Li Xiao, Ph.D. - US grants

The primary objective of this proposal is to enable first-principle simulations of liquid atomization involving turbulence in a realistic atomizer, which is at least an order of magnitude more complicated than is possible today with current computing capabilities. Achieving this objective will facilitate the fundamental understanding needed of liquid atomization. It will also enable the develop of a theory-based method for design, control, and optimization of atomizers for ultra efficient, clean, and stable combustion in automotive, aircraft, and rocket engines, for spray forming that creates structurally perfect materials, for liquid jets that will not atomize until reaching high elevations for fighting fires as well as other applications.

Broader Impacts of the Proposed Activity
The proposed research will greatly enhance the university's infrastructure for performing high-performance scientific computing of liquid atomization through parallel network RAM and the scientific software to be developed.
In terms of education and outreach, the results of this research will be disseminated via annual CFD short course and a workshop to be held at Michigan State University (MSU).
There will also be a summer internship program for undergraduate students each summer at MSU, where students will work with the PIs of this proposal in carrying out the research. For this internship program, students from under represented groups will be aggressively recruited.
Finally, undergraduate and graduate courses will be developed in liquid atomization in order to educate students in this area of scientific computing.

Collaborative Research: Foundations of Solving Large Direct and Inverse Scattering Problems --- Algorithm Analysis and System Support



Summary

We propose to develop and implement new computational methods on large cluster-based high-end systems for solving the direct and inverse problems in electromagnetics motivated by industrial and military applications. We will address two sets of important and closely related technical issues for this high-end scientific computing project. First, the targeted problems for the proposed project are large-scale direct and inverse scattering problems that can be only solved in high-end systems. These problems involve
particularly electromagnetic wave propagation with high wave numbers. A major difficulty for solving the inverse problems by an optimization method is the ill-posedness and the presence of many local minima. We propose a novel approach for solving the inverse medium scattering problem of Maxwell's equations in three dimensions. Crucial to the approach will be the development of an efficient regularized iterative linearization algorithm (recursive linearization with respect to the wave number). A challenge in
developing our numerical methods is to deal with large and structured data sets. The second set of technical issues for solving the targeted problems is concerned with the lack of system support in high-end architectures to maintain high sustained performance of
computing due to increasingly high speed gap between the CPU and the memory and the I/O storage. This challenge can also be found in many other large scientific computation problems on high-end systems. An equivalently important objective to the scientific computing in this proposal is to design and build effective system support by effectively allocating both CPU and memory resources, by establishing a global network RAM system in high-end architecture, and by providing exceptional system handlers to deal with dynamic and unexpectedlylarge memory demands from applications.

Intellectual merits of this proposal come from several aspects. (1) Our proposed numerical methods will address several scientific challenges in applied mathematics including electromagnetic wave propagation with high wave numbers, ill-posedness for inverse problems, and management of large data sets in multiple dimensions. (2) Processors and high-end systems have become increasingly complex, which makes the understanding of execution behavior more and more difficult. Our proposed system support based on both hardware counters and a system kernel instrumentation tool will address the system complexity issue, and provide insightful runtime system information for resource management systems with low overhead. (3) In order to effectively support high sustained performance and high productivity computing in clusters, our system support aims for several important resource management objectives, such as high memory utilization, low communication latency, and fast response time. (4) Although our system will be mainly tested by solving the large direct and inverse problems, it is also our aim to build it as a general purpose system so that it will become a fundamental software system infrastructure for many other large scientific applications in high-end systems.

Broader impact of this proposal will be: (1) Due to the fast development of high performance systems, computational electromagnetics has become a fundamental, vigorously growing technology in diverse science and engineering disciplines, such as
microwaves, millimeter waves, optics, and acoustics. Our computational models and cluster system support will provide an inexpensive and easily controllable ``virtual prototype" of the structures/media as opposed to costly, time-consuming physical
prototyping. (2) The proposed system resource management tools and system prototype will be disseminated in the high-end computing and systems community for a wide usage. (3) The research results will be timely introduced to both undergraduate and graduate curriculum development of scientific computing, parallel computing, and operating systems.

One important problem in peer-to-peer (P2P) overlay distributed systems is
to enforce the trust of the data stored in the system, and protect the
security and privacy of the peers. However, current research addresses
this problem in two directions with independent efforts: privacy and
security. In the privacy direction, researchers focus on proposing
anonymity protocols to hide peers' identities and on how to prevent the
attackers from revealing peers' identities. In the security direction, one
major research focus is to detect the identities of peers who are
suspicious to disturb normal operations in P2P systems intentionally or
unintentionally. If a peer can absolutely hide his/her identity, the
anonymity protocol can be possibly abused to threat the Internet security.
Increasing peers' privacy means increasing the difficulties to ensure
security. In this project, the PI conducts research to understand and
provide solutions on balancing the tradeoffs between privacy and security
in P2P systems. Specifically, (1) the PI develops anonymity communication
protocols to provide privacy protection for P2P users with security
insurance; (2) taking overlay distributed denial-of-service and file
integrity as case studies for security issues, the PI develops technical
security solutions without conflict with privacy requirements; (3) the
research investigates a wide range of security issues by developing
security solutions to balance tradeoffs of privacy and security. This
project is integrated to a Microsoft sponsored undergraduate course
development, where important topics of networking security and privacy,
design of decentralized P2P systems are being studied.
----------------------------------------------------------

