Study on Quality Control Method of Using the Kalman Filtering nder the conditions of small batch, there are

many problems in the applications of statistical process control (SPC), which has limitation in the course of mode rn industry. The parameter sigma must maintain invariably which comes from its s upposition process white noise N (0 , sigma ), or else the control chart is unte nable. This paper proposed to carry on the quality control using the Kalman filt ering. Using the Kalman filtering as the basic theory of the industry quality co ntrol process can fully reflect the content of control. It can completely elimin ate inadaptation of the using with SPC in the course of modern industry quality control process. Through the example research, the paper indicates that the Kalm an filtering has the compatibility of the quality control compared to SPC, the e stimating error of which is small. Neural Networks Training Based on Sequential Extended Kalman Filtering for Singl e Trial EEG Classification The nonstationary nature of the brain signals provides a rather unstable input r esulting in uncertainty and complexity in the control. Intelligent processing al gorithms adapted to the task are a prerequisite for reliable BCI applications. T his work presents a novel intelligent processing strategy for the realization of an effective BCI which has the capability to improved classification accuracy a nd communication rate as well. A neural networks training based on sequential ex tended Kalman filtering analysis for classification of extracted EEG signal is p roposed. A statistically significant improvement was achieved with respect to th e rates provided by raw data. Compensation for two specific types of artifact in pulse wave using a Kalman fil ter An algorithm is proposed to interpolate the pulse wave during the measurement of SpO2 by using Kalman filter identification of autoregression (AR) coefficients, Kalman filter interpolation, and a fixed-interval smoother. This combined algor ithm picks up the pulse rate smoothly from pulse waves contaminated by two speci fic artifact types: respiration artifact and sporadic body movement artifact. Detection of oscillations in power systems using Kalman filtering techniques This paper addresses online estimation of electromechanical oscillatory modes in power systems using dynamic data such as currents, voltages and angle differenc es measured on-line across transmission lines. These measurements are provided b y newly installed phasor measurement units (PMUs). Here, the collected data are evaluated with the main objective to detect the dominant frequency and damping o f oscillations during normal operation of the power system. The goal is to give the transmission operator qualitative and quantitative information about actual oscillations in the power system as well as about related properties, such as st ability margins. The approach is demonstrated on a model of a real power system. Several problems and limitations of the approach are pointed out as well. Application of Kalman Filter for the estimation of human head tissue conductivit ies n this study Extended Kalman Filtering is proposed for the estimation of human h ead tissue conductivities by using EEG data. The proposed method first linearize s the relationship between the tissue conductivities and surface potentials (EEG measurements) and then iteratively estimates the tissue conductivities. In the study the mathematical background of the proposed method is presented and then p erformance of the proposed method is investigated by a simulation study. In the simulation study a three layered realistic head model (composed of scalp, skull and brain compartments) obtained from MR images of a real patient is used. The s urface potential is calculated by using an arbitrarily chosen conductivity distr ibution. Then conductivity estimation is iteratively performed by using the calc ulated potentials and at each iteration relative error rates are calculated by c omparing the original conductivities and estimated ones. It is found that the re lative error rates decrease below of 1% after five iterations.

