Superimposed Training and Blind Channel Estimation

Abstract

Superimposed training (ST) is one of the most promising channel estimation techniques, which provides an even better performance than pilot‐symbol‐assisted modulation (PSAM). ST techniques superimpose pilot symbols over the data signal, so that pilots and data share the same resources, thus avoiding loss in data rate. This feature makes them very attractive for future 5G systems where high data rates are expected. ST techniques suffer from interference between pilots and data, which requires more sophisticated channel estimation and data detection algorithms. Apart from pilot‐based strategies, channel estimation can be accomplished without relying on training signals. Then, blind channel estimation can obtain channel information based on the analysis of the received signal. Blind techniques, similarly to ST schemes, do not waste resources for channel estimation, but their performance is not as precise as pilot‐based schemes. Nowadays, there is a great interest in ST and blind channel estimation due to its spectral efficiency.