Robert L. Heath - US grants

This is a request to obtain funds to acquire an infrared thermal imaging system. The equipment will be used by five investigators as part of several ongoing studies of plant and animal physiology. The specific projects described in this proposal utilize thermal imaging to monitor cutaneous vasomotor changes in dogs, rats and rabbits with respect to heat, exercise and hydration; to study thermo regulation in desert insects and Adelie penguins; to study heat and pollutant effects on gas exchange in plants; to measure the influence of solar radiation on amphibian thermal regulation.

The Water Resources Research Institute (WRRI) of Kent State University (KSU) seeks to establish a computer facility for the analysis and distribution of graphical data for research in water resources. The facility will consist of a central file and compute server, a network of graphical workstations for processing and displaying three-dimensional data and PowerPC workstations for processing and displaying two-dimensional data, and the necessary input-output support for high quality image acquisition (scanners, digital cameras, video cameras) and output (color and black-andwhite printers, slide maker). Environmental research requires a collaborative, interdisciplinary approach because of the complexity of natural systems. Field observations must be integrated with laboratory experiments. Knowledge from one discipline must be applied to interpret data in another. The sheer size of some systems (spatial and temporal) requires data sets too large for a single group of investigators to acquire or to evaluate. The proposed facility is specifically intended to enhance collaborative research related to water in the environment, including on-campus, in-state and out-of-state collaborations. The network will simplify transfer of graphical data among WRRI members and between KSU and off-campus workers. The file/compute server will provide on-campus access to large environmental and climate-modeling databases and will facilitate the numerically intensive calculations required for modeling complex systems. Graphical workstations will enable 3D visualization of climatological, hydrological and chemical structures in real time. The facility will be administered by WRRI, an interdisciplinary organization created to foster interdepartmental, collaborative research on water resource problems. Current research in WRRI which would be supported by this facility includes projects on water quality, climate and global change, ecosystem health, and the application of molecular level tools to macroscopic environmental problems. All the PI's are engaged in water quality research, with specific interests in developing biological and chemical methodologies (Leff, Cabaniss), system level studies of chemical and biological perturbations of ecosystems (Heath, Leff, Smith), modeling of catchment-basin flow and pollutant transport (Carlson, Maurice), and mapping of water quality on a regional or national scale (Carlson, Lee). Drs. Maurice and Smith (Geology) are also examining the role of water in climate and global change. Drs. Carlson, Heath, Leff (Biological Sciences) and Lee (Geography) are examining ecosystem health on microscopic and macroscopic levels. Drs. Cabaniss (Chemistry), Heath (Biological Sciences) and Maurice are exploring the application of molecular-level tools- molecular spectroscopy, biochemical assays, atomic force microscopy- to environmental problems.

9513967 Heath The Department of Botany and Plant Sciences, University of California, Riverside (UCR) has held a symposium for the past 17 years on timely topics in plant biology. The 18th Annual Sympoisum in Plant Biology, entitled "How Plants Use the Light Environment to Regulate Growth and Development," will be held January 18-20, 1996. The objective of the 1996 Symposium is to bring together scientists working in the area of plant growth and development as influenced by light, with the intention that the presentations, and the discussion arising from them, will lead to a greater understanding of the cellular, physiological, and genetic aspects of the photo-environment. It will cover the major electromagnetic bands which influence plants on earth, including the phytochrome and blue light responses and the UV spectrum which will be increasing due to the weakening of the ozone layer shield. The five sections will be: 1) properties of the photoreceptors; 2) light regulation of gene expression; 3) signal transduction; 4) the use of mutants; and 5) the whole plant response, to also including circadian rhythms. Speakers from abroad and from other institutions, including young investigators who have introduced innovative molecular approaches, will be invited to participate. Poster presentations will also be included in the program. The presented papers, as well as poster abstracts, will be published rapidly and inexpensively. Riverside is an ideal location for this symposium because there are many researchers within California and the western states.

