The Application of Rate Adaptation with Finite Alphabet in MIMO-OFDM

To maximize the spectral efficiency of MIMO-OFDM under a certain target bit error rate (BER) in an environment where alphabet is discrete and upper bounded, it is necessary to apply adaptive power control and rate adaptation to always find the maximal alphabet size it can support. Although water filling (WF) can achieve the capacity of MIMO fading channel theoretically, it is not a preferable power allocation strategy in practice when alphabet is finite. To get higher spectral efficiency, a modified power control based on water filling can be utilized. In this paper we will compare these bit-, power-loading algorithms in the scenario where instantaneous channel state information (CSI) is known to the transmitter. Besides, the space-time block coding (STBC) with rate adaptation is implemented and it turns out that STBC can even achieve higher spectral efficiency in some cases