无线电工程
無線電工程
무선전공정
RADIO ENGINEERING OF CHINA
2012年
10期
16-19
,共4页
集员滤波%仿射投影算法%可变数据重用因子%稳态失调%收敛速度
集員濾波%倣射投影算法%可變數據重用因子%穩態失調%收斂速度
집원려파%방사투영산법%가변수거중용인자%은태실조%수렴속도
set-membership fihering%affine projection algorithm%variable data-reuse factor%steady-state misadjustment%conver-gence speed
为了解决传统集员滤波仿射投影(SM—AP)算法收敛速度与稳态失调和计量复杂度之间的矛盾,提出一种新的数据选择性仿射投影算法。此算法在传统SM—AP算法的基础上,引入可变阶数(也称数据重用因子),称为基于可变数据重用因子的集员滤波仿射投影(VDRF—SM—AP)算法。通过利用步长提供的信息,此算法可以自动地分配数据重用因子,实现了征初始阶段数据重用因子大,收敛后数据重用因子小的目标,从而既保证了收敛速度又降低了稳态失调。通过理论分析和仿真验证,新算法的整体复杂度比其他传统的SM-AP算法低很多,同时保留了传统的SM—AP算法的快速收敛特性,但是却能达到更小的稳态失调。
為瞭解決傳統集員濾波倣射投影(SM—AP)算法收斂速度與穩態失調和計量複雜度之間的矛盾,提齣一種新的數據選擇性倣射投影算法。此算法在傳統SM—AP算法的基礎上,引入可變階數(也稱數據重用因子),稱為基于可變數據重用因子的集員濾波倣射投影(VDRF—SM—AP)算法。通過利用步長提供的信息,此算法可以自動地分配數據重用因子,實現瞭徵初始階段數據重用因子大,收斂後數據重用因子小的目標,從而既保證瞭收斂速度又降低瞭穩態失調。通過理論分析和倣真驗證,新算法的整體複雜度比其他傳統的SM-AP算法低很多,同時保留瞭傳統的SM—AP算法的快速收斂特性,但是卻能達到更小的穩態失調。
위료해결전통집원려파방사투영(SM—AP)산법수렴속도여은태실조화계량복잡도지간적모순,제출일충신적수거선택성방사투영산법。차산법재전통SM—AP산법적기출상,인입가변계수(야칭수거중용인자),칭위기우가변수거중용인자적집원려파방사투영(VDRF—SM—AP)산법。통과이용보장제공적신식,차산법가이자동지분배수거중용인자,실현료정초시계단수거중용인자대,수렴후수거중용인자소적목표,종이기보증료수렴속도우강저료은태실조。통과이론분석화방진험증,신산법적정체복잡도비기타전통적SM-AP산법저흔다,동시보류료전통적SM—AP산법적쾌속수렴특성,단시각능체도경소적은태실조。
To solve the confliction between convergence speed and steady-state misadjustment or computational complexity,a new data-selective affine projection (AP) algorithm is proposed. The algorithm which is called affine projection algorithm with variable order or data-reuse factor based on set-membership filtering (VDRF-SM-AP) generalizes the ideas of the conventional set-membership affine -projection (SM-AP) algorithm to include a variable data-reuse factor. By utilizing the information provided by step size,the data-reuse factor can be automatically adjusted. At the beginning of the algorithm a large data-reuse factor is achieved while a small one is achieved after the algorithm has converged. Simulations show that a significant reduction in the overall complexity can be obtained with the pro- posed algorithm as eompared with the conventional SM-AP algorithm. In addition,the proposed algorithm retains the fast convergence of the conventional SM-AP algorithm,but obtains the lower steady-state misadjustment.