This award provides funding to initiate a Research Experience for Undergraduates (REU) site that will provide a eight-week summer research experience in limnology and environmental science for a total of five students. The REU site will be run by Kent State University and will focus on the environmental issues in Lake Erie, its tributaries and its coastal wetlands. The REU program is based on the theory that mentoring and networking are the most effective methods for recruiting students into a profession and for training young researchers. This program will provide students with an in-depth understanding of the ecology and limnology of the Great Lakes, particularly Lake Erie. Plans for the site include all the essential elements for an excellent REU site. A team of faculty mentors will help the students develop research projects on the freshwater ecology, microbiology, geology and environmental chemistry of the Great Lakes. Students will participate in a series of lectures focused on scientific methods, ethics in science, and an interdisciplinary view of the Great Lakes, and they will receive one-on-one mentoring from researchers. They will also participate in field trips in the Lake Erie watershed. At the end of the program, students will develop a presentation on their research project and will be encouraged to present their results at a major scientific conference. This program fills an important niche in the continuum of program supported under the REU program by the Division of Ocean Sciences. It's focus on the ecology and limnology of the Lake Erie is unique. The program will advance the awareness of students concerning important environmental issues in the coastal regions and watersheds of the Great Lakes. The program also has the potential to recruit minority students to the Geosciences.

This collaborative project (with Dandekar at Drexel, proposal 03-22795), performing an experimental investigation of Multiple-Input Multiple-Output (MIMO) systems, aims at designing and testing novel antenna systems for wideband digital wireless communication systems to provide designers with new data points. The experimental program is divided between two testbeds. The first testbed (implemented at Drexel) focuses on the electromagnetic propagation characteristics of the MIMO channel, while the second testbed (implemented at UT-Austin) makes use or the electromagnetic characterization to examine the channel impact of the higher network layers. Systems with MIMO communication links, fueled by the increasing demand for high-speed and high-quality mobile communication, use multiple antenna arrays, one at the transmitter and one at the receiver, to take advantage of the spatial dimension of the propagation channel. When properly designed, multi-array communication links can provide multi-fold increases in link throughput in addition to reductions in channel variation which in turn may be used to provide higher rates to single users, lower delay links, or to allow multiple users to coexist in the spatial channel. Because of this advantages, MIMO capability is being considered for indoor local area networks (LAN)s, cellular multimedia data networks, and broadband wireless access. However, without careful design, the results of electromagnetic interaction, such as mutual coupling, may correlate the channel coefficients, thus resulting in a loss in terms of capacity or error probability. Real experimental propagation and experimental data are needed to correlate the tradeoffs between different antenna designs, antenna placements, scattering models, channel bandwidth, and transceiver algorithms. These research involves designing and building two experimental platforms for wideband MIMO propagation channel measurement and real-time prototyping. The prototype constructed at Drexel will be used for the low-level electromagnetic characterization of the MIMO propagation channel, while the UTA prototype will be used to demonstrate the impact of this characterization on higher layers. Thus the work involves
Estimating and analyzing wideband MIMO propagation channels,
Verifying performance loss due to mutual coupling and electromagnetic interactions, and
Studying advanced MIMO architectures in which the antennas are distributed throughout the environment.
The research plan involves a synergy of field experimentation with prototype hardware backed by computational electromagnetic (CEM) modeling and communication theory.

Broader impacts are addressed in several ways:
Introducing jointly concepts of communications and applied electromagnetics to advanced undergraduate and graduate courses (rather than as separate courses),
Deploying interdisciplinary course modules jointly taught at both universities, generating a lab manual containing a series of experiments and a set of lecture notes,
Providing the data records collected that illustrate concepts critical to design, and
Enabling instructors at other universities to incorporate some experimental methods into their teaching.

Wireless communication and networking is increasingly prevalent in everyday life. This project focuses on exploiting the substantial number of degrees of freedom that are increasingly available at the physical layer to design and build multihop wireless networks with enhanced capacity, robustness, and usability. Our central objective is to identify how to best leverage these tradeoffs and experiment with complementary medium access control (MAC) protocols that are capable of doing so.

Our approach is an integrated one that focuses on cross-layer algorithm design and implementation. The focus is three fold: one, the development of a multi-hop wireless network prototype testbed - Hydra; two, the design and implementation of adaptive physical layer (PHY) modulation and coding schemes that significantly increase PHY flexibility; and three, the design and implementation of novel medium access control (MAC) algorithms that efficiently exploit the flexibility offered by next-generation PHYs. Our approach crosses the boundaries between both the physical layer and the networking layers and between theory and practice.

Perhaps the greatest impact of our work comes from designing and building a working prototype. This allows us to generate results that are extremely well grounded in physical reality, taking us away from the simulation-oriented approach that currently dominates wireless networking. These results will help to enable more effective wireless networks, especially those where data takes multiple wireless hops. We will disseminate our results through the production of students as well as the usual publication venues.

