系统工程与电子技术
繫統工程與電子技術
계통공정여전자기술
SYSTEMS ENGINEERING AND ELECTRONICS
2010年
2期
279-283,307
,共6页
杨明磊%张守宏%陈伯孝%朱守平
楊明磊%張守宏%陳伯孝%硃守平
양명뢰%장수굉%진백효%주수평
多载频多输入多输出雷达%幅相误差校正%子空间拟合法%最大似然法%Cramer-Rao界
多載頻多輸入多輸齣雷達%幅相誤差校正%子空間擬閤法%最大似然法%Cramer-Rao界
다재빈다수입다수출뢰체%폭상오차교정%자공간의합법%최대사연법%Cramer-Rao계
multi-carrier-frequency MIMO radar%gain and phase error calibration%subspace fitting method%maximum likelihood method%Cramer-Rao bound
建立了多载频(multi-carrier-frequency, MCF) MIMO雷达的幅相误差模型,并针对其发射和接收阵列的幅相误差耦合到一起的特点,提出了对信号预处理后等效阵列的联合幅相误差进行整体估计来实现误差校正的思想.针对单辅助信源的情况完成了两种误差估计方法:子空间拟合(subspace fitting, SF)法和最大似然(maximum likelihood, ML)法,子空间拟合法利用信号子空间与阵列流型张成空间的对应系列方程求解,而最大似然法利用似然函数最大来得到幅相误差.推导了幅相误差估计的Cramer-Rao界(CRB),并仿真分析了两种方法的估计性能与信噪比、快拍数的关系以及辅助信源的定位误差对估计性能的影响.
建立瞭多載頻(multi-carrier-frequency, MCF) MIMO雷達的幅相誤差模型,併針對其髮射和接收陣列的幅相誤差耦閤到一起的特點,提齣瞭對信號預處理後等效陣列的聯閤幅相誤差進行整體估計來實現誤差校正的思想.針對單輔助信源的情況完成瞭兩種誤差估計方法:子空間擬閤(subspace fitting, SF)法和最大似然(maximum likelihood, ML)法,子空間擬閤法利用信號子空間與陣列流型張成空間的對應繫列方程求解,而最大似然法利用似然函數最大來得到幅相誤差.推導瞭幅相誤差估計的Cramer-Rao界(CRB),併倣真分析瞭兩種方法的估計性能與信譟比、快拍數的關繫以及輔助信源的定位誤差對估計性能的影響.
건립료다재빈(multi-carrier-frequency, MCF) MIMO뢰체적폭상오차모형,병침대기발사화접수진렬적폭상오차우합도일기적특점,제출료대신호예처리후등효진렬적연합폭상오차진행정체고계래실현오차교정적사상.침대단보조신원적정황완성료량충오차고계방법:자공간의합(subspace fitting, SF)법화최대사연(maximum likelihood, ML)법,자공간의합법이용신호자공간여진렬류형장성공간적대응계렬방정구해,이최대사연법이용사연함수최대래득도폭상오차.추도료폭상오차고계적Cramer-Rao계(CRB),병방진분석료량충방법적고계성능여신조비、쾌박수적관계이급보조신원적정위오차대고계성능적영향.
A gain and phase error model of the multi-carrier-frequency (MCF) MIMO radar is constructed and a method to calibrate the array error by estimating the combined gain and phase error of the equivalent array after the signal preprocessing is proposed based on the coupling characteristics of gain and phase error of the transmit and receive array. Two estimation methods, subspace fitting (SF) method and maximum likelihood (ML) method, are proposed to estimate the gain and phase error when there is a single auxiliary source at a certain position. The SF method estimates the error by the corresponding equation of the signal subspace and the subspace expanded by array manifold while the ML method obtains it by maximizing the likelihood function. The Cramer-Rao bound(CRB) of gain and phase error estimation is derived, and the relationship between the estimation performance of these two methods and both signal to noise ratio (SNR) and snap number as well as the influence of the angle error of assistant source on the estimation performance are studied and simulated.