نوع مقاله : مقاله پژوهشی
نویسندگان
1 استادیار،دانشگاه جامع امام حسین(ع)، تهران، ایران
2 استاد،دانشگاه جامع امام حسین(ع)، تهران، ایران
3 دانشیار،دانشگاه جامع امام حسین(ع)، تهران، ایران
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
Tracking methods based on Bayesian algorithms are the most common methods in wireless sensor networks due to tracking accuracy and appropriate scalability. But, on the other hand, due to the high telecommunication overhead, these methods do not have the necessary efficiency in terms of bandwidth and energy. Due to the limitation of the energy source of each node, in this article a combined method called QDCAC based on dynamic clustering and multimode particle filter for target tracking is proposed. In this method, using the dynamic clustering based on the posterior Cramer-Rao lower band, the extracted position is used as the input of the tracking filter to estimate the position and speed of the maneuvering target and uses the estimated location of the target to determine the next master node and wake up the sensor nodes effective in tracking. It can be seen that in the proposed method, despite the non-linearity of the observation quantization algorithm and reducing the accuracy of the sent samples by 50% (4 bits) in order to reduce the information overhead and reduce the power consumption of the network, the accuracy level in the tracking algorithm is better than 1.7 meters, which is a desirable value in the 8,000 square meter test field.
کلیدواژهها [English]
)