Surpassing the fundamental limits that govern all electromagnetic structures, such as reciprocity and the delay-bandwidth-size limit, will have a transformative impact on all applications based on electromagnetic circuits and systems. For instance, violating principles of reciprocity enables non-reciprocal components such as isolators and circulators, which find application in full-duplex wireless radios, radar, biomedical imaging, and quantum computing systems. Overcoming the delay-bandwidth-size limit enables ultra-broadband yet extremely-compact devices whose size is not fundamentally related to the wavelength at the operating frequency. The focus of this dissertation is on using time-variance as a new toolbox to overcome these fundamental limits and re-imagine circuit and system design. Traditional non-reciprocal components are realized using ferrite materials that loose their reciprocity under the application of external magnetic bias. However, the sheer volume, cost and weight of these magnet based non-reciprocal components coupled with their inability to be fabricated in conventional semiconductor processes, have limited their application to bulky and large-scale systems. Other approaches such as active-biased and non-linearity based non-reciprocity are compatible with semiconductor processes, however, they suffer from other poor linearity and noise performance. In this dissertation, using passive transistor switch as the modulating element, we have proposed the concept of spatio-temporal conductivity modulation and have demonstrated a gamut of non-reciprocal devices ranging from gyrators to isolators and circulators. Through novel circuit topologies, for the first time, we have demonstrated on-chip circulators with multi-watt input power handling, operation at high millimeter-wave frequencies, and tailor made circulators for emerging technologies such as simultaneous-transmit-and-receive MRI and quantum computing. Delay-bandwidth-size trade-off is another fundamental electromagnetic limit, that constrains the delay imparted by a medium or a device within a fixed footprint to be inversely proportional to the signal bandwidth. It is this limit that governs the size of any microwave passive devices to be inversely proportional to its operating frequency. As a part of this dissertation, through intelligent clocking of switched capacitor networks we overcame the delay-bandwidth-size limit, thus resulting in infinitesimal, yet broadband microwave devices. Here we proposed a new paradigm in wave propagation where the properties such as the propagation delay and characteristic impedance does not depend on the constituent elements/materials of the medium, but rather heavily rely on the user-defined modulation scheme, thereby opening huge opportunities for realizing highly-reconfigurable passives. Leveraging these concepts, we demonstrated wide range of reciprocal an non-reciprocal devices including ultra-compact delay elements, highly-reconfigurable microwave passives, ultra-wideband circulators with infinitesimal form-factors and dispersion-free chip scale floquet topological insulators. Application of these devices have also been evaluated in real-world systems through our demonstrations of wideband, full-duplex receivers leveraging switched capacitors based true-time-delay interference cancelers and floquet topological insulator based antenna interfaces for full-duplex phased-arrays and ultra-wideband beamformers. Furthermore, to cater the growing RF and microwave needs of future, large-scale quantum computing systems, we demonstrated a low-cryogenic, wideband circulator based on time modulation of superconducting devices. This superconducting circulator is expected to operate alongside the superconducting qubits, inside a dilution refrigerator at 10mK-100mK, thus enabling a tightly integrated quantum system. We also presented the design and implementation of a cryogenic-CMOS clock driver chip that will generate the clocks required by the superconducting circulator. Finally, we also demonstrated the design and implementation of a low-noise, low power consumption, 6GHz - 8GHz cryogenic downconversion receiver at 4K for cryogenic qubit readout.

No description available.

In this dissertation, we study energy efficient transmission in wireless communication networks. The general problem of energy efficient transmission over wireless networks is formulated into optimization problems for the following distinct (but inter-related) settings: Problem Setting 1: The minimization of energy (power) consumption given a system throughput and other constraints, and Problem Setting 2: The maximization of system throughput given energy (power) constraints. Under Problem Setting 1, we focus on energy efficient transmission problems over wideband channels. The first result we obtain is as follows: We consider a two user multiple access channel. In this multiple access channel, previous research shows that cooperation with respect to block error rate is only possible if two transmitters share their sources completely. However, we find that a modified pulse position modulation with synchronization enables cooperation without complete sharing of their sources between two transmitters if we replace a block error rate requirement with a normalized error rate constraint. Normalized error rate, a quantity that resembles bit error rate, is developed in this work as an error metric that is of value in practical communication systems. We show full cooperation between two transmitters without sharing their sources by deriving that the minimum energy per bit required for reliable transmission reduces by quarter compared with the minimum energy per bit required for point-to-point channels. Next, we generalize this analysis to a cognitive communication framework with a wideband cognitive transmitter, which can causally sense signal levels over multiple frequency bands, and a cognitive receiver. We assume that multiple legitimate users already exist in the system and each one transmits in its own non-overlapping frequency band. In this setting, from order statistical analysis, we show that the wideband cognitive transmit-receive pair is able to communicate reliably with minimum energy as if the legitimate users were absent from the system, while causing negligible interference to bandlimited legitimate users. The wideband cognitive transmit-receive pair employs a strategy defined as opportunistic group orthogonal signaling to achieve the minimum energy per bit. Under Problem Setting 2, we investigate the impact of correlation and transmit and receive strategies on the throughput of multiple antenna broadcast channels in cellular networks. With perfect channel state information at the transmitter, it is well known that dirty paper coding (DPC) is the optimal multi-user broadcast transmission method. However, with partial channel state information at the transmitter, the picture changes significantly. Specifically, since multi-user transmission is unable to employ DPC perfectly, singleuser transmission strategies can have a better performance than multi-user transmission strategies when we have a small number of users and correlated antenna gains. We explore the trade-offs between the single-user and multiuser MIMO transmission strategies. Order statistical analysis provides us with both analytical expressions and insights about these trade-offs. We verify that the analytical framework that we develop is accurate by checking the values obtained against numerical results. From this analysis, we confirm that 'mode switching' between single-user and multi-user MIMO transmission schemes is necessary for maximizing throughput for emerging MIMO solutions. Finally, we suggest an adaptive mode switching algorithm between single-user and multi-user MIMO transmission strategies based on this analytical framework. / text

