An optimized multi-tone calibration signal for quadrature receiver communication systems

Communication systems are subject to stringent image rejection requirements. Thus, accurate and regular field calibration is important. Regression techniques are effective in the calibration of quadrature receiver systems. These techniques require the transmission or injection of a calibration signal to estimate potentially frequency-dependent errors. Existing regression-based methods use nonlinear models with signals that calibrate only one frequency at a time. This paper recasts the problem in terms of linear regression and develops an optimized multi-tone calibration signal for quadrature receiver communication systems. Linear regression ensures closed-form solutions that can be computed in real-time by using adaptive filtering techniques. Simulations demonstrate the advantages of the multi-tone signal: simultaneous multi-frequency calibration and minimal interference with information bearing communication channels. At the same time, the benefits of regression-based calibration are also realized: modest model assumptions, effective performance assessment, and accommodation of non-uniformly sampled or missing calibration data.