中文摘要：

无线传感器网络地理位置路由算法需要克服本地最小化现象.已有的算法均以“右手规则”为基础来克服这一现象,但是,“右手规则”会损失链路且难以在路由过程中动态调整所选路由,对自适应多路由选择不利.通过深入分析本地最小化现象,发现合理限定贪心算法的选路区域可以克服本地最小化现象,由此提出Clockwise Rule,该规则不损失链路,可动态调整所选路由,耗费低.在此基础上,提出了自适应并发多路由算法APMR（Adaptive Parallel Multi-path Routing）.算法是分布式的,可在任意中继节点处动态调整路由策略,优化路由过程.仿真结果表明APMR算法对网络性能有很好的改善.

英文摘要：

Geographic muting in wireless sensor networks suffers from the local minimum phenomenon. To help packets get out of local minimum phenomenon,the long-known right-hand rule is used, which needs the graphs planarizing. It will lost many links and doesn＇ t suit adaptive routing. Through studying the local minimum phenomenon deeply, we find out that greedy forwarding in restricting routing regions could deal with the local minimum phenomenon. Based on the observation, a new method Clockwise Rule was proposed, which does not lost any links and more suits adaptive routing. We also proposed an adaptive parallel multipath routing algorithm （APMR） using Clockwise Rule, which can provide various routing strategies and switch these strategies at relay nodes freely. Simulations show that APMR can significantly improve the performance of wireless sensor networks.