The design and performance of a neural network for predicting turbodecoding error with application to hybrid ARQ protocols

It is shown that a neural network can be trained to observe the cross entropy of the outputs of component decoders in a turbo error control system and to accurately predict the presence of errors in the decoded data. The neural network can be used as a trigger for retransmission requests at either the beginning or the conclusion of the decoding process, providing improved reliability and throughput performance at a lower average decoding complexity than turbo decoding with cyclic redundancy check error detection     

基于神经网络广义逆两电机同步控制研究

基于神经网络广义逆系统,提出二自由度的内模控制。来改善两电机同步系统的解耦控制性能和鲁棒性能。提出先对原来系统的数学模型进行广义逆的存在性分析,进而推出原系统广义逆的数学模型,再用神经网络逼近广义逆,接在原系统前组成具有等价效果的伪线性系统,来实现系统的解耦线性化。有利于系统的综合。然后对广义伪线性系统引入二自由度内模控制,以保证系统的鲁棒稳定性。     

A neural network for predicting decoder error in turbo decoders

It is shown that a neural network can be trained to predict the presence of errors in turbo-decoded data. The inputs to the network are samples of the cross entropy of the component decoder outputs at two or more time instants. Such a neural network can be used as a trigger for retransmission requests at either the beginning or at the conclusion of the decoding process, providing improved reliability performance and lower average decoding complexity than turbo decoding with CRC error detection     

Neural network based uniformity profile control of linear chemical-mechanical planarization

In this paper, a neural network based uniformity controller is developed for the linear chemical-mechanical planarization (CMP) process. The control law utilizes the metrology measurements of the wafer uniformity profile and tunes the pressures of different air-bearing zones on Lam linear CMP polishers. A feedforward neural network is used to self-learn the CMP process model and a direct inverse control with neural network is utilized to regulate the process to the target. Simulation and experimental results are presented to illustrate the control system performance. Compared with the results by using statistical surface response methods (SRM), the proposed control system can give more accurate uniformity profiles and more flexibility.     

Statistical analysis of neural network modeling and identificationof nonlinear systems with memory

The paper presents a statistical analysis of neural network modeling and identification of nonlinear systems with memory. The nonlinear system model is comprised of a discrete-time linear filter H followed by a zero-memory nonlinear function g(.). The system is corrupted by input and output independent Gaussian noise. The neural network is used to identify and model the unknown linear filter H and the unknown nonlinearity g(.). The network architecture is composed of a linear adaptive filter and a two-layer nonlinear neural network (with an arbitrary number of neurons). The network is trained using the backpropagation algorithm. The paper studies the MSE surface and the stationary points of the adaptive system. Recursions are derived for the mean transient behavior of the adaptive filter coefficients and neural network weights for slow learning. It is shown that the Wiener solution for the adaptive filter is a scaled version of the unknown filter H. Computer simulations show good agreement between theory and Monte Carlo estimations     

BPNN辅助KF的MEMS陀螺仪数据处理方法

针对微机电系统(MEMS)陀螺仪数据误差建模不精确或无法给出模型的情况,提出了误差反馈（BP）神经网络辅助卡尔曼滤波对陀螺仪数据进行降噪处理的方法。分析卡尔曼滤波器的系统噪声方差Q矩阵可知,当模型不精确时可通过Q补偿。基于BP神经网络优化Q值原理,首先把采集到的MEMS陀螺仪数据输入卡尔曼滤波器得到Q;再把新息、滤波增益、量测噪声方差输入神经网络,把Q作为神经网络的输出,神经网络优化系统噪声协方差矩阵得到Q*;最后将Q*作为卡尔曼滤波算法系统噪声方差矩阵。实验结果表明,在建模不精确的情况下该方法也能有效提高陀螺仪的精度。     

Turbo processing for an OFDM-based MIMO system

We consider improving the overall system performance of an orthogonal frequency-division multiplexing-based multiple-input multiple-output (MIMO) wireless local area network system. We use a combined iterative detection/decoding and channel updating method, referred to herein as turbo processing, to improve performance. First, we improve a recently proposed list sphere decoder-based iterative MIMO soft-detector by constraining the value of the a priori information from a soft-in soft-out channel decoder. Second, we propose a channel updating scheme using the decoded packet data to improve the channel estimation accuracy. Simulation results show that turbo processing can be used to significantly improve the performance of the system considered.     

