复杂网络波动扩展衍射欠定采样预测模型

复杂网络环境中对网络波动的准确预测可以有效监测网络环境,防范网络入侵和拥堵。由于在复杂网络受到干扰的可能性更大,其网络波动具有扩展衍射特征,不可预测性强。传统方法中采用自回归移动平均模型进行复杂网络波形预测算法设计,在波动信号的时频重叠调制过程中未能纳入杂波先验信息,波动序列的扩展衍射特征形成欠定采样,预测效果不好。提出基于空间扩展自回归移动平均模型的复杂网络波动欠定预测算法,采用LTE线性均衡滤波,进行降噪去除杂波干扰,提取波动序列的扩展衍射特征形成欠定采样样本序列,设计网络波动时空序列扩展衍射点阵,准确预测网络波动的参数信息。以病毒入侵,网络监听和拥塞堵塞等波动产生模型为实例,进行仿真实验,结果表明该算法具有较高预测精度,监测点波动误差较小,实现复杂网络波动状态的动态跟踪和评估。

Complex Network Fluctuation Extension Diffraction Underdetermined Sampling Prediction Model

Complex network environment of network fluctuation can predict the effective monitoring network environment,to guard against network intrusion and congestion. Due to the disturbance of complex networks is more likely,the network fluctuation has extended the diffraction characteristic, unpredictability is strong. In the traditional method using autoregressive moving average model to complex network waveform prediction algorithm design,fluctuation signal in the time -frequency overlapped modulation process not included in the clutter a priori information,volatility form underdetermined sampling sequence is the extension of the diffraction characteristics,prediction result is bad. A kind of based on the autoregressive moving average model of the extended space complex network fluctuation underdetermined prediction algorithm is put forward,by using linear equalization filter,LTE tech-on lattice type structure noise reduction,getting rid of clutter interference,extending the diffraction characteristics of extracting volatility sequences form underdetermined sampling sample sequence,desiging the network fluctuation sequence of time and space lattice extension diffraction, accurately predicting network fluctuations extension distance,wave amplitude,frequency and phase deviation of fluctuation signal parameters,such as information. To viral invasion,the network listening and congestion jams volatility model as an example,the simulation experiment,the results show that the algorithm has higher prediction accuracy,monitoring fluctuations error is small,a complex network of fluctuations of dynamic tracking and evaluation.