卫星认知无线电检测门限与功率分配联合优化算法

Joint Optimization Algorithm of Detection Threshold and Power Allocation for Satellite Underlay Cognitive Radio

针对窄带卫星通信中频谱利用率不足的情况,以最大化带内数据传输量为目标,提出了基于卫星Underlay认知无线电的上行链路中信道检测门限与功率分配联合优化(JDPO)算法。首先根据检测误差、功率向量与数据传输量之间的运算关系构建了卫星Underlay认知无线电接入模型,之后将目标函数分解为检测门限与功率分配2个子问题分别进行优化,以加窗粒子群优化算法逼近了最优检测门限,根据库恩-塔克条件求解了最优功率分配向量。通过引入中间量使2个子算法反复迭代,最终得到了检测门限与功率分配的联合最优解。仿真结果表明:存在多个次要用户时,JDPO算法可以获得更多的带内数据传输量;与传统方法相比,JDPO算法的数据传输量最大可提高50%。

A joint detection threshold and power allocation optimization （JDPO） based on underlay cognitive radio is proposed to improve spectrum utilization in narrowband satellite systems. The joint scheme maximizes throughput and focuses on uplink. A corresponding model is established using the operation relationship among detection error, power allocation and throughput. Then, the optimization problem is partioned into two sub-problems： optimization of detection threshold and optimization of power allocation. A windowed particle swarm optimization is proposed to solve the former sub-problem, and the Kuhn-Tucker condition is applied to find the solution of the latter one. An intermediate variable is introduced to repeatedly perform iteration between both the sub-solutions to get the joint-optimal solutions for the detection threshold and power allocation. Simulation results and comparisons with the traditional satellite communication show that the JDPO scheme gets about 50% throughput increasing.