WO2022236969A1 - Signal detection method for underwater acoustic fbmc system - Google Patents

Signal detection method for underwater acoustic fbmc system Download PDF

Info

Publication number
WO2022236969A1
WO2022236969A1 PCT/CN2021/109272 CN2021109272W WO2022236969A1 WO 2022236969 A1 WO2022236969 A1 WO 2022236969A1 CN 2021109272 W CN2021109272 W CN 2021109272W WO 2022236969 A1 WO2022236969 A1 WO 2022236969A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
hfm
fbmc
symbol
channel
Prior art date
Application number
PCT/CN2021/109272
Other languages
French (fr)
Chinese (zh)
Inventor
瞿逢重
陆雪松
朱江
魏艳
陈鹰
涂星滨
吴叶舟
Original Assignee
浙江大学
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 浙江大学 filed Critical 浙江大学
Publication of WO2022236969A1 publication Critical patent/WO2022236969A1/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/26—Systems using multi-frequency codes
    • H04L27/2601—Multicarrier modulation systems
    • H04L27/2647—Arrangements specific to the receiver only
    • H04L27/2655—Synchronisation arrangements
    • H04L27/2657—Carrier synchronisation
    • H04L27/266—Fine or fractional frequency offset determination and synchronisation
    • 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

Definitions

  • the invention belongs to the field of multi-carrier communication, relates to an underwater acoustic multi-carrier signal detection method, in particular to a signal detection method for an underwater acoustic FBMC system.
  • Orthogonal Frequency Division Multiplexing In the existing multi-carrier underwater acoustic communication technology, Orthogonal Frequency Division Multiplexing (OFDM) technology is mostly used. Compared with the single-carrier underwater acoustic communication technology, OFDM can deal with it with lower receiver complexity. multipath channel conditions. However, in the OFDM system, since each sub-carrier needs to be orthogonal, it is very sensitive to inter-carrier interference. Intercarrier interference usually comes from Doppler shift and carrier frequency offset (CFO). The Doppler frequency shift is caused by the relative movement between the transceivers, and the CFO is mainly caused by the inconsistency of the crystal oscillator frequencies at the transceivers.
  • CFO carrier frequency offset
  • FBMC filter bank based multi-carrier
  • FBMC technology uses the idea of non-orthogonal sub-carriers to reduce the sensitivity of FBMC systems to inter-sub-carrier interference.
  • spectrum efficiency unlike OFDM, FBMC does not need to add a cyclic prefix or guard interval before each multi-carrier symbol, and the method adopted in the present invention only needs to insert a preamble before each frame for the underwater environment where spectrum resources are scarce. It only needs to perform channel estimation and frequency offset estimation, and each frame can contain multiple FBMC symbols to improve spectral efficiency.
  • the present invention aims at the problems in the FBMC signal detection technology, and proposes a compromise between the detection complexity and detection performance.
  • the signal utilization rate is further improved, that is, a FBMC signal detection method for an underwater acoustic channel environment is provided.
  • the advantages of this method include: 1) multiplexing the synchronization head pulse to realize signal synchronization and Doppler estimation, and the positive and negative HFM signals used have Doppler invariance, which can effectively reduce the synchronization error; 2) multiplexing the preamble signal , to achieve CFO estimation and channel estimation; 3) Construct a joint transmission matrix, which can fully consider inter-carrier interference, inter-symbol interference and inherent interference; 4) Adopt the sliding window equalization method, which can reduce the computational complexity.
  • the present invention aims to provide a FBMC system signal detection method for underwater acoustic environment aiming at the deficiency of the existing underwater acoustic multi-carrier communication technology. Reduce the influence of Doppler and CFO on the system, and improve the spectral efficiency of the system.
  • the FBMC data frame in the described underwater acoustic FBMC system is made up of a pair of hyperbolic frequency modulation (HFM) synchronous signal, some guard intervals and Several data blocks are formed, and the method includes the following steps:
  • Signal synchronization The purpose of signal synchronization is to find out when the signal arrives at the receiving end. Using the cross-correlation method, a pair of positive and negative frequency-modulated hyperbolic FM HFM signals in the frame header of the data frame is matched and filtered to obtain a pair of correlation peaks, and the middle position of the maximum value of the two correlation peaks can be used as the data frame. The arrival time of the signal, and using this time as the starting time of the subsequent detection operation of the frame signal, the synchronization of the signal is realized;
  • step (1) Doppler estimation and compensation: according to the matched filtering result of step (1), judge the translation time length of HFM signal, estimate Doppler factor, and adopt the method for resampling to carry out Doppler compensation to whole frame signal;
  • step (3) Carrier frequency offset estimation and compensation: After the whole frame signal undergoes Doppler compensation in step (2), combined with the pilot symbols in the preamble of the data block, the time-frequency offset two-dimensional search method is used to perform carrier frequency offset (CFO) estimate, and make CFO compensation for the data block;
  • CFO carrier frequency offset
  • T the joint transmission matrix
  • Ts the joint transmission matrix
  • z the transmitted data symbol and the received data symbol
  • ⁇ the noise, which is a column vector of MN ⁇ 1
  • m and p are the mth and pth subcarriers in an FBMC multicarrier symbol
  • q and n are the qth and nth FBMC multicarrier symbols in the time dimension in the data block
  • the value range of q and n is [0,N-1]
  • N represents the total number of FBMC multi-carrier symbols
  • h k represents the kth channel impulse response the value of the tap;
  • Channel equalization divide the sub-matrix along the main diagonal direction of the joint transmission matrix obtained in step (5), and use the minimum mean square error equalizer to equalize each sub-matrix in turn to obtain all the symbol channels to be detected
  • the equalization adopts the sliding window equalization method, as follows:
  • step (1) is divided into the following three steps:
  • the synchronous signal is composed of a pair of positive and negative frequency modulated HFM signals, and a group of local positive and negative frequency modulated HFMs are used to perform cross-correlation calculations with the received signal respectively, and the arrival times of the two correlation peaks are t up and t down ;
  • step (2) is divided into the following three steps:
  • step (4) the channel estimation method in step (4) is specifically as follows:
  • 0 represents the pilot symbol on the mth subcarrier in the preamble
  • L represents the baseband symbol length of the prototype filter
  • g is the prototype filter function
  • l represents the lower point of the lth point in the prototype filter with a length of L
  • k represents the kth tap subscript of the channel
  • m and p are the mth and pth subcarriers respectively
  • M is the total number of subcarriers
  • sync head signal used in the present invention is a pair of positive and negative frequency modulated HFM pulse signals, compared with the traditional single linear frequency modulation (LFM) pulse signal, it has better Doppler stability and improves the synchronization accuracy.
  • the present invention multiplexes the synchronization head signal to realize Doppler estimation and reduces system overhead; and the estimated value of Doppler can be obtained before the data signal arrives, which has better real-time performance and reduces the consumption of system storage resources.
  • the two-dimensional search method in the present invention can realize CFO estimation and time fine synchronization at the same time, and has better real-time performance.
  • pilot symbols are multiplexed to reduce system overhead; a transfer matrix is constructed, a linear relationship between channel vectors and received symbols is established, and channel estimation complexity is reduced.
  • the joint transmission matrix constructed in the present invention fully considers the inherent interference between prototype filters, channels, and different subcarriers and the interference between symbols, is suitable for the equalization of different system parameters, and can easily adjust the value range of parameters, Use different complexity requirements.
  • Fig. 1 is a sliding window equalization flowchart in the signal detection method of the underwater acoustic FBMC system of the present invention
  • Fig. 2 is a synchronous frame structure and a schematic diagram of matched filtering results
  • Figure 3 is a comparison of LFM and HFM autocorrelation characteristics when there is a Doppler frequency shift; in the figure, (a) represents the comparison of LFM correlation characteristics with or without Doppler, and (b) represents the HFM with or without Doppler relevant characteristics;
  • Figure 4 is the block structure of FBMC
  • Fig. 5 is a schematic structural diagram of a joint transmission matrix and a sliding window.
  • Fig. 6 is a flowchart of the signal detection method of the underwater acoustic FBMC system according to the present invention.
  • the FBMC data frame in the underwater acoustic FBMC system is by a pair of hyperbolic frequency modulation (HFM) synchronous signal, some guards interval and several data blocks, and the method includes the following steps:
  • phase and instantaneous frequency of the modulated HFM signal are designed to satisfy:
  • the other a in the first HFM signal is positive, that is, positive frequency modulation, and a in the second HFM signal is negative, that is, negative frequency modulation.
  • the specific value of a is determined by the modulation signal bandwidth, t represents time, and f 0 represents HFM the initial frequency of .
  • the positions of the two correlation peaks are shown by the dotted line in Figure 2. If the transceiver moves in the opposite direction, the relative positions of the correlation peaks will shift, as shown by the solid line in Figure 2. Conversely, if the transceiver ends move in opposite directions, the relative position of the correlation peak will still shift, but will be closer, but the shift time length ⁇ t is the same, so take the middle value to end the class to get the accurate signal arrival time, and use this time as the frame signal The starting moment of the subsequent detection operation realizes the synchronization of the signal.
  • HFM has better autocorrelation characteristics than traditional LFM waveforms in the case of Doppler, as shown in Figure 3, when the LFM signal has Doppler, the correlation peak is more blurred, so the HFM signal has better synchronization accuracy sex.
  • the method of resampling is used to do Doppler compensation to the received signal, that is, according to the received signal y(t), the
  • step (2) After the whole frame signal undergoes Doppler compensation in step (2), the time-frequency offset two-dimensional search method is used to realize CFO estimation combined with the known pilot symbols in the preamble, And use the estimated frequency offset value to perform CFO compensation on the received signal;
  • ⁇ and ⁇ represent the search range of frequency offset and time respectively, and * represents the conjugate transpose.
  • R( ⁇ , ⁇ ) takes the maximum value,
  • the value of is the estimated time delay and frequency shift value, namely
  • the frame structure of the transmitted signal is designed as shown in Figure 4, and a guard interval of not less than 3 symbol lengths is reserved between the preamble and the data symbols, so that the interference of the data symbols on the preamble can be ignored.
  • 0 represents the pilot symbol on the mth subcarrier in the preamble
  • L represents the baseband symbol length of the prototype filter
  • g is the prototype filter function
  • l represents the lower point of the lth point in the prototype filter with a length of L
  • k is the subscript of the kth tap of the channel
  • m and p are the mth and pth subcarriers respectively
  • M is the total number of subcarriers.
  • m and p are the mth and pth subcarriers in an FBMC multicarrier symbol
  • q and n are the qth and nth FBMC multicarrier symbols in the time dimension in the data block
  • the value range of m and p It is [-M/2,M/2-1]
  • the value range of q and n is [0,N-1]
  • N represents the total number of FBMC multi-carrier symbols
  • h k represents the kth channel impulse response The value of the tap.
  • the channel equalization adopts the sliding window equalization method as follows:

