NEDG: A New Systematic Framework for Cross-layer Optimization

无线网络中的跨层设计问题通常非常复杂，因为它们需要同时优化大量的算法和参数。大多数现有的跨层优化解决方案依赖于启发式过程来解决这个问题。然而，为了获得无线用户的最优效用，跨层优化应严格制定为一个顺序决策问题，考虑到各层自主预测其经验动态的能力，并进行前瞻性适应，同时坚持现有的分层网络体系结构。为了应对这一挑战，研究人员研究了一种新的跨层优化系统框架，该框架允许每层对所选择的传输策略（例如协议参数和算法）做出自主和有远见的决策，同时通过最佳地确定层之间应该交换哪些必要的信息来合作最大化无线用户的效用。本研究主要涉及两个重点：(a)开发一种具有层间消息交换的新颖跨层优化框架，其中每层根据自身经验动态和与其他层交换的信息优化自己的协议参数和算法，以协同最大化无线用户？在坚持分层网络架构的同时，前瞻性地提供S实用程序；(b)为跨层优化设计分层在线学习算法，使每一层可以交互地从其他层学习经验动态和其他必要的信息，从而实现跨层策略的协同优化。这项研究为设计无线网络、系统和应用程序提供了一种全新的方法，通过从设备与环境的相互作用中学习，设备会随着时间的推移而进化并变得更智能。

Cross-layer design problems in wireless networks are usually very complex, since they require simultaneously optimizing a large amount of algorithms and parameters. Most existing solutions for cross-layer optimization rely on heuristic procedures to solve this problem. However, to obtain an optimal utility for the wireless user, cross-layer optimization should be formulated rigorously as a sequential decision problem that takes into account the capability of the various layers to autonomously make forecasts about their experienced dynamics, and perform foresighted adaptation, while adhering to the existing layered network architecture. To address this challenge, the investigators study a new, systematic framework for cross-layer optimization that allows each layer to make autonomous and foresighted decisions on the selected transmission strategies (e.g. protocol parameters and algorithms), while cooperatively maximizing the utility of the wireless user by optimally determining what necessary information should be exchanged among layers.This research involves two main thrusts: (a) Develop a novel cross-layer optimization framework with message exchange among layers in which each layer optimizes its own protocol parameters and algorithms based on its own experienced dynamics and the information exchanged with other layers, in order to cooperatively maximize the wireless user?s utility in a foresighted manner while adhering to the layered network architecture; (b) Design layered on-line learning algorithms for the cross-layer optimization to allow each layer to interactively learn the experienced dynamics and other necessary information from other layers, such that the cross-layer strategies can be optimized cooperatively. This research leads to a fundamentally new way for designing wireless networks, systems and applications, where devices evolve and become smarter, by learning from their interactions with the environment over time.