you can get H_better. Where H_ref is the frequency domain response of the channel. Generally speaking, the length of the multipath channel does not exceed the length pnlen of the pn sequence, otherwise it will cause interference ISI between DATA. Therefore, the noise of H_ref can be filtered out by filtering to obtain H_better.
  • this function can be implemented by designing a filter.
  • the bandwidth of the designed filter is pnlen; it can also be implemented by the fft transform method, and the fft transform is essentially a filter.
  • the method of fft transform filtering is as follows:
  • the length of H_better and h2 are both fftlen.
  • H_init, H_better and H_ref have been calculated, and the channel estimation H_flt can be obtained by adaptive filtering.
  • H_init and H_better are input signals
  • H_ref is a reference signal
  • the final output signal is generated by adaptive filtering.
  • LMS Least Mean Square
  • RLS Recursive Least Squares
  • the following is an example of LMS.
  • LMS Least Mean Square
  • RLS Recursive Least Squares
  • the usual LMS is divided into two parts, filtering (generating H_updt) and coefficient updating (updating init_coeff and bet_coeff). In the present embodiment, on the basis of these two parts, it is preferable to add a new part, and new data is generated (generating H_flt).
  • the array init_idx indicates which Symbol's H_init is used.
  • the array bet_idx indicates which Symbol H_better is used to participate in the filtering. Since H_better cannot be generated earlier than H_ref, the value in bet_idx must be less than zero.
  • the coefficients can be updated with the following six formulas.
  • the H_init_array in the following formula is an array of H_init of multiple Symbols.
  • H_init_array[n,m] means the channel response H_init[m] of the subcarrier m generated by the signal of the nth Symbol.
  • H_better_array and H_better can be formed.
  • H_updt[i] H_updt1[i]+H_updt2[i];
  • Stepsize is the step size of the coefficient update, which can be set a priori as needed.
  • the latest init_coeff and bet_coeff can be utilized to generate the channel estimate H_flt for the next Symbol n+1 .
  • the formula for generating H_flt of subcarrier i is as follows:
  • H_flt[i] H_flt1[i]-H_flt2[i]
  • the present invention provides a single carrier channel estimation method, which uses channel estimation and input current frame signal to obtain a sequence code and a signal to noise ratio of a current frame signal by equalization, and obtains a reference channel H_ref and preferably.
  • Channel H_better through the correlator, obtain the initial channel estimate H_init; then use the initial channel estimate H_init and the preferred channel H_better, use the current H_ref as a reference value, perform adaptive channel estimation, update the coefficient, and obtain the channel estimate of the next frame.
  • the technical solution of the present invention realizes that the channel estimation can be calculated for any channel environment (whether it is stationary or a channel environment moving at different speeds), thereby improving the performance of the receiver.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Power Engineering (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
  • Noise Elimination (AREA)

Abstract

本发明涉及信道编码和调制技术领域，尤其涉及一种单载波信道估计方法，包括：对输入的当前帧信号以及所述当前帧信号所在信道进行载波均衡，得到所述当前帧信号的序列码，发送序列判决值和信噪比估计值；根据所述当前帧信号、所述序列码和所述信噪比估计值，计算得到初始信道估计值；根据所述当前帧信号、所述发送序列判决值和所述信噪比估计值，计算得到参考信道；对所述参考信道进行滤波，得到滤波后的信道值；以所述参考信道为参照值，对所述初始信道估计值和所述滤波后的信道值进行自适应滤波，得到下一帧信号的信道估计值，实现了对于任何信道环境，均能计算得到最佳的信道估计。