Abstract

Disclosed in the present invention is a signal detection method for an underwater acoustic FBMC communication system. The method relates to filter bank multi-carrier (FBMC) underwater acoustic communication technology. The underwater acoustic FBMC communication signal detection method provided by the present method is a method for an underwater acoustic delay-Doppler dual-extended-channel model. The method comprises the steps of: realizing signal synchronization and Doppler estimation by using a group of positive and negative frequency modulation HFM signals; realizing carrier frequency offset (CFO) estimation by using a known pilot symbol; realizing channel estimation by using a pilot symbol subjected to CFO compensation; constructing a joint transmission matrix according to a channel, which is obtained by means of estimation; and segmenting the joint transmission matrix into a plurality of sub-matrices, and successively performing channel equalization on each sub-matrix, so as to obtain a symbol to be tested. The present invention has the advantages that multiplexing HFM and a preamble reduces system overheads; a constructed joint transmission matrix includes all non-additive interferences received by each receiving symbol, thereby improving an equalization performance; and by means of an equalization method for a sliding window, the computational complexity of a system is effectively reduced.

Description

一种用于水声FBMC系统的信号检测方法A signal detection method for underwater acoustic FBMC system 技术领域technical field
本发明属于多载波通信领域，涉及一种水声多载波信号检测方法，特别涉及一种用于水声FBMC系统的信号检测方法。The invention belongs to the field of multi-carrier communication, relates to an underwater acoustic multi-carrier signal detection method, in particular to a signal detection method for an underwater acoustic FBMC system.
背景技术Background technique
在现有的多载波水声通信技术中，多使用正交频分复用(Orthogonal Frequency Division Multiplexing，OFDM)技术，相对于单载波水声通信技术，OFDM可以以较低的接收机复杂度应对多径信道情况。但是，在OFDM系统中，由于需要各个子载波之间是正交的，因此对载波间干扰很敏感。载波间干扰通常来自于多普勒频移和载波频偏(CFO)。多普勒频移是由收发端之间的相对运动引起的，CFO主要是由收发端的晶振频率不一致造成的。因此在信号检测中需要准确进行信号同步、多普勒估计及CFO估计，并补偿。在无线电通信系统中，由于收发端的相对速度远远低于电磁波的传播速度，多普勒因子通常＜10 -6，几乎可以忽略，但是在水中，声速较慢(约为1400～1600米/秒)，这就导致多普勒因子通常在10 -3级别，不可忽略。虽已有技术着力于解决OFDM中的载波间干扰问题，但是在复杂多变的水声信道环境中，依然难以做到较好性能的多普勒补偿。此外，每个OFDM符号前都需要循环前缀或保护间隔，以避免符号间干扰，降低了频谱效率。 In the existing multi-carrier underwater acoustic communication technology, Orthogonal Frequency Division Multiplexing (OFDM) technology is mostly used. Compared with the single-carrier underwater acoustic communication technology, OFDM can deal with it with lower receiver complexity. multipath channel conditions. However, in the OFDM system, since each sub-carrier needs to be orthogonal, it is very sensitive to inter-carrier interference. Intercarrier interference usually comes from Doppler shift and carrier frequency offset (CFO). The Doppler frequency shift is caused by the relative movement between the transceivers, and the CFO is mainly caused by the inconsistency of the crystal oscillator frequencies at the transceivers. Therefore, signal synchronization, Doppler estimation and CFO estimation must be performed accurately during signal detection, and compensation must be performed. In a radio communication system, since the relative speed of the transceiver end is much lower than the propagation speed of electromagnetic waves, the Doppler factor is usually <10 -6 , which can be almost ignored, but in water, the speed of sound is relatively slow (about 1400-1600 m/s ), which leads to the Doppler factor usually at the level of 10 -3 , which cannot be ignored. Although existing technologies focus on solving the problem of inter-carrier interference in OFDM, it is still difficult to achieve better performance of Doppler compensation in complex and changeable underwater acoustic channel environments. In addition, a cyclic prefix or guard interval is required before each OFDM symbol to avoid inter-symbol interference and reduce spectral efficiency.
为了应对OFDM的这些缺点，基于滤波器组的多载波(Filter Bank Multi-carrier，FBMC)技术得到关注与研究。在对抗载波间干扰方面，FBMC技术应用非正交子载波的思想，降低了FBMC系统对子载波间干扰的敏感度。在频谱效率方面，不同于OFDM，FBMC无需每个多载波符号前都添加循环前缀或保护间隔，而且本发明所采用的方法，针对频谱资源稀缺的水下环境，只需要每帧前插入一个preamble进行信道估计和频偏估计即可，而每帧可包含多个FBMC符号，提高频谱效率。然而，FBMC系统所应用的非正交思想也引入了一定的固有干扰，因此本发明针对FBMC信号检测技术中的问题，在检测复杂度和检测性能上，提出了一种折中的选择，并进一步提高了信号利用率，即，提供了一种用于水声信道环境的FBMC信号检测方法。该方法的优势包括：1)复用同步头脉冲，实现信号同步和多普勒估计，且所使用的正负HFM信号具有多普勒不变性，可有效降低同步误差；2)复用preamble信号，实现CFO估计和信道估计；3)构造联合传输矩阵，该矩阵能全面考虑载波间干扰、符号间干扰和固有干扰；4)采用滑动窗均衡方法，可以降低计算复杂度。In order to deal with these shortcomings of OFDM, filter bank based multi-carrier (Filter Bank Multi-carrier, FBMC) technology has received attention and research. In terms of combating inter-carrier interference, FBMC technology uses the idea of non-orthogonal sub-carriers to reduce the sensitivity of FBMC systems to inter-sub-carrier interference. In terms of spectrum efficiency, unlike OFDM, FBMC does not need to add a cyclic prefix or guard interval before each multi-carrier symbol, and the method adopted in the present invention only needs to insert a preamble before each frame for the underwater environment where spectrum resources are scarce. It only needs to perform channel estimation and frequency offset estimation, and each frame can contain multiple FBMC symbols to improve spectral efficiency. However, the non-orthogonal idea used in the FBMC system also introduces certain inherent interference. Therefore, the present invention aims at the problems in the FBMC signal detection technology, and proposes a compromise between the detection complexity and detection performance. The signal utilization rate is further improved, that is, a FBMC signal detection method for an underwater acoustic channel environment is provided. The advantages of this method include: 1) multiplexing the synchronization head pulse to realize signal synchronization and Doppler estimation, and the positive and negative HFM signals used have Doppler invariance, which can effectively reduce the synchronization error; 2) multiplexing the preamble signal , to achieve CFO estimation and channel estimation; 3) Construct a joint transmission matrix, which can fully consider inter-carrier interference, inter-symbol interference and inherent interference; 4) Adopt the sliding window equalization method, which can reduce the computational complexity.
发明内容Contents of the invention
本发明旨在于针对现有水声多载波通信技术的不足，提供了一种用于水声环境的FBMC系统信号检测方法。降低系统受多普勒和CFO的影响，并提高系统的频谱效率。The present invention aims to provide a FBMC system signal detection method for underwater acoustic environment aiming at the deficiency of the existing underwater acoustic multi-carrier communication technology. Reduce the influence of Doppler and CFO on the system, and improve the spectral efficiency of the system.
为实现上述目的，本发明提供的一种用于水声环境的FBMC系统信号检测方法，所述水声FBMC系统中的FBMC数据帧由一对双曲调频(HFM)同步信号、若干保护间隔和若干数据块组成，且该方法包括以下步骤：In order to achieve the above object, a kind of FBMC system signal detection method for underwater acoustic environment provided by the invention, the FBMC data frame in the described underwater acoustic FBMC system is made up of a pair of hyperbolic frequency modulation (HFM) synchronous signal, some guard intervals and Several data blocks are formed, and the method includes the following steps:
(1)信号同步：信号同步的目的是找到信号何时到达接收端。采用互相关的方法，对数据帧帧头部分的一对正负调频的双曲调频HFM信号进行匹配滤波，得到一对相关峰，这两个相关峰最大值的中间位置即可作为该数据帧信号的到达时刻，并以这个时刻，作为该帧信号后续检测操作的起始时刻，实现了信号的同步；(1) Signal synchronization: The purpose of signal synchronization is to find out when the signal arrives at the receiving end. Using the cross-correlation method, a pair of positive and negative frequency-modulated hyperbolic FM HFM signals in the frame header of the data frame is matched and filtered to obtain a pair of correlation peaks, and the middle position of the maximum value of the two correlation peaks can be used as the data frame The arrival time of the signal, and using this time as the starting time of the subsequent detection operation of the frame signal, the synchronization of the signal is realized;
