A new look at maximal ratio combining

The performance of ideal maximal ratio diversity combining cannot be achieved in a practical system because of channel gain estimation errors. The accuracy of channel gain estimation is in turn a function of signal-to-noise ratio. An upper bound to the performance of maximal ratio diversity accounting for channel gain estimation errors and their dependence on signal-to-noise ratio is derived. Importantly, it is shown that the traditional theoretical performance of maximal ratio combining in slow fading cannot be achieved when the channel gain estimates are based on finite channel gain estimation signal-to-noise ratio (finite number of past symbols). It is shown that the advantages of maximal ratio/coherent reception over square-law/noncoherent reception in slow fading may be much less than previously believed