CIF: Medium: Collaborative Research: Multihop Multiflow Wireless Networks: A Treasure Hunt

无线网络是人们连接互联网、交流、做生意和获取信息的主要媒介。对更高数据速率不断增长的需求使传统的网络设计方法面临相当大的压力。为了满足这一需求，未来的无线网络正迅速向高密度、用户部署、异构基础设施发展，其特点是在较短的通信范围内具有非常高的数据速率，可能存在严重的干扰。在这种情况下，多流和多跳通信有望发挥越来越重要的作用。然而，由于对这种新型通信范式的基本理解仍处于起步阶段，目前关于如何设计这种网络的设想受到了限制。本研究以新思想、新见解和新工具的形式解决了关于多跳多流网络的关键问题，对未来网络的设计产生了变革性影响。该项目分为三个共生研究重点：可扩展性、侧信息和算法。在可扩展性的推力下，研究人员对原型网络模型的多跳多流网络的容量有了新的认识。侧信息推力超越了理想化的模型假设，并在多个空间和时间尺度上纳入了不完美或不确定的条件。算法的第三个重点是将前两个重点中发展的新关键思想转化为网络大小具有多项式复杂度的显式编码和信号处理算法，用于资源高效的多流多跳通信。这项研究可以产生新的想法，为未来几代无线网络的设计提供动力，从而广泛地影响无线行业。该项目还包括一项专门的教育计划，在这一领域征聘和培训人数不足的少数民族。

Wireless networks are the dominant medium through which people connect to the Internet, communicate, do business and access information. The ever-growing demand for higher data rates places the conventional approaches to network design under considerable stress. To meet this demand, future wireless networks are rapidly evolving towards high-density, user-deployed, heterogeneous infrastructure characterized by very high data rates at shorter communication range, with possibly harsh interference. In this context, multiple flow and multiple hop communication is expected to play an increasingly important role. Yet, current visions of how such networks should be designed are limited by the fact that the fundamental understanding of such novel communication paradigms is still in its infancy. This research addresses key questions about multihop multiflow networks in the form of new ideas, insights and tools, with a transformative impact on the design of future networks. The project is organized into three symbiotic research thrusts on extensibility, side information and algorithms. In the extensibility thrust, the investigators develop a new understanding of the capacity of multihop multiflow networks of prototypical network models. The side information thrust moves beyond idealized model assumptions and incorporates imperfect or uncertain conditions at multiple spatial and temporal scales. The third thrust on algorithms translates the new key ideas developed in the first two thrusts into explicit coding and signal processing algorithms with polynomial complexity in the network size, for resource-efficient multiflow multihop communication. The research can broadly impact the wireless industry by generating new ideas that will power the design of future generations of wireless networks. The project also includes a dedicated educational plan to recruit and train under-represented minorities in this field.