Self-contained calibration determination by jointly solving the attitude estimation and calibration problem in the steering vector domain

Antenna calibration is a mandatory requirement to determine the attitude of a multi-antenna receiver. Moreover, knowing the antenna attitude can vastly improve detection and mitigation of spoofers and / or Radio Frequency Interferences (RFIs). In todays receivers the time-varying phase offset introduced by the active elements like amplifiers, downconverters as well as various filters and unequal cable lengths are usually calibrated using a pilot signal, which is injected in the antenna signal. This method, however, does not only complicate the manufacturing process, but also increases the complexity of the receiver architecture and power consumptions. New approaches developed recently use GNSS live signals for calibration. These methods are established based on the fact that for each incoming GNSS signal the direction of arrival (DOA) is exactly known by means of the ephemeris data, which is transmitted from each satellite. In order to estimate the calibration phases, the DOAs of the received satellite signals are compared with their corresponding expected DOAs, which are determined based on the ephemeris data. The problem, however, lies in the fact, that the expected DOAs are given in the East-North-Up (ENU) coordinate frame, whereas the measured DOAs are given in the local coordinate frame and are additionally affected by the unknown phase offsets to be calibrated. The proposed algorithm therefore solves the attitude estimation and calibration problem jointly. This paper proposes a novel algorithm and validates it using measurement data.