噪声相关的一步滞后无序量测递推融合算法传感技术学报.pdf

第 22 卷 　第 1 期 传 感 技 术 学 报 Vol . 22 　N o . 1 2009 年 1 月 CHIN ES E J OU RNAL O F S EN SORS AND AC TUA TORS J an . 2009 Recursive Fusion Algorithm with Correlated Noises and OneStep OutofSequence Measurements 1 2 1 1 GE Q u anbo , M A Guoj i n , TA N G X i anf eng , W E N Chengl i n 1. I ns t it ute of I nf orm at ion an d Cont rol , H ang z hou D i anz i Uni vers ity , H angz hou 3 100 18 , Chi na; 2 . S chool of Elect ronic I nf orm at ion . H angz hou D i anz i Uni vers ity , H ang z hou 3 100 18 , Chi na Abstract :Becau se of t he sp ecial co mmunication mo de s of sen sor net wor k s an d t he different dat a of p rep ro ce ssin g time of each sen sor no de , t he or dered mea surement s f ro m t he same t ar get oft en ar rive in t he f u sion cent er out of sequence acro ss sen sor net wor k s , named a s “outofsequence ”mea surement (OO SM) p rob lem . A nd mo st of f u sion algorit hm s are p re sent ed for t he in dep en dent noi se s a ssump tion . Thi s p ap er re searche s t he OO SM s Sen sor net wor k s co mpo sed of multip le sub sy st em s an d each of t hem ha s t wo sen sor s w here one ha s t he samp lin g rat e and samp lin g time ,and anot her i s a synchronou s wit h t he f u sion cent er . For t he ca se wit h correlat ed sen sor and sy st em noi se s , a recur sive di st ribut ed weight ed OO SM s f u sion al gorit hm , w hich can realtimely e stimat e t ar get st at e variable s and i s op timal in minimizing t he t race of t he er ror covariance mat rix ,