飞机自动着陆的一种非线性鲁棒控制器设计

将一种直接基于非线性模型的带神经网络补偿信号的逆系统方法用于具有强非线性和受不确定扰动干扰的飞机自动着陆控制,并对神经网络补偿的方式进行了改进。采用神经网络补偿动态逆反馈线性化后伪系统的逆误差,使得非线性系统在参数受到摄动或外部扰动的情况下仍能保持良好的控制效果。可以证明该方法在理论上是收敛的,对于任意的状态初值和给定的期望输出信号,能控制到指定的精度。神经网络的权值是局部收敛的,同时大量仿真表明,经过较少的神经网络离线训练,即能够获得很好的控制效果。最后通过飞机着陆下滑段的仿真验证表明,该方法具有强的鲁棒性和良好的跟踪精度。     

Turbo decoding for robust multiuser detection in impulsive noisechannels

A more robust version of the nonlinear M-decorrelating detector through iterative (turbo) decoding is proposed. The turbo code performance in the linear decorrelating detector is also studied. Simulation results indicate that combined nonlinear signal processing and turbo decoding can achieve highly robust performance in impulsive noise     

Near-optimal decoding for coded and power-controlled DS-CDMA downlink

We first present the traditional decoding approach that employs the common pilot-channel-based maximal ratio combining and the Viterbi or iterative decoding cannot achieve the optimal error-rate performance for downlink direct-sequence code-division multiple-access (CDMA) signals when a fast power control technique is applied together with a convolutional or turbo coding. Then, as an efficient method to realize a nearly optimal decoding, we propose a branch metric power readjustment (BMPR) technique, where the downlink power control command generated by the mobile station is used not only to adjust the base station power in the transmitter side, but also to readjust the decoder input branch metric power in the receiver side. Numerical results show that the BMPR technique applied to the IMT-2000 wideband-CDMA system can improve the transmit power utilization by up to 0.4 dB for the block-error rate of 10/sup -2/.     

Stochastic analysis of turbo decoding

This paper proposes a stochastic framework for dynamic modeling and analysis of turbo decoding. By modeling the input and output signals of a turbo decoder as random processes, we prove that these signals become ergodic when the block size of the code becomes very large. This basic result allows us to easily model and compute the statistics of the signals in a turbo decoder. Using the ergodicity result and the fact that a sum of lognormal distributions is well approximated using a lognormal distribution, we show that the input-output signals in a turbo decoder, when expressed using log-likelihood ratios (LLRs), are well approximated using Gaussian distributions. Combining the two results above, we can model a turbo decoder using two input parameters and two output parameters (corresponding to the means and variances of the input and output signals). Using this model, we are able to reveal the whole dynamics of a decoding process. We have discovered that a typical decoding process is much more intricate than previously known, involving two regions of attraction, several fixed points, and a stable equilibrium manifold at which all decoding trajectories converge. Some applications of the stochastic framework are also discussed, including a fast decoding scheme     

Accurate contour control of mechatronic servo systems usingGaussian networks

This paper presents a method of contour control of mechatronic servo systems by using neural networks. The neural network learns the inverse dynamics of the mechatronic servo system. The input data for the mechatronic servo systems are modified from objective trajectories by using the neural network. The Gaussian network is adopted to construct the inverse dynamics of the mechatronic servo system because the Gaussian function is well defined, and its structure and initial parameters can be systematically selected such that the initial network approximates the inverse dynamics of the mechatronic servo system. The actual input/output data of the mechatronic servo system are used for the learning of the Gaussian network. Effectiveness of the proposed method is assured by experimental results of contour control of an X-Y table     

一种改善Turbo乘积编码OFDM系统性能的新方法

本文提出一种利用对译码器软信息限幅来改善多径衰落信道中Turbo乘积编码OFDM(TPC-OFDM)系统性能的新方法。通过对不同多径衰落信道中QPSK映射和16QAM映射的TPC-OFDM系统性能的数值仿真，结果表明在10^-5误比特率下，这种新方法比传统的迭代译码大约有6～10dB的改进，对严重多径环境下TPC-OFDM系统的错误平底也有明显的改进。     

Reducing the power of wireless terminals by adaptive baseband processing

In this paper, we illustrate specific power savings obtained from exploiting a reconfigurable mobile terminal under the 3GPP LTE standard. Building on traditional link adaptation towards maximum throughput and extended towards minimal power consumption, we add two flexible baseband components: the turbo decoder and the multiple input multiple output (MIMO) detector. Optimizing their configuration leads to larger power savings when compared to non-flexible systems only performing link adaptation. The gain observed strongly depends on the scenario. For low-activity set-ups with a few minutes of voice per day, the idle power dominates and the active data rate is relatively low. This makes analog front-end and time-domain processing dominant given their constant power consumption while MIMO detection and turbo decoding that scale with data rate play a smaller role. Still, because of its ability to improve the system spectral efficiency and hence reduce its duty cycle, an advanced MIMO detector can save 10% in power consumption, on the condition that the network requires to use MIMO. Otherwise single input single output is more power-efficient in downlink. In high-throughput scenarios, larger gains are obtained. The flexible MIMO detector can save up to 35% of average power consumption. The turbo decoding also brings some gain, saving up to 12% of power when the full bandwidth is allocated to a single user.     

