采用宽窄巷结合的LAMBDA解算方法的北斗多频差分定位技术

Multi-frequency Beidou differential positioning using combination of wide and narrow lane for LAMBDA resolution

目前,常用的多频整周模糊度解算方法有TCAR(Three-Carrier Ambiguity Resolution)、CIR(Cascading Integer Resolution)和LAMBDA(Least-squares Ambiguity Decorrelation Adjustment)这3种.其中TCAR和CIR方法均利用载波相位的宽巷组合易于整周模糊度解算和窄巷组合可减小噪声的特点进行整周模糊度的解算.但是它们的整周模糊度解算性能相对于LAMBDA而言较弱,而且需要预设相关矩阵.针对这一问题,本文提出一种宽窄巷结合的LAMBDA整周模糊度解算方法用于北斗多频的差分定位.该方法在采用LAMBDA算法的基础上,充分利用宽巷组合和窄巷组合的特点,通过使用宽巷组合快速有效解算出的整周模糊度来约束窄巷组合求得的整周模糊度,以达到整周模糊度的快速准确解算的目的,同时通过窄巷组合定位模型的低噪声特点实现较高的定位精度,可有效的提高北斗多频差分定位中整周模糊度的解算效率和定位解算的精度.本文通过在北京航空航天大学楼顶的实际测试数据进行验证.结果表明,采用北斗三频进行定位解算时,宽窄巷结合的LAMBDA整周模糊度解算方法能在比窄巷组合收敛时间减小一半的情况下,实现与窄巷组合相同的3mm定位精度.

Until recently,TCAR( Three-Carrier Ambiguity Resolution),CIR( Cascading Integer Resolution) and LAMBDA( Least-squares Ambiguity Decorrelation Adjustment) are the wildly used method for integer ambiguity resolution. TCAR and CIR,both taking advantages of the wide-lane combination which is easier to fix the integer ambiguity,and the narrow-lane combination which can reduce the noise more efficiently,are currently used as two common integer ambiguity resolutions. However,either TCAR or CIR performs a lower probability of resolution success in comparison with LAMBDA,even requires a pre-set admissible ambiguity transformation. To deal with this problem,a method based on a combination of wide and narrow lane LAMBDA integer ambiguity resolution,in terms of Beidou differential positioning problem,is proposed. On the basis of LAMBDA,this method employs the advantages of both the wide-lane combination and the narrow-lane combination. In this method,the integer ambiguity of narrow-lane combination is constrained by using the wide-lane combinations,which conducts a fast integer ambiguity resolution,while a high accuracy positioning is obtained by using the narrow combination,which suffers low noisy affect. The performance of the method was evaluated in a test site on the roof of Beihang University using tri-frequency data from Bei Dou. The result shows that this method can achieve a 3 mm positioning accuracy at a shorter convergence time when compared with the convergence time of narrow-lane combination.