(2)多普勒估计与补偿：根据步骤(1)的匹配滤波结果，判断HFM信号的平移时长，估计多普勒因子，并采用重采样的方法对整帧信号进行多普勒补偿；(2) Doppler estimation and compensation: according to the matched filtering result of step (1), judge the translation time length of HFM signal, estimate Doppler factor, and adopt the method for resampling to carry out Doppler compensation to whole frame signal;
(3)载波频偏估计与补偿：在整帧信号经过步骤(2)的多普勒补偿后，结合数据块的preamble中的导频符号，采用时间-频偏二维搜索方法进行载波频偏(CFO)估计，并对该数据块做CFO补偿；(3) Carrier frequency offset estimation and compensation: After the whole frame signal undergoes Doppler compensation in step (2), combined with the pilot symbols in the preamble of the data block, the time-frequency offset two-dimensional search method is used to perform carrier frequency offset (CFO) estimate, and make CFO compensation for the data block;
(4)信道估计：建立信道脉冲响应和导频符号之间的线性模型，结合步骤(3)中补偿过CFO的导频符号，采用加权最小二乘方法，实现信道脉冲响应的估计；(4) Channel estimation: set up a linear model between the channel impulse response and the pilot symbols, in conjunction with the pilot symbols compensated for CFO in step (3), adopt the weighted least squares method to realize the estimation of the channel impulse response;
(5)构造联合传输矩阵：根据FBMC系统中的原型滤波器函数以及步骤(4)中估计得到的信道脉冲响应，构造联合传输矩阵；联合传输矩阵的构建方法具体如下：(5) Constructing the joint transmission matrix: according to the prototype filter function in the FBMC system and the channel impulse response estimated in step (4), the joint transmission matrix is constructed; the construction method of the joint transmission matrix is as follows:
定义MN×MN维的联合传输矩阵T满足z＝Ts+η，其中z和s分别表示发送的数据符号和接收的数据符号，η为噪声，均为MN×1的列向量，T中的每个元素定义为Define the joint transmission matrix T of MN×MN dimensions to satisfy z=Ts+η, where z and s respectively represent the transmitted data symbol and the received data symbol, and η is the noise, which is a column vector of MN×1, and each in T elements are defined as
Figure PCTCN2021109272-appb-000001
Figure PCTCN2021109272-appb-000001
其中m和p为一个FBMC多载波符号中的第m和第p个子载波，q和n是数据块中，时间维度上的第q和第n个FBMC多载波符号，m和p的取值范围是[-M/2,M/2-1]，q和n的取值范围是[0,N-1]，N表示FBMC多载波符号的总数量，h k表示信道脉冲响应的第k个抽头的值； Where m and p are the mth and pth subcarriers in an FBMC multicarrier symbol, q and n are the qth and nth FBMC multicarrier symbols in the time dimension in the data block, and the value range of m and p It is [-M/2,M/2-1], the value range of q and n is [0,N-1], N represents the total number of FBMC multi-carrier symbols, h k represents the kth channel impulse response the value of the tap;
(6)信道均衡：沿步骤(5)中得到的联合传输矩阵的主对角线方向，分割子矩阵，并采用最小均方误差均衡器依次对每个子矩阵进行均衡，得到全部待检测符号信道均衡采用滑动窗均衡方法，具体如下：(6) Channel equalization: divide the sub-matrix along the main diagonal direction of the joint transmission matrix obtained in step (5), and use the minimum mean square error equalizer to equalize each sub-matrix in turn to obtain all the symbol channels to be detected The equalization adopts the sliding window equalization method, as follows:
(6.1)滑动窗的作用是对矩阵T进行截取，初始化滑动窗的角标变量p w＝1，设定滑动窗口大小为3N×3N； (6.1) The function of the sliding window is to intercept the matrix T, initialize the index variable p w =1 of the sliding window, and set the size of the sliding window to be 3N×3N;
(6.2)设定滑动窗起点为
Figure PCTCN2021109272-appb-000002
并按此起点截取子矩阵
Figure PCTCN2021109272-appb-000003
并截取接收符号向量z的片段
Figure PCTCN2021109272-appb-000004
(6.2) Set the starting point of the sliding window as
Figure PCTCN2021109272-appb-000002
And intercept the sub-matrix according to this starting point
Figure PCTCN2021109272-appb-000003
And intercept the fragment receiving the symbol vector z
Figure PCTCN2021109272-appb-000004
(6.3)执行最小均方误差(MMSE)均衡得到
Figure PCTCN2021109272-appb-000005
的估计值：
Figure PCTCN2021109272-appb-000006
其中σ 2表示噪声方差，I表示单位矩阵； (6.3) Perform minimum mean square error (MMSE) equalization to get
Figure PCTCN2021109272-appb-000005
Estimated value of :
Figure PCTCN2021109272-appb-000006
where σ2 represents the noise variance, and I represents the identity matrix;
(6.4)如果p w＝1，则待估计符号
Figure PCTCN2021109272-appb-000007
接着执行步骤(6.6)；如果p w＞1，则执行步骤(6.5)； (6.4) If p w =1, the symbol to be estimated
Figure PCTCN2021109272-appb-000007
Then execute step (6.6); if p w > 1, then execute step (6.5);
(6.5)如果p w＜M-2，则待估计符号
Figure PCTCN2021109272-appb-000008
接着执行步骤(6.6)；如果p w＝M-2，则转执行步骤(6.7)； (6.5) If p w <M-2, the symbol to be estimated
Figure PCTCN2021109272-appb-000008
Then execute step (6.6); if p w =M-2, then turn to execute step (6.7);
(6.6)p w＝p w+1，跳转回步骤(6.2)； (6.6) p w = p w +1, jump back to step (6.2);
(6.7)待估计符号
Figure PCTCN2021109272-appb-000009
完成全部数据检测并结束。 (6.7) Symbol to be estimated
Figure PCTCN2021109272-appb-000009
Complete all data detection and end.
进一步地，步骤(1)中的信号同步方法分为以下三步：Further, the signal synchronization method in step (1) is divided into the following three steps:
(1.1)同步信号由一对正负调频的HFM信号构成，用本地的一组正负调频的HFM分别与接收信号做互相关运算，得到两个相关峰的到达时刻为t up和t down； (1.1) The synchronous signal is composed of a pair of positive and negative frequency modulated HFM signals, and a group of local positive and negative frequency modulated HFMs are used to perform cross-correlation calculations with the received signal respectively, and the arrival times of the two correlation peaks are t up and t down ;
(1.2)计算两个相关峰到达时刻的中间值t a＝(t up+t down)/2； (1.2) Calculate the median value t a =(t up +t down )/2 of the arrival time of the two correlation peaks;
(1.3)计算FBMC符号信号的到达时刻t FBMC＝t a+0.5t HFM+t gap，其中，t HFM为HFM的脉冲周期，t gap为HFM与信号之间保护间隔时长。 (1.3) Calculate the arrival time of the FBMC symbol signal t FBMC =t a +0.5t HFM +t gap , where t HFM is the pulse period of the HFM, and t gap is the duration of the guard interval between the HFM and the signal.
进一步地，步骤(2)中的多普勒估计与补偿方法分为以下三步：Further, the Doppler estimation and compensation method in step (2) is divided into the following three steps:
(2.1)计算正负调频的HFM相关峰平移时长：Δt＝(t down-t up)-t HFM； (2.1) Calculate the HFM correlation peak translation duration of positive and negative frequency modulation: Δt=(t down -t up )-t HFM ;
(2.2)根据
Figure PCTCN2021109272-appb-000010
计算得到多普勒因子的估计值
Figure PCTCN2021109272-appb-000011
其中a和f 0分别表示HFM的调制系数和中心频率； (2.2) According to
Figure PCTCN2021109272-appb-000010
Calculate the estimated value of the Doppler factor
Figure PCTCN2021109272-appb-000011
where a and f 0 represent the modulation coefficient and center frequency of HFM, respectively;
(2.3)根据估计得到的
Figure PCTCN2021109272-appb-000012
采用重采样的方法对接收信号做多普勒补偿。 (2.3) According to the estimated
Figure PCTCN2021109272-appb-000012
The method of resampling is used to do Doppler compensation to the received signal.
进一步地，步骤(4)中的信道估计方法具体如下：Further, the channel estimation method in step (4) is specifically as follows:
(4.1)建立接收导频符号向量z 0与信道脉冲响应h之间模型，满足：z 0＝Λh+η 0，其中，η 0为噪声项，Λ是M p×N k维度的变换矩阵，M p和N k分别表示导频符号数量和信道抽头数，
Figure PCTCN2021109272-appb-000013
其中行向量λ m定义为 (4.1) Establish a model between the received pilot symbol vector z 0 and the channel impulse response h, satisfying: z 0 =Λh+η 0 , wherein, η 0 is a noise term, and Λ is a transformation matrix of M p ×N k dimensions, M p and N k represent the number of pilot symbols and the number of channel taps, respectively,
Figure PCTCN2021109272-appb-000013
where the row vector λ m is defined as
Figure PCTCN2021109272-appb-000014
Figure PCTCN2021109272-appb-000014
其中
Figure PCTCN2021109272-appb-000015
s m,0表示preamble中第m个子载波上的pilot符号，L表示原型滤波器的基带符号长度，g是原型滤波器函数，l表示长度为L的原型滤波器中的第l个点的下标，k表示信道的第k个抽头下标，m和p分别为第m和第p个子载波，M是总子载波数； in
Figure PCTCN2021109272-appb-000015
s m, 0 represents the pilot symbol on the mth subcarrier in the preamble, L represents the baseband symbol length of the prototype filter, g is the prototype filter function, l represents the lower point of the lth point in the prototype filter with a length of L where k represents the kth tap subscript of the channel, m and p are the mth and pth subcarriers respectively, and M is the total number of subcarriers;
(4.2)采用加权最小二乘方法估计得到信道时域脉冲响应
Figure PCTCN2021109272-appb-000016
其中C η0为噪声η 0的协方差矩阵，Λ H表示Λ的共轭转置。 (4.2) Using the weighted least squares method to estimate the channel time domain impulse response
