基于深度神经网络的正交频分复用波形外辐射源雷达参考信号重构

针对正交频分复用(OFDM)波形外辐射源雷达的参考信号获取问题,基于"解调-再调制"的重构方法结合了波形优势,能获得更为纯净的参考信号.该文在此基础上提出一种联合OFDM解调、信道估计、信道均衡和星座点逆映射的深度神经网络(DNN)重构方法,建立了基于DNN的参考信号重构方案,通过网络学习自适应深度挖掘从时域接收符号到传输码元之间的映射关系、隐式地估计信道响应,从而提高解调精度和重构性能.该文首先研究了仿真数据集的获取问题、DNN的搭建和训练问题,接着对基于DNN方法在导频数目减少、循环前缀的移除、存在符号定时偏差、存在载波频偏、对高峰值平均功率比信号进行时域加窗滤波等情况下的参考信号重构性能进行了仿真分析,仿真结果表明该方法对参考信号重构的有效性.     

LTE-A物理层下行参考信号的研究

LTE-A（改进的长期演进）OFDM系统接收端采用相干解调技术恢复原始数据信息,相干解调必须利用信道的相关信息,获取信道相关信息接收端需采用先进的信道估计技术实现。LTE-A OFDM系统是基于导频信号的信道估计获取信道的脉冲响应特性。本文主要对LTE-A物理层下行CRS（小区特定参考信号）进行了深入的分析研究。     

基于虚导频和功控导频的OFDM信道估计

提出了一种基于虚导频和功控导频的OFDM信道估计方法。该方法通过选择可靠的解调数据符号作为虚导频，同时对导频采用功率控制，从而提高了信道估计的质量。计算机仿真结果显示，相对于传统方法，可以用较少的导频获得性能的改善。     

采用BICM-ID的PP-OFDM系统及后缀幅度优化

该文设计了一种用于正交频分复用(OFDM)系统的新型保护间隔信号-导频后缀(Pilot Postfix, PP),并构建了采用比特交织编码调制-迭代解码(BICM-ID)技术的PP-OFDM系统。PP由OFDM符号中的频域导频符号进行逆傅氏变换(IFFT)生成,在接收端可与OFDM符号中的导频相干合并,从而提高信道估计性能。该文给出了相应的信道估计、均衡和BICM-ID算法,并通过研究信道估值误差带来的接收信号有效信噪比变化,给出了PP幅度优化设计方法。仿真表明:选取适当的PP大小后,PP-OFDM比循环前缀/补零后缀-OFDM(CP/ZP-OFDM)的信道估计性能更好,误包率更低。     

基于信道分布空间的OFDM信道半盲估计方法

信道分布空间的构建将OFDM系统中发送信号的有限符号特性和信道响应的统计相关特性集中到信道分布平面中.算法使用了一个初始OFDM 导频符号在信道分布平面中建立信道频域响应参考曲线,信道的估计是利用相邻OFDM符号对应的信道响应有相关延续特性,通过在更新的信道分布平面上更新参考曲线来实现.计算机仿真表明,算法的性能和基于最小均方误差的信道估计算法相当,由于避免了矩阵求逆运算,该方法在工程实现中有较大的应用价值.     

基于信道估计的限幅OFDM信号检测性能分析

大峰均比是正交频分复用(OFDM)系统的主要缺陷之一，对时域OFDM信号限幅是通常采用的一种抑制发送信号峰值功率的方法，但是限幅产生的非线性干扰，使系统的信道估计和信号检测性能降低。本文分析了限幅干扰对信道估计以及基于估计信道信息的OFDM数据信号检测性能的影响，推导出了基于信道估计信号检测的信噪比与限幅参数之间的关系。利用信噪比和高斯信道的误符号率及误比特率公式，本文对频率选择性瑞利衰落信道环境限幅OFDM系统中信号的误符号率和误比特率进行估算。计算机仿真和估算的结果相吻合。     

OFDM系统中基于导频的时变信道估计

该文对OFDM系统中基于导频的时变信道估计进行了研究,采用分组导频对一个OFDM符号间隔内某些时刻的信道冲击响应值进行估计,提出一种利用低通插值估计一个符号间隔内其他时刻的信道值的方法。仿真结果表明该文采用的低通插值方法所得到的估计均方误差要小于用线性插值或直接用信道模型产生的系数进行插值所得到的估计均方误差。     

OFDM/OQAM系统中联合迭代信道估计和信号检测

与基于复数域空间正交条件的传统正交频分复用系统(OFDM with Cyclic Prefix, CP-OFDM)有所不同,基于交错正交调制的正交频分复用系统(OFDM/Offset QAM, OFDM/OQAM)满足实数域空间严格正交条件。因此在多径衰落信道条件下,CP-OFDM系统中的信道估计方法会导致OFDM/OQAM系统严重的字符间干扰和载波间干扰。该文结合OFDM/OQAM系统结构特点,提出了一种基于迭代信道估计和信号检测算法。该算法通过信道估计器和接收到的信号互相交换信息,消除导频序列中的字符间干扰和载波间干扰,提高信道估计和信号检测的准确度。仿真分析结果表明,经过一定次数迭代处理后,OFDM/OQAM迭代信道估计性能趋近于理想信道估计性能。     