Application of extended Kalman filtering to hyperthermia The success of treating cancerous tissue with heat depends on the temperature el evation, the amount of tissue elevated, and the length of time that the tissue i s heated. The temperature of most of the treatment volume is unknown because onl y a small number of temperature sensors can be inserted into the issue. A state space model based on a finite difference approximation of the bioheat transfer e quation is developed for identification purposes. The extended Kalman filter is applied to estimate both the unknown blood perfusion parameters and the unmeasur ed temperature states. Simulation results show that the success of the estimatio n scheme depends on the number and location of the temperature sensors An hour ahead wind speed prediction by Kalman filter This paper proposes a trustworthy and practical approach to predict wind speed, which could be used as input to predict wind power generation. In the prediction model, we applied the variable speed wind as input of this model. Then the Gaus sian noise is added to the input due to existing of noise in measuring of wind s peed. Kalman filtering is used as the prediction method in this study. The param eters of shaping filter are calculated using power spectral analysis and Gauss-M arkov theory. Results under different time horizon of prediction are compared wi th actual data and the comparison shows that the results obtained using Kalman f ilter is reliable. The proposed method works well for an hour ahead prediction a nd for short-time prediction works even better. The application of Kalman filter on wind speed prediction is implemented in MATLAB software and results are prov ided in this paper. Application of Kalman Filtering in the Detection of Evoked Potentials A method is proposed for de-noising and extracting non-stationary electroencepha logram (EEG) signals. Kalman filtering is an optimal recursive data processing a lgorithm. In this paper Kalman filtering is used to estimate evoked potentials ( EP) from large background noise of electroencephalogram (EEG). The Waveforms of before filtering and after filtering is simulated and compared. The results show that the method can extract EP from the stationary random noise signals, and th e filtering effect is more satisfied Application of hybrid adaptive filters for stock market prediction Prediction of stock market trends has been an area of great interest both to tho se who wish to profit by trading stocks in the stock market and for researchers attempting to uncover the information hidden in the stock market data. Tradition al techniques such technical analysis and signal processing techniques such as m oving averages and regression have had limited success in predicting markets, wh ich could be attributed to the dynamic behavior of the markets. In signal proces sing, adaptive filters have been widely used for efficient filtering of signals. However, the utilization of adaptive filters for prediction, especially of fina ncial signals, has not received much attention in literature. In this study, hyb rid adaptive filters are introduced for prediction to obtain highly accurate res ults. The hybrid filters used are DCT-LMS, DCT-NLMS, DCT-RLS and Kalman filters. The proposed method is used to predict the values of five of the largest stock markets, namely, BSE100, NASDAQ, NIKKEI225, S&P NIFTY, and FTSE100. The performa nce of hybrid adaptive filters is compared against the conventional filters like autoregressive (AR), Moving Average (MA) filters and adaptive filters like LMS, NLMS etc. The base technique considered is the Random Walk (RW) process which a cts as the benchmark technique. The results show a high degree of prediction acc uracy for the hybrid adaptive filters, which is very high when compared to conve ntional filters, thus indicating that hybrid adaptive filters can be successfull y used for stock market prediction. Applications of adaptive filtering to ECG analysis: noise cancellation and arrhy thmia detection Several adaptive filter structures are proposed for noise cancellation and arrhy

thmia detection. The adaptive filter essentially minimizes the mean-squared erro r between a primary input, which is the noisy electrocardiogram (ECG), and a ref erence input, which is either noise that is correlated in some way with the nois e in the primary input or a signal that is correlated only with ECG in the prima ry input. Different filter structures are presented to eliminate the diverse for ms of noise: baseline wander, 60 Hz power line interference, muscle noise, and m otion artifact. An adaptive recurrent filter structure is proposed for acquiring the impulse response of the normal QRS complex. The primary input of the filter is the ECG signal to be analyzed, while the reference input is an impulse train coincident with the QRS complexes. This method is applied to several arrhythmia detection problems: detection of P-waves, premature ventricular complexes, and recognition of conduction block, atrial fibrillation, and paced rhythm. Application of Adaptive Filtering and Linear Regression for Determining T-Wave L imits in Neonatal ECG Two algorithms have been identified, namely the adaptive filtering method and ta ngent method, which can be used to automatically detect the end of T-wave in neo natal ECG for determining the QT and QTc duration. By modifying the initial cond itions and parameters in the algorithms, both algorithms have been successfully applied in real time to process real neonatal ECGs with various noise levels. Th e accuracy and the precision of both algorithms were examined by comparing the r esult with manual readings done by a cardiologist. The algorithms can be used to analyze neonatal ECG for long QT syndrome in real time and automatically to aid early diagnosis, subjected to further test and verification. MARKET DATA PREDICTION WITH AN ADAPTIVE KALMAN FILTER http://www.haikulabs.com/mdpwakf.htm#timeseries