Figure PCTCN2021109272-appb-000016
where C η0 is the covariance matrix of noise η 0 , and Λ H represents the conjugate transpose of Λ.
本发明的有益效果：Beneficial effects of the present invention:
1.本发明中同步头信号使用的是一对正负调频的HFM脉冲信号，相比于传统的单个线性调频(LFM)脉冲信号，具有更好的多普勒稳定性，提高同步精度。1. What the sync head signal used in the present invention is a pair of positive and negative frequency modulated HFM pulse signals, compared with the traditional single linear frequency modulation (LFM) pulse signal, it has better Doppler stability and improves the synchronization accuracy.
2.本发明复用同步头信号实现多普勒估计，降低系统开销；且在数据信号到达前即可得到多普勒的估计值，实时性更好，降低系统存储资源消耗。2. The present invention multiplexes the synchronization head signal to realize Doppler estimation and reduces system overhead; and the estimated value of Doppler can be obtained before the data signal arrives, which has better real-time performance and reduces the consumption of system storage resources.
3.本发明中二维搜索方法可同时实现CFO估计和时间精同步，实时性更好。3. The two-dimensional search method in the present invention can realize CFO estimation and time fine synchronization at the same time, and has better real-time performance.
4.本发明中复用了导频符号，降低系统开销；构造了传递矩阵，建立了信道向量与接收符号之间的线性关系，降低了信道估计复杂度。4. In the present invention, pilot symbols are multiplexed to reduce system overhead; a transfer matrix is constructed, a linear relationship between channel vectors and received symbols is established, and channel estimation complexity is reduced.
5.本发明中构造的联合传输矩阵全面考虑了原型滤波器、信道、不同子载波间的固有干扰和符号间的干扰，适用于不同系统参数的均衡，且可方便调整参数的取值范围，使用不同的复杂度需求。5. The joint transmission matrix constructed in the present invention fully considers the inherent interference between prototype filters, channels, and different subcarriers and the interference between symbols, is suitable for the equalization of different system parameters, and can easily adjust the value range of parameters, Use different complexity requirements.
6.本发明中使用滑动窗均衡方法，MMSE均衡的复杂度与矩阵纬度的三次方成正比，通过滑动窗的方法降低了单次MMSE估计的矩阵纬度，大大降低了均衡的计算复杂度。6. use sliding window equalization method among the present invention, the complexity of MMSE equalization is proportional to the cube of matrix latitude, the matrix latitude of single MMSE estimation has been reduced by the method for sliding window, greatly reduced the computational complexity of equalization.
附图说明Description of drawings
图1是本发明所述的水声FBMC系统的信号检测方法中滑动窗均衡流程图；Fig. 1 is a sliding window equalization flowchart in the signal detection method of the underwater acoustic FBMC system of the present invention;
图2是同步帧结构与的匹配滤波结果示意图；Fig. 2 is a synchronous frame structure and a schematic diagram of matched filtering results;
图3是存在多普勒频移情况时LFM与HFM自相关特性比较；图中，(a)表示有无多普勒时的LFM相关特性对比，(b)表示有无多普勒时的HFM相关特性；Figure 3 is a comparison of LFM and HFM autocorrelation characteristics when there is a Doppler frequency shift; in the figure, (a) represents the comparison of LFM correlation characteristics with or without Doppler, and (b) represents the HFM with or without Doppler relevant characteristics;
图4是FBMC的block结构；Figure 4 is the block structure of FBMC;
图5是联合传输矩阵和滑动窗的结构示意图。Fig. 5 is a schematic structural diagram of a joint transmission matrix and a sliding window.
图6是本发明所述的水声FBMC系统的信号检测方法流程图。Fig. 6 is a flowchart of the signal detection method of the underwater acoustic FBMC system according to the present invention.
具体实施方式Detailed ways
为了使本发明的技术方案及优点更加清楚明白，下面结合附图与具体实施方式对本方面作进一步描述。应当理解，此处所描述的具体实施方法仅仅用以解释本发明，并不用于限定本发明，尤其是参数的选取，仅用于举例。In order to make the technical solutions and advantages of the present invention clearer, the following will further describe this aspect in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific implementation methods described here are only used to explain the present invention, and are not intended to limit the present invention, especially the selection of parameters is only for example.
参考图1和图6，描述了本发明所提供的一种用于水声FBMC系统的信号检测方法，水声FBMC系统中的FBMC数据帧由一对双曲调频(HFM)同步信号、若干保护间隔和若干数据块组成，且该方法包括以下步骤：With reference to Fig. 1 and Fig. 6, described a kind of signal detection method that is used for underwater acoustic FBMC system provided by the present invention, the FBMC data frame in the underwater acoustic FBMC system is by a pair of hyperbolic frequency modulation (HFM) synchronous signal, some guards interval and several data blocks, and the method includes the following steps:
(1)信号同步方法。采用互相关方法，对一组正负调频的HFM信号进行匹配滤波：(1) Signal synchronization method. Using the cross-correlation method, a group of positive and negative frequency-modulated HFM signals are matched and filtered:
具体地，设计所调制HFM信号的相位和瞬时频率分别满足：Specifically, the phase and instantaneous frequency of the modulated HFM signal are designed to satisfy:
Figure PCTCN2021109272-appb-000017
Figure PCTCN2021109272-appb-000017
其中，第一个HFM信号中另a为正，即正调频，第二个HFM信号中a为负，即负调频，a的具体数值由调制信号带宽所决定，t表示时间，f 0表示HFM的初始频率。 Among them, the other a in the first HFM signal is positive, that is, positive frequency modulation, and a in the second HFM signal is negative, that is, negative frequency modulation. The specific value of a is determined by the modulation signal bandwidth, t represents time, and f 0 represents HFM the initial frequency of .
用本地的一组，具有相同调制方式的正负调频的HFM信号，分别与接收信号做互相关，得到两个相关峰，相关峰时刻为t up和t down，如图2。 Use a local group of positive and negative frequency-modulated HFM signals with the same modulation mode to perform cross-correlation with the received signal respectively to obtain two correlation peaks. The correlation peak times are t up and t down , as shown in Figure 2.
计算两个相关峰到达时刻的中间值t a＝(t up+t down)/2。 Calculate the median value t a =(t up +t down )/2 of the arrival times of the two correlation peaks.
计算FBMC符号信号的到达时刻t FBMC＝t a+0.5t HFM+t gap，其中，t HFM为HFM的脉冲周期，t gap为HFM与信号之间保护间隔时长。 Calculate the arrival time t FBMC of the FBMC symbol signal =t a +0.5t HFM +t gap , where t HFM is the pulse period of the HFM, and t gap is the guard interval duration between the HFM and the signal.
如果收发端无相对距离的变化，则两个相关峰的位置如图2中虚线所示，如果收发端反向运动，相关峰的相对位置会发生偏移，如图2中实线所示。反之，如果收发端相向运动，相关峰相对位置依然会偏移，但会靠近，但偏移时间长度Δt相同，因此取中间值结课得到准确信号到达时刻，并以这个时刻，作为该帧信号后续检测操作的起始时刻，实现信号的同步。If there is no change in the relative distance of the transceiver, the positions of the two correlation peaks are shown by the dotted line in Figure 2. If the transceiver moves in the opposite direction, the relative positions of the correlation peaks will shift, as shown by the solid line in Figure 2. Conversely, if the transceiver ends move in opposite directions, the relative position of the correlation peak will still shift, but will be closer, but the shift time length Δt is the same, so take the middle value to end the class to get the accurate signal arrival time, and use this time as the frame signal The starting moment of the subsequent detection operation realizes the synchronization of the signal.
另外由于HFM在有多普勒情况下，自相关特性优于传统的LFM波形，如图3所示，LFM信号在存在多普勒时，相关峰更模糊，因此HFM信号具有更好的同步准确性。In addition, because HFM has better autocorrelation characteristics than traditional LFM waveforms in the case of Doppler, as shown in Figure 3, when the LFM signal has Doppler, the correlation peak is more blurred, so the HFM signal has better synchronization accuracy sex.
(2)根据(1)的匹配滤波结果，判断HFM信号的平移时长，估计多普勒因子，并对接收信号做重采样，对整帧信号进行多普勒补偿：(2) According to the matched filtering result of (1), judge the translation duration of the HFM signal, estimate the Doppler factor, and resample the received signal, and perform Doppler compensation on the entire frame signal:
具体地，计算正负调频的HFM相关峰平移时长：Δt＝(t down-t up)-t HFM。 Specifically, calculate the HFM correlation peak translation duration of positive and negative frequency modulation: Δt=(t down −t up )−t HFM .
根据
Figure PCTCN2021109272-appb-000018
计算得到多普勒因子的估计值
Figure PCTCN2021109272-appb-000019
其中a和f 0分别表示HFM的调制系数和中心频率。 according to
Figure PCTCN2021109272-appb-000018
Calculate the estimated value of the Doppler factor
Figure PCTCN2021109272-appb-000019
Where a and f 0 represent the modulation coefficient and center frequency of HFM, respectively.
根据估计得到的
Figure PCTCN2021109272-appb-000020
采用重采样的方法对接收信号做多普勒补偿，即根据接收信号y(t)得到
