Smart antenna has been regarded as one promising technology to enhance the performance of CDMA mobile communication systems. However, when applied to 3G systems, the performance of traditional adaptive arrays may degr...Smart antenna has been regarded as one promising technology to enhance the performance of CDMA mobile communication systems. However, when applied to 3G systems, the performance of traditional adaptive arrays may degrade due to code and time multiplexing in dedicated physical channels. A novel semiblind adaptive array approach is proposed to solve this problem. It overcomes the selfinterfering problem by introducing a quasidespreading technique for the control channel, and contains a timemultipexed structure to utilize both pilot symbols and unknown control symbols within the control channel. The blind part of the proposed approach is based on the competition of two parallel branches with different reference sequences. Simulation results show significant improvement can be achieved by the proposed approach in a fastfading WCDMA environment.展开更多
Contrary to the other multi-carrier modulation systems, the coherent optical orthogonal frequency division multiplexing communication system with an offset quadrature amplitude modulation (CO-OFDM-OQAM) possesses inhe...Contrary to the other multi-carrier modulation systems, the coherent optical orthogonal frequency division multiplexing communication system with an offset quadrature amplitude modulation (CO-OFDM-OQAM) possesses inherent imaginary interference (IMI). This has an important impact on the channel estimation process. Currently, a variety of frequency-domain channel estimation methods have been proposed. However, there are various problems that still exist. For instance, in order to reduce the influence of IMI, it is necessary to insert more guard intervals between the training sequence and the payload, leading to the occupation of excessive spectrum resources. In order to address this problem, this work designs a high spectral efficient frequency-domain channel estimation method for the polarization-division-multiplexing CO-OFDM-OQAM systems. First, the working principle of the proposed method is described in detail. Then, its spectral efficiency, power peak-to-average ratio, and channel estimation performance are studied based on simulations. The simulation results show that the proposed method improves the spectral efficiency without worsening the power peak-to-average ratio. The channel estimation capability of this method is verified in three scenarios of long-distance transmissions, including back-to-back, 100 km, and 200 km transmissions. .展开更多