Description

一种单载波信道估计方法 技术领域
本发明涉及信道编码和调制技术领域，尤其涉及一种单载波信道估计方法。
背景技术
在信道编码和调制技术领域，对信道的估计方法是将输入信号通过相关器，或者通过频域除法得到初始的信道估计；之后对多个帧的初始的信道估计，用固定系数的滤波器进行滤波，得到信道估计；对重构的信道，用阿尔法(alpha)滤波的方式得到信道估计。
这种信道估计方法所得到的初始的信道估计性能不好；且由于固定系数滤波，不能得到最佳的信道估计，因为信道的特性是未知的，滤波器的带宽由多普勒决定，而多普勒是未知的，导致估计得到的最佳信道性能不佳。
发明内容
鉴于上述技术问题，本发明旨在提供一种新型的信道估计方法，通过信道重构，得到较好的过去的初始信道估计；通过相关器或者频域除法，可以得到未来的初始信道估计；对初始信道估计做自适应滤波，得到最佳的信道估计。
本发明解决上述技术问题的主要技术方案为：
一种单载波信道估计方法，其特征在于，包括：
对输入的当前帧信号以及所述当前帧信号所在信道进行载波均衡，得到所述当前帧信号的序列码，发送序列判决值和信噪比估计值；
根据所述当前帧信号、所述序列码和所述信噪比估计值，计算得到初始信道估计值；
根据所述当前帧信号、所述发送序列判决值和所述信噪比估计值，计算得到参考信道；
对所述参考信道进行滤波，得到滤波后的信道值；
以所述参考信道为参照值，对所述初始信道估计值和所述滤波后的信道值进行自适应滤波，以得到下一帧信号的信道估计值。
优选的，上述的单载波信道估计方法中，利用单载波均衡器对所述当前帧信号以及所述当前帧信号所在信道进行载波均衡。
优选的，上述的单载波信道估计方法中，所述初始信道估计值为信道的频域响应，通过对所述当前帧信号、所述序列码和所述信噪比估计值进行时域相关算法或频域除法计算得到。
优选的，上述的单载波信道估计方法中，所述时域相关算法包括线性相关算法或循环相关算法。
优选的，上述的单载波信道估计方法中，所述发送序列判决值包括序列码和发送序列估计值。
优选的，上述的单载波信道估计方法中，所述参考信道为信道的频域响应，通过滤波器对所述参考信道进行滤波以得到所述滤波后的 信道值，且所述滤波器的带宽等于所述序列码的长度。
优选的，上述的单载波信道估计方法中，通过傅里叶变换对所述参考信道进行滤波以得到所述滤波后的信道值。
优选的，上述的单载波信道估计方法中，所述自适应滤波包括：
以所述参考信道为参照值，对所述初始信道估计值和所述滤波后的信道值进行滤波，以生成子滤波；
对所述子滤波进行系数更新，以生成最新初始信道估计值和最新滤波后的信道值；
利用所述最新初始信道估计值和所述最新滤波后的信道值，计算生成所述下一帧信号的信道估计值。
优选的，上述的单载波信道估计方法中，所述当前帧信号为时域信号，并且
所述时域信号是对经模数转换后的信号进行下变频、滤波、采样率变换之后转换成的基带信号，且所述基带信号的速率为单倍的符号率。
优选的，上述的单载波信道估计方法中，每一帧信号包括序列码和连接在所述序列码尾部的单载波数据，且每一帧信号的所述序列码一致，以构成循环相关。
上述技术方案具有如下优点或有益效果：本发明提供的一种单载波信道估计方法，用信道估计和输入的当前帧信号，通过均衡，得到当前帧信号的序列码和信噪比，并且得到参考信道H_ref和滤波后的信道值H_better(H_better也可称为较佳信道估计值，下文中均将 H_better称作较佳信道估计值)；通过相关器，得到初始信道估计H_init；然后把初始信道估计H_init和滤波后的信道值H_better，用当前的H_ref作参考值，进行自适应信道估计，更新系数，得到下一帧的信道估计。本发明的技术方案实现了对于任何信道环境(不管是静止还是以不同速度移动的信道环境)，均能计算得到最佳的信道估计，从而提高接收机的性能。
附图说明
参考所附附图，以更加充分的描述本发明的实施例。然而，所附附图仅用于说明和阐述，并不构成对本发明范围的限制。
图1是本发明DTMB系统的帧结构；
图2是本发明单载波的信道估计和均衡；
图3是本发明自适应滤波的示意图。
具体实施方式
在下文的描述中，给出了大量具体的细节以便提供对本发明更为彻底的理解。当然除了这些详细描述外，本发明还可以具有其他实施方式。
本发明的单载波信道估计方法，对输入的当前帧信号eqin以及eqin所在信道H_flt进行载波均衡，得到eqin的pn码，发送序列判决值slicer_out，和信噪比估计值snr_esti；然后根据eqin，pn和snr_esti，计算得到初始信道估计值H_init；根据eqin，slicer_out和snr_esti， 计算得到参考信道H_ref；对参考信道H_ref进行滤波，得到滤波后的信道值H_better；再以参考信道H_ref为参照值，对初始信道估计值H_init和滤波后的信道值H_better进行自适应信道估计，更新系数，得到下一帧信号的信道估计值。
