A Novel Sidelobe Suppression Technique for OFDM-Based Cognitive Radio Transmission

In this paper, we propose a novel two-step sidelobe suppression technique for OFDM-based cognitive radios designed to minimize the amount of out-of-band emissions that could potentially interfere with incumbent licensed transmissions in adjacent frequency bands. The proposed technique combines both constellation expansion (CE) and cancellation subcarrier (CC) sidelobe reduction approaches in order to achieve lower out- of-band interference levels relative to their individual application. Being a two-step process, the first step of the proposed technique consists of performing CE, which exploits the combination of different random symbol sequences across the subcarriers in order to yield lower sidelobe levels. The second step is to insert several CCs on either side of contiguous blocks of data-bearing subcarriers such that they combine destructively with the data- bearing subcarrier sidelobe levels, yielding a reduction in out- of-band emissions. Simulation results show that with the CE of four symbols mapped to an 8-PSK signal constellation and with one CC employed on each side of a 32 subcarrier OFDM spectrum, there is a reduction of as much as 16 dB on the peak- interference causing sidelobe. From a probabilistic point of view, this translates to reducing 99.9% of the sidelobe power levels from around -5 dB to around -22 dB.