Figure PCTCN2021109272-appb-000021
based on estimates
Figure PCTCN2021109272-appb-000020
The method of resampling is used to do Doppler compensation to the received signal, that is, according to the received signal y(t), the
Figure PCTCN2021109272-appb-000021
(3)载波频偏估计与补偿：在整帧信号经过步骤(2)的多普勒补偿后，采用时间-频偏二维搜索的方法，结合preamble中的已知导频符号实现CFO估计，并用估计得到的频偏值对接收信号进行CFO补偿；(3) Carrier frequency offset estimation and compensation: After the whole frame signal undergoes Doppler compensation in step (2), the time-frequency offset two-dimensional search method is used to realize CFO estimation combined with the known pilot symbols in the preamble, And use the estimated frequency offset value to perform CFO compensation on the received signal;
具体地，定义发送已知的preamble为x pre(t)，定义接收信号为y(t)，计算模糊度函数 Specifically, define the known preamble to be sent as x pre (t), define the received signal as y(t), and calculate the ambiguity function
Figure PCTCN2021109272-appb-000022
Figure PCTCN2021109272-appb-000022
其中τ,ε分别表示频偏和时间的搜索范围， *表示共轭转置。当R(τ,ε)取最大值时，
Figure PCTCN2021109272-appb-000023
的取值即估计得到的时延和频移值，即 Among them, τ and ε represent the search range of frequency offset and time respectively, and * represents the conjugate transpose. When R(τ,ε) takes the maximum value,
Figure PCTCN2021109272-appb-000023
The value of is the estimated time delay and frequency shift value, namely
Figure PCTCN2021109272-appb-000024
Figure PCTCN2021109272-appb-000024
接着，对接收信号做CFO补偿：
Figure PCTCN2021109272-appb-000025
Next, perform CFO compensation on the received signal:
Figure PCTCN2021109272-appb-000025
(4)信道估计：建立信道脉冲响应和接收符号之间的线性模型，结合步骤(3)中补偿过CFO的导频符号，采用最小二乘方法，实现信道脉冲响应的估计；(4) Channel estimation: set up a linear model between the channel impulse response and the received symbols, in conjunction with the pilot symbols compensated for CFO in step (3), adopt the least squares method to realize the estimation of the channel impulse response;
具体地，设计发送信号的帧结构如图4所示，在preamble和数据符号之间保留不小于3个符号长度的保护间隔，使得数据符号对preamble的干扰可以忽略。Specifically, the frame structure of the transmitted signal is designed as shown in Figure 4, and a guard interval of not less than 3 symbol lengths is reserved between the preamble and the data symbols, so that the interference of the data symbols on the preamble can be ignored.
建立接收导频符号向量z 0与信道脉冲响应h之间模型，满足：z 0＝Λh+η 0，其中，η 0为噪声项，Λ是M p×N k维度的变换矩阵，M p和N k分别表示导频符号数量和信道抽头数，
Figure PCTCN2021109272-appb-000026
其中行向量λ m定义为 Establish a model between the received pilot symbol vector z 0 and the channel impulse response h, satisfying: z 0 =Λh+η 0 , where η 0 is a noise term, Λ is a transformation matrix of M p ×N k dimensions, M p and N k represent the number of pilot symbols and the number of channel taps, respectively,
Figure PCTCN2021109272-appb-000026
where the row vector λ m is defined as
Figure PCTCN2021109272-appb-000027
Figure PCTCN2021109272-appb-000027
其中
Figure PCTCN2021109272-appb-000028
s m,0表示preamble中第m个子载波上的pilot符号，L表示原型滤波器的基带符号长度，g是原型滤波器函数，l表示长度为L的原型滤波器中的第l个点的下标，k表示信道的第k个抽头下标，m和p分别为第m和第p个子载波，M是总子载波数。采用加权最小二乘方法：
Figure PCTCN2021109272-appb-000029
估计得到信道时域脉冲响应
Figure PCTCN2021109272-appb-000030
其中
Figure PCTCN2021109272-appb-000031
为噪声η 0的协方差矩阵，Λ H表示Λ的共轭转置。 in
Figure PCTCN2021109272-appb-000028
s m, 0 represents the pilot symbol on the mth subcarrier in the preamble, L represents the baseband symbol length of the prototype filter, g is the prototype filter function, l represents the lower point of the lth point in the prototype filter with a length of L where k is the subscript of the kth tap of the channel, m and p are the mth and pth subcarriers respectively, and M is the total number of subcarriers. Using weighted least squares method:
Figure PCTCN2021109272-appb-000029
Estimated channel time domain impulse response
Figure PCTCN2021109272-appb-000030
in
Figure PCTCN2021109272-appb-000031
is the covariance matrix of noise η 0 , and Λ H represents the conjugate transpose of Λ.
(5)构造联合传输矩阵：根据FBMC系统中的原型滤波器函数以及步骤(4)中估计得到的信道脉冲响应，构造联合传输矩阵；(5) Constructing a joint transmission matrix: constructing a joint transmission matrix according to the prototype filter function in the FBMC system and the channel impulse response estimated in step (4);
具体地，定义MN×MN维的联合传输矩阵T满足z＝Ts+η，其中z和s分别表示发送的数据符号和接收的数据符号，η为噪声，均为MN×1的列向量，如图5所示，T的形式如下：Specifically, the joint transmission matrix T of MN×MN dimensions is defined to satisfy z=Ts+η, where z and s respectively represent the transmitted data symbol and the received data symbol, and η is noise, both of which are column vectors of MN×1, such as As shown in Figure 5, the form of T is as follows:
Figure PCTCN2021109272-appb-000032
Figure PCTCN2021109272-appb-000032
T中的每个元素定义为Each element in T is defined as
Figure PCTCN2021109272-appb-000033
Figure PCTCN2021109272-appb-000033
其中m和p为一个FBMC多载波符号中的第m和第p个子载波，q和n是数据块中，时间维度上的第q和第n个FBMC多载波符号，m和p的取值范围是[-M/2,M/2-1]，q和n的取值范围是 [0,N-1]，N表示FBMC多载波符号的总数量，h k表示信道脉冲响应的第k个抽头的值。 Where m and p are the mth and pth subcarriers in an FBMC multicarrier symbol, q and n are the qth and nth FBMC multicarrier symbols in the time dimension in the data block, and the value range of m and p It is [-M/2,M/2-1], the value range of q and n is [0,N-1], N represents the total number of FBMC multi-carrier symbols, h k represents the kth channel impulse response The value of the tap.
(6)沿步骤(5)中得到的联合传输矩阵的主对角线方向，分割子矩阵，并采用最小均方误差均衡器依次对每个子矩阵进行均衡，得到全部待检测符号。信道均衡采用滑动窗均衡方法具体如下：(6) Divide the sub-matrix along the main diagonal direction of the joint transmission matrix obtained in step (5), and use the minimum mean square error equalizer to equalize each sub-matrix in turn to obtain all symbols to be detected. The channel equalization adopts the sliding window equalization method as follows:
(6.1)滑动窗的作用是对矩阵T进行截取，初始化滑动窗的角标变量p w＝1，设定滑动窗口大小为3N×3N； (6.1) The function of the sliding window is to intercept the matrix T, initialize the index variable p w =1 of the sliding window, and set the size of the sliding window to be 3N×3N;
(6.2)设定滑动窗起点为
Figure PCTCN2021109272-appb-000034
并按此起点截取子矩阵
Figure PCTCN2021109272-appb-000035
如图5所示的，实线框中所框出的区域，即“sliding window”。当p w＝2时，滑动窗如图5中虚线框所框出的区域。截取接收符号向量z的片段
Figure PCTCN2021109272-appb-000036
(6.2) Set the starting point of the sliding window as
Figure PCTCN2021109272-appb-000034
And intercept the sub-matrix according to this starting point
Figure PCTCN2021109272-appb-000035
As shown in FIG. 5 , the area framed by the solid line box is the "sliding window". When p w =2, the sliding window is the area framed by the dotted line box in FIG. 5 . Intercept a segment of the received symbol vector z
Figure PCTCN2021109272-appb-000036
(6.3)执行最小均方误差(MMSE)均衡得到
Figure PCTCN2021109272-appb-000037
的估计值：
Figure PCTCN2021109272-appb-000038
其中σ 2表示噪声方差，I表示单位矩阵； (6.3) Perform minimum mean square error (MMSE) equalization to get
Figure PCTCN2021109272-appb-000037
Estimated value of :
Figure PCTCN2021109272-appb-000038
where σ2 represents the noise variance, and I represents the identity matrix;
(6.4)如果p w＝1，则待估计符号
Figure PCTCN2021109272-appb-000039
接着执行步骤(6.6)；如果p w＞1，则执行步骤(6.5)； (6.4) If p w =1, the symbol to be estimated
Figure PCTCN2021109272-appb-000039
Then execute step (6.6); if p w > 1, then execute step (6.5);
(6.5)如果p w＜M-2，则待估计符号
Figure PCTCN2021109272-appb-000040
接着执行步骤(6.6)；如果p w＝M-2，则转执行步骤(6.7)； (6.5) If p w <M-2, the symbol to be estimated
Figure PCTCN2021109272-appb-000040
Then execute step (6.6); if p w =M-2, then turn to execute step (6.7);
(6.6)p w＝p w+1，跳转回步骤(6.2)； (6.6) p w = p w +1, jump back to step (6.2);
(6.7)待估计符号
Figure PCTCN2021109272-appb-000041
完成全部数据检测并结束。 (6.7) Symbol to be estimated
Figure PCTCN2021109272-appb-000041
Complete all data detection and end.
上述实施例用来解释说明本发明，而不是对本发明进行限制，在本发明的精神和权利要求的保护范围内，对本发明作出的任何修改和改变，都落入本发明的保护范围。The above-mentioned embodiments are used to illustrate the present invention, rather than to limit the present invention. Within the spirit of the present invention and the protection scope of the claims, any modification and change made to the present invention will fall into the protection scope of the present invention.

