Cooperative source number estimation for cognitive radio networks

In this paper, we consider a cognitive radio (CR) network operating on a spectrum reserved for licensed primary user (PU) by employing spatial spectrum sharing (SS). When the secondary users (SUs) are equipped with multiple antennas, they can perform SS without interfering with the PU by orthogonalizing the secondary transmission with the interference channel . In order to avoid interference and to maximize the capacity of the secondary transmissions, it is essential that the SUs determine the dimension of the orthogonal transmission space. In CR networks where no cooperation is provided by the PU, this information can be obtained by estimating the number of primary transmit streams at the secondary terminals. We study this issue when the secondary network performs cooperative estimation by using the minimum description length (MDL) algorithm to obtain the independent estimates and propose a decision fusion rule which takes into account the estimation bias of MDL. We will show that the proposed method can outperform conventional decision fusion rule and that it provides additional protection for the PU in the case of estimation error.