A Limited Feedback Algorithm Sacred from Feedback Error for Slowly Time-varying Channel

Abstract

The Grassmannian predictive codebook algorithm is an effective method for limited feedback in slowly time-varying channels. But a drawback of this algorithm is feedback error propagation that can cause errors in lots of channel states. This paper proposed an algorithm that can detect feedback error by comparing the choral distance between two adjacent predicting channels to a predefined threshold. Numerical results show that the algorithm can correct feedback error at the expense of little performance loss in tracking channel evolution.