一种符号重叠PCC-OFDM盲时偏估计算法

针对符号重叠PCC OFDM系统,提出了一种盲时偏估计算法。利用并行干扰删除方法消除系统固有的严重的符号间干扰,利用解调信号方差进行判决,获得精确的时偏估计。该算法不需要训练序列、导频信号或周期前缀。在高斯白噪声和Ricean信道下对算法进行了链路仿真,仿真结果证明,该算法能够有效地实现符号重叠PCC OFDM盲时偏估计。     

基于整体最小二乘的联合信道估计及OFDM信号检测算法

针对现有的联合信道估计及OFDM信号检测算法性能不高的问题,该文提出一种基于整体最小二乘的联合信道估计及OFDM信号检测算法。该算法首先通过导频估计初始的信道信息,在此基础上不断地采用整体最小二乘进行OFDM信号检测及信道估计,有效缓解迭代模型误差的影响,加快了算法迭代的收敛速度,提高了信道估计的精度,从而降低了OFDM系统的误码率。该文推导的信道估计克拉美罗界及仿真结果均表明所提出的算法在时变信道环境下优于现有的联合信道估计及OFDM信号检测算法。     

OFDM系统频偏自适应跟踪算法的研究

为了减小正交频分复用系统受频率偏差的影响,在OFDM系统中设计了一种新的载波频偏估计算法.频偏估计包括跟踪和捕获两个阶段,文中采用变步长自适应算法来进行跟踪,这个跟踪算法的捕捉范围是整个信号带宽的一半.通过此方法,不但可以利用前向纠错的方式来避免反馈,而且可以缩减复数乘法的次数以降低计算复杂度和系统功耗.该方法运算复杂度低,并具有较好的收敛性,能改善ODFM的解调性能.从计算机的仿真效果可以看出无论是在AWGN信道上还是在多径信道上都表现出了卓越的性能.     

Robust channel estimation and ISI cancellation for OFDM systems with suppressed features

A feature-suppressed orthogonal frequency-division multiplexing (OFDM) system and the corresponding channel estimation and intersymbol interference (ISI) mitigation techniques are investigated in this paper. Cyclic prefix (CP) and pilot tones, which are commonly used in civilian OFDM systems for ISI mitigation and channel estimation, create distinctive waveform features that can be easily used for synchronization and channel estimation purposes by intercepting receivers. As a result, CP and pilot tones are eliminated in the proposed feature suppressed OFDM system to reduce the interception probability. Instead, a set of specially designed OFDM symbols, driven by different pseudorandom sequences, are employed as preambles to avoid unique spectral signature. These preambles are inserted into the OFDM data symbol stream periodically and in a round-robin manner. In addition, a random frequency offset is introduced to each preamble to further mask the multicarrier signature. New challenges arising from these feature suppression efforts are studied, including robust channel estimation and demodulation techniques in the presence of frequency offset and severe interference. Based on our interference analysis, an iterative ISI and intercarrier interference (ICI) estimation-cancellation-based technique is proposed for both channel estimation and OFDM data demodulation. Our channel estimator performs joint frequency offset and channel impulse response estimation based on the maximum-likelihood (ML) principle. To reduce its complexity, we employ a number of techniques, which include approximation of the ML metrics, as well as fast Fourier transform pruning. The performances and feasibility of the proposed feature suppressed OFDM system and the channel estimator are analyzed and verified through numerical simulations.     

OFDM同步跟踪阶段的频偏估计算法

ＯＦＤＭ对栽频偏差是很敏感的，因此不仅在同步的捕获阶段，而且在同步的跟踪阶段，精确的频偏估计是很重要的。文献［３］提出了一种简化的相干判决反馈的迭代算法来估计同步跟踪阶段的载频偏差，对于１２８个子载波、ＱＰＳＫ调制的ＯＦＤＭ系统，在多径衰落信道下，当频偏小于子载波间隔的１５％时，频偏估计的标准方差小于１％，但是这个结果是在假设信道频率响应、定时偏差、载波相位等参数已知的条件下取得的，而当这些参数的估计存在误差时，由仿真结果可看出，频偏估计性能恶化了，因此，提出了采用差分判决反馈的频偏估计算法，虽然只能估计小于子载波间隔 ８％的频偏，但差分判决反馈算法不需要知道信道的参数，由于在同步跟踪阶段，频偏一般都较小，因此采用差分判决反馈算法可以很好地估计出同步跟踪阶段的频率偏差。     