Claims (4)

  1. 一种用于水声FBMC系统的信号检测方法，其特征在于，所述水声FBMC系统中的FBMC数据帧由一对双曲调频(HFM)同步信号、若干保护间隔和若干数据块组成，且该方法包括以下步骤：A kind of signal detection method for underwater acoustic FBMC system, it is characterized in that, the FBMC data frame in described underwater acoustic FBMC system is made up of a pair of hyperbolic frequency modulation (HFM) synchronous signal, some guard intervals and some data blocks, and The method includes the following steps:
    (1)信号同步：信号同步的目的是找到信号何时到达接收端。采用互相关的方法，对数据帧帧头部分的一对正负调频的双曲调频HFM信号进行匹配滤波，得到一对相关峰，这两个相关峰最大值的中间位置即可作为该数据帧信号的到达时刻，并以这个时刻，作为该帧信号后续检测操作的起始时刻，实现信号的同步；(1) Signal synchronization: The purpose of signal synchronization is to find out when the signal arrives at the receiving end. Using the cross-correlation method, a pair of positive and negative frequency-modulated hyperbolic FM HFM signals in the frame header of the data frame is matched and filtered to obtain a pair of correlation peaks, and the middle position of the maximum value of the two correlation peaks can be used as the data frame The arrival time of the signal, and use this time as the starting time of the subsequent detection operation of the frame signal to realize the synchronization of the signal;
    (2)多普勒估计与补偿：根据步骤(1)的匹配滤波结果，判断HFM信号的平移时长，估计多普勒因子，并采用重采样的方法对整帧信号进行多普勒补偿；(2) Doppler estimation and compensation: according to the matched filtering result of step (1), judge the translation time length of HFM signal, estimate Doppler factor, and adopt the method for resampling to carry out Doppler compensation to whole frame signal;
    (3)载波频偏估计与补偿：在整帧信号经过步骤(2)的多普勒补偿后，结合数据块的preamble中的导频符号，采用时间-频偏二维搜索方法进行载波频偏(CFO)估计，并对该数据块做CFO补偿；(3) Carrier frequency offset estimation and compensation: After the whole frame signal undergoes Doppler compensation in step (2), combined with the pilot symbols in the preamble of the data block, the time-frequency offset two-dimensional search method is used to perform carrier frequency offset (CFO) estimate, and make CFO compensation for the data block;
    (4)信道估计：建立信道脉冲响应和导频符号之间的线性模型，结合步骤(3)中补偿过CFO的导频符号，采用加权最小二乘方法，实现信道脉冲响应的估计；(4) Channel estimation: set up a linear model between the channel impulse response and the pilot symbols, in conjunction with the pilot symbols compensated for CFO in step (3), adopt the weighted least squares method to realize the estimation of the channel impulse response;
    (5)构造联合传输矩阵：根据FBMC系统中的原型滤波器函数以及步骤(4)中估计得到的信道脉冲响应，构造联合传输矩阵；联合传输矩阵的构建方法具体如下：(5) Constructing the joint transmission matrix: according to the prototype filter function in the FBMC system and the channel impulse response estimated in step (4), the joint transmission matrix is constructed; the construction method of the joint transmission matrix is as follows:
    定义MN×MN维的联合传输矩阵T满足z＝Ts+η，其中z和s分别表示发送的数据符号和接收的数据符号，η为噪声，均为MN×1的列向量，T中的每个元素定义为Define the joint transmission matrix T of MN×MN dimensions to satisfy z=Ts+η, where z and s respectively represent the transmitted data symbol and the received data symbol, and η is the noise, which is a column vector of MN×1, and each in T elements are defined as
    Figure PCTCN2021109272-appb-100001
    Figure PCTCN2021109272-appb-100001
    其中m和p为一个FBMC多载波符号中的第m和第p个子载波，q和n是数据块中，时间维度上的第q和第n个FBMC多载波符号，m和p的取值范围是[-M/2,M/2-1]，q和n的取值范围是[0,N-1]，N表示FBMC多载波符号的总数量，h k表示信道脉冲响应的第k个抽头的值； Where m and p are the mth and pth subcarriers in an FBMC multicarrier symbol, q and n are the qth and nth FBMC multicarrier symbols in the time dimension in the data block, and the value range of m and p It is [-M/2,M/2-1], the value range of q and n is [0,N-1], N represents the total number of FBMC multi-carrier symbols, h k represents the kth channel impulse response the value of the tap;
    (6)信道均衡：沿步骤(5)中得到的联合传输矩阵的主对角线方向，分割子矩阵，并采用最小均方误差均衡器依次对每个子矩阵进行均衡，得到全部待检测符号；信道均衡采用滑动窗均衡方法，具体如下：(6) Channel equalization: along the main diagonal direction of the joint transmission matrix obtained in step (5), split the sub-matrix, and adopt the minimum mean square error equalizer to equalize each sub-matrix in turn to obtain all symbols to be detected; The channel equalization adopts the sliding window equalization method, as follows:
    (6.1)滑动窗的作用是对矩阵T进行截取，初始化滑动窗的角标变量p w＝1，设定滑动窗口大小为3N×3N； (6.1) The function of the sliding window is to intercept the matrix T, initialize the index variable p w =1 of the sliding window, and set the size of the sliding window to be 3N×3N;
    (6.2)设定滑动窗起点为T((p w-1)N+1,(p w-1)N+1)，并按此起点截取子矩阵
    Figure PCTCN2021109272-appb-100002
    并截取接 收符号向量z的片段
    Figure PCTCN2021109272-appb-100003
    (6.2) Set the starting point of the sliding window as T((p w -1)N+1, (p w -1)N+1), and intercept the sub-matrix according to this starting point
    Figure PCTCN2021109272-appb-100002
    And intercept the fragment receiving the symbol vector z
    Figure PCTCN2021109272-appb-100003
    (6.3)执行最小均方误差(MMSE)均衡得到
    Figure PCTCN2021109272-appb-100004
    的估计值：
    Figure PCTCN2021109272-appb-100005
    其中σ 2表示噪声方差，I表示单位矩阵；     (6.3) Perform minimum mean square error (MMSE) equalization to get
    Figure PCTCN2021109272-appb-100004
    Estimated value of :
    Figure PCTCN2021109272-appb-100005
    where σ2 represents the noise variance, and I represents the identity matrix;
    (6.4)如果p w＝1，则待估计符号
    Figure PCTCN2021109272-appb-100006