下面结合具体的实施例以及附图详细阐述本发明的单载波信道估计方法。
如图1DTMB(数字电视地面广播传输系统帧结构、信道编码和调制)系统的帧结构所示，本实施例的每一帧信号(以下简称Symbol)包括帧头和帧体，帧头为序列码(PN)，帧体为数据块(DATA)，各个帧信号之间的连接方式为：前一帧信号的DATA与后一帧信号的PN连接，例如图1中前一帧信号Symbol0的DATA0与后一帧信号Symbol1的PN1连接，SymbolM-1的DATAM-1与SymbolM的PNM连接，以此类推，此处不进行赘述。
本实施例的序列码PN有三种可能，分别为PN420，PN595和PN945，对应的PN的长度分别为420，595和945；各个Symbol的PN可能是不变的，也有可能是变化的，本实施例只描述不变的情况。本实施例的DATA有两种可能，分别对应着单载波和多载波，其长度为3780。如果DATA是单载波的话，各个Symbol中的PN都是一致的，它们构成了循环相关。本文中所有涉及到的傅里叶变换(FFT)和傅里叶逆变换(IFFT)运算的长度fftlen，都是PN的长度+3780。
图2是单载波的信道估计，分为四个部分。分别为：①初始信道估计值(H_init)的获取；②参考信道(H_ref)的获取；③较佳信道 估计值(H_better)的获取；④自适应滤波。
图2中的单载波均衡器，通过已知的当前帧信号(eqin)和eqin的信道估计(H_flt)，可以选择任意的方法来完成均衡，得到发送序列的判决值(slicer_out，该判决值包含了序列码(pn)和发送序列的估计)以及当前帧信号eqin的信噪比(SNR)估计(snr_esti)。其中，eqin是一时域信号，是对模数变换(ADC)之后的信号，进行下变频、滤波、采样率变换之后转换成的基带信号，该信号的速率为单倍的符号率，在DTMB系统中，为7.56MHz。
下面对上述四个部分进行详细阐述。
①初始信道估计值(H_init)的获取
H_init是信道估计的一个输入，它是信道的频域响应，可以用时域相关(包括线性相关，或者循环相关)的算法得到，也可以用频域除法得到。
1.1时域相关
对单个symbol来说，H_init的长度为fftlen。以循环相关为例，
Figure PCTCN2017080944-appb-000001
Figure PCTCN2017080944-appb-000002
H_init＝fft(hcir))
其中，mod(a，b)为a对b求模。
h1为循环相关得到的时域信道响应；
hcir在h1的基础上补零，以得到fftlen的频域信道响应H_init。
h_len是循环相关的长度，对于PN420，PN595和PN945，其长度分别是165，595，434。
对于PN420，PN595和PN945，corrlen分别是255，595，512。
对于PN420，PN595和PN945，pnlen是PN的长度，分别是420，595，945。
gdlen为pn中保护间隔的长度，对于PN420，PN595和PN945，分别是165，0，434。
当ck0的pn码是0时，pn[ck0]＝1+1i
当ck0的pn码是1时，pn[ck0]＝﹣1－1i
pnAMP是PN的幅度，是个已知值。
1.1频域除法
已知输入eqin，pn和snr_esti，可以通过频域的除法来完成信道估计。具体为：把pn补零到fftlen的长度，然后做fft，可以得到X。如下公式：
Y＝fft(eqin)
Figure PCTCN2017080944-appb-000003
Figure PCTCN2017080944-appb-000004
Figure PCTCN2017080944-appb-000005
Figure PCTCN2017080944-appb-000006
其中，Y为输入信号eqin经过fft变换之后的频域信号，其长度 为fftlen。
noisePow为估计得到的噪声功率；
cof用于修正整个除法的能量损失，使得得到的H_init是一个无偏的MMSE估计。
H_init为估计得到的频域信道响应，其长度为fftlen。
②参考信道(H_ref)的获取
这个模块是在均衡器后面实现的，它的输入是eqin，均衡器输出的判决值slicer_out以及信噪比估计snr_esti。H_ref的获取方法与上述的频域除法非常类似，只是把pn换成了slicer_out。该判决值slicer_out包含了pn和发送序列的估计。具体公式如下：
X＝fft(slicer_out)
Y＝fft(eqin)
Figure PCTCN2017080944-appb-000007
Figure PCTCN2017080944-appb-000008
