MULTI-SCALE MODELING BASED ON IRREGULAR TREE FOR A STOCHASTIC PROCESS
There is a sort of dynamic system with multiple sensors, which are usually used to measure the same target and possess different sampling frequency and characteristic. To fuse the data arising from these sensors in this dynamic system, the new concept irregular tree is put forward, a new multiscale representation is presented by use of the Markov statistical characteristics in most phenomena or processes, and the corresponding multiscale model is proposed based on a irregular q th tree. In addition, these parameters in the model are derived. Simulation illustrates the performance and effectiveness of this new method. These efforts and achievements are development and force to resolve effectively the data fusion problem for multi-source asynchronous information describing the same target.
Irregular tree multi-scale representation and modeling Markov stochastic process data fusion
CHENG-LIN WEN JUN-JIE SHI ZHU-LIN ZHENG
Henan Institute of Science and Technology, Xinxiang 453003, China Department of computer and information engineering, Henan University, Kaifeng 475001, China
国际会议
2006 International Conference on Machine Learning and Cybernetics(IEEE第五届机器学习与控制论坛)
大连
英文
3406-3411
2006-08-13（万方平台首次上网日期，不代表论文的发表时间）