Khushal Khan
, Abdul Qadar, Abdul Hadi
State estimates play a dominant role in almost all fields of engineering and technology, unambiguously. It plays a great role in many physical systems, where system measurements are uncertain. Instead, there are the highest requirements for the development of powerful algorithms that can lead to limited evaluation errors when the loss of the package occurs at the exit of the system, say, e.g., Data loss is the major issue of control and different areas of engineering which degrade the efficiency of the system[1-2]. The minimum mean square formula is used to minimize the error that occurred. Kalman filter is an iterative process it predicts the system state and will update its state after each step. Kalman filter is mostly used in challenging problems of data loss to overcome the effects of loss therefore it updates after estimating the actual state, infrequently it is quite challenging if the input is lost for a known duration of time. Data loss in the systems states is quite challenging and degrade the efficiency of communication and control systems. The most dominant method for the recovery of lost samples in the case of estimating the state of the system are OLKF and CLKF[3-4]. The CCLKF utilize the 3 strategies, Normal Equation, LDA, and LGA, using AR model, ARMA, and ARMAX(Auto regressive moving average with exogenous input) If the input is lost for a known time The effective technique is AR(where only previous measurements are used), ARMA (previous measurements and moving average), ARMAX(modal has more parameters for executing data loss i.e input, the noise we consider its results is best ) The accuracy of ARMAX must be higher than ARMA and AR Model but this technique is computationally expensive so there must be a trade-off between precision and simulation time In many systems the parameters increase the accuracy of the system increases but computational time also increase. Computation of this extra information bears an observable increase in Computational time. It will be verified after simulation that ARMAX will recover more efficiently as compared to AR and ARMA because the model parameter will increase in the case of (ARMAX) model (i.e exogenous input, noise, and regression to previous data) It is considered that ARMA model is Efficient as compere to AR but computationally expensive to overcome the problem of efficiency and computational time we will evaluate the linear prediction coefficients of ARMAX model and compere the results with AR, ARMA and ARMAX model using open-loop and compensated closed loop Kalman filter.
Khushal Khan Abdul Qadar Abdul Hadi “Performance Evaluation of Data Loss Recovery Techniques using Compensated Closed Loop International Journal of Engineering Works Vol. 9 Issue 01 PP. 17-21 January 2022 https://doi.org/10.34259/ijew.22.9011721.
[1] S. Fekri N. Khan and D. Gu. Improvement on state estimation for discrete time ltt systems with measurement loss. 43:1609– 1622, December 2010. Clerk Maxwell, A Treatise on Electricity and Magnetism, 3rd ed., vol. 2. Oxford: Clarendon, 1892, pp.68–73.
[2] P.D. Allison. Missing Data. sage Publications” 1st edition, 2001K. Elissa, “Title of paper if known,” unpublished.
[3] Naeem Khan. Linear Prediction Approaches to Compensation of Missing Measurements in Kalman Filtering. Phd thesis, Uni- versity of Leicester, November 2011
[4] P.P. Vaidyanathan. The theory of linear prediction,morgan and claypool publishers,. 2008.
[5] Sung Ho Cho Faheem Khan and Naeem Khan. Integration of linear prediction techniques with kalman ?lter estimation. KSII Transaction on internet and information system, 1:1–16, 2014.
[6] M. Athans S.Fakhri and A.Pascoal. Robust multiple model adaptive control (rmmac): A case study. International jour- nal of Adaptive Control and Signal Processing, pages 21:1–30, November 2007.
[7] Conference on Decision and Control, pages 4180 – 4186, December 2004
[8] Khattak and D-W. Gu§. Implementation of linear prediction techniques in state estimation. pages 1–9, 2015.
[9] F. M. Mirzaei and S. I. Roumeliotis. A Kalman Filter-Based Algorithm for IMU-Camera Cal-ibration: Observability Analysis and Performance Evaluation. IEEE Transactions on Robotics,24 (5):1143 – 1156, October 2008.
[10] X. Liu and A. Goldsmith. Kalman Filtering with Partial Observation Loss. In 43rd IEEE Conference on Decision and Control, pages 4180 – 4186, December 2004.
[11] V. Kodri´c, editor. Kalman Filter. Intech, India, 2010.
[12] T. Kobayashi and D. L. Simon. Evaluation of an Enhanced Bank of Kalman Filters for In-Flight Aircraft Engine Sensor Fault Diagnostics.Journal of Engineering for Gas Turbines and Power,127:497 – 504, July 2005.
[13] Y. Kim, D. Gu, and I. Postlethwaite. Fault-tolerant Cooperative Target Tracking in Distributed UAV Networks. In Proceedings of the 17th IFAC World Congress. Korea, July 2008.
[14] N. Xiao and L. Xie. Peak Covariance Stability of Kalman Filter with Bounded Markovian Packet Losses. In The 7th World Congress on Intelligent and Automation, China, June 2008
[15] E. R. B. Xiaoping Yun, Mariano Lizarraga and R. B. McMGhee. An Improved Quaternion
[16] S. K. Yang. An experiment of state estimation for predictive maintenance using Kalman filter on a DC motor. Reliability Engineering & System Safety, 75:103 – 111, August 200