针对当前广义频分复用(Generalized Frequency Division Multiplexing,GFDM)系统时变信道估计精度低的问题,提出基于稀疏贝叶斯学习的GFDM系统联合信道估计与符号检测算法.具体地,采用无干扰导频插入的GFDM多重响应信号模型,在稀疏贝叶...针对当前广义频分复用(Generalized Frequency Division Multiplexing,GFDM)系统时变信道估计精度低的问题,提出基于稀疏贝叶斯学习的GFDM系统联合信道估计与符号检测算法.具体地,采用无干扰导频插入的GFDM多重响应信号模型,在稀疏贝叶斯学习框架下,结合期望最大化算法(Expectation-Maximization,EM)和卡尔曼滤波与平滑算法实现块时变信道的最大似然估计;基于信道状态信息的估计值进行GFDM符号检测,并通过信道估计与符号检测的迭代处理逐步提高信道估计与符号检测的精度.仿真结果表明,所提算法能够获得接近完美信道状态信息条件下的误码率性能,且具有收敛速度快、对多普勒频移鲁棒性高等优点.展开更多
针对正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)水声通信中常用的相干和非相干通信分别面临的对多普勒敏感和频谱效率低的问题,提出一种高阶幅度键控调制的半相干通信技术,将OFDM符号时频帧结构中全部频点采用高...针对正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)水声通信中常用的相干和非相干通信分别面临的对多普勒敏感和频谱效率低的问题,提出一种高阶幅度键控调制的半相干通信技术,将OFDM符号时频帧结构中全部频点采用高阶幅度键控调制方式,并利用信号幅度信息完成半相干信道估计。通过两种基于深度学习的算法优化半相干信道估计这一非线性过程,较非相干通信有效提高了频谱效率,较一定信噪比下的相干通信提高了鲁棒性,降低了误比特率和系统复杂度,并利用元学习算法降低深度学习算法对训练数据的依赖。最后,提取海试信道数据,完成OFDM半相干水声通信系统仿真,验证了所提方法在频谱效率和系统误比特率性能方面较非相干和相干通信的优势,当信道长度改变时,基于元学习的算法依然可以获得较好的性能。展开更多
针对现有正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)系统信道估计和迭代检测算法中频谱效率低和鲁棒性差等问题,提出了一种基于酉近似消息传递和叠加导频的信道估计与联合检测方法。首先,在软调制/解调中叠加导频...针对现有正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)系统信道估计和迭代检测算法中频谱效率低和鲁棒性差等问题,提出了一种基于酉近似消息传递和叠加导频的信道估计与联合检测方法。首先,在软调制/解调中叠加导频对正交幅度调制的星座点进行预处理,检测时将叠加的导频作为频域符号的先验分布,利用置信传播算法进行调制和解调,实现检测模型的简化。然后,应用因子图-消息传递算法对OFDM传输系统和信道进行建模和全局优化,引入酉变换加强信道估计算法的鲁棒性。最后,建立OFDM仿真环境对现有方法进行仿真分析。仿真结果表明,相对于现有的独立导频类算法,所提算法能够以相同复杂度显著提升OFDM系统的频谱效率和鲁棒性。展开更多
正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)在现代通信系统中发挥着重要的作用,而信道估计是OFDM系统接收机的关键环节。针对传统导频信道估计算法需要大量的导频开销,降低信道带宽利用率,以及已有的盲信道估计算...正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)在现代通信系统中发挥着重要的作用,而信道估计是OFDM系统接收机的关键环节。针对传统导频信道估计算法需要大量的导频开销,降低信道带宽利用率,以及已有的盲信道估计算法估计性能差且算法复杂度高等不足,该文提出了一种基于改进聚类算法的盲信道估计算法。该算法在系统接收端把接收到信号看作一幅时频二维平面,通过使用时频窗口以迭代方式遍历该平面,在遍历过程中用改进的K-means算法对时频窗口内的符号数据做聚类分析,并根据符号先验信息以及信道时频相关性实现信道估计与均衡的目的。通过计算机仿真实验结果表明,改进K-means的盲信道估计算法的误比特率整体上低于基于简单线性预编码的盲估计算法,当信噪比大于10 dB时,改进K-means的盲信道估计算法的误比特率比子空间方法更低。在高信噪比情况下估计性能与最小均方误差(Minimum Mean Square Error,MMSE)估计算法相当。因此,相较于已有的盲信道估计算法,改进K-means的盲信道估计算法提高了估计精度,降低了算法复杂度。展开更多
针对脉冲噪声导致正交频分复用(orthogonal frequency division multiplexing,OFDM)水声通信系统误码率性能降低的问题,提出一种迭代脉冲噪声抑制与信道估计方法。首先,利用对称α稳定(symmetricαstable,SαS)分布对水下噪声进行建模,...针对脉冲噪声导致正交频分复用(orthogonal frequency division multiplexing,OFDM)水声通信系统误码率性能降低的问题,提出一种迭代脉冲噪声抑制与信道估计方法。首先,利用对称α稳定(symmetricαstable,SαS)分布对水下噪声进行建模,并通过实测噪声数据进行拟合验证。其次,基于脉冲噪声幅值较大的特性,利用切比雪夫不等式对基带接收信号实部和虚部分别进行脉冲噪声检测和抑制。同时,通过分析最小二乘(least squares,LS)信道估计算法估计误差的统计特性,提出自适应门限LS信道估计算法,减轻残余噪声对信道估计的影响。最后,结合估计的信道和译码结果重构噪声并进行脉冲噪声估计,实现迭代脉冲噪声抑制与信道估计。仿真结果表明,在不同程度脉冲噪声和实测噪声场景下,所提方法能够有效抑制脉冲噪声,显著降低OFDM系统误码率。展开更多
为了降低水声信道强多径效应、大多普勒频移等特性对水声通信系统的影响,提出一种基于极化码的正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)的水声通信系统。选择极化(Polar)码作为信道编码方案,通过量化信道输出直...为了降低水声信道强多径效应、大多普勒频移等特性对水声通信系统的影响,提出一种基于极化码的正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)的水声通信系统。选择极化(Polar)码作为信道编码方案,通过量化信道输出直接统计信道转移概率,替代通过信道估计得到的信道冲激响应,再计算转移概率,以降低通信系统的复杂度。在特性随时不变、时变、快时变的3种类型水声信道模型下,分别基于蒙特卡罗算法构造Polar码。仿真结果表明,Polar码在3种信道上的编码增益优于同码长同码率的低密度奇偶校验(Low Density Parity Check,LDPC)码和Turbo码,且信道条件越复杂,取得的编码增益越大,所提系统能够明显降低信道多径效应、多普勒频移对水声通信系统的扰乱,提高系统的可靠性。展开更多
