The rapid development of Industrial Internet of Things (IIoT) technologies has not only enabled new applications, but also presented new challenges for reliable communication with limited resources. In this work, we define a deceptively simple novel problem that can arise in these scenarios, in which a set of sensors need to communicate a joint observation. This observation is shared by a random subset of the nodes, which need to propagate it to the rest of the network, but coordination is complex: as signaling constraints require the use of random access schemes over shared channels, each sensor needs to implicitly coordinate with others with the same observation, so that at least one of the transmissions gets through without collisions. Unlike existing medium access control schemes, the goal here is not to maximize total goodput, but rather to make sure that the shared message gets through, regardless of the sender. The lack of any signaling, aside from an acknowledgment or lack thereof from the rest of the network, makes determining the optimal collective transmission strategy a significant challenge. We analyze this coordination problem theoretically, prove its hardness, and provide low-complexity solutions. While a low-complexity clustering-based approach is shown to provide near-optimal performance in certain special cases, for the general scenarios, we model each sensor as a multi-armed bandit (MAB), and provide a learning-based solution. Numerical results show the effectiveness of this approach in a variety of cases.
翻译:在这项工作中,我们界定了在这种情景中可能出现的一个简单而明了的新问题,在这些情景中,传感器需要进行联合观测。这一观测由随机的节点分组共享,这些节点需要将其传播到网络的其余部分,但协调是复杂的:由于信号限制要求使用共享频道的随机接入计划,每个传感器需要与其它有相同观察的系统进行隐性协调,这样至少一个传输能够不受碰撞地通过。与现有的中等访问控制系统不同,这里的目标是确保共享信息能够通过,而不管发送者是谁。除了网络其余部分的认可或缺乏,缺乏任何信号,因此确定最佳的集体传输战略是一项重大挑战。我们从理论上分析这一协调问题,证明其硬性,并提供低兼容性解决方案。尽管基于低兼容性组合的模型方法与现有的中等有效,但这里的目标是确保共享信息能够通过,而不管发送者是谁。除了网络其余部分的认可或缺乏信号之外,缺乏任何信号,使得确定最佳的集体传输战略成为一项重大挑战。我们从理论上分析这一协调问题,证明其硬性,并且提供低度的解决方案。在每种特殊情况下,一种基于低频段式的模型方法提供了一种我们感官学习的一种特殊的情景。