This project, validating and extending work in sensor location and security, and MANET address allocation and reconfiguration, aims at building an infrastructure to service the development and evaluation of new algorithms and systems for

-Localization of sensor nodes within an obstructed sensor network environment,
-Security of sensor nodes and the maintenance of functional module integrity,
-Autoconfiguration of network addresses within MANETs, and
-Reconfiguration and address handoff of MANET addresses as nodes move and independent MANETs merge.

The planning work concentrates mainly in the localization, autoconfiguration and reconfiguration, while the security of sensor nodes and the maintenance of functional module integrity of the original work will be minimized. The researchers have obtained previous results via mathematical analysis and simulation and eventually would like to validate it on the originally proposed sensor and ad hoc research infrastructure, but at present will continue to simulate a large scaled sensor and ad hoc network.

Broader Impact: Adding real system advances the research quality and attracts new students. Both graduate and undergraduate students will be able to use the infrastructure in their research endeavors. The PIs plan to work with the Diversity office to attract underrepresented students in these projects.
Moreover, wireless and mobile ad hoc networks have significant potential to contribute to numerous societal advancements, including environmental monitoring, precision agriculture, vehicle-to-vehicle communication, emergency rescue operations, etc.

In deployment of sensor networks for various applications, sensors need to be placed in the areas of difficult terrain and natural obstacles. In such settings, many existing algorithms may perform poorly or may have high overhead while inefficiently consuming energy. One approach is to utilize mobile sensors in these situations, such as sensors with wheels. However, mobile wheeled sensors may not be able to move to the desired locations in the areas of difficult terrain with obstacles. Wheeled sensors can also be very expensive. A hopping sensor is a type of mobile sensor with a bionic mobility design that is inspired by creatures, such as grasshoppers. These sensors are still at an interesting concept stage. This proposal addresses the design, prototyping, and evaluation of hopping sensors and efficient algorithms for sensor deployment in difficult areas and rugged terrain. We focus on four research issues that are critical to the effective deployment and management of large scale sensor networks in such settings: robust and power-efficient hopping sensors, sensor localization, sensor coverage, and deployment of a self-adaptive all terrain sensor networks equipped with the hopping sensors and the proposed algorithms, which is called LEAPNet. Most future sensor networks are likely to be deployed in our targeted environments, and hence this research will benefit real-world applications such as monitoring ecosystems, disaster relief, and military reconnaissance. Strong collaborative efforts will be made to provide efficient and practical solutions with solid engineering designs and strong algorithmic foundations for this purpose. In addition, an innovative integration of the proposed research and education program will provide students with analytical skills and hands-on experiences by emerging technologies, which will better prepare them for strong technical careers in this rapidly growing and changing area.

The objectives of this project include the design of networked mobile robotic sensor nodes, which are
inspired by insects within the family gerridae, that can monitor water quality in lakes and rivers, and jump
repeatedly to enhance communication capability to overcome the effect of water on radio signal propagation.
This work will also develop energy efficient algorithms that carefully select sensor nodes to jump to gain the
benefit of longer distance communication while managing the energy required to jump. The robotic sensor
can harvest solar energy in a outdoor environment to maintain long term performance. New methods for
sensor localization as the sensor floats within a lake or river will be developed as well as efficient algorithms to collect sensor data and deliver it to the cyber infrastructure that could be used for hydrology modeling and prediction.

The results of this research effort should significantly enhance the capability of networked mobile sensor
systems for water quality monitoring in response to water pollution or other environmental conditions. These
concerns can be from oil spills, issues of bacterial levels, or even detection of levels of radiation from nuclear power plant disasters. This research should provide a cost-effective approach for water quality monitoring that can protect health and emergency management such as advising people to vacate areas of environmental damage. Furthermore, new sensor technology should be developed that provides enhanced opportunities for business and job opportunities to industries and society and provides students with a platform to learn and understand all levels of sensor development and operation.

