Array calibration with modified Iterative HOS-SOS (MIHOSS) algorithm

Joint direction-of-arrival (DOA) and sensor position estimation for randomly deployed sensors is introduced in Iterative HOS-SOS (IHOSS) algorithm [1]. IHOSS algorithm exploits the advantages of both higher-order-statistics (HOS) and second-order-statistics (SOS) with an iterative algorithm using two reference sensors. The iterative algorithm is guaranteed to converge. IHOSS algorithm solves the position ambiguity by using source signals observed at multiple frequencies and hence it is applicable for wideband signals. In this paper, we propose Modified-IHOSS (MIHOSS) algorithm to solve the same problem for narrowband signals. In MIHOSS, it is assumed that the nominal sensor positions are known. It is shown that ambiguity problem is solved effectively without any assumption on the position perturbations. The upper bound of perturbations for unambiguous sensor position estimation is presented. The performance of MIHOSS approaches to the Cramér-Rao bound (CRB) for both DOA and position estimation.