一种基于双编码遗传算法的机动微波接力网组网方法

针对机动微波接力网组网及优化需要,提出一种基于双编码遗传算法(DMGA)的机动微波接力网组网方法.以网络价值最大化为优化目标,综合考虑节点价值、吸引系数、衰落概率等条件,实现了对每个节点接力设备的智能分配.通过研究网络价值与设备数的非线性关系,引入最优配置点的概念,为微波接力设备的编配数量提供量化依据.在遗传算法中使用邻接矩阵和邻接表的双编码法,其中邻接矩阵的优势是基因改变一定不会产生重复、自环链路,邻接表的优势是基因改变不会影响链路的总数.两种编码法在变异、交叉运算中交替运用,使约束条件与染色体形态特征优势匹配,避免了为满足约束进行的循环操作,提高了运算效率.仿真算例表明,双编码算法与单编码相比,计算时耗大大降低.

Mobile microwave relay network construction method based on double coding genetic algorithm

To facilitate the construction and optimization of mobile microwave relay network, a method of mobile microwave relay network construction based on the double coding genetic algorithm(DMGA) is proposed. Taking the maximization of network value as the optimization goal, considering the node value, attraction coefficient and fading probability, the intelligent distribution of relay device for each node is realized. By studying the non-linear relationship between the network value and the number of devices, the concept of optimal configuration points is introduced, which provides a quantitative basis for the allocation of microwave relay devices. In the genetic algorithm, the double coding strategy of the adjacency matrix and the adjacency list is used. The advantage of the adjacency matrix is that the change of gene will not produce repeated and self-loop links, and the advantage of the adjacency list is that the change of gene will not affect the total number of links. The two coding strategies are alternately used in mutation and cross operation, which makes the constraint condition and chromosome morphological characteristics match. This coding method avoids the iterative operation which is used to meet the constraints, and improves the operation efficiency. The simulation results show that compared with the single coding, the proposed double coding strategy reduces the computation time greatly.