MC-CDMA系统中的PPIC多用户接收机研究

研究了上行链路MC-CDMA系统中的部分干扰抵消多用户接收机方案,提出了初级采用线性MMSE检测的多级PIC多用户接收机。为了减小初级判决不精确带来的误差传播,每级乘以一个不同的部分相消因子以实现部分消除。在不考虑信道估计误差的情况下,仿真了不同用户数下该方案与传统解相关和最小均方误差多用户检测的性能比较。仿真结果表明提出的PPIC接收机具有更好的误码率性能。     

WCDMA系统中的差分最小均方误差多用户检测算法研究

针对现有自适应多用户检测算法需要训练比特、不能适应快衰落信道的问题,提出适用于WCDMA系统的基于差分最小均方误差（DMMSE）准则的自适应多用户检测算法。该算法结合二级扩频体制的特点,根据具有相同综合扩频码的相邻接收符号幅度变化的比率变化自适应调整横向滤波器的权系数,由于该方法更能刻画多址干扰的变化规律,因此可有效地抑制多址干扰。仿真结果表明,在多普勒频移的传播环境下,DMMSE算法的误码性能优于现有自适应MMSE算法。     

Efficient groupwise multiuser detection with iterative soft interference cancellation for multi-rate MC-CDMA

This paper presents an effective multiuser detector (MUD) for the uplink of multi-rate multi-carrier code division multiple access (MC-CDMA) systems. The MUD considered is an iterative receiver which utilizes the soft information to refine the estimation of the interference to enhance the interference cancellation capability. More specifically, users are first classified into separate groups according to their transmission rates. In each iteration, these groups of users are detected sequentially based on a set of group detectors with the removal of multiple access interferences (MAI) group by group, where the estimated interferences in each group, incurred from the users either in the same rate group or in the other rate groups, are refined successively with the assistance of the soft information in the symbol detection process. Moreover, for practical low-complexity implementations, the users in each group are further partitioned into smaller subgroups based on their effective channel correlations and then detected in parallel by a bank of minimum mean-squared error (MMSE) soft detectors to further reduce the computational load. Conducted simulations show that the proposed MUD can effectively suppress the MAI to render superior performance but with reduced computations compared to previous works.     

CDMA系统中基于SVM的多用户检测算法

针对多用户检测的特点将支持向量机（SVM）与多用户检测相结合,构成一种多用户检测算法,应用在异步DS—CDMA系统。该算法能用较少的训练样本获得更佳的分类器,因此它的推广能力较好．且对训练要求的样本数也大大下降。仿真结果表明：该算法的误码性能优于传统匹配滤波检测器以及MMSE多用户检测器。     

A novel initial ranging algorithm for OFDMA systems using smart antenna

To improve the capacity of initial ranging （IR） users simultaneously accessing into network and enhance the performance of the IR, a novel IR algorithm is presented for the orthogonal frequency division multiple access systems with the smart antenna. The beamforming weight vectors provided by the antenna array are introduced for the IR users, and the procedures of the multiuser parameter estimation and the interference cancellation are designed. The proposed method can improve the accuracy of the channel estimation of active paths and efficiently reduce the residual multiple access interference. Simulation results show that the proposed method has a much better performance than the available successive multiuser detection and interference cancellation algorithm and accommodates more active ranging users simultaneously accessing into a cell.     

Simplified MMSE Detectors for Turbo Receiver in BICM MIMO Systems

In this article, two methods adopting simplified minimum mean square error (MMSE) filter with soft parallel interference cancellation (SPIC) are discussed for turbo receivers in bit interleaved coded modulation (BICM) multiple-input multiple-output (MIMO) systems. The proposed methods are utilized in the non-first iterative process of turbo receiver to suppress residual interference and noise. By modeling the components of residual interference after SPIC plus the noise as uncorrelated Gaussian random variables, the matrix inverse for weighting vector of conventional MMSE becomes unnecessary. Thus the complexity can be greatly reduced with only slight performance deterioration. By introducing optimal ordering to SPIC, performance gap between simplified MMSE and conventional MMSE further narrows. Monte Carlo simulation results confirm that the proposed algorithms can achieve almost the same performance as the conventional MMSE SPIC in various MIMO configurations, but with much lower computational complexity.     

