New methods to estimate the observed noise variance for an ARMA model

Wang, Kedong; Wu, Yuxia; Gao, Yifeng; Li, Yong

doi:10.1016/j.measurement.2016.12.029

Cited by 14 publications

(11 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Conventional statistical methods include Allan variance (AV) analysis [6], wavelet denoising (WD) [7], empirical mode decomposition (EMD) [8], Kalman filter (KF) [3], autoregressive movement mean value (ARMA) modeling [9]- [11] and autoregressive conditional heteroscedasticity (ARCH) modeling [11]- [13], etc. In particular, ARMA modeling is the use of traditional time series analysis methods to actually estimate and compensate the random drift time of the gyroscope.…”

Section: Introductionmentioning

confidence: 99%

Research on Ship Motion Prediction Algorithm Based on Dual-Pass Long Short-Term Memory Neural Network

2021

View full text Add to dashboard Cite

When forecasting ship movements, the random errors of the inertial navigation system (INS) seriously affect the accuracy of general prediction methods. In actual measurement, the main causes of the random errors are electrostatic bias and micro-electric disturbance. In response to this problem, a novel type of dual-pass Long Short-Term Memory (LSTM) neural network architecture is developed, on the basis of regular LSTM neural network. In the designed dual-pass LSTM neural network, the random drift and the noise residual of the INS are regarded as a autoregressive moving average (ARMA) and generalized autoregressive conditional heteroskedasticity (GARCH) model. Through dual-pass layers, the prediction of drift and the correction of residual errors are realized respectively in the same time. The simulation of ship heave motion was carried out on the ship motion simulation platform, and the real-time datas which are measured by the INS are inputted to the trained dual-pass LSTM netural network. The experiment proved that, when training the same source datas offline, the average Root Mean Squared Error (RMSE) percentage of conventional LSTM network was 3.94%, but when training different source datas or training online, the prediction accuracy obvious decline. In contrast, the average RMSE percentage of the dual-pass LSTM neural network was 1.05% when training offline and 1.12% when training online. Compared with conventional LSTM networks, the dual-pass LSTM network is more targeted and has better adaptability in the field of ship-motion prediction, and this network restores the motion prediction to the actual trajectory of a ship more accurately. INDEX TERMS Dual-pass LSTM neural network, generalized autoregressive conditional heteroskedasticity (GARCH) model, inertial navigation system (INS).

show abstract

Section: Introductionmentioning

confidence: 99%

Research on Ship Motion Prediction Algorithm Based on Dual-Pass Long Short-Term Memory Neural Network

2021

View full text Add to dashboard Cite

show abstract

“…This type of error directly affects the stability of the output signals and is difficult to be processed directly through device calibration [8]. Therefore, the modeling and compensation schemes of the nonlinear error components are widely studied and two mainstream research schemes [9][10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26] are formed, namely, (1) establishing a statistical model and performing error compensating and (2) error compensation schemes based on machine learning or deep learning.…”

Section: Introductionmentioning

confidence: 99%

“…Through power spectral density function analysis [12], the nondeterministic error can be modeled as Gaussian white noise or colored noise, and error compensated methods are analyzed [13]. In order to improve the accuracy of modeling, the empirical model decomposition [14], and autoregressive-moving-average (ARMA) time series model [15][16][17] are introduced into the error compensation schemes. Based on the Kalman filtering algorithm, the error compensation scheme can achieve better results and improve the accuracy of the statistical model [18,19].…”

Section: Introductionmentioning

confidence: 99%

Online Nonlinear Error Compensation Circuit Based on Neural Networks

Gao

Zhou

et al. 2021

Machines

View full text Add to dashboard Cite

Nonlinear errors of sensor output signals are common in the field of inertial measurement and can be compensated with statistical models or machine learning models. Machine learning solutions with large computational complexity are generally offline or implemented on additional hardware platforms, which are difficult to meet the high integration requirements of microelectromechanical system inertial sensors. This paper explored the feasibility of an online compensation scheme based on neural networks. In the designed solution, a simplified small-scale network is used for modeling, and the peak-to-peak value and standard deviation of the error after compensation are reduced to 17.00% and 16.95%, respectively. Additionally, a compensation circuit is designed based on the simplified modeling scheme. The results show that the circuit compensation effect is consistent with the results of the algorithm experiment. Under SMIC 180 nm complementary metal-oxide semiconductor (CMOS) technology, the circuit has a maximum operating frequency of 96 MHz and an area of 0.19 mm2. When the sampling signal frequency is 800 kHz, the power consumption is only 1.12 mW. This circuit can be used as a component of the measurement and control system on chip (SoC), which meets real-time application scenarios with low power consumption requirements.

show abstract

“…Forecasting network traffic helps service providers to enhance their services. In order to predict the district network traffic for ANUB, various prediction techniques have been proposed in the literatures, such as auto regressive and moving average (ARMA) [12], auto regressive integrated moving average (ARIMA) [13], [14] and fractionally auto regressive integrated moving average (FARIMA) [15]. However, time series data do not always have the same characteristics.…”

Section: Introductionmentioning

confidence: 99%

Density-Ratio Peak Based Semi-Supervised Algorithm for Access Network User Behavior Analysis

Zhang

Yao

et al. 2019

IEEE Access

View full text Add to dashboard Cite

In order to improve the prediction accuracy of the access network user behavior (ANUB), we propose a novel density-ratio peak (DRP)-based semi-supervised algorithm. It first rescales the given dataset with non-uniform density clusters by density-ratio estimation (DRE) and conducts the subscriber detailed classification by using the density peak (DP) algorithm. The proposed DRP algorithm can identify all clusters in a dataset with greatly varying densities. Then, a semi-supervised algorithm evolves three regression prediction methods, namely, an auto regressive and moving average (ARMA), an auto regressive integrated moving average (ARIMA), and a fractionally auto regressive integrated moving average (FARIMA), as typical representatives to generate an accurate predictor for the ANUB and establish a prediction model for each subcluster. The behaviors of access network users in the same subcategory share similarities and the statistics of the behaviors of all access network users in a district can emerge the model of the district network properties, such as the prediction of the district network traffic that is more detailed than the direct prediction. The proposed model is evaluated through the dataset of the ANUB collected from China Telecom, and the obtained results show that the integrated model is an effective way to improve the accuracy of prediction achieved by the DRP clustering, compared with the conventional ones.INDEX TERMS Access network user behavior, density-ratio estimation, density peak clustering, semisupervised model.

show abstract

New methods to estimate the observed noise variance for an ARMA model

Cited by 14 publications

References 17 publications

Research on Ship Motion Prediction Algorithm Based on Dual-Pass Long Short-Term Memory Neural Network

Research on Ship Motion Prediction Algorithm Based on Dual-Pass Long Short-Term Memory Neural Network

Online Nonlinear Error Compensation Circuit Based on Neural Networks

Density-Ratio Peak Based Semi-Supervised Algorithm for Access Network User Behavior Analysis

Contact Info

Product

Resources

About