《電子技術(shù)應(yīng)用》
您所在的位置:首頁(yè) > 通信與網(wǎng)絡(luò) > 設(shè)計(jì)應(yīng)用 > 基于訓(xùn)練序列的OFDM頻率同步改進(jìn)算法
基于訓(xùn)練序列的OFDM頻率同步改進(jìn)算法
2017年電子技術(shù)應(yīng)用第3期
王思撥,,馬社祥,,孟 鑫,王俊峰
天津理工大學(xué) 計(jì)算機(jī)與通信工程學(xué)院,,天津300384
摘要: 針對(duì)現(xiàn)有正交頻分復(fù)用(OFDM)通信系統(tǒng)中頻率同步算法存在的同步精度不高,、估計(jì)范圍較小等問(wèn)題,設(shè)計(jì)了一種訓(xùn)練序列結(jié)構(gòu),,并給出了相應(yīng)的估計(jì)函數(shù),,提出了基于訓(xùn)練序列的頻率同步改進(jìn)算法。頻率同步改進(jìn)算法是在準(zhǔn)確的符號(hào)定時(shí)之后,,根據(jù)時(shí)域訓(xùn)練序列結(jié)構(gòu)的共軛對(duì)稱性,,利用新的度量函數(shù)求出小數(shù)倍頻偏;根據(jù)接收端頻域訓(xùn)練序列的循環(huán)移位估計(jì)整數(shù)倍頻偏,。理論分析和仿真結(jié)果表明,,改進(jìn)算法頻偏估計(jì)的均方誤差較低,擴(kuò)大了小數(shù)倍頻偏估計(jì)范圍,,提高了整數(shù)倍頻偏的正確檢測(cè)概率,。
中圖分類號(hào): TN919.3
文獻(xiàn)標(biāo)識(shí)碼: A
DOI:10.16157/j.issn.0258-7998.2017.03.026
中文引用格式: 王思撥,馬社祥,,孟鑫,,等. 基于訓(xùn)練序列的OFDM頻率同步改進(jìn)算法[J].電子技術(shù)應(yīng)用,2017,,43(3):104-107.
英文引用格式: Wang Sibo,,Ma Shexiang,Meng Xin,,et al. Improved frequency synchronization method for OFDM systems based on training sequence[J].Application of Electronic Technique,,2017,43(3):104-107.
Improved frequency synchronization method for OFDM systems based on training sequence
Wang Sibo,Ma Shexiang,,Meng Xin,,Wang Junfeng
School of Computer and Communication Engineering,Tianjin University of Technology,,Tianjin 300384,,China
Abstract: In this paper, a new training sequence is designed and an improved frequency synchronization method is proposed for OFDM systems, aim to overcome the low synchronization precision and the small estimation range. The proposed method is based on training sequence to estimate, and the corresponding carrier frequency offset(CFO) estimation function is presented. After accurate timing synchronization, the fractional frequency offset(FFO) can be obtained by using the new metric function, according to the conjugate characteristic of the training sequence structure in the time domain. And the integer frequency offset(IFO) can be estimated by the cyclic shift of the training sequence of the received signal in the frequency domain. Theoretical analysis and simulation results show that the mean square error(MSE) of the frequency offset estimation is relatively low, and the range of fractional frequency offset estimation is expanded, the correct detection probability of the integer frequency offset is higher than other methods.
Key words : OFDM; frequency synchronization; training sequence; MSE; correct detection probability

0 引言

    2013年5月,在國(guó)際電信聯(lián)盟無(wú)線電通信工作組(ITU-R Working Party 5B)會(huì)議上,,多國(guó)政府和國(guó)際航標(biāo)協(xié)會(huì)(IALA)提出了VDES(VHF Data Exchange System),。VDES是針對(duì)AIS加強(qiáng)和升級(jí)的系統(tǒng),由AIS(Automatic Identification System),、ASM(Application Specific Messages)以及VDE(VHF Data Exchange) 所組成的集成系統(tǒng),。VDES允許多于一種編碼速率和調(diào)制方式,可以在系統(tǒng)規(guī)模和業(yè)務(wù)可用性上提供更多的靈活性,。VDE采用自適應(yīng)調(diào)制和編碼來(lái)使頻譜效率和吞吐量最大,。2015年10月,國(guó)際電信聯(lián)盟無(wú)線電通信第5研究組建議VDES采用ITU-RM.2092-0建議書(shū)[1],。在VDE-SAT上/下行鏈路50 kHz專用信道中,,可以采用OFDM技術(shù)。在VDES中的VDE-SAT上/下行鏈路通信時(shí),,衛(wèi)星通信的環(huán)境給OFDM技術(shù)的使用造成了同步問(wèn)題,。

