It is estimated that there are more wireless devices in use today than people on the planet. This explosive growth has created a capacity crisis for mobile operators whose impact goes well beyond the inconveniences experienced by casual users. Wireless communication is now a commodity, like roads, water, and electricity, on which the nation's economic development and government services rely. Emergency services, medical information systems, environmental monitoring, smart-grid energy distribution, smart transportation systems - all of these technologies depend on reliable access to high speed broadband wireless. It is clear that dramatic improvements in spectral access and capacity will be needed soon to accommodate the multitude of users who need broadband services anytime, anywhere. This proposal aims to address this critical need by determining how to harness potential gains from three new technologies: (1) the use of very high frequency bands, 10-20 times higher than those used in today's wireless networks, (2) decreasing the size of wireless "cells" to allow frequencies to be reused in a more dense fashion, and (3) the adoption of massive arrays of hundreds of antennas in order to handle the inevitable increase in interference and need for user selectivity. There appears to be considerable symbiosis among these new technologies that could potentially be exploited to achieve transformative gains in network speed and connectivity - increasing them by a factor of 1000 - although numerous challenges would have to first be vercome. Given the ubiquitous need for improved broadband access in industry, government, medicine and the home, the progress made by this research in overcoming these challenges will have a broad impact on all walks of life.
A potential remedy for the shortage of spectrum is the use of millimeter wave frequency bands, which offer significantly more bandwidth than today's systems. Other approaches for increasing wireless capacity are also rekindling interest in the potential use of millimeter wave frequencies for cellular communications, such as using pico- and femto-cells with ranges on the order of 10-200 meters to increase frequency reuse, and the idea of basestations equipped with a very large number of antennas that can simultaneously accommodate many co-channel users, an idea referred to as massive multi-input multi-output wireless. These ideas work well in combination with each other: smaller cells are attractive for operation at millimeter wave frequencies where radio frequency path loss is significantly higher, the shorter wavelength associated with higher frequencies is appealing for massive multi-input multi-output wireless designs since the physical dimensions of the antenna array and associated electronics are reduced, the large beamforming gains achievable with very large antenna arrays can extend coverage to help overcome the high millimeter wave path loss, and the reduction in channel coherence time at millimeter wave frequencies is offset by the lower mobility and hence the higher channel coherence bandwidth due to operation in small cells. If the individual gains in capacity offered by these approaches could be achieved in combination with one another, then one could envision realizing the orders-of-magnitude increases in throughput that are predicted to be required in coming years. The goal of this proposal is to investigate the feasibility implementing a millimeter wave massive multi-input multi output system incorporating in excess of 100 antennas, addressing issues associated with millimeter wave signal propagation, communication system design, the impact of communication and signal processing requirements on the hardware design, and how practical hardware limitations affect achievable performance. Ultimately, this research effort will demonstrate the degree to which millimeter wave small-cell massive multi-input multi-output wireless systems can achieve their potential for dramatically enhanced access to the radio spectrum. No prior work has fully addressed the interdisciplinary issues associated with implementing a complete millimeter wave massive multi-input multi-output wireless transceiver, covering crucial aspects such as the impact of antenna array geometries, demodulation, baseband processing, sampling and multichannel data aggregation. Prior efforts have been restricted to scenarios with a small number of antennas, lower frequencies, or hybrid architectures in which full beamforming flexibility is not available. There are many interdisciplinary challenges that must be addressed before a system that jointly exploits millimeter wave frequencies, massive multi-input multi-output wireless and small cells could be analyzed and its potential for providing enhanced spectrum access quantified. The enormous gains in capacity and spectral efficiency that could be provided by a millimeter wave massive multi-input multi-output wireless system could have a revolutionary effect on wireless applications. This impact includes not only consumer applications, but those involving emergency services, medical information systems, environmental monitoring, smart-grid energy distribution, smart transportation systems - all technologies that rely on access to high-speed broadband wireless services.
