Performance evaluation of sensor allocation algorithm based on covariance control

1 .INTRODUCTIONSensing platforms for target tracking have becomein-creasingly complex as more and more types of sensorshave been added to them. Sensor management tech-niques can provide automatic or semi-automatic con-trol of a suite of sensors to i mprove the performanceof sensing and give an effective usage of sensor re-sources. Traditionally, sensor allocation approachesselect sensor sets which mini mize a cost function ormaxi mize an efficiency function consisting of sensor-target pair…

Performance evaluation of sensor allocation algorithm based on covariance control

The covariance control capability of sensor allocation algorithms based on covariance control strategy is an important index to evaluate the performance of these algorithms. Owing to lack of standard performance metric indices to (evaluate) covariance control capability, sensor allocation ratio, etc, there are no guides to follow in the design procedure of sensor allocation algorithm in practical applications. To meet these demands, three quantified performance metric indices are presented, which are average covariance misadjustment quantity (ACMQ), average sensor allocation ratio (ASAR) and matrix metric influence factor (MMIF), where ACMQ, ASAR and MMIF quantify the covariance control capability, the usage of sensor resources and the robustness of sensor allocation algorithm, respectively. Meanwhile, a covariance adaptive sensor allocation algorithm based on a new objective function is proposed to improve the covariance control capability of the algorithm based on information gain. The experiment results show that the proposed algorithm have the advantage over the preceding sensor allocation algorithm in covariance control capability and robustness.