Abstract—Kalman Filter is used in system estimation applications today like state estimation, load flow analysis, harmonic estimation, digital signal processing, sensor integration, Navigational Systems, etc. In using a Kalman Filter the user has to give the parameters relating the estimates of process and measurement noise along with system state modeling. The values of process and measurement noise covariance are usually not available beforehand and have to be estimated, usually by hit or trial method. This involves heavy computation, as two variables have to be estimated for optimal filtering independently. For multi-state systems this value further increases the computation time. This paper presents the application of Kalman Filter to a simple one state problem. This paper, through using simulations, finds relationships between the two different parameters Q (Process Noise Covariance) and R (Measurement Noise Covariance). This results in reduction of computation time. The proposed scheme’s low complexity and robustness makes it practical for real implementations.
Index Terms—Estimator, kalman error analysis, kalman filter, kalman optimization, measurement noise covariance, state estimation, process noise covariance.
T. Singhal, A. Harit, and D. N. Vishwakarma are with the Electrical Engineering Dept., Indian Institute of Technology, Banaras Hindu University, Varanasi, India (e-mail: email@example.com;firstname.lastname@example.org; email@example.com).
Cite: Toshak Singhal, Akshat Harit, and D. N. Vishwakarma, "State Estimation and Error Analysis of a Single State Dynamic System with Sensor Data Using Kalman Filter," International Journal of Information and Electronics Engineering vol. 3, no. 4, pp. 399-402, 2013.