基于校验孪生神经网络的有线通信网络故障点定位方法研究

由于有线通信网络中设备或者链路的承受能力有限,一旦发生故障将带来严重的负面影响,研究基于校验孪生神经网络的有线通信网络故障点定位方法。采集有线通信网络中的正常流量和故障流量数据,对采集到的流量数据进行清洗、变换等预处理操作,构建一个自校验的孪生神经网络模型,输入流量数据后故障点定位结果。实验结果表明,设计方法下有线通信网络故障点定位结果的准确度为98.65%,证实该方法的定位精度。

校本教研中实施网络教研的意义与实施措施

信息技术在课堂教学中应用,以及网络教研已成为当前学校课堂教学改革和校本教研的主要研究方向。为了提高信息技术在我校教学和教研中的应用。我校教师在教研处的组织下对信息技术在我校教学和教研中的应用进行了深入的探讨和研究。本文主要讨论在学校实施校本教研时,网络手段在校本教研中的应用,网络教研对于校本教研的意义以及在校本教研中实现网络教研的具体实施措施。

基于CNN与多尺度特征融合的城市交通流预测模型

随着出租车和网约车的日益普及,GPS数据生成大量的时空视频流数据,对城市交通流预测提供坚实的数据价值。传统城市流量预测方法存在精度低,目标区域受周围区域影响等问题。卷积神经网络在交通流预测上表现出色,但仍存在目标区域受全局信息的干扰、低层网络的特征表征能力弱及高层下采样损失过多特征等问题。该文提出一种基于卷积神经网络(Convolutional Neural Network,CNN)与多尺度融合机制的交通流预测模型MS-RSCNet(Multi-scale Residual Self Checking Network)。该模型采用了一种残差自校验网络(Residual Self Checking Network,RSCNet)结构,并引入融合多尺度特征的双向门控循环单元设计方案。通过公开数据集对交通流预测性能进行测试验证,相较于ST-ResNet、ARIMA、STAR等模型,MS-RSCNet模型具有更优的交通流预测性能。

Modified Benes network architecture for WiMAX LDPC decoder

A modified Benes network is proposed to be used as an optimal shuffle network in worldwide interoperability for microwave access （WiMAX） low density parity check （LDPC） decoders, When the size of the input is not a power of two, the modified Benes network can achieve the most optimal performance. This modified Benes network is non-blocking and can perform any sorts of permutations, so it can support 19 modes specified in the WiMAX system. Furthermore, an efficient algorithm to generate the control signals for all the 2 × 2 switches in this network is derived, which can reduce the hardware complexity and overall latency of the modified Benes network. Synthesis results show that the proposed control signal generator can save 25.4% chip area and the overall network latency can be reduced by 36. 2%.

Practical Polar Code Construction For Degraded Multiple-Relay Networks

In this paper, based on the characteristics of polar codes, a new decode-and-forward strategy called generalized partial information relaying protocol is proposed for degraded multiple-relay networks with orthogonal receiver components(MRNORCs). In such a protocol, with the help of partial information from previous nodes, each relay node tries to recover the received source message and re-encodes part of the decoded message for transmission to satisfy the decoding requirements for the following relay node or the destination node. In order to construct practical polar codes, the nested structures are developed based on this protocol and the information sets corresponding to the partial messages forwarded are also calculated. The proposed scheme is proved to be capable of achieving the theoretical capacity of the degraded MRN-ORCs while still retains the low-complexity feature of polar codes. We perform simulations to testify the practicability of the proposed scheme and compare polar codes by using successive-cancellation list decoder(SCLD) with traditional low-density parity-check(LDPC) codes. The results show that the obtained polar codes provide significant gain.

Construction of Quasi-Cyclic Low-Density Parity-Check Codes for Simplifying Shuffle Networks in Layered Decoder

Offset Shuffle Networks(OSNs) interleave a-posterior probability messages in the Block Row-Layered Decoder(BRLD) of QuasiCyclic Low-Density Parity-Check(QC-LDPC)codes.However,OSNs usually consume a significant amount of computational resources and limit the clock frequency,particularly when the size of the Circulant Permutation Matrix(CPM)is large.To simplify the architecture of the OSN,we propose a Simplified Offset Shuffle Network Block Progressive Edge-Growth(SOSNBPEG) algorithm to construct a class of QCLDPC codes.The SOSN-BPEG algorithm constrains the shift values of CPMs and the difference of the shift values in the same column by progressively appending check nodes.Simulation results indicate that the error performance of the SOSN-BPEG codes is the same as that of the codes in WiMAX and DVB-S2.The SOSNBPEG codes can reduce the complexity of the OSNs by up to 54.3%,and can improve the maximum frequency by up to 21.7%for various code lengths and rates.

JOINT LLR-CRC ERROR MITIGATION FOR TWO-WAY ADAPTIVE DENOISE-AND-FORWARD NETWORK CODING

Network Coding (NC) brings correlation between the coded signals from different sources, which makes the system more vulnerable to the decode error at relay. Conventional Cyclic Redundancy Code (CRC) has been implemented for error bit detection. However, its error correction is simply ignored. To fully exploit this feature, this paper proposes a novel joint Log-Likelihood Ratio (LLR) CRC error mitigation for NC two way relay channel. Specific thresholds are designed to estimate the error number of data block and identify those which can be recovered if the number is within the error correction scope of CRC. We examine two modes of the thresholds, one based on the average Bit Error Rate (BER) of source-relay link, while the other based on that of instantaneous one. We provide the full analysis for the Pair-wise Error Probability (PEP) performance of the scheme. A variety of numerical results are presented to reveal the superiority of the proposed scheme to conventional CRC NC under independent Rayleigh fading channels. Moreover, the efficiencies of the proposed thresholds are also validated.