    OFDM同步技術(shù)主要是對(duì)OFDM系統(tǒng)中符號(hào)定時(shí)偏差(Symbol Timing Offset,STO)和載波頻率偏差(Carrier Frequency Offset,,CFO)進(jìn)行估計(jì),,同步方法可分為兩類:(1)非數(shù)據(jù)輔助方法[2],主要是利用循環(huán)前綴(Cyclic Prefix,,CP)的盲估計(jì)算法,;(2)數(shù)據(jù)輔助的估計(jì)方法[3-7]。數(shù)據(jù)輔助類算法中最經(jīng)典的是由SCHMIDL M等人[2]提出的同步算法(以下簡(jiǎn)稱S&C算法),,但循環(huán)前綴導(dǎo)致這種算法存在定時(shí)平臺(tái),,造成定時(shí)模糊現(xiàn)象。MINN H及PARK B等人[4,,5]對(duì)S&C算法進(jìn)行了改進(jìn),,雖然有效地防止了定時(shí)估計(jì)的峰值平臺(tái)問(wèn)題,但是在糾正小數(shù)倍載波偏差時(shí)誤差較大,,且要求訓(xùn)練序列的偶數(shù)位是等比數(shù)列,,這樣才能糾正整數(shù)倍載波偏差,算法復(fù)雜度高[8,,9],。文獻(xiàn)[6]中雖然利用CAZAC(恒包絡(luò)零自相關(guān))序列設(shè)計(jì)的訓(xùn)練符號(hào)可以準(zhǔn)確定時(shí),,但在定時(shí)同步時(shí)都使用了本地序列,存在計(jì)算量大的缺點(diǎn),,并且載波頻偏估計(jì)的性能沒(méi)有得到改進(jìn),。Guo Yi等人在文獻(xiàn)[7]中提出使用實(shí)訓(xùn)練序列,利用時(shí)域訓(xùn)練序列的共軛特性,,準(zhǔn)確定時(shí),。但是在對(duì)小數(shù)倍頻偏估計(jì)時(shí),由于訓(xùn)練序列中沒(méi)有重復(fù)結(jié)構(gòu),,只能用基于循環(huán)前綴的ML算法進(jìn)行估計(jì),,估計(jì)范圍較小,低信噪比下估計(jì)性能較差,,整數(shù)倍頻偏估計(jì)算法復(fù)雜。本文利用Guo Yi的符號(hào)定時(shí)方法準(zhǔn)確定時(shí)后,,對(duì)頻率同步算法改進(jìn),,根據(jù)時(shí)域訓(xùn)練序列結(jié)構(gòu)的共軛對(duì)稱性,利用新的度量函數(shù)求出小數(shù)倍頻偏,;根據(jù)接收端頻域訓(xùn)練序列的循環(huán)移位估計(jì)整數(shù)倍頻偏,。本文頻率同步改進(jìn)算法在不降低估計(jì)精度的前提下擴(kuò)大了小數(shù)倍頻偏估計(jì)的范圍,提高了整數(shù)倍頻偏的正確檢測(cè)概率,。

1 OFDM系統(tǒng)模型

    定義發(fā)送信號(hào)為x(n),,接收信號(hào)為y(n),帶有循環(huán)前綴的發(fā)送端信號(hào)可以表示為:

tx4-gs1-2.gif

2 訓(xùn)練序列結(jié)構(gòu)的設(shè)計(jì)

    在發(fā)送端,,在頻域設(shè)計(jì)新的訓(xùn)練序列,,偶數(shù)位發(fā)送:

    tx4-gs3.gif

    奇數(shù)位發(fā)送:

tx4-gs4-6.gif

tx4-t1.gif

    在接收端,時(shí)域接收信號(hào)可以表示為:

tx4-gs7-12.gif

    可根據(jù)式(12)估計(jì)出準(zhǔn)確符號(hào)定時(shí)位置,,如圖2所示,。估計(jì)出符號(hào)定時(shí)偏差后,對(duì)信號(hào)進(jìn)行補(bǔ)償,,實(shí)現(xiàn)符號(hào)定時(shí)同步,。

tx4-t2.gif

3 改進(jìn)的頻偏估計(jì)算法

    經(jīng)符號(hào)同步算法估計(jì)出符號(hào)起始點(diǎn),在時(shí)域上進(jìn)行小數(shù)倍載波頻偏估計(jì),;補(bǔ)償后,,將信號(hào)從時(shí)域變換至頻域,在頻域進(jìn)行整數(shù)倍載波頻偏估計(jì),。

3.1小數(shù)倍載波頻偏估計(jì)

    在接收端準(zhǔn)確定時(shí)后,,在時(shí)域上對(duì)小數(shù)倍頻偏估計(jì),時(shí)域接收信號(hào)可以表示為: 

tx4-gs13.gif

tx4-t3.gif

tx4-gs14-19.gif

3.2 整數(shù)倍載波頻偏估計(jì)

