What semi-blind channel estimation brings in terms of throughput gain ?

In most wireless communication systems, channel estimation is required for equalization and symbol detection. It is currently achieved via dedicated pilot symbols consuming a significant part of the throughput and global resources especially for future systems. This paper shows that semi-blind identification techniques reduces significantly this overhead. By using the Cramer Rao Bound tool, we compare the estimation error variance of the pilot based and semi-blind based techniques for a SIMO-OFDM system. The comparisons illustrate that, depending on the context, the semi-blind approach can lead up to 90% reduction of the pilot sequence size without affecting the channel estimation quality.