The increasing spectrum demand from wireless applications cannot be satisfied with existing fixed spectrum allocation techniques. To provide dynamic spectrum access, auction based spectrum trading has received considerable attention. In spectrum trading, licensed users rent their unused spectrum to unlicensed users in exchange for money. The availability of unused spectrum depends on the usage pattern of licensed users and varies over time, frequency, and spatial domain. Unlicensed users can bid for this spectrum according to their needs which again vary by time and location. Thus, the auction scenario has to deal with the dynamic user demand and dynamic spectrum supply. Existing research on spectrum trading consider either a static auction environment or at most one side of the dynamics of the auction environment. Therefore, for improving spectrum unitization, it is critical to consider an auction scenario where bidders with heterogeneous demand arrive dynamically and bid for dynamically available spectrum units.

This project studies a dynamic auction framework and undertakes the development of auction rules and regulations designed to achieve an efficient spectrum allocation. The study consists of four major components. (1) Developing the dynamic auction framework with dynamic arrival of users with variable lifetime and dynamic spectrum supply. (2) Tackling the design challenges associated with the diversity in demand and supply in terms of units and duration and devising a truthful auction mechanism. (3) Designing a generic auction framework to fit the auction rules and regulations and define the communication method using a secure communication protocol. (4) Developing a testbed to implement, experiment and validate the auction framework. This research will enhance radio spectrum sharing and utilization, providing more spectrum opportunities for existing and new applications. This project provides interdisciplinary research opportunities for graduate students working on the proposed research. This project incorporates research results into courses and to recruit students from underrepresented backgrounds.

Dynamic spectrum access brings great benefits for users in overcrowded radio spectrum to opportunistically harvest spectrum resources, which is especially useful for high date rate applications such as video conferencing and multimedia streaming. Since available spectrum may be fragmented, carrier aggregation becomes a critical enabling technique. Many recent and emerging protocols (e.g., 802.11ac WiFi, 802.11af TV whitespace networks, 3GPP LTE unlicensed, and 802.11ay millimeter-wave networks) have been incorporating carrier aggregation mechanisms. However, these mechanisms are still preliminary and heavily rely on conventional MAC/PHY which is designed for nodes sharing the same contiguous channel. As carrier aggregation networks evolve to new generations, three inherent problems will become the fundamental barrier to their efficiency, fairness, and scalability: (i) a transmitter is unable to sense medium state during transmission, (ii) more users and high spectrum dynamics worsen the medium access contention, and (iii) the need of fast spectrum agreement to aggregate multiple fragmented spectrum chunks.

The objective of this research is to explore general primitives to effectively realize carrier aggregation with three research components: (i) Migrate the spectrum sensing paradigm from "detect and then transmit" to "simultaneously transmit and detect". The end goal is to overcome the efficiency and fairness problems in standard spectrum sensing mechanisms to enable asynchronous, out-of-band OFDM spectrum sensing and multi-antenna based full-duplex sensing. (ii) Address the contention issue in wideband fragmented spectrum sharing with low collision probabilities and low overhead. A spectrum arbitration mechanism is proposed, enabled by a parallel bitwise signaling scheme and a novel compound packet preamble design. (iii) Design a light-weight in-band mechanism that achieves fast synchronization and spectrum agreement, so as to overcome the substantial control channel overhead for spectrum agreement under high spectrum dynamics. The success of the proposed research will enhance radio spectrum sharing, and inspire standardization bodies to incorporate the proposed new protocol primitives for much higher performance. A continuous effort will be made to integrate the proposed research to education program, to mentor undergraduate researchers, and recruit women and minority students.

Emerging mobile applications are increasingly focused on technology interacting with and augmenting the real-world environment the user occupies. Augmented reality is a technology that places virtual objects on a user?s view of the real world with a wide range of applications such as navigation, gaming, and education. Augmented reality as a technology is inherently extremely computation-heavy, leading to latency, accuracy, and energy-consumption issues on resource-constrained smartphones. Image-recognitio- based augmented reality compounds this issue by requiring the computation of the entire image-recognition pipeline. In addition, mobile hardware is not designed with augmented reality and heavy image-based computations in mind. Mobile caching, multicores, GPUs and other mobile architectures are not being utilized to their full potential to help resolve the issues plaguing mobile augmented reality. This project explores methods to utilize the unique mobile architecture of off-the-shelf smartphones in new ways to realize augmented reality on a wide variety of mobile devices. Specifically, augmented reality has become an important tool for educators at all levels from K-12 all the way through collegiate and post-graduation education. This project will allow for this new educational technology to be more widely utilized in the world.

The objective of this project is to enable smartphones to support augmented reality via efficiently and seamlessly computing image-recognition and world-tracking tasks simultaneously, with three research components. (1) Investigate the foundational issues of smartphone-based augmented-reality through approximate-tracking to provide high-quality object tracking at reduced computational and energy loads. (2) Research software-defined caching techniques to utilize the unique nature of mobile augmented reality to provide caching specially designed for highly collaborative device-to-device augmented reality. (3) Explore hardware-based techniques relating to GPGPU computation and cache management to facilitate fast and efficient image-recognition and augmented-reality tasks.

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.