Joint turbo decoding and estimation of hidden Markov sources

We describe a joint source-channel scheme for modifying a turbo decoder in order to exploit the statistical characteristics of hidden Markov sources. The basic idea is to treat the trellis describing the hidden Markov source as another constituent decoder which exchanges information with the other constituent decoder blocks. The source block uses as extrinsic information the probability of the input bits that is provided by the constituent decoder blocks. On the other hand, it produces a new estimation of such a probability which will be used as extrinsic information by the constituent turbo decoders. The proposed joint source-channel decoding technique leads to significantly improved performance relative to systems in which source statistics are not exploited and avoids the need to perform any explicit source coding prior to transmission. Lack of a priori knowledge of the source parameters does not degrade the performance of the system, since these parameters can be jointly estimated with turbo decoding     

Iterative multiuser detection for turbo-coded FHMA communications

An iterative receiver structure Is proposed for turbo-coded frequency-hop multiple access (FHMA) systems. In FHMA systems, the adjacent channel interference (ACI) is the major contributor of multiple access interference (MAI) if orthogonal hopping patterns are used. The ACI is a function of the tone spacings of the adjacent subchannels and the rolloff factor of the pulse-shaping filter. The calculation of the ACI for a square-root raised-cosine pulse-shaping filter in an FHMA system is presented in this paper. In addition, a low complexity iterative multiuser detector is developed to mitigate the degradation caused by ACI in the FHMA systems. The iterative receiver structure is based on a modified turbo decoding algorithm which makes use of the a posteriori log-likelihood ratio (LLR) information of the systematic bits to obtain the a posteriori information of the turbo-encoded parity bits. Iterations of the receiver/decoder are used as the mechanism to estimate and mitigate the MAI in the FHMA system. The properties of both soft and hard interference suppressors based on the modified turbo decoding algorithm are examined and an efficient recursive implementation is derived. Compared to maximum-likelihood multiuser detection, the proposed system is more practical and its complexity is only a linear function of the number of users. Simulation results show that the proposed iterative receiver structure offers significant performance gain in bandwidth efficiency and the required signal-to-noise ratio (SNR) for a target bit-error rate (BER) over the noniterative receiver structure. Moreover, the single user performance can be achieved when imperfect power control exists     

Self-tuning adaptive neurocontroller for brushless DC motors

This paper describes a self-tuning adaptive neurocontroller for brushless DC motors. Nonlinear and unknown motor dynamics are identified by using a multilayer neural network and the control input for the motor is derived from the identified model. The effect of the load torque on the control system is damped by filtering the control input. Simulation and experimental results show that the self-tuning adaptive neurocontrol has a good tracking performance but needs an adaptive filter and a parallel PI controller in the case of disturbances.     

基于外信息收敛的Turbo码译码优化算法

目前Turbo码译码算法运算较复杂,无法满足LTE及LTE-Advanced的高速吞吐要求,因此,研究LTE环境下的Turbo码译码算法具有理论意义与工程价值。从降低译码延时考虑,评估各种迭代停止准则对性能及译码效率的影响,在分析外信息统计特性收敛的基础上,设计基于外信息收敛的双门限停止准则。仿真结果表明,改进的迭代停止准则在不影响译码性能的前提下,能较好地提高译码速率。     

压电陶瓷迟滞逆模型的前馈PID控制

为减小压带陶瓷迟滞特性对系统跟踪精度的影响,在Preisach模型的基础上对压电陶瓷迟滞性进行建模。借助Matlab软件对实验数据进行拟合和采用逆控制思想,在迟滞逆模型的基础上提出前馈PID(即比例、积分、微分)控制算法。实验结果表明,压电陶瓷最大迟滞性控制在2.2%内,输入、输出具有较好的线性关系,带前馈的PID控制具有良好的控制性能。     

Iterative turbo decoder analysis based on density evolution

We track the density of extrinsic information in iterative turbo decoders by actual density evolution, and also approximate it by symmetric Gaussian density functions. The approximate model is verified by experimental measurements. We view the evolution of these density functions through an iterative decoder as a nonlinear dynamical system with feedback. Iterative decoding of turbo codes and of serially concatenated codes is analyzed by examining whether a signal-to-noise ratio (SNR) for the extrinsic information keeps growing with iterations. We define a “noise figure” for the iterative decoder, such that the turbo decoder will converge to the correct codeword if the noise figure is bounded by a number below zero dB. By decomposing the code's noise figure into individual curves of output SNR versus input SNR corresponding to the individual constituent codes, we gain many new insights into the performance of the iterative decoder for different constituents. Many mysteries of turbo codes are explained based on this analysis. For example, we show why certain codes converge better with iterative decoding than more powerful codes which are only suitable for maximum likelihood decoding. The roles of systematic bits and of recursive convolutional codes as constituents of turbo codes are crystallized. The analysis is generalized to serial concatenations of mixtures of complementary outer and inner constituent codes. Design examples are given to optimize mixture codes to achieve low iterative decoding thresholds on the signal-to-noise ratio of the channel