Figure PCTCN2017080944-appb-000009
Figure PCTCN2017080944-appb-000010
需要注意的是，本实施例仅仅列举了一种获得H_ref的方法，于实际运用中还可使用其他类似算法获得H_ref，这并不构成对本发明的限制。
③较佳信道估计值(H_better)的获取
H_better是自适应滤波器的另外一个输入，对H_ref做滤波，就可以得到H_better。其中，H_ref是信道的频域响应，一般来说，多径信道的长度不会超过pn序列的长度pnlen，不然会引起DATA之间的干扰ISI。因此，可以通过滤波，把H_ref的噪声滤除，以得到H_better。
具体的，可以通过设计一个滤波器来实现这个功能，所设计的滤波器的带宽为pnlen；也可以通过fft变换的方法来实现，fft变换本质上也是一个滤波器。其中，fft变换滤波的方法如下：
h1＝ifft(H_ref)
Figure PCTCN2017080944-appb-000011
H_better＝fft(h2)
其中，H_better和h2的长度都是fftlen。
④自适应滤波
到目前为止，已经计算得到H_init，H_better和H_ref，通过自适应滤波的方法，可以得到信道估计H_flt。如图3所示，H_init和H_better为输入信号，H_ref为参考信号，通过自适应滤波来生成最后的输出信号。自适应滤波有多种方式，例如最小均方(LMS)算法、递归最小二乘(RLS)算法等等，下面以LMS为例来说明。通常的LMS分为两个部分，滤波(生成H_updt)和系数更新(更新init_coeff和bet_coeff)。在本实施例中，在这两个部分的基础上，优选的还增加一个新的部分，新数据生成(生成H_flt)。
在下面的公式中，数组init_idx指示了采用哪些Symbol的H_init 来参与滤波，其取值范围没有限制；数组bet_idx指示了采用哪些Symbol的H_better来参与滤波。因为H_better不可比H_ref生成得早，因此bet_idx中的值必然小于0。
对于所有的子载波i(i的取值为0<＝i<fftlen-1)，可以用如下六个公式进行系数的更新。下面公式中的H_init_array为多个Symbol的H_init所构成的数组。H_init_array[n,m]的意思是第n个Symbol的信号所产生的子载波m的信道响应H_init[m]。同理也可以构成H_better_array和H_better的对应关系。
假设对Symboln进行完均衡之后，得到第n个Symbol的H_better和H_ref，然后开始进行系数更新，如下：
Figure PCTCN2017080944-appb-000012
Figure PCTCN2017080944-appb-000013
H_updt[i]＝H_updt1[i]+H_updt2[i]；
err[i]＝H_ref[i]－H_updt[i]；
对所有的init_coeff进行系数更新：
init_coeff[init_idx]
＝init_coeff[init_idx]+stepsize*err[i]
*H_init_array[init_idx+n，i]*
对所有的bet_coeff进行系数更新：
bet_coeff[bet_idx]
＝bet_coeff[bet_idx]+stepsize*err[i]
*H_better_array[bet_idx+n，i]*
Stepsize为系数更新的步长，可以根据需要先验设定。
在对所有的子载波都更新了系数之后，可以利用最新的init_coeff和bet_coeff来生成下一个Symboln+1的信道估计H_flt。子载波i的H_flt的生成公式如下:
Figure PCTCN2017080944-appb-000014
Figure PCTCN2017080944-appb-000015
H_flt[i]＝H_flt1[i]－H_flt2[i]
综上所述，本发明提供的一种单载波信道估计方法，用信道估计和输入的当前帧信号，通过均衡，得到当前帧信号的序列码和信噪比，并且得到参考信道H_ref和较佳信道H_better；通过相关器，得到初始信道估计H_init；然后把初始信道估计H_init和较佳信道H_better，用当前的H_ref作参考值，进行自适应信道估计，更新系数，得到下一帧的信道估计。本发明的技术方案实现了对于任何信道环境(不管是静止还是以不同速度移动的信道环境)，均能计算得到最佳的信道估计，从而提高接收机的性能。
对于本领域的技术人员而言，阅读上述说明后，各种变化和修正无疑将显而易见。因此，所附的权利要求书应看作是涵盖本发明的真实意图和范围的全部变化和修正。在权利要求书范围内任何和所有等价的范围与内容，都应认为仍属本发明的意图和范围内。

