Fast convergence outer loop link adaptation with infrequent updates in steady state

In wireless communications, link adaptation is used to select a suitable modulation and coding scheme. The purpose of link adaptation is to adapt to varying channel and interference conditions and to aim for a specified block error rate or to maximize the throughput. In support of link adaptation, there will be estimates of signal to interference ratios, path gain or transmit powers. These estimates can contain systematic and random errors, that may affect the performance of link adaptation. To correct for such errors, there is an outer loop performing link adaptation, usually based on feedback of the bit error rate. We investigate commonly used outer loop link adaptation algorithms and propose a new scheme based on sequential hypothesis testing. The new scheme is shown to converge faster at initialization and after disturbances and to have good performance in steady state.