Mobile wireless communication is becoming the enabler of Internet access. Unfortunately, mobile networks are not equipped to provide the high-speed access that Internet users expect due to fundamental problems associated with wireless communication: scarce bandwidth and poor quality. To overcome these problems, this research develops a family of algorithms that allow the transmitter to respond to changes in the propagation channel. The major innovation in this work is the development of new methods for compressing information about the propagation channel. This information is sent from the receiver back to the transmitter to help the transmitter adjust the transmitted signal.

The approach to this research is to develop a framework for source coding with a new source: the wireless channel. This differs from traditional source coding in that the objective is to improve communication theoretic system performance as opposed to improving the fidelity of the reconstruction. The objectives of the research are to determine what channel state information should be quantized; develop algorithms for quantizing the essential parameters of the channel; derive suitable communication theoretic notions of fidelity of the quantization such as mutual information and bit error rate; characterize the tradeoff between feedback rate and network performance; and confront practical issues introduced by estimation error, errors and delay in the feedback channel, and implementation constraints.

The broader impacts of the work are expected in diverse areas including research in the form of new algorithms, theoretical results, and insights; industry through mobile network applications developed by industry partners; and education through better trained engineers, research experiences for minority students, mentoring, and enhanced learning worldwide thanks to publicly available courseware.

NeTS-ProWiN: Practical Use of Channel Information in Multihop Wireless Networks

The explosion of cellular communications has lead to a tremendous body of work on sophisticated wireless communication techniques many of which require knowledge of wireless channel state information (CSI). However, very little of this work has been applied to wireless networks where nodes communicate over multiple hops and where distributed rather than centralized solutions are required.

Taking advantage of this opportunity requires a marriage of physical layer, medium access control, and networking expertise to use CSI to improve network capacity, throughput, and reliability, and to reduce delay. New algorithms and protocols are being developed that span the protocol stack to provide effective sharing of CSI. These techniques are being validated by using a novel, open source, multihop, multiantenna, wireless network testbed - Hydra.

Contributions will include:
. Development of strategies for using feedback to effectively share CSI.
. Development of multiuser channel models, both to aid in the dissemination of CSI and also to facilitate accurate network level simulation of feedback based systems.
. Demonstrating the use of CSI to allow practical exploitation of the relay channel at the level of the network.

In addition to the usual dissemination of research results through the production of students and scholarly papers, the Hydra prototype itself will be available to other researchers. Since Hydra is open source and based on inexpensive hardware, these outside groups will be able to test and build upon the results from this research program.

The demand for wireless communication is only growing, as wireless becomes the dominant means of Internet access. To meet this demand, wireless systems may employ new concepts such as multiple antennas, multiple user processing, transmitter coordination, and interference alignment to improve system capacity. Unfortunately, implementing these communication techniques requires substantial feedback from the receiver to the transmitter and more complex signal processing to reap the promised capacity gains. Fortunately, there is special manifold structure in these signal processing problems that is not yet exploited.
Manifolds are generalizations of surfaces to higher dimensions and can be used to capture structure in signals. Special manifolds like the Stiefel and Grassmann manifolds are of particular interest in wireless communication systems that use multiple antennas. This research involves developing a suite of signal processing techniques for analyzing, filtering, predicting, and optimizing signals with curved manifold structure.

This project applies new manifold signal processing tools to three emerging problems in wireless communication: multiple user multiple antenna communication, interference aligned transmission, and coordinated base station transmission. Each problem requires progressively more sophisticated manifold signal processing techniques to enable their eventual application in commercial and military wireless communication systems. The most immediate impact of this research will be to improve the quality and capacity of wireless communication links thus impacting their design, implementation and deployment. The long range impact will be tools and analytical techniques that influence other disciplines including control theory, optimization, image and video processing, data mining and manifold learning. Broader impacts of this research program will occur in education through the training of undergraduate and graduate students and in industry through rapid dissemination of research results through electronic preprints.

Cellular system infrastructure deployment is becoming more heterogeneous including random deployment locations and more kinds of infrastructure. Using a mixture of macro, pico, and femto base stations, as well as fixed relay stations and distributed antennas, heterogeneous networks have the potential to break through the capacity bottleneck. While holding great promise, this potential has yet to be realized since the new infrastructure creates a different and challenging distributed interference environment.

This research develops new mathematical tools to enable a simplified analysis of cellular systems exploiting concepts from stochastic geometry and random shapes, to model the impact of random interference source locations, accounting for important environmental considerations. It uses these tools to analyze and develop interference management strategies for different proposed technologies including small cell networks with heterogeneous interference and large cell networks with many antennas. A main theme in the analysis and algorithms is the use of antennas to avoid, cancel, align, and otherwise mitigate the effects of interference -- even with limited coordination possible among transmitters.