Claims (10)

  1. 一种单载波信道估计方法，其特征在于，包括：
    对输入的当前帧信号以及所述当前帧信号所在信道进行载波均衡，得到所述当前帧信号的序列码、发送序列判决值和信噪比估计值；
    根据所述当前帧信号、所述序列码和所述信噪比估计值，计算得到初始信道估计值；
    根据所述当前帧信号、所述发送序列判决值和所述信噪比估计值，计算得到参考信道；
    对所述参考信道进行滤波，得到滤波后的信道值；
    以所述参考信道为参照值，对所述初始信道估计值和所述滤波后的信道值进行自适应滤波，以得到下一帧信号的信道估计值。
  2. 如权利要求1所述的单载波信道估计方法，其特征在于，利用单载波均衡器对所述当前帧信号以及所述当前帧信号所在信道进行载波均衡。
  3. 如权利要求1所述的单载波信道估计方法，其特征在于，所述初始信道估计值为信道的频域响应，通过对所述当前帧信号、所述序列码和所述信噪比估计值进行时域相关算法或频域除法计算得到。
  4. 如权利要求3所述的单载波信道估计方法，其特征在于，所述时域相关算法包括线性相关算法或循环相关算法。
  5. 如权利要求1所述的单载波信道估计方法，其特征在于，所述发送序列判决值包括序列码和发送序列估计值。
  6. 如权利要求1所述的单载波信道估计方法，其特征在于，所述参考信道为信道的频域响应，通过滤波器对所述参考信道进行滤波以得到所述滤波后的信道值，且所述滤波器的带宽等于所述序列码的长度。
  7. 如权利要求1所述的单载波信道估计方法，其特征在于，通过傅里叶变换对所述参考信道进行滤波以得到所述滤波后的信道值。
  8. 如权利要求1所述的单载波信道估计方法，其特征在于，所述自适应滤波包括：
    以所述参考信道为参照值，对所述初始信道估计值和所述滤波后的信道值进行滤波，以生成子滤波；
    对所述子滤波进行系数更新，以生成最新初始信道估计值和最新滤波后的信道值；
    利用所述最新初始信道估计值和所述最新滤波后的信道值，计算生成所述下一帧信号的信道估计值。
  9. 如权利要求1所述的单载波信道估计方法，其特征在于，所述当前帧信号为时域信号，并且
    所述时域信号是对经模数转换后的信号进行下变频、滤波、采样率变换之后转换成的基带信号，且所述基带信号的速率为单倍的符号率。
  10. 如权利要求1所述的单载波信道估计方法，其特征在于，每一帧信号包括序列码和连接在所述序列码尾部的单载波数据，且每一帧信号的所述序列码一致，以构成循环相关。
PCT/CN2017/080944 2016-07-01 2017-04-18 一种单载波信道估计方法 WO2018000914A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US15/560,139 US10270623B1 (en) 2016-07-01 2017-04-18 Single-carrier channel estimation method

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610507985.6A CN106059967B (zh) 2016-07-01 2016-07-01 一种单载波信道估计方法
CN201610507985.6 2016-07-01

Publications (1)

Publication Number Publication Date
WO2018000914A1 true WO2018000914A1 (zh) 2018-01-04

Family

ID=57200548

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/080944 WO2018000914A1 (zh) 2016-07-01 2017-04-18 一种单载波信道估计方法

Country Status (3)

Country Link
US (1) US10270623B1 (zh)
CN (1) CN106059967B (zh)
WO (1) WO2018000914A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106059967B (zh) * 2016-07-01 2020-06-05 晶晨半导体(上海)股份有限公司 一种单载波信道估计方法
CN109412984B (zh) * 2018-11-15 2021-01-01 中国电子科技集团公司第三十八研究所 一种基于Aitken加速法的多天线场景下盲信噪比估算方法
CN115865579B (zh) * 2023-03-02 2023-05-16 武汉能钠智能装备技术股份有限公司 一种基于深度学习的数字调制信号的滚降系数盲估计方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1684450A (zh) * 2004-04-16 2005-10-19 株式会社Ntt都科摩 接收机、发送机和无线通信系统及信道估计方法
CN101588327A (zh) * 2008-05-19 2009-11-25 清华大学 用于ofdm系统的数据指导信道估计方法及其实现装置
CN101692666A (zh) * 2009-10-13 2010-04-07 深圳市力合微电子有限公司 一种滤除时域信道响应噪声的方法
CN104394105A (zh) * 2014-11-25 2015-03-04 清华大学 Tds-ofdm信道估计均衡方法及系统
CN106027431A (zh) * 2016-05-06 2016-10-12 晶晨半导体(上海)有限公司 一种信道估计方法
CN106059967A (zh) * 2016-07-01 2016-10-26 晶晨半导体(上海)有限公司 一种单载波信道估计方法