    當(dāng)頻偏ε只存在小數(shù)倍頻偏時(shí),,可用上述方法準(zhǔn)確估計(jì),;當(dāng)頻偏中存在整數(shù)倍頻偏時(shí),,經(jīng)過(guò)小數(shù)倍頻偏估計(jì)且補(bǔ)償后,接收信號(hào)中僅存在偶數(shù)倍頻偏[3],。在接收端補(bǔ)償小數(shù)倍頻偏后,,只存在整數(shù)倍頻偏εi,且εi為偶數(shù),,時(shí)域接收信號(hào)可以表示為:tx4-gs20-s1.gif利用FFT對(duì)接收信號(hào)解調(diào)后,,Y(k)=X(k-εi),Y(k)為X(k)的循環(huán)移位,,得到Y(jié)(k)的偶數(shù)位序列:

     tx4-gs20.gif

    通過(guò)求解碼后的訓(xùn)練序列能量值最大處,,定位原偶數(shù)位訓(xùn)練序列的非零訓(xùn)練序列起始位置,該位置的移位個(gè)數(shù)即為估計(jì)的整數(shù)倍頻偏:

tx4-gs21.gif

4 仿真結(jié)果及分析

    在理論分析的基礎(chǔ)上進(jìn)行仿真實(shí)驗(yàn),,對(duì)本文算法的性能進(jìn)行測(cè)試,。在仿真過(guò)程O(píng)FDM系統(tǒng)的主要參數(shù)為:FFT/IFFT的變換長(zhǎng)度為128,循環(huán)前綴長(zhǎng)度為32,,子載波采用16QAM調(diào)制方式,,仿真次數(shù)為100次。

    圖4為各算法在不同的CFO下MSE的性能對(duì)比,。如圖4(a)所示,,當(dāng)ε=0.3時(shí),本文算法MSE性能優(yōu)于MINN H算法和PARK B算法,,在較低信噪比下Guo Yi算法MSE性能明顯降低,,而本文能保持良好的MSE性能。在圖4(b)中,,當(dāng)ε=2.3時(shí),,實(shí)際頻偏值超出了MINN H算法和PARK B算法的估計(jì)范圍, MINN H算法和PARK B算法不能用于頻偏估計(jì),本文算法比Guo Yi算法有更好的估計(jì)性能,。

tx4-t4.gif

    圖5中比較了本文算法與Guo Yi算法的整數(shù)倍頻偏估計(jì)性能,。從圖5中可以看出,本文算法在信噪比較低時(shí)整數(shù)倍頻偏正確檢測(cè)的概率較高,,具有更好的同步性能,。

tx4-t5.gif

tx4-5-s1.gif

tx4-t6.gif

5 結(jié)論

    本文在符號(hào)定時(shí)準(zhǔn)確的基礎(chǔ)上,對(duì)OFDM系統(tǒng)的載波頻率偏差進(jìn)行估計(jì),,提出了一種基于新型訓(xùn)練序列結(jié)構(gòu)的OFDM頻率同步的改進(jìn)算法,。根據(jù)接收端的時(shí)域接收信號(hào)中存在部分共軛特性進(jìn)行小數(shù)倍頻偏估計(jì),在頻域根據(jù)訓(xùn)練符號(hào)的移位估計(jì)出整數(shù)倍的頻偏,。理論推導(dǎo)和仿真結(jié)果都表明,,本文算法擴(kuò)大了小數(shù)倍頻偏的估計(jì)范圍,小數(shù)倍頻偏估計(jì)在低信噪比下有更好的MSE性能,,提高了整數(shù)倍頻偏的正確檢測(cè)概率,,具有很好的估計(jì)性能,。

參考文獻(xiàn)

[1] International Telecommunication Union.Technical characteristics for a VHF data exchange system in the VHF maritime mobile band[S].Recommendation ITU-RM.2092-0,October 2015.

[2] VAN DE BEEK J J,,SANDELL M,,BRJESSON P O.ML estimation of timing and frequency offset in OFDM systems[J].IEEE Trans on.Signal Processing,1997,,45(7):1800-1805.

[3] SCHMIDL T M,,COX D C.Robust frequency and timing synchronization for OFDM[J].IEEE Trans.Commun.,1997,,45(12):1613-1621.

[4] MINN H,,ZENG M,BHARGAVA V K.On timing offset estimation for OFDM systems[J].IEEE Commun.Lett.2000,,4(7):242-244.

[5] PARK B,,CHOEN H,KO E,,et al.A novel timing estimation method for OFDM systems[J].IEEE Commun.Lett.,,2003,7(5):53-55.

[6] 嚴(yán)春林,,李少謙,唐友喜,,等.利用CAZAC序列的OFDM頻率同步方法[J].電子與信息學(xué)報(bào),,2006,28(1):139-142.

[7] Guo Yi,,Liu Gang,,Ge Jianhua.A novel time and frequency synchronization scheme for OFDM systems[J].IEEE Trans on Consumer Electronics,2008,,54(2):321-325.

[8] CHO Y S,,KIM J,YANG W Y,,et al.MIMO-OFDM wireless communication with MATLAB[M].1st ed.John Wiley and Sons(Asia) PteLtd,,2010.

[9] NISHAD P K,SINGH P.Carrier frequency offset estimation in OFDM Systems[C].Proceedings of 2013 IEEE Conference on Information and Communication Technonlogies(ICT2013),2013:885-889.



作者信息:

王思撥,,馬社祥,,孟  鑫,王俊峰

(天津理工大學(xué) 計(jì)算機(jī)與通信工程學(xué)院,,天津300384)

此內(nèi)容為AET網(wǎng)站原創(chuàng),,未經(jīng)授權(quán)禁止轉(zhuǎn)載。