Efficient Spectrum Allocation and Time of Arrival Based Localization in Cognitive Networks

Position of the primary user is significant to the transmission among secondary users in cognitive networks (CN). In this paper, orthogonal frequency division multiplexing (OFDM)-based spectrum allocation and time of arrival-based localization are proposed for the devised CN. Crame?CRao lower bound for range estimation is theoretically derived in terms of the proposed spectrum allocation in CN, compared with that of the nonoverlapped allocation in CN and OFDM-based static spectrum allocation in noncognitive networks (NCN). The 2D localization accuracy is investigated based on horizontal dilution of precision (HDOP). The theoretical minimum HDOP is explored and the corresponding network topology to attain the minimum HDOP is provided. Theoretical analysis and simulation results demonstrate that the proposed spectrum allocation in CN exhibits a much better ranging and localization accuracy, and a better data transmission rate than the nonoverlapped spectrum allocation in CN, no matter the designed spectrum with nonoverlapped allocation is a sinc function or a rectangle. Also, the proposed spectrum allocation in CN is demonstrated to have a better ranging and localization accuracy than OFDM-based static spectrum allocation in NCN.