Research On Frequent Pattern Based Spectrum Sensing Scheduling Algorithms For Cognitive Radio

Nowadays, with the dramatically increased requirement of wireless communication, the static spectrum allocation policy leads to a scarcity and under-utilization of spectrum resource. Cognitive radio has been extensively studied in recent years to solve this issue by allowing secondary users to access licensed frequency bands when the primary users are absent. Under the condition of protecting the primary users from harmful interference, cognitive radio has enhanced the spectrum utilization.Most of existed works have paid attention to maximize the throughput of the secondary users, or to minimize the interference to the primary users caused by incorrect spectrum detection of the secondary users. These works do not take account of the energy consumption of the secondary users. Firstly, a spectrum sensing scheduling algorithm is proposed for a cognitive radio transmitter to allocate sensing and data transmission time adaptively to achieve high throughput with limited energy. Furthermore, a frequent pattern prediction based spectrum sensing scheduling algorithm is proposed. The secondary users could perform data transmission without sensing the channels in time slots which are predicted to be idle. Finally, we propose an improved frequent pattern prediction method to make the secondary users predict the channel state in more time slots.Simulation results show that the proposed spectrum sensing scheduling algorithm can achieve a high throughput with less energy consumption. By joint the spectrum sensing scheduling algorithm with the assist of frequent pattern prediction algorithm, the throughput to energy consumption ratio of the secondary user is further improved.