Performance analysis of multicell multiuser MIMO MC DS/CDMA system with MMSE user-ordered SIC technique

In this paper, we study the performance of co-channel interference (CCI) cancellation technique based on minimum mean square error (MMSE) user-ordered successive interference cancellation (UOSIC) for multi input multi output (MIMO) multicarrier (MC) direct sequence code division multiple access (DS/CDMA) system for multi cell setting. Though MC DS/CDMA possesses several advantages, interference limits the capacity of the system. Mitigating the interference can directly swell the system capacity. In this contribution, we consider K co-channel users arbitrarily distributed in an L cell uplink and downlink communication system with both the base stations (BSs) and the mobile stations (MSs) equipped with two antennas each. Also, we assume that, both the BSs and MSs employ space time block code (STBC) based on the transmission matrix (G2). Our simulation study shows that, MMSE UOSIC provides better achievable bit-error rate (BER) than MMSE maximum likelihood (ML) and MMSE successive interference cancellation (SIC) technique due to optimal ordering among users and iterative interference cancellation (IC).     

WCDMA系统中的差分最小均方误差多用户检测算法

宽带码分多址(Wideband code division multiple access,WCDMA)系统中现有的自适应多用户检测算法需要训练比特,不能适应快衰落信道。本文根据WCDMA系统中特有的二级扩频体制的特点,提出基于差分最小均方误差(Differential minimum mean square error,DMMSE)准则的自适应多用户检测算法。该算法根据相邻接收符号幅度变化的比率变化自适应调整横向滤波器的权系数,从而不需要通过训练比特开销跟踪信道状态信息的方式来抑制多址干扰。仿真结果表明,在存在多普勒频移的传播环境下,DMMSE算法的误码性能优于现有的自适应最小均方误差(Minimum mean square error,MMSE)算法。     

基于子空间跟踪的修正MMSE多用户检测算法

在研究基于子空间跟踪的最小均方误差（MMSE）多用户检测器（MUD）的基础上，为解决原算法因引入特征值估计误差而导致检测性能下降的问题，设计了一种改进的算法——修正的MMSE多用户检测器．采用正交性能优良的OPAST算法跟踪子空间，提出一种基于OPAST的修正MMSE多用户检测算法．仿真结果显示,与基于OPAST的MMSE MUD相比，基于OPAST的修正MMSE MUD算法收敛速度快，输出信干比和误码率性能逼近SVD MUD算法，并且计算复杂度较低．     

非理想信道条件下的V-BLAST系统检测算法

针对信道相关条件下同时存在信道估计误差的V-BLAST系统,提出了一种基于MMSE准则的检测算法,推导了接收均衡矩阵的计算式,同时也给出了相应的最佳排序策略。仿真结果表明,本文提出的算法比传统算法以及仅考虑信道估计误差影响的算法性能更优,当信噪比较高以及收发天线数目较多时,优势更明显。该算法要求接收端已知信道估计误差的方差以及收发天线之间的相关矩阵等参数信息,但系统性能对这些参数的估计值有较强的鲁棒性,当参数估计不准确时,该算法性能优势仍然可以在一个较大的容错范围内得以保持。     

Two-stage cascaded classification approach based on genetic fuzzy learning for speech/music discrimination

Automatic discrimination of speech and music is an important tool in many multimedia applications. The paper presents a robust and effective approach for speech/music discrimination, which relies on a two-stage cascaded classification scheme. The cascaded classification scheme is composed of a statistical pattern recognition classifier followed by a genetic fuzzy system. For the first stage of the classification scheme, other widely used classifiers, such as neural networks and support vector machines, have also been considered in order to assess the robustness of the proposed classification scheme. Comparison with well-proven signal features is also performed. In this work, the most commonly used genetic learning algorithms (Michigan and Pittsburgh) have been evaluated in the proposed two-stage classification scheme. The genetic fuzzy system gives rise to an improvement of about 4% in the classification accuracy rate. Experimental results show the good performance of the proposed approach with a classification accuracy rate of about 97% for the best trial.     

