应急实时音视频混合智能路由优化方法

打开文本图片集

中图分类号：TN915；TP393 文献标识码：A 文章编号：2096-4706（2026）04-0001-08

Emergency Real-time Audio-Video Hybrid Intelligent Routing Optimization Method

CHEN Kun

（People's Public SecurityUniversityofChina，Beijinglooo38，China）

Abstract： Aiming at the chalenges of strong network dynamics， multi-objective conflictsand poor Qualityof Service （QoS）guarante faced byreal-time audio-video transmision inemergency communication scenarios，a hybrid intellgent routingmechanismintegratingDepReinforcementLearming（DRL）andmulti-populationhierarchicalAntColonyOptimization （ACO）is proposed.Firstly，this paper establishesastochastictime-varyingmodel of emergency networks，uses Markovchains tocharacterize linkavailabilityevolution，and describes performance parameterfluctuationcharacteristics trough first-order autoregressive processes.Secondlyitconstructsatwo-layercollaborativearchitectureofDeepDuelingDoubleQ-Network （D3QN）andmulti-populationantcolonyalgorithm，andrealizes deepfusionofvalueassessmentandpath search through Q -value driven pheromone modulationmechanism.Furthermore，adual-gate stabilitycontrol mechanismis designed to efectively suppressfrequent path switching while maintaining rapid response.The experimental results show thatthe proposed method reduces the average end-to-end delay，packet loss rate，and path switching frequency by 43.2% 65.4% ，and 67.5% respectively compared to the shortest path algorithm， while increasing bandwidth utilization by 78.9% ，demonstrating excellent adaptability androbustnessinburst congestion andlink failure scenarios.

Keywords： emergency communication; Deep Reinforcement Learning; Ant Colony Optimization; routing algorithm

0 引言

在重大自然灾害、突发公共安全事件以及大型应急处置任务中，实时可靠的音视频通信是指挥调度、态势感知与资源协同的关键支撑[]。（剩余10953字）