Joint invariant estimation of RF impairments for reconfigurable Radio Frequency(RF) front-end

To support a multi-standard platform using Software Defined Radio (SDR), reconfigurable Radio Frequency (RF) front-ends have recently been proposed as a more reliable front-end than the currently used fixed wide-band RF front-ends. With these reconfigurable front-ends, each possible configuration must be characterized in terms of its RF impairments, to decide its suitability for a particular communication standard. RF impairment estimation has previously been studied for fixed RF front-ends, with respect to the impairment most likely to affect a particular standard. However, in a reconfigurable RF system, several RF impairments may simultaneously be present, since only a few of the many possible configurations will be suitable as a front-end. In this paper, we show how several RF parameters can simultaneously be estimated using the baseband signal. The RF parameters chosen are the ones that are most likely to affect the heterodyne front-ends that are popular in current SDRs; namely nonlinearity, frequency offset, noise figure and gain. The invariance of each parameter's estimator to the other parameters is accomplished by using a subspace-based approach based on a carefully chosen bilinear model. Simulation results demonstrate that the designed invariant estimators are more accurate than previous estimators, since the latter are designed for single impairment estimation.