Ad Hoc网络SPMA接入协议研究

Research on SPMA Protocol in Ad Hoc Network

为了解决现有介质访问控制子层(Media Access Control,MAC)协议无法较好地满足Ad Hoc网络资源接入的需求,提出了一种统计优先级多址接入(Statistic Priority-based Multiple Access,SPMA)协议,该协议是一种无中心节点控制的随机接入协议,适合节点移动性强、拓扑频繁变化的场景。首先介绍了SPMA协议的运行机制,重点研究了SPMA协议的信道负载统计方法,提出了MAC层信道负载统计方法,与物理层脉冲统计方法相比实现更简单;其次分析了在多跳场景下存在的问题,提出了多跳信道负载统计方法,同时为了克服周期性统计方法带来的滞后性问题,引入了基于长短期记忆网络(Long Short-Term Memory,LSTM)的时间序列预测模型。实验结果表明,所提方法能够准确反映信道真实状态,提高网内节点信息传输的实时性和可靠性。

To address the problem that the existing MAC(Medium Access Control)protocols cannot meet the needs of Ad Hoc network resource access well,a SPMA(Statistic Priority-based Multiple Access)protocol is proposed.The protocol is a random access protocol without central node control,suitable for scenarios with strong node mobility and frequent topology changes.First,this paper introduces the operating mechanism of SPMA protocol,focuses on the channel load statistics method of SPMA protocol,and proposes the MAC layer channel load statistics method,which is simpler to implement compared with the physical layer pulse statistics method.Then,it analyzes the problems in multi-hop scenarios and proposes a multi-hop channel load statistics method.At the same time,in order to overcome the lag problem caused by periodic statistics methods,the paper introduces a time series prediction model based on LSTM(Long Short-Term Memory)networks.Experimental results indicate that the proposed method can accurately reflect the true state of the channel and improve the real-time and reliability of the information transmission among nodes in the network.