Broader impacts of the proposed theory and algorithms are expected in diverse areas. The mathematical tools and fundamental theory will impact the understanding and design of communication systems taking into account the network geometry. The signal processing algorithms will pave the way for a new understanding of multiple antenna communication techniques. Industry impact will occur through the WICAT / Wireless Networking and Communications Group industrial affiliates incorporating research results into their wireless networking technologies. The project will foster the training of graduate students in course projects and will reach out to the community through public demonstrations.

Cellular networks are challenged by the limited spectrum available at microwave frequencies. In the past several years, various technologies have been proposed to achieve ultra-high levels of spectral efficiency involving the use of multiple antennas, serving multiple users, and shrinking cell sizes. Many of these technologies have significant potential but do not provide the high data rates due to the limited spectrum available. An alternative is to use the millimeter wave (mmWave) band between 3-300 GHz to provide high bandwidth communication channels. Realizing mmWave cellular communication, however, requires addressing the fundamental difference between mmWave and microwave communication. mmWave networks will require high gain directional antennas at the transmitter and receiver to overcome path loss at higher frequencies and to ensure sufficient signal-to-noise-ratio at the receiver. These directional antennas must be implemented using a combination of analog and digital signal processing techniques, due to power hungry mixed signal hardware. The cellular systems must be designed to support highly directional transmission and reception from their inception, and must also deal with the increased impact of signal blockages.

This research project will establish the potential of millimeter wave cellular networks by incorporating the key features and constraints of mmWave communication. To understand the potential of such networks, this project will develop mathematical tools to analyze the performance of large-scale millimeter wave cellular networks, explicitly incorporating directional antennas and blockages. To enable robust communication, this project will create communication strategies that are simultaneously suitable for the harsh millimeter wave propagation environment and the limited capabilities of millimeter wave hardware. To leverage potential coexistence with lower frequency signals, this project will investigate the design and analysis of a hybrid-access system that exploits the coexistence of microwave and mmWave cellular systems. A main theme in the research thrusts is to incorporate key features of mmWave communication into the algorithms and analysis.

Broader impacts of the project's theory, algorithms, and architectures are expected in diverse areas. The mathematical tools developed in this research will impact the design and realization of cellular networks with directional communication. The signal processing algorithms based on array processing and stochastic geometry will pave the way for a new understanding of millimeter wave wireless communication. Industry impact will occur through the Wireless Networking and Communications Group (WNCG) industrial affiliates incorporating the results of the research into their wireless networking technologies. The project will foster the training of graduate students in course projects and will reach out to the community through public demonstrations.

As communication systems embrace ever wider bandwidths, analog-to-digital converters (ADCs) struggle to meet rate, resolution, and power requirements. The problem is exacerbated by the massive antenna arrays under consideration for next-generation wireless, which imply tens or even hundreds of receive channels. In response, this project develops new communication systems that operate with very low-resolution (e.g., 1-3 bit) ADCs. In particular, the investigators derive fundamental limits on quantized many-antenna communication and design advanced communications strategies to approach those limits.

The use of low-resolution ADCs radically changes both the theory and practice of communication, motivating a thorough re-examination of capacity bounds, modulation and coding designs, receiver processing algorithms, and limited-feedback strategies. The investigators address these topics in the context of multiple-input multiple-output systems with massive arrays. In particular, they derive theoretical capacity (or achievable rate) bounds that account for coarse ADC quantization, imperfect channel state information (CSI) at the receiver, partial CSI at the transmitter, and multi-user interference. They also characterize the mean-squared error achievable for sparse-channel estimation and the symbol-error rate achievable via equalization, both in the large-system limit. In addition, they develop limited-feedback strategies to provide transmitter CSI to the transmitter, transmitter-precoding designs that exploit that CSI, and optimized training sequences. Furthermore, they develop channel-estimation algorithms that learn and exploit channel sparsity; equalization (and/or multi-user detection) algorithms that take channel-estimation error into account; and efficient strategies for joint channel-estimation, equalization, and decoding. Finally, they develop optimization and adaptation strategies for the ADC thresholds, and blind calibration strategies that account for ADC imperfections. The research methodology leverages recent results from wireless communication theory, one-bit compressed sensing, and approximate message passing.