文摘Smart antenna has been regarded as one promising technology to enhance the performance of CDMA mobile communication systems. However, when applied to 3G systems, the performance of traditional adaptive arrays may degrade due to code and time multiplexing in dedicated physical channels. A novel semiblind adaptive array approach is proposed to solve this problem. It overcomes the selfinterfering problem by introducing a quasidespreading technique for the control channel, and contains a timemultipexed structure to utilize both pilot symbols and unknown control symbols within the control channel. The blind part of the proposed approach is based on the competition of two parallel branches with different reference sequences. Simulation results show significant improvement can be achieved by the proposed approach in a fastfading WCDMA environment.
文摘Contrary to the other multi-carrier modulation systems, the coherent optical orthogonal frequency division multiplexing communication system with an offset quadrature amplitude modulation (CO-OFDM-OQAM) possesses inherent imaginary interference (IMI). This has an important impact on the channel estimation process. Currently, a variety of frequency-domain channel estimation methods have been proposed. However, there are various problems that still exist. For instance, in order to reduce the influence of IMI, it is necessary to insert more guard intervals between the training sequence and the payload, leading to the occupation of excessive spectrum resources. In order to address this problem, this work designs a high spectral efficient frequency-domain channel estimation method for the polarization-division-multiplexing CO-OFDM-OQAM systems. First, the working principle of the proposed method is described in detail. Then, its spectral efficiency, power peak-to-average ratio, and channel estimation performance are studied based on simulations. The simulation results show that the proposed method improves the spectral efficiency without worsening the power peak-to-average ratio. The channel estimation capability of this method is verified in three scenarios of long-distance transmissions, including back-to-back, 100 km, and 200 km transmissions. .
文摘针对当前广义频分复用(Generalized Frequency Division Multiplexing,GFDM)系统时变信道估计精度低的问题,提出基于稀疏贝叶斯学习的GFDM系统联合信道估计与符号检测算法.具体地,采用无干扰导频插入的GFDM多重响应信号模型,在稀疏贝叶斯学习框架下,结合期望最大化算法(Expectation-Maximization,EM)和卡尔曼滤波与平滑算法实现块时变信道的最大似然估计;基于信道状态信息的估计值进行GFDM符号检测,并通过信道估计与符号检测的迭代处理逐步提高信道估计与符号检测的精度.仿真结果表明,所提算法能够获得接近完美信道状态信息条件下的误码率性能,且具有收敛速度快、对多普勒频移鲁棒性高等优点.
