Efficient Pilot Pattern for OFDM-based Cognitive Radio Channel Estimation - Part 1

Nowadays research in the area of Cognitive Radio has grown tremendously. In the application of spectrum pooling in OFDM-based Cognitive Radio system there will be subcarriers that are deactivated due to the access of Licensed Users. The unloaded subcarriers that are caused by Licensed Users access create problems in performing channel estimation. The unloaded subcarriers restrict putting the pilot symbols in time and frequency lattice. This issue should be considered in the designing pilot symbols grid. In the part I of this paper, new pilot patterns for OFDM-based Cognitive Radio channel estimation are proposed. These new pilot patterns are the modification of rectangular pilot patterns by shifting the pilot subcarricrs into the adjacent subcarriers which are free from the interference of Licensed Users. Using the proposed pilot patterns, the channel estimation based on two cascaded one dimensional (2 times 1-D) Wiener filter is investigated by considering the appearance of narrowband and wideband Licensed Users. Results have shown that the proposed pilot pattern outperforms the regular pilot pattern by around 6.5 dB gain at Mean Square Error value of 2 times 10^-3 in the case of narrowband interference.