Cellular communication networks have become one of society's most important and complex technologies. Mobile data usage has been approximately doubling every year since about 2008, which corresponds to a 1,000 times increase over a decade. To meet this insatiable demand, much more bandwidth is required for cellular systems. This requires going to (much) higher frequencies, since there is very little unused spectrum below about 10 GHz, especially at peak times in urban areas. Significant amounts of lightly used spectrum is available in the "millimeter wave" (mmW) spectrum, defined here as being above 25 GHz. There are numerous fundamental technical challenges in mobile data communication above 25 GHz, and the goal of this project is to explore these fundamental challenges and build new tools to overcome them. The research undertaken should considerably impact the telecommunications industry and society as a whole, by enabling an entirely new paradigm for cellular communication networks. In addition to theoretical contributions, the investigators will pursue an aggressive technology transition plan through their numerous government and industry partners.

This project will develop new mathematical and analysis tools that uncover the potential of mmW cellular networks. The research agenda is built on the belief that the wealth of tools developed for lower frequency systems are insufficient to capture key mmW signal propagation features, specifically high directionality and blockage. The research is structured as three inter-related thrusts, each with several proposed research tasks: mmW signal strength, mmW interference, and the mmW network connectivity. The research tasks are unified around several technical themes that cut across all three thrusts: (1) modeling mmW networks in 3D, including the obstacles for which statistical blocking models will be developed and validated with real building data; (2) accounting for signal and interference correlation in performance analysis, which will be significant in mmW due to their main randomizing factors being blocking and beam alignment rather than fading and shadowing; (3) accounting for and addressing the possibly severe effects of mobility on beam alignment and network connectivity.

The developed theories will be used to devise useful models, parametrized by real building data to facilitate fast performance evaluation. Features of practical mmW transceivers like beam adaptation, mobility, and interference cancellation will be included and used to study key design tradeoffs. The new mathematical framework developed in this proposal will allow transparent and comprehensive performance analysis of mmW cellular systems, and enable the development and fair comparison of new communication techniques.

Automotive and aerial vehicles are being equipped with more sensors to enhance automated driving/flying as well as to perform general sensing tasks. Despite increasing data rates from cameras, LIDAR, and RADAR, along with higher levels of onboard computation, such vehicles do not have the means to share such high rate sensor data to enhance their situational awareness or improve cooperation. Millimeter-wave communication is one solution for high data rate vehicular communications. Achieving the highest rates with millimeter-wave, though, requires frequent link reconfiguration in mobile environments. This research is aimed at using the information derived from sensors that operate in some band other than the communication band, to help configure the communication link. The emphasis is placed on theoretical development and experimental validation of frameworks of using sensor information to permit link reconfiguration in mobile environments. Broader impacts include teaching, mentoring, participation of underrepresented groups, community outreach through demos and videos, and technology transfer through collaborations with industry partners.

This research project addresses the fundamentals of communication using side information derived from various sensing techniques. The project leverages mathematical tools from array signal processing and machine learning and will focus on the applications related to beam configuration. On the theoretical side, this project develops mathematical tools for understanding the frequency of beam realignment, a framework for relating radar and communication paths, and fundamental limits for sensor-aided beam alignment in millimeter-wave communication. On the practical side, this project tests critical hypothesis about statistical correlations between sensor measurements and communication link performance in a series of testbeds. The testbeds support low-cost radar derived from WiFi signals and higher cost automotive radars, coupled with custom developed millimeter-wave communication link capability. Outcomes of this research include (i) an understanding of how much overhead can be reduced by using information derived from potentially very different communication bands, and (ii) rapidly reconfigurable and robust millimeter-wave communication links.

This project will develop new energy-efficient techniques to co-use wireless spectrum for both communications and sensing. The setting considered will be that of emerging cellular networks consisting of a dense deployment of base-stations (wireless infrastructure on cell towers) providing overlapping wireless coverage. By enabling co-use of spectrum, this research will result in both better connectivity and improved data rates to users, and improved sensing through radar technologies for locating and tracking mobile users. These technologies will thus enable new smart city services and applications. On the educational front, the project will lead to development of new educational materials. Outreach activities to middle and high school students will occur through widely-accessible lecture series. The proposed activities will pay special attention to promoting diversity and nurturing student talent. Finally, the results will be shared with industry through the industrial affiliates program of the Wireless Networking and Communications Group (WNCG) at The University of Texas at Austin.