文摘针对正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)水声通信中常用的相干和非相干通信分别面临的对多普勒敏感和频谱效率低的问题,提出一种高阶幅度键控调制的半相干通信技术,将OFDM符号时频帧结构中全部频点采用高阶幅度键控调制方式,并利用信号幅度信息完成半相干信道估计。通过两种基于深度学习的算法优化半相干信道估计这一非线性过程,较非相干通信有效提高了频谱效率,较一定信噪比下的相干通信提高了鲁棒性,降低了误比特率和系统复杂度,并利用元学习算法降低深度学习算法对训练数据的依赖。最后,提取海试信道数据,完成OFDM半相干水声通信系统仿真,验证了所提方法在频谱效率和系统误比特率性能方面较非相干和相干通信的优势,当信道长度改变时,基于元学习的算法依然可以获得较好的性能。
文摘针对现有正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)系统信道估计和迭代检测算法中频谱效率低和鲁棒性差等问题,提出了一种基于酉近似消息传递和叠加导频的信道估计与联合检测方法。首先,在软调制/解调中叠加导频对正交幅度调制的星座点进行预处理,检测时将叠加的导频作为频域符号的先验分布,利用置信传播算法进行调制和解调,实现检测模型的简化。然后,应用因子图-消息传递算法对OFDM传输系统和信道进行建模和全局优化,引入酉变换加强信道估计算法的鲁棒性。最后,建立OFDM仿真环境对现有方法进行仿真分析。仿真结果表明,相对于现有的独立导频类算法,所提算法能够以相同复杂度显著提升OFDM系统的频谱效率和鲁棒性。
文摘正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)在现代通信系统中发挥着重要的作用,而信道估计是OFDM系统接收机的关键环节。针对传统导频信道估计算法需要大量的导频开销,降低信道带宽利用率,以及已有的盲信道估计算法估计性能差且算法复杂度高等不足,该文提出了一种基于改进聚类算法的盲信道估计算法。该算法在系统接收端把接收到信号看作一幅时频二维平面,通过使用时频窗口以迭代方式遍历该平面,在遍历过程中用改进的K-means算法对时频窗口内的符号数据做聚类分析,并根据符号先验信息以及信道时频相关性实现信道估计与均衡的目的。通过计算机仿真实验结果表明,改进K-means的盲信道估计算法的误比特率整体上低于基于简单线性预编码的盲估计算法,当信噪比大于10 dB时,改进K-means的盲信道估计算法的误比特率比子空间方法更低。在高信噪比情况下估计性能与最小均方误差(Minimum Mean Square Error,MMSE)估计算法相当。因此,相较于已有的盲信道估计算法,改进K-means的盲信道估计算法提高了估计精度,降低了算法复杂度。
文摘针对脉冲噪声导致正交频分复用(orthogonal frequency division multiplexing,OFDM)水声通信系统误码率性能降低的问题,提出一种迭代脉冲噪声抑制与信道估计方法。首先,利用对称α稳定(symmetricαstable,SαS)分布对水下噪声进行建模,并通过实测噪声数据进行拟合验证。其次,基于脉冲噪声幅值较大的特性,利用切比雪夫不等式对基带接收信号实部和虚部分别进行脉冲噪声检测和抑制。同时,通过分析最小二乘(least squares,LS)信道估计算法估计误差的统计特性,提出自适应门限LS信道估计算法,减轻残余噪声对信道估计的影响。最后,结合估计的信道和译码结果重构噪声并进行脉冲噪声估计,实现迭代脉冲噪声抑制与信道估计。仿真结果表明,在不同程度脉冲噪声和实测噪声场景下,所提方法能够有效抑制脉冲噪声,显著降低OFDM系统误码率。
文摘为了降低水声信道强多径效应、大多普勒频移等特性对水声通信系统的影响,提出一种基于极化码的正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)的水声通信系统。选择极化(Polar)码作为信道编码方案,通过量化信道输出直接统计信道转移概率,替代通过信道估计得到的信道冲激响应,再计算转移概率,以降低通信系统的复杂度。在特性随时不变、时变、快时变的3种类型水声信道模型下,分别基于蒙特卡罗算法构造Polar码。仿真结果表明,Polar码在3种信道上的编码增益优于同码长同码率的低密度奇偶校验(Low Density Parity Check,LDPC)码和Turbo码,且信道条件越复杂,取得的编码增益越大,所提系统能够明显降低信道多径效应、多普勒频移对水声通信系统的扰乱,提高系统的可靠性。