基于超越数论的无线传感器网络时空编码方法

Spatio-Temporal Coding Method for Wireless Sensor Networks Based on Transcendence Number Theory

在无线传感器网络(WSN)规模逐渐增大与传感器逐渐微型化的背景下,全局信息收集的持续性和实时性的要求与无线传感节点受限能力之间的矛盾日益严峻。传统方法使用压缩、融合、聚合等方式降低数据传输量,并通过优化路由增加网络能力,但越来越难以应对上述挑战。为此,考虑利用计算能力克服存储、传输瓶颈,通过本地化计算范式实现全局信息的感知,并基于超越数论和非定域感知方法,提出一种面向大规模分布式WSN的信息存算与通信一体化方法。通过对网络进行建模,将网络每时每刻产生的信息以去中心化本地计算的方式融合到常数量级的可计算编码中。该方法通过节点邻居之间周期性地交换搭载时空编码的Beacon消息。根据时空编码在相空间中构造具有确定性和因果性的相空间轨迹来存储和交换信息,避免直接存储和传输庞大的原始数据,从而降低计算、通信、存储等开销。实验结果表明,该方法能够实现O(1)的存储和通信开销,具有毫秒级的收敛速率,相较现有WSN存储方法,在通信开销方面具有明显优势。

In recent years,the scale of Wireless Sensor Network(WSN)has expanded,with sensors trending toward miniaturization.This expansion has intensified the tension between the need for continuous,real-time global information collection and the limited capabilities of wireless sensor nodes.Historical strategies,such as data compression,fusion,and aggregation,have been used to reduce data transmission volume and enhance network capacity through optimized routing.However,addressing emerging demands with these techniques is increasingly challenging.This study explores the utilization of computational power to overcome storage and transmission bottlenecks and proposes that global information perception can be achieved via localized computing paradigms.Drawing inspiration from transcendence number theory and non-local sensing methodologies,this study introduces a comprehensive approach to information storage and communication tailored for expansive distributed wireless sensor networks.The proposed strategy capitalizes on the periodic exchange of Beacon messages enriched with spatiotemporal encoding between neighboring nodes.By crafting deterministic and causal trajectories in phase space via spatiotemporal coding,information is encapsulated and relayed without the direct storage or transmission of extensive raw data,significantly reducing computational,communication,and storage costs.Experimental findings validate that the proposed method realizes a storage and communication overhead of O(1),converging rapidly within milliseconds,this approach offer significant advantages over existing WSN storage-centric techniques.