Spectrum sensing is the key and premise of cognitive radio( CR). Current parallel cooperative spectrum sensing strategies have some problems,such as large number of cooperative secondary users and lack of consideration for the sensing overhead and the transmission gain. To solve those problems,an optimized parallel cooperative spectrum sensing strategy based on iterative KuhnMunkres( KM) algorithm was proposed. To maximize the total system profit,it considers the tradeoff between the sensing overhead and the transmission gain. Iterative KM algorithm was applied to obtaining the optimal assignment,which indicated when and which channels secondary users should sense. Furthermore,the required detection probability was introduced to avoid unnecessary waste when the accuracy met the system requirement. Monte Carlo simulations show that the proposed strategy can obtain higher total system profit with fewer cooperative secondary users.
With the objective of taking full use of channel resource, we proposed two utility based dynamic subcarrier allocation （DSA） algorithms for the single carrier frequency division multiple access （SC-FDMA） system, which are the proportional fair frugality constrained （PF-FC） algorithm and the weighted proportional fair frugality constrained （WPF-FC） algorithm. The two proposed algorithms are designed under the frugality constraint （FC） control consideration so as to avoid service rate waste and improve the spectrum efficiency. Moreover, the queuing buffer model in this paper is established on a finite size structure rather than the traditional infinite queuing manner, which is more consistent with the practical transmission condition. Simulation results indicate that the two proposed algorithms can both achieve significantly better system rate-sum capacity and quality of service （QoS） performance than their primary algorithms, and are more applicable for the heterogeneous traffic.