The proposed research is geared at developing a new class of energy efficient cross-layer resource allocation algorithms for joint management of base-station and spectrum resources in dense wireless networks. At the same time, mathematical models and tools based on stochastic geometry will be introduced enabling one to characterize and optimize energy-performance for dense deployments. The intellectual merit can be summarized along four thrusts: (a) Infrastructure and spectrum co-use for communications and radar/sensing, where the focus will be on algorithms for handling interference from a mixture of communications and radar waveforms; (b) Stochastic geometry models and tools for characterizing heterogeneous spectrum use over space, where Central Limit Theorems over space will be developed to characterize SINR and radar distortion performance; (c) Algorithms for mode switching and base-station activation, where load-dependent algorithms for network management will be developed that are cognizant of hysteresis costs associated with changing modes and activation states of base-stations; and (d) Managing user dynamics through directed association to reduce the frequency of change in the network, which will result in improved convergence behavior of management algorithms at various spatial scales.

Wearable electronic devices for advanced applications like augmented or virtual reality typically generate large amounts of data. Using wireless links to deliver such data avoids the need for cumbersome cables, yet conventional wireless technologies do not support the gigabits per second throughput envisioned by these emerging applications. Millimeter wave communication is one solution to provide wearable electronic devices with higher data rates and lower latency. This research is aimed at providing a rigorous and systematic design of the key aspects of a millimeter wave wearable network. The emphasis is placed on designing important aspects of the system that will permit operation in crowded areas with many people using wearable devices. Broader impacts include teaching, mentoring, participation of underrepresented groups, community outreach through demonstrations and videos, and technology transfer through collaborations with industry partners.

This research project addresses the fundamentals of communication in millimeter wave wearable links at the physical layer and medium access control layers. On the theoretical side, this project develops new mathematical tools for estimating millimeter wave channels and devises new algorithms for communicating in millimeter wave multiple antenna channels incorporating array constraints. It also creates new protocols, which are designed from the ground up to include multiple antenna capabilities not found in existing millimeter wave medium access control protocol work. On the practical side, this project creates channel models that incorporate array geometry, orientation, and blockage effects tailored to the wearable setting. It also implements key aspects of the algorithms in a testbed to validate the theoretical hypotheses and provide refinements to models. Outcomes of this research include rapidly re-configurable and robust high bandwidth millimeter wave communication links for wearable devices.

Cellular communication systems continue to incorporate new multiple-antenna technologies. In particular, third, fourth and fifth generation cellular systems saw advancements in the use of multiple antennas at the base-station infrastructure and multiple antennas in the devices. A main application of these antennas was to support multiple-input multiple-output (MIMO) communication, which is known to increase spectral efficiency and thus the data rates that can be achieved by devices in a given bandwidth. The numbers of antennas and the ways the antennas are used can vary across device models even from the same manufacturer. At the same time, the types of devices supported in cellular systems is growing beyond smartphones to include other highly mobile platforms like aerial vehicles, automobiles, and robots. The differences in the hardware between devices, coupled with the high device mobility, makes it challenging to configure the antennas to provide MIMO communication with the highest performance. This project develops machine learning-inspired solutions to empower devices to learn optimal configurations collaboratively. <br/><br/>System-wide Operation via Learning In-device Dissimilarities is a cooperation among experts in wireless communications at North Carolina State University (NC State) and Tampere University (TAU). The overall objective of the proposal is to employ machine-learning-assisted collaborative solutions for MIMO beam prediction and codebook optimization in a large-scale dynamic system. The key challenge of such networks is the extreme diversity of the devices’ hardware (e.g., antenna designs and configurations). The existing distributed ML approaches do not explicitly include this type of client heterogeneity and do not fully support the temporal and spatial heterogeneity of data, network resources, and deployments. The project team will develop a novel integrated-learning and wireless-networking framework, which will enable the design and optimization of advanced MIMO beam-management solutions specifically tailored to the highly diverse and dynamic system. This project will result in new algorithms for collaborative device-centric beam management for 5G+/pre-6G MIMO communications in non-stationary environments with highly mobile and heterogeneous agents. The specific technical contributions occur in several directions: (a) Distributed user-centric learning for optimizing codebook-based MIMO communications; (b) Novel representation of device heterogeneity in an ML-friendly way; and (c) Network-resource optimization to facilitate distributed learning. The immediate impact will be improved communication efficiency in 5G+/pre-6G networks. The longer-term impact will be the establishment of the core principles for designing fast and reliable methods of distributed ML training deployed over wireless systems with diverse hardware and resources.<br/><br/>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.