认知网络中协作感知吞吐量的优化

为了提高集中式认知网络的吞吐量,提出了基于信任度的吞吐量优化算法.该算法在主用户充分保护的前提下,以认知用户的吞吐量为目标函数,融合中心采用双门限值对本地感知结果进行融合.从理论上证明了吞吐量是全局漏检概率的增函数,当全局漏检概率等于门限值时,吞吐量达到最大值.并利用牛顿迭代法求出单节点概率,然后采用遍历法可得到认知用户吞吐量最大值.仿真结果表明,当信噪比为-14 dB时认知用户融合优化算法相对"AND准则"OR准则"以及"HALF准则"归一化吞吐量分别提高了0.62、0.3和0.09.

Throughput optimization of cooperative spectrum sensing in cognitive radio networks

In order to improve the throughput of secondary user ( SU ) in centralized cooperative spectrum sensing cognitive radio networks,a throughput optimization algorithm based on reliable data combing is pro-posed. The throughput is taken as the objective function and the fusion center uses double threshold to make final decision when the primary users are sufficiently protected. It is proved that the throughput is an increasing function of the missed detection probability. The throughput reaches its maximum value while the general missed probability is equal to threshold value. The Newton iteration method is proposed to cal-culate the probability of detection of a single node,and then traversal method is used to obtain the maxi-mum throughput. Simulation results show that the normalized throughput of SU fusion scheme algorithm is increased by 0 . 62 ,0 . 3 and 0 . 09 compared with that of"AND rule" "OR rule" and "K out of N rule"when singal-to-noise ratio( SNR) is -14 dB.