    接着执行步骤(6.6)；如果p w＞1，则执行步骤(6.5)；     (6.4) If p w =1, the symbol to be estimated
    Figure PCTCN2021109272-appb-100006
    Then execute step (6.6); if p w > 1, then execute step (6.5);
    (6.5)如果p w＜M-2，则待估计符号
    Figure PCTCN2021109272-appb-100007
    接着执行步骤(6.6)；如果p w＝M-2，则转执行步骤(6.7)；     (6.5) If p w <M-2, the symbol to be estimated
    Figure PCTCN2021109272-appb-100007
    Then execute step (6.6); if p w =M-2, then turn to execute step (6.7);
    (6.6)p w＝p w+1，跳转回步骤(6.2)； (6.6) p w = p w +1, jump back to step (6.2);
    (6.7)待估计符号
    Figure PCTCN2021109272-appb-100008
    完成全部数据检测并结束。     (6.7) Symbol to be estimated
    Figure PCTCN2021109272-appb-100008
    Complete all data detection and end.
  2. 根据权利要求1所述的一种用于水声FBMC系统的信号检测方法，其特征在于，所述步骤(1)中的信号同步方法具体如下：A kind of signal detection method for underwater acoustic FBMC system according to claim 1, is characterized in that, the signal synchronization method in described step (1) is specifically as follows:
    (1.1)同步信号由一对正负调频的HFM信号构成，用本地的一组正负调频的HFM分别与接收信号做互相关运算，得到两个相关峰的到达时刻为t up和t down； (1.1) The synchronous signal is composed of a pair of positive and negative frequency modulated HFM signals, and a group of local positive and negative frequency modulated HFMs are used to perform cross-correlation calculations with the received signal respectively, and the arrival times of the two correlation peaks are t up and t down ;
    (1.2)计算两个相关峰到达时刻的中间值t a＝(t up+t down)/2； (1.2) Calculate the median value t a =(t up +t down )/2 of the arrival time of the two correlation peaks;
    (1.3)计算FBMC符号信号的到达时刻t FBMC＝t a+0.5t HFM+t gap，其中，t HFM为HFM的脉冲周期，t gap为HFM与信号之间保护间隔时长。 (1.3) Calculate the arrival time of the FBMC symbol signal t FBMC =t a +0.5t HFM +t gap , where t HFM is the pulse period of the HFM, and t gap is the duration of the guard interval between the HFM and the signal.
  3. 根据权利要求1所述的一种用于水声FBMC系统的信号检测方法，其特征在于，所述步骤(2)中的多普勒估计与补偿方法具体如下：A kind of signal detection method that is used for underwater acoustic FBMC system according to claim 1, is characterized in that, the Doppler estimation and compensation method in described step (2) are specifically as follows:
    (2.1)计算正负调频的HFM相关峰平移时长：Δt＝(t down-t up)-t HFM； (2.1) Calculate the HFM correlation peak translation duration of positive and negative frequency modulation: Δt=(t down -t up )-t HFM ;
    (2.2)根据
    Figure PCTCN2021109272-appb-100009
    计算得到多普勒因子的估计值
    Figure PCTCN2021109272-appb-100010
    其中a和f 0分别表示HFM的调制系数和中心频率；     (2.2) According to
    Figure PCTCN2021109272-appb-100009
    Calculate the estimated value of the Doppler factor
    Figure PCTCN2021109272-appb-100010
    where a and f 0 represent the modulation coefficient and center frequency of HFM, respectively;
    (2.3)根据估计得到的
    Figure PCTCN2021109272-appb-100011
    采用重采样的方法对接收信号做多普勒补偿。     (2.3) According to the estimated
    Figure PCTCN2021109272-appb-100011
    The method of resampling is used to do Doppler compensation to the received signal.
  4. 根据权利要求1所述的一种用于水声FBMC系统的信号检测方法，其特征在于，所述步骤(4)中的信道估计方法具体如下：A kind of signal detection method for underwater acoustic FBMC system according to claim 1, is characterized in that, the channel estimation method in described step (4) is specifically as follows:
    (4.1)建立接收导频符号向量z 0与信道脉冲响应h之间模型，满足：z 0＝Λh+η 0，其中，η 0为噪声项，Λ是M p×N k维度的变换矩阵，M p和N k分别表示导频符号数量和信道抽头数，
    Figure PCTCN2021109272-appb-100012
    其中行向量λ m定义为     (4.1) Establish a model between the received pilot symbol vector z 0 and the channel impulse response h, satisfying: z 0 =Λh+η 0 , wherein, η 0 is a noise term, and Λ is a transformation matrix of M p ×N k dimensions, M p and N k represent the number of pilot symbols and the number of channel taps, respectively,
    Figure PCTCN2021109272-appb-100012
    where the row vector λ m is defined as
    Figure PCTCN2021109272-appb-100013
    Figure PCTCN2021109272-appb-100013
    其中
    Figure PCTCN2021109272-appb-100014
    s m,0表示preamble中第m个子载波上的pilot符号，L表示原型滤波器的基带 符号长度，g是原型滤波器函数，l表示长度为L的原型滤波器中的第l个点的下标，k表示信道的第k个抽头下标，m和p分别为第m和第p个子载波，M是总子载波数；     in
    Figure PCTCN2021109272-appb-100014
    s m, 0 represents the pilot symbol on the mth subcarrier in the preamble, L represents the baseband symbol length of the prototype filter, g is the prototype filter function, l represents the lower point of the lth point in the prototype filter with a length of L where k represents the kth tap subscript of the channel, m and p are the mth and pth subcarriers respectively, and M is the total number of subcarriers;
    (4.2)采用加权最小二乘方法估计得到信道时域脉冲响应
    Figure PCTCN2021109272-appb-100015
    其中
    Figure PCTCN2021109272-appb-100016
    为噪声η 0的协方差矩阵，Λ H表示Λ的共轭转置。     (4.2) Using the weighted least squares method to estimate the channel time domain impulse response
    Figure PCTCN2021109272-appb-100015
    in
    Figure PCTCN2021109272-appb-100016
    is the covariance matrix of noise η 0 , and Λ H represents the conjugate transpose of Λ.
PCT/CN2021/109272 2021-05-08 2021-07-29 Signal detection method for underwater acoustic fbmc system WO2022236969A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110500461.5 2021-05-08
CN202110500461.5A CN113259295B (en) 2021-05-08 2021-05-08 Signal detection method for underwater acoustic FBMC system