Subcarrier and power allocation for OFDMA-based multicast cellular networks using a coalitional game

The subcarrier and power allocation problem for orthogonal frequency-division multiple access (OFDMA)-based multicast cellular networks is investigated in this paper. Due to the intrinsic heterogeneity of the channel gains experienced by different multicast users, the conventional multicast schemes are highly conservative and spectrally inefficient. To address this issue, a novel multicast formulation is proposed where multiuser diversity inherent in OFDMA can be adaptively exploited by clustering users within a multicast group into smaller subgroups based on their channel gains. Subcarriers and power are then dynamically allocated to these subgroups to maximize the total multicast rate of the system. Coalitional game theory is adopted to model the group formation in which users can autonomously form coalitions with other users to compete for network resources. A low-complexity algorithm is proposed for the multicast coalitional game to reach multicoalitional equilibrium in which a sub-optimal performance can be obtained. Simulation results demonstrate that the proposed scheme outperforms the conventional unicast and multicast schemes while achieving a sub-optimal performance comparable to the exhaustive search scheme.     

上行多用户MIMO系统检测算法及抗远近效应研究

上行多用户MIMO通信系统必须抑制用户间的共信道干扰,本文提出了一种基于最大信号干扰噪声比准则的多用户干扰抑制算法,并且给出了其在存信在道估计误差时的改进算法。与BD算法相比,该算法考虑了噪声的影响,性能有较大的提升。然后提出了一种信号重组方法将Alamouti空时编码用于上行多用户系统中,以进一步提升系统的检测性能。最后针对存在远近效应时算法的局限性,提出了一种分组优化算法,以保证各用户均能达到所期望的容量。仿真结果表明本文所提算法具有优异的性能。     

MMSE SQRD based SISO detection for coded MIMO-OFDM systems

In this paper, we propose a minimum mean square error （MMSE） sorted QR decomposition （SQRD） based soft-input soft-output （SISO） detection algorithm for coded multiple-input multiple-output orthogonM frequency division multiplexing （MIMO-OFDM） systems. The proposed detection is derived from the SISO MMSE detection, which is a popular detection strategy for iterative receivers. For each transmitted symbol in the proposed detection, a soft successive interference cancellation （SIC） is performed based on a posteriori probabilities of past detected symbols. Simulation results show that the proposed detection, while needing less computationM efforts, achieves significant performance gain compared with the SISO MMSE detection.     

规则低密度校验码的比特翻转解码算法研究

本文提出一种规则低密度校验码的比特翻转迭代解码算法。在解码算法的每一次迭代运算过程中,解码运算可以从总体上分为两个阶段：首先,满足可靠性要求的校验节点从与其相邻接的信息节点中选择一个信息比特作为翻转候选比特;然后,解码算法采用投票的方法对于这些候选翻转比特进行进一步的筛选。本算法由于对于最终翻转比特的选择结果是通过两次筛选而得到的,从而极大地降低了误翻的概率,加快了迭代解码算法的收敛速度,提高了系统的性能。另外,在第一阶段的比特选择过程中,我们综合校验节点所提供的校验检测信息和信道输出所提供的可靠性信息,提出了新的翻转比特选择标准。仿真结果表明,本文所提出的解码算法有着较好的性能,在解码运算复杂度和纠错性能之间提供了另外一个均衡。     

Polarimetric multiple-radar architectures with distributed antennas for discriminating between radar targets and deception jamming

The potential abilities of polarimetric multiple-radar architectures with distributed antennas in electronic counter-countermeasures (ECCMs) are analyzed based on differences between target echoes and deception jamming in scattering coefficients properties. According to the differences in generation mechanism between the target scatter and jamming, a signal model of real targets and deception jamming for two-dimensional vector sensors is established. By applying the Neyman-Pearson (NP) criterion and a hybrid NP-Maximum Posteriori (MAP) criterion, a two-stage detection/discrimination algorithm is proposed. In our proposed algorithm, the targets and jamming are detected in the first stage from a Gaussian noise background. Then, a discriminator based on the generalized likelihood ratio test is designed by exploiting the property of polarization discrimination. A theoretical analysis is also given to evaluate the performance of the second stage of the algorithm. Moreover, the discrimination performance can be improved by optimizing the transmitter polarization parameters. Simulation results indicate that the proposed ECCM strategy is effective against DRFM jamming, especially when JNR is high and transmit polarization is optimally designed.     

