Intelligent Tactical Wireless Networks for Challenging Environments

近年来，由机器学习（ML）实现的人工智能（AI）已经彻底改变了许多技术，并将其性能和智能功能推向了超出预期的水平。因此，旨在增加无线网络的鲁棒性、可靠性、吞吐量和其他特征的新的智能无线解决方案正在出现。这些解决方案对科技行业具有吸引力，该行业一直在寻求设计突破性的产品，这些产品在每一项性能指标上都表现出色，特别是在民防、安全、应急和战术任务等救生技术领域。 该项目将ML引入战术无线网络（TWN），这是一种不同形式的商业通信网络，其特点是在不断变化的战术情况和不同的环境中具有鲁棒性，没有预先存在的基础设施。TWN应用程序的敏感性迫使公司独立开发自己的技术，以确保安全性和可靠性标准不受损害。因此，战术通信行业的竞争非常激烈，需要公司大力研发，以保持其产品的创新性和竞争力。在这方面，Ultra Electronics TCS和ETS之间的合作，旨在将ML的最新突破引入战术通信，对于保持公司在市场上的主要参与者之一至关重要。 该项目的长期目标是开发一个主动、自我感知、自适应和预测的战术无线网络，利用ML和优化工具自动化其部署、运营，并随后在具有挑战性的环境中提高其性能。 为TWN添加这种智能功能的设想方法是在三个网络级别设计新颖的ML方案，节点级别用于环境分类，链路级别用于两个节点之间的决策，以及集群级别用于MAC和路由决策。 该研究项目将有助于巩固加拿大作为人工智能和无线网络领导者的地位。

In recent years, artificial intelligence (AI), enabled by machine learning (ML), has revolutionized many technologies and pushed their performance and intelligent capabilities beyond expectations. Consequently, new intelligent wireless solutions aiming to increase the robustness, reliability, throughput, and other feature of the wireless networks are emerging. These solutions are appealing to the tech-industry which is continuously seeking to design ground-breaking products that excel in every performance measure, especially when it comes to lives-saving technologies such as in fields of civil defense, security, emergency, and tactical missions. This project introduces ML to tactical wireless networks (TWN), a different form of commercial communication networks, which is characterized by its robustness in constantly changing tactical situations and varying environments, with no pre-existing infrastructure. The sensitive nature of TWN applications forces the companies to independently develop their own technologies to ensure that the security and reliability standards are uncompromised. Thus, the tactical communications industry is highly competitive and requires strong R&D efforts from companies to keep their products innovative and competitive. In this respect, the collaboration between Ultra Electronics TCS and ETS, with the aim of introducing the recent breakthroughs in ML to tactical communications, is vital to keep the company among the major players in the market. The long-term objective of this project is to develop a proactive, self-aware, self-adaptive, and predictive tactical wireless networks that exploit ML and optimization tools to automate its deployment, operations and subsequently enhance its performance in challenging environments. The envisioned approach for adding such intelligent capabilities to the TWN is designing novel ML schemes at three network levels, the node-level for environment classification, link-level for decisions between two nodes, and cluster-level for the MAC and routing decisions. This research project will contribute to consolidating the stature of Canada as a leader in AI and wireless nets.