论文部分内容阅读
在移动自组织网络中,需要通过移动学习控制,全面监控移动计算设备和网络上的各种应用进程和学习进程,提高网络通信性能和学习交互能力。传统的移动自组织网络学习控制算法采用基于半监督学习的移动学习控制算法,半监督学习方法对网络自组织特征的检测率较高,但对数据的内容感知性能不好。提出一种基于内容感知的同频率震荡移动学习控制算法。构建移动自组织网络模型,结合协同量子群算法,模仿粒子的学习行为,设计同频震荡移动学习控制策略,通过创建种群基因库实现了子群间的信息共享,实现对学习内容的最优适应度感知,实现控制算法的改进。仿真结果表明,该控制算法能通过学习内容的感知,结合协同搜索策略与粒子学习行为,实现了学习过程的同频率震荡控制,达到收敛速度与收敛精度间的平衡区域平衡,具有更良好的搜索性能,具有较好的移动学习控制能力和寻优能力。
In mobile ad hoc networks, mobile learning controls need to be used to fully monitor various application processes and learning processes on mobile computing devices and networks, and improve network communication performance and learning interaction. Traditional mobile self-organizing network learning control algorithm based on semi-supervised learning-based mobile learning control algorithm, semi-supervised learning method for network self-organization features a high detection rate, but the content of the data perception performance is not good. This paper proposes a concentric frequency oscillation learning algorithm based on content perception. The model of mobile ad-hoc network is built, and collaborative quantum-subgroup algorithm is used to simulate the learning behavior of particles. The control strategy of co-oscillating mobile learning is designed. By creating a population gene library, information sharing among sub-groups is realized, Degree of perception, to achieve the improvement of control algorithm. The simulation results show that the control algorithm can realize the same frequency concussion control in the learning process by learning the content perception, combining the collaborative search strategy and particle learning behavior, and achieve a balanced regional balance between the convergence speed and the convergence precision, and has a better search Performance, better mobile learning control ability and optimization ability.