一种OFDM系统同步参数盲估计方法

首先提出一种扩大周期平稳信号循环周期的方法，然后在此基础上提出一种基于信号周期平稳特性的盲同步参数快速估计算法，最后通过分析算法获得一种可以估计非整数倍样点周期时延和扩大频偏估计范围的方法，从而实现整个OFDM基带带宽范围内的频偏估计。理论分析和仿真结果表明，该算法仅用很少的OFDM符号就可以实现同步参数的估计、有好的抗噪特性、衰落信道适应性和高的估计精度。     

固定波形时间保护间隔OFDM系统的信道估计

针对固定波形时间保护间隔OFDM系统信号构成的特点,本文提出了一种利用系统中周期出现的两种伪随机序列(PN和pn)相互配合以完成信道估计和跟踪的算法。理论分析和计算机仿真结果表明,在相邻两个OFDM符号的信道特性变化较小的工作环境中,本算法能够很好的对信道状态信息(CSI)进行估计和跟踪,实现OFDM系统的信道均衡。     

Deep Learning for OFDM Channel Estimation in Impulsive Noise Environments

Impulsive noise suppression is essential in orthogonal frequency division multiplexing (OFDM) systems, since impulsive noise may cause a serious decline in channel estimation performance. To solve this problem, a channel estimator based on denoising autoencoder-deep neural network (DAE-DNN) is proposed in this paper. The proposed method is based on a data-driven deep learning framework. Firstly, DAE preprocesses signals to learn damaged data and recover the complete signal are used in the presence of impulsive noise. Then, the transmitted data processed by DAE are used to train the DNN in the offline training process. Finally, the estimated channel state information (CSI) is offered by the proposed DNN model in the online working process. The simulation results demonstrate that the proposed method improves OFDM channel estimation performance significantly. As expected, the proposed method has a better performance than existing ones, such as least squares, minimum mean square error and orthogonal matching pursuit algorithms. Moreover, the proposed method is robust under impulsive noise environments.

     

Burst coherent demodulation with combined symbol timing, frequencyoffset estimation, and diversity selection

A low-overhead burst coherent demodulation method that jointly estimates symbol timing and carrier frequency offset and then performs diversity selection is studied. It coherently demodulates individual bursts of TDMA (time division multiple access) symbols by operating solely on random data within the burst without requiring training sequences. Its performance is robust against frequency offset between transmitter and receiver, thereby eliminating the need for a highly stable frequency reference. The performance of this demodulation method in a fading channel can be further improved by using a diversity selection technique based on a quality measure derived as part of the joint timing/frequency offset estimation process. Simulations and experiments have confirmed that two-branch diversity using this method can provide reliable speech communication using TDMA with a transmission rate of 450 kb/s for a portable radio channel with an RMS delay spread of 555 ns or less     

Blind Sequential Monte Carlo Joint Tracking of Channel State and Frequency Offset in OFDM Systems

Orthogonal frequency division multiplexing (OFDM) is a multi-carrier modulation scheme, which has been adopted for several wireless standards. In order to fully exploit the benefits of an OFDM system, estimation of the channel state information must be performed. Moreover a well-known problem of OFDM is its sensitivity to frequency offset between the transmitted and received carrier frequencies. This frequency offset introduces inter-carrier interference in the OFDM signal. In this paper we address the problem of jointly tracking the channel and frequency offset based on a Sequential Monte Carlo filtering approach. The proposed algorithm works in a decision-directed way, thus does not require the use of pilot symbols, providing a worth-mentioned increase in the useful data rate. Through simulations we demonstrate the efficiency of this approach against a similar approach where the Extended Kalman Filter is used. Moreover our method is compared with two recently proposed pilot-based methods.     

A NOVEL INTEGER FREQUENCY OFFSET ESTIMATOR FOR OFDM

One of the principal disadvantages of Orthogonal Frequency Division Multiplexing （OFDM） is very sensitive to carrier frequency offset. The integer frequency offset has no effect on the orthogonality among the subcarriers, but it causes a circular shift and phase rotation of the received data symbols sequence, resulting in a Bit Error Rate（BER） of 0.5. In this paper, a novel integer frequency offset estimator for OFDM is derived based on maximum likelihood estimation technique and exploration of the differential relation between two consecutive OFDM data symbol sequences in frequency domain. Its performance is compared with the conventional method by computer simulations for the additive white Gaussian noise channel and a multipath fading channel. Simulation results show that the performance of the proposed estimator is better than the conventional estimator.     

一种基于OFDM信号周期平稳特性的系统时延和频偏的盲估计算法

该文首先提出一种基于信号周期平稳特性的OFDM系统时延和频偏的盲估计算法,然后分析算法,获得一种可以估计非整数倍样点周期的时延和扩大频偏估计范围,实现整个OFDM系统带宽范围内的频偏估计的方法,并指出本算法可以在信道未知的情况下获得好的估计性能,最后给出相应的仿真结果。理论分析和仿真结果表明,本算法有很好的抗噪特性、强的变化信道适应性,与同类方法比较,具有更高的估计精度、估计非整数倍样点周期时延和扩大频偏估计范围的能力。