论文部分内容阅读
以多径估计延迟锁定环路(MEDLL)为代表的参量式基带抗多径算法以其能够同时估计直达信号和多径信号各参数而受到广泛关注,然而该类算法估计精度的提高是以计算复杂度的增加为代价的。为了在保证估计精度的同时降低资源消耗并加快搜索速度,提出了一种MEDLL最大似然(ML)估计搜索方法,将粗搜索后得到的参数估计值作为下一级搜索的先验信息,缩小搜索范围后减小搜索步进,完成逐级精细的参数估计。推导了原始MEDLL及树状分级结构的MEDLL(T-MEDLL)算法计算复杂度表达式;其中在存在一路多径信号的情况下,对于0.01 chip的多径延迟分辨率,复杂度可以降低为传统盲搜索的30%~50%。进一步基于自研的数字中频(IF)GPS信号模拟器和软件GPS接收机平台对不同多径情况下盲搜索法与分级搜索算法的参数估计精度及码伪距多径误差等性能进行了对比,验证了更低计算复杂度的T-MEDLL可以达到与传统MEDLL算法同样的跟踪精度。
The parametric baseband anti-multipath algorithm represented by multipath estimation delay locked loop (MEDLL) has received much attention for its ability to estimate the parameters of both direct signal and multipath signal simultaneously. However, the improvement of the accuracy of this kind of algorithm is based on the calculation Complexity increases at the expense of. In order to reduce the resource consumption and speed up the search while ensuring the accuracy of the estimation, a MEDLL maximum likelihood (ML) estimation search method is proposed, which estimates the parameters obtained after the rough search as prior information of the next level search After the search range to reduce the search step to complete the level-by-level parameter estimation. The computational complexity expression of the MEDLL (T-MEDLL) algorithm for the original MEDLL and the tree hierarchy is deduced. In the presence of multipath signals, the complexity can be reduced to the traditional multipath delay resolution of 0.01 chip Blind search of 30% to 50%. Based on the self-developed digital IF signal simulator and software GPS receiver platform, the performance of blind search method and hierarchical search algorithm based on self-researched parameters and precision of pseudo-multi-path error are compared. T-MEDLL which verifies the lower computational complexity can achieve the same tracking accuracy as the traditional MEDLL algorithm.