Sachin Katti - US grants

Wireless networking assumes that radios are half-duplex. On a given frequency, a half-duplex radio can either transmit or receive, but not both at the same time. The project disproves this long-held assumption; it shows how a radio that can transmit and receive simultaneously on the same frequency can be built using commodity off-the-shelf components. The design is based on two key ideas. First, is the design of analog circuits that can perform adaptive signal inversion, i.e. take an input RF signal and produce its exact inverse, and programmatically adapt the attenuation and delay on the inverted signal to match the self-interference experienced by the received signal. This enables the design of wideband full duplex radios that can handle transmit powers upto 20dBm. Second, the project exploits the full duplex primitive to design a real-time bidirectional channel for control and data, as well as more complex patterns such as chains. By interspersing control and data information in a message, nodes can dynamically react to channel changes in real-time.

The above two primitives - full-duplex operation and a real-time bidirectional control channel - can help solve many long-standing fundamental problems in wireless networks, including hidden terminals, bitrate adaptation, network congestion, resource allocation and unfairness. The project will produce a prototype full duplex radio for WiFi style networks and show experimentally how it can improve their performance, further all designs will also be made public through research publications and open-source hardware designs.

The goal of this proposal is a fundamental redesign of wireless networks to systematically exploit interference to increase network capacity. Traditionally interference is considered harmful, hence current designs strive to avoid interference by scheduling concurrent transmissions in separate frequencies/time slots. Hence the only way to add more capacity is to add more spectrum. However spectrum that can be used for building wireless networks has mostly been allocated and is in use, thus imposing hard limit on the scalability of the current network design. This proposal makes a fundamental shift: instead of avoiding interference, it designs techniques that systematically encourage and exploit interference to increase network capacity. The key insight is that interference is not random noise, but has structure since it is a synthetic signal created by another transmitter. If transmitters and receivers are aware of the interference structure, they can exploit it to actively shape/code interference and better decode their own transmissions to cancel interference, and thus greatly increase capacity. The proposed research will produce techniques that can each have substantial impact of the design of wireless networks. This project will design single channel full duplex radios, a technical feat that has hitherto been considered impossible. Second, this research will produce rateless codes that can decode constituent packets from collisions, obviating the need for complex scheduling primitives and thus simplify PHY/MAC design. Finally, it will produce smart radios that can adaptively operate in dense radio neighborhoods and maximize throughput, and thus coexist in environments with a variety of interfering radios.

Modern datacenter networks and private wide area networks underpin cloud computing. Today, due to the lack of flexibility, operators end up over provisioning their networks significantly to avoid performance bottlenecks. This project will make these networks more efficient and more tuned to the application's needs. Specifically, this investigation explores a weighted transport abstraction as a flexible and robust substrate for systems that optimize a network's bandwidth allocation. The proposed research revisits a classic theoretical framework in this space, Network Utility Maximization (NUM), and develops a novel and practical distributed algorithm for NUM that is significantly faster than prior approaches, and is thus applicable to modern high speed networks such as datacenter fabrics. This project seeks to design and build a flexible transport architecture.

The proposed architecture has two main technical components:

1. Weighted Transport: Instead of using flow rates to control the bandwidth allocation, this research develops a transport based on weights. To tune the bandwidth allocation, flows adapt a weight field in their packet headers; each link then divides its bandwidth among contending flows in proportion to their weights.

2. Fast Utility Maximization: This project leverages the weighted transport to design a network fabric that can be dynamically tuned for different bandwidth allocation objectives such as minimizing flow/coflow completion time, or service-level fairness.

The PIs plan to interact closely with companies that can influence standards and build commercial systems using the proposed ideas.
The education plan includes the incorporation of this research's findings into the undergraduate and graduate curricula and offers an opportunity to take a "top-down" approach to teaching transport architectures with a focus on key bandwidth allocation objectives and how they affect real applications. The course material will be made widely available through MIT OpenCourseWare and on the MITx MOOC.

This award is a planning grant for the Spectrum Innovation Initiative: National Center for Wireless Spectrum Research (SII-Center). The focus of a spectrum research SII-Center goes beyond 5G, IoT, and other existing or forthcoming systems and technologies to chart out a trajectory to ensure United States leadership in future wireless technologies, systems, and applications in science and engineering through the efficient use and sharing of the radio spectrum. The radio spectrum should be utilized to the greatest public benefit at national and global scales. Spectrum shortages, both real and perceived, are leading to conflicts between existing users and anticipated new uses ? some of which were not imagined when existing spectrum allocations were made decades ago. Many stakeholders seek to protect and advance their interests, with aspects of these interests overlapping and conflicting with each other. Hence, the public debate over the optimal model for managing spectrum is a complex interplay of technology, economics, law and regulation, policy, and the history of past successes and failures.

This project is aimed at the development of a comprehensive plan for an SII-Center which would help maintain and extend US leadership in future wireless technologies, systems, and applications in science and engineering through the efficient use and sharing of radio spectrum. The project team is led by the University of Notre Dame, with partners from Northwestern University, Clemson University, University of California, Berkeley, University of California, Los Angeles, New York University, and Stanford University. The team has an extensive record of successful research, as well as significant and relevant industry and implementation experience. The project seeks to develop plans for a multi-disciplinary center that emphasizes instrumentation of the radio spectrum; collecting and sharing accurate regulatory, usage, and economic data; and developing data-rich system designs and regulatory policies for more efficient spectrum utilization.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.