State Estimation over a Semi-Markov Model Based Cognitive Radio System

Cognitive radio is a very popular research area in the communication community as cost and bandwidth can be saved by sensing the available licensed spectrum for unlicensed users. This paves the way for the application of cognitive radio in control systems over wireless communication links. A typical control system comprises a sensor interconnected with an actuator, a plant and controller. In this paper, it is assumed that the sensor and estimator communicate through a cognitive radio link. The state estimator needs to adjust to this new communication media as it is affected by interruptions from primary users, resulting in packet losses. The cognitive radio link is modeled as multiple semi-Markov processes each of which can capture the channel dynamics. Two different cases are considered. The first case assumes that acknowledgement of packet arrivals is available at the estimator while the second case does not. For the first case, sufficient conditions are derived for the stability of the peak covariance process which guarantees the stability of the estimator covariance. For the second case, an estimator is designed. Numerical examples are given to show the performance of the proposed estimators. Several applications of the proposed work are also discussed.