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8345733B2 (en) * 2005-09-13 2013-01-01 At&T Intellectual Property I, Lp Method and apparatus for equalizing signals
CN102045285B (zh) * 2009-10-14 2013-09-11 华为技术有限公司 信道估计方法、装置以及通信系统
CN103166878B (zh) * 2011-12-08 2016-02-03 联芯科技有限公司 一种信道估计的方法和装置
WO2013115666A1 (en) * 2012-01-30 2013-08-08 Intel Corporation Multi-mode frequency domain equalization with adaptation to varying channel conditions
CN103812806B (zh) * 2014-02-24 2017-10-10 北京大学 一种基于时域信道估计的信道均衡方法和系统
CN105323201B (zh) * 2015-03-23 2020-10-16 济南山芯电子有限公司 一种应用于多载波接收机的抗多径干扰系统及方法
CN104967581B (zh) * 2015-07-02 2018-08-24 山东航天电子技术研究所 低能量弥散多径信道下单载波频域均衡的信道估计方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1684450A (zh) * 2004-04-16 2005-10-19 株式会社Ntt都科摩 接收机、发送机和无线通信系统及信道估计方法
CN101588327A (zh) * 2008-05-19 2009-11-25 清华大学 用于ofdm系统的数据指导信道估计方法及其实现装置
CN101692666A (zh) * 2009-10-13 2010-04-07 深圳市力合微电子有限公司 一种滤除时域信道响应噪声的方法
CN104394105A (zh) * 2014-11-25 2015-03-04 清华大学 Tds-ofdm信道估计均衡方法及系统
CN106027431A (zh) * 2016-05-06 2016-10-12 晶晨半导体(上海)有限公司 一种信道估计方法
CN106059967A (zh) * 2016-07-01 2016-10-26 晶晨半导体(上海)有限公司 一种单载波信道估计方法

Also Published As

Publication number Publication date
US10270623B1 (en) 2019-04-23
US20190116067A1 (en) 2019-04-18
CN106059967A (zh) 2016-10-26
CN106059967B (zh) 2020-06-05

Similar Documents

Publication Publication Date Title
Zakharov et al. OFDM transmission without guard interval in fast-varying underwater acoustic channels
US5285474A (en) Method for equalizing a multicarrier signal in a multicarrier communication system
US6996199B2 (en) Approach for processing data received from a communications channel to reduce noise power and optimize impulse response length to reduce inter-symbol interference and inter-channel interference
TWI382672B (zh) 累進平行干擾消除器及其方法與其接收機
US10103909B2 (en) Channel estimation method
IL145090A (en) Approach for processing data received from a communication channel in finite precision arithmetic applications
US7224725B2 (en) Method for determining coefficients of an equalizer and apparatus for determining the same
US7031379B2 (en) Time domain equalizer for DMT modulation
CN108712353A (zh) 软迭代信道估计方法
JP3909123B2 (ja) 搬送波を使わない振幅位相変調用周波数領域復調器
WO2018000914A1 (zh) 一种单载波信道估计方法
CN112087266B (zh) 一种基于emd-wfft时变宽带多普勒补偿方法
CN108650199B (zh) 无线通信接收端iq不平衡和信道联合估计方法及装置
CN112953653A (zh) 一种单载波多用户水声通信方法
WO2011044846A1 (zh) 信道估计方法、装置以及通信系统
CN109361631B (zh) 稀疏度未知的水声正交频分复用信道估计方法及装置
CN113381954B (zh) 一种基于广义近似消息传递的频域均衡方法
Zakharov et al. Doppler effect compensation for cyclic-prefix-free OFDM signals in fast-varying underwater acoustic channel
Alaghbari et al. Robust precoder for mitigating inter-symbol and inter-carrier interferences in coherent optical FBMC/OQAM
CN113411277B (zh) 一种多载波超奈奎斯特系统接收机的构建方法
Ogundile et al. Improved reliability information for OFDM systems on time-varying frequency-selective fading channels
Anupriya et al. Channel estimation for MIMO-OFDM systems using superimposed training In VLSI
Al-Shuwaili et al. Ball’s-Based Adaptive Channel Estimation Scheme Using RLS Family-Types Algorithms
Schlamann et al. Lattice filter-based frequency-domain equalizer with multiple frequency offset compensation for MIMO SDMA OFDM systems
CN117640326A (zh) 基于载波频偏估计补偿及循环前缀重构的cp-ufmc迭代接收方法

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: 17818911

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: 17818911

Country of ref document: EP

Kind code of ref document: A1