Publications (1)

Publication Number Publication Date
WO2022236969A1 true WO2022236969A1 (en) 2022-11-17

Family

ID=77222183

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/109272 WO2022236969A1 (en) 2021-05-08 2021-07-29 Signal detection method for underwater acoustic fbmc system

Country Status (2)

Country Link
CN (1) CN113259295B (en)
WO (1) WO2022236969A1 (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114095321B (en) * 2021-10-27 2023-04-07 华南理工大学 Hyperbolic frequency modulation-based communication perception integrated design method for capsule endoscope
CN116016076A (en) * 2021-12-23 2023-04-25 江苏科技大学 Filter bank combined subcarrier multidimensional index modulation underwater acoustic communication system and method
CN114553647B (en) * 2022-01-12 2023-09-05 扬州大学 High-precision Doppler frequency offset estimation method based on FBMC-OQAM system under high dynamic environment
CN114337982B (en) * 2022-03-16 2022-07-15 深圳市华普微电子有限公司 Chirp pilot signal generation and time-frequency estimation method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080240308A1 (en) * 2007-03-27 2008-10-02 Samsung Electronics Co. Ltd. Apparatus and method for estimating channel using sliding windows in a broadband wireless communication system
CN102185822A (en) * 2011-05-31 2011-09-14 电子科技大学 OFDM/OQAM (Orthogonal Frequency Division Multiplexing/Offset Quadrature Amplitude Modulation) system and time frequency synchronization method thereof
CN107426120A (en) * 2017-07-24 2017-12-01 哈尔滨工程大学 A kind of underwater sound OFDM MFSK channel equalization methods based on least mean-square error
CN109474304A (en) * 2018-11-29 2019-03-15 中国海洋大学 Adaptive multi-standard underwater sound communication system and method

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8660216B2 (en) * 2010-05-29 2014-02-25 Marvell International Ltd. Method and apparatus for mitigating the residual CFO effect on channel estimation for OFDM receivers
CN102546511B (en) * 2012-01-12 2014-05-07 浙江大学 Underwater sound communication device and method based on time reversal and orthogonal frequency division multiplexing (OFDM) combined treatment
WO2016092323A1 (en) * 2014-12-11 2016-06-16 University Of Surrey Estimating data symbols from a filter bank multicarrier (fbmc) signal
CN110581813B (en) * 2019-08-12 2020-10-30 华中科技大学 Method for transmitting pilot signal of multi-carrier system
CN111740934B (en) * 2020-05-21 2022-08-19 江苏科技大学 Underwater sound FBMC communication signal detection method based on deep learning
CN112003809B (en) * 2020-10-28 2021-01-29 鹏城实验室 Multi-carrier underwater acoustic communication Doppler estimation and compensation method based on motion platform

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080240308A1 (en) * 2007-03-27 2008-10-02 Samsung Electronics Co. Ltd. Apparatus and method for estimating channel using sliding windows in a broadband wireless communication system
CN102185822A (en) * 2011-05-31 2011-09-14 电子科技大学 OFDM/OQAM (Orthogonal Frequency Division Multiplexing/Offset Quadrature Amplitude Modulation) system and time frequency synchronization method thereof
CN107426120A (en) * 2017-07-24 2017-12-01 哈尔滨工程大学 A kind of underwater sound OFDM MFSK channel equalization methods based on least mean-square error
CN109474304A (en) * 2018-11-29 2019-03-15 中国海洋大学 Adaptive multi-standard underwater sound communication system and method

Also Published As

Publication number Publication date
CN113259295A (en) 2021-08-13
CN113259295B (en) 2022-07-15

Similar Documents

Publication Publication Date Title
WO2022236969A1 (en) Signal detection method for underwater acoustic fbmc system
WO2019119778A1 (en) Anti-interference signal detection and synchronization method for wireless broadband communication system
CN107438038B (en) Pilot design and synchronous channel estimation method of FBMC/OQAM
CN101771657B (en) Multicarrier underwater acoustic communication method
CN101414990B (en) Method for capturing carrier frequency bias and time delay of single carrier frequency domain equalizing system
CN102185822B (en) OFDM/OQAM (Orthogonal Frequency Division Multiplexing/Offset Quadrature Amplitude Modulation) system and time frequency synchronization method thereof
CN102377712A (en) Method and apparatus for detection on secondary synchronization signal
CN105516045B (en) A kind of OFDM training sequence structures and synchronous method
CN113259291B (en) Phase compensation method realized by dynamic Doppler tracking of underwater sound continuous signals
CN103905351A (en) Method and system for DVB-T2 channel estimation
CN107086974A (en) A kind of OFDM synchronous method under high dynamic environment
CN101729479B (en) Blind channel estimation method based on cyclostationarity of OFDM signals
CN104796372A (en) LTE (long term evolution) system downward auxiliary synchronous signal detection method with low implementation complexity
CN106330251B (en) Underwater sound communication system doppler spread estimation method based on zero correlation band sequence
James et al. Channel estimation for OFDM systems
CN103078819B (en) Fine symbol timing synchronization method and device thereof
CN102065035B (en) Channel estimation method of multi-band orthogonal frequency-division multiplexing ultra-wideband system
CN107959649B (en) Symbol timing synchronization simplification method in OFDM system receiver
Wang et al. Low-complexity carrier frequency offset estimation algorithm in TD-LTE
CN106101042B (en) A kind of CFO estimation method based on more noises
CN111800366B (en) OFDM symbol timing synchronization method and receiving device under complex multipath environment
CN107276654B (en) Signal processing method and system
CN102821078B (en) Blind channel estimation method based on OFDM (orthogonal frequency division multiplexing) signal cyclostationary features
CN101431492B (en) Method for timing estimation of OFDM communication system signal
CN110691055A (en) Time-frequency offset joint estimation method in OQAM/OFDM

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: 21941538

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE