Exploiting Interference in Wireless Networks: Protocol Design, Analysis and Implementation

无线网络中的主要性能限制因素是干扰。当前的无线网络是基于干扰避免的原理设计的，该原理提倡消除并发的附近传输以避免冲突，因为冲突导致浪费稀缺的无线带宽。关键问题是，盲目防止干扰的网络设计是次优的，并导致带宽有限的无线网络效率严重低下。随着无线网络物理层的最新进展，如果控制得当，可以成功解码附近链路的并发传输。如果发射机和接收机知道干扰结构，它们可以利用它来更好地解码它们的传输，从而大大提高网络性能。 该提案从无线网络中的传统协议设计进行了根本性的转变：它设计了系统地鼓励和利用干扰来提高网络性能的网络协议。具体地说，它利用三个物理层机制利用干扰：多包传输的叠加编码的基础上，多包接收的连续干扰消除的基础上，和合作通信设计无线网络协议（具体地说，在链路层调度和网络层路由），把干扰变成一个优势。 拟议的研究将产生技术，可以对未来的无线网络的设计产生重大影响。该提案的范围包括协议设计，网络建模和分析，并在实验测试平台上实施选定的协议。除了对新的无线技术的理论和实践的潜在影响，拟议的研究是一个很好的培训基地，本科生和研究生在无线网络的各个方面。

A major performance limiting factor in wireless networks is interference. Current wireless networks are designed based on the principle of interference avoidance, which advocates for eliminating concurrent nearby transmissions to avoid collisions, as collisions result in waste of scarce wireless bandwidth. The key issue is that a network design that blindly prevents interference is sub-optimal and results in severe inefficiencies in bandwidth-limited wireless networks. With recent advances at the physical layer of wireless networks, concurrent transmissions of nearby links can be successfully decoded if controlled properly. If transmitters and receivers are aware of the interference structure, they can exploit it to better decode their transmissions, and thus greatly improve network performance. This proposal makes a fundamental shift from the traditional protocol design in wireless networks: it designs network protocols that systematically encourage and exploit interference to improve network performance. Specifically, it utilizes three physical layer mechanisms for exploiting interference: multi-packet transmission based on superposition coding, multi-packet reception based on successive interference cancellation, and cooperative communication to design wireless network protocols (specifically, scheduling at link layer and routing at network layer) that turn interference into an advantage. The proposed research will produce techniques that can have substantial impact on the design of future wireless networks. The scope of the proposal includes protocol design, network modeling and analysis, and implementation of selected protocols on an experimental testbed. In addition to potential impacts on the theory and practice of new wireless technologies, the proposed research is an excellent training ground for undergraduate and graduate students in all aspects of wireless networks.