提出了一种基于蚁群的新型按能需求路由算法（ARADE），针对移动网络节点剩余能量分布不均而导致网络生命周期短和蚁群算法收敛速度慢这两种缺陷做了相应的改进。存有路由信息的反向信息素表的引入简化了路由请求数据包的数据结构，将节点剩余能量、路径消耗能量以及跳数通过相应的影响因子归一化成一个参数作为路由性能评价值，以此来达到均衡整个网络节点剩余能量的目的。此外，迭代过程中信息素的最大值Γhigh可以当作参数进行一次自我迭代，使得当前节点维护在反向信息素表中的信息素值得到正反馈变化，从而加快整个算法的收敛速度。整个实验仿真是在NS2平台上进行，随着数据包的传输速率的增加，它的生命周期、平均能量消耗和端到端延迟都会优于AODV。

Presented is an antbased routing algorithm demanding on energy(ARADE), to improve the two situations of short lifetime and slow convergence speed of ant colony algorithm caused by unbalanced energy distribution. The inversed pheromone table which stores routing information simplifies the data structure of routing request packet. The residual energy of the node, energy consumption of the path and the number of hops are normalized to a parameter as a performance evaluation value, which can balance the residual energy of the entire network node. What’s more, the maximum pheromone value during the iteration will be used in the selfiteration to make the pheromone value get a positive feedback. As a result, the convergence speed of the algorithm can be accelerated. The simulation of the experiment is performed on NS2 platform. With the increase of the transmission rate of the data packet, its lifetime, average energy consumption and endtoend delay will be better than AODV.