A constrained total least squares approach for sensor position calibration and direction finding

In this paper we present a new maximum likelihood array signal processing algorithm that uses self-calibration and the total least squares (TLS) technique for improved direction finding when the sensor (or antenna) positions are not known exactly. The TLS method is a generalized data fitting method that is appropriate for cases when the system model contains error or is not known. In order to apply TLS techniques to array signal processing, we propose constrained-TLS (C-TLS) and show that the solution can be represented by combining the constraint and the least squares (LS) approach. The algorithm performs calibration and direction estimation iteratively using both TLS and C-TLS techniques