基于信号子空间的MMSE盲多用户检测

修正的最小均方误差(MMSE)多用户检测器解决了传统MMSE因引入特征值估计误差而导致检测性能下降的问题。基于修正MMSE多用户检测器,采用正交性能良好的改进PASTd算法跟踪子空间,提出一种基于改进PASTd的修正MMSE多用户检测算法,改进后的MMSE盲多用户检测误码率低,收敛速度快。仿真结果验证了该算法的有效性和可行性。     

基于卷积神经网络的左心室超声图像特征点定位

针对传统级联卷积神经网络（CNN）在左心室超声图像中定位准确度较低的问题，提出一种融合更快速区域卷积神经网络（Faster-RCNN）模型提取区域的级联卷积神经网络，实现对超声图像中左心室心内膜和心外膜轮廓特征点的定位。首先，采用两级级联的方式改进传统级联卷积神经网络的网络结构，第一级网络利用一个改进的卷积网络粗略定位左心室心内膜和心外膜联合的特征点，第二级网络使用四个改进的卷积网络分别对心内膜特征点和心外膜特征点进行位置微调，之后定位输出左心室心内膜和心外膜联合的轮廓特征点位置；然后，将改进的级联卷积神经网络与目标区域提取融合，即利用Faster-RCNN模型提取包含左心室的目标区域并将目标区域送入改进的级联卷积神经网络；最后，由粗到细对左心室轮廓特征点进行定位。实验结果表明，与传统级联卷积神经网络相比，所提方法在左心室超声图像上的定位效果更好，更逼近真实值，在均方根误差的评价标准下，特征点定位准确度提升了32.6个百分点。     

Accurate signal detection for BPSK-OFDM systems in time-varying channels

Binary phase shift keying (BPSK) orthogonal frequency division multiplexing (OFDM) has been adopted in the fourth-generation wireless communications. However, the real-valued property of BPSK-OFDM signals has not yet been utilized for channel equalization. In this work, this property is employed to develop a new equalizer based on the minimum mean square error (MMSE) criterion, leading to considerable enhancement in signal detection. In particular, the proposed equalizer is able to achieve 3 dB improvement over the existing MMSE equalizer in large signal-to-noise ratio region, and is more robust against channel estimation errors. Moreover, successive detection technique employed in the devised method is able to make use of diversity in time-varying channels, resulting in further performance enhancement. Additionally, the computational complexity of the proposed approach is analyzed. Simulation results are included to illustrate its performance superiority and computational attractiveness.     

Multiuser detection algorithm based on belief propagation in multiple-input multiple-output systems

In MIMO (multiple-input, multiple-output) systems, signals from differenttransmitting antennas interfere at each receiving antenna and multiuser detection (MUD)algorithms may be adopted to improve the system performance. This paper proposes anovel multiuser detection algorithm in MIMO systems based on the idea of "beliefpropagation" which has achieved great accomplishment in decoding of low-densityparity-check codes. The proposed algorithm has a low computation complexityproportional to the square of transmitting/receiving antenna number. Simulation resultsshow that under low signal-to-noise ratio (SNR) circumstances, the proposed algorithmoutperforms the traditional linear minimum mean square error (MMSE) detector while itencounters a "floor' of bit error rate under high SNR circumstances. So the proposedalgorithm is applicable to MIMO systems with channel coding and decoding. Although inthis paper the proposed algorithm is derived in MIMO systems, obviously it can be appliedto ordinary code-division m