The Community Broadband Investment Act, introduced in the Georgia legislature in 2012 and 2013 is a reflection of a broader national debate over the role of government investment in broadband infrastructure. The bill would limit local governments' ability to invest in broadband infrastructure for their communities because of arguments that government entry into the telecommunications sector crowds-out private competition, does not serve the public adequately, and lacks a comprehensive business model or best practices. A closer look at the history of utility regulation and various economic perspectives on the proper government role in utility provision reveals that government has had an extensive historical role in utility infrastructure investment and regulation, and several economic doctrines support the conclusion that government can be helpful in facilitating effective broadband service to their communities. Case studies of different models of municipal broadband networks in Georgia reveal that government entry can facilitate private sector competition, often provides quality service, and has a set of best practices. The success of municipal broadband reveals an evolution in the approach to telecommunications regulation from a regulated monopoly approach to a public-private cooperation approach which considers public participation with private entities on a dynamic scale. Government-entry into the broadband market was a rational decision for several Georgia communities due to their unique set of circumstances, and while municipal broadband may not be the answer to many communities' problems with meeting public demand, communities should be allowed to maintain flexibility in their decision-making about how to best serve their residents, effectively allowing them to decide which combination of public and private advantages they can leverage to meet the demand of their communities in relation to their unique local characteristics.

published_or_final_version / Business Administration / Master / Master of Business Administration

published_or_final_version / abstract / Electrical and Electronic Engineering / Doctoral / Doctor of Philosophy

published_or_final_version / Electrical and Electronic Engineering / Doctoral / Doctor of Philosophy

Approved for public release; distribution is unlimited / The need for broadband network access is experiencing rapid growth, but what is currently available is not sufficient. Copper-based technologies cannot address the requirements of today's bandwidth-intensive Internet applications. End-users in the "last mile" demand access speeds equivalent to those supported by fiber optics backbone networks, although, the cost and time associated with its installation are prohibitive factors for bringing fiber to every home and business. This results in the well-known "last mile access problem," which prevents the Internet from reaching its full potential, and has paved the way for the development of many innovative technologies. Driven by demands for more bandwidth, wireless broadband technologies have been proposed. This thesis provides an investigation of two candidates to address the lack of adequate bandwidth in the "last mile," Free Space Optics (FSO), and the IEEE 802.11 Wireless Local Area Networking (WLAN) standard. FSO uses optical signals to deliver information at extremely high data rates, more quickly and cost-effectively than fiber systems. The IEEE 802.11 standard uses radio technology to transfer data. They both use license-free frequency bands for transmission through the atmosphere. They both are quickly deployable, easily scalable, and cheaper than wired solutions, characteristics able to support applications requiring high bandwidth and a high degree of mobility. / Lieutenant Commander, Hellenic Navy

by Chi-yin Tse. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1995. / Includes bibliographical references (leaves 118-[121]). / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Video Compression and Transport --- p.2 / Chapter 1.2 --- VBR-CBR Adaptation of Video Traffic --- p.5 / Chapter 1.3 --- Research Contributions --- p.7 / Chapter 1.3.1 --- Spatial Smoothing: Video Aggregation --- p.8 / Chapter 1.3.2 --- Temporal Smoothing: A Control-Theoretic Study。 --- p.8 / Chapter 1.4 --- Organization of Thesis --- p.9 / Chapter 2 --- Preliminaries --- p.13 / Chapter 2.1 --- MPEG Compression Scheme --- p.13 / Chapter 2.2 --- Problems of Transmitting MPEG Video --- p.17 / Chapter 2.3 --- Two-layer Coding and Transport Strategy --- p.19 / Chapter 2.3.1 --- Framework of MPEG-based Layering --- p.19 / Chapter 2.3.2 --- Transmission of GS and ES --- p.20 / Chapter 2.3.3 --- Problems of Two-layer Video Transmission --- p.20 / Chapter 3 --- Video Aggregation --- p.24 / Chapter 3.1 --- Motivation and Basic Concept of Video Aggregation --- p.25 / Chapter 3.1.1 --- Description of Video Aggregation --- p.28 / Chapter 3.2 --- MPEG Video Aggregation System --- p.29 / Chapter 3.2.1 --- Shortcomings of the MPEG Video Bundle Scenario with Two-Layer Coding and Cell-Level Multiplexing --- p.29 / Chapter 3.2.2 --- MPEG Video Aggregation --- p.31 / Chapter 3.2.3 --- MPEG Video Aggregation System Architecture --- p.33 / Chapter 3.3 --- Variations of MPEG Video Aggregation System --- p.35 / Chapter 3.4 --- Experimental Results --- p.38 / Chapter 3.4.1 --- Comparison of Video Aggregation and Cell-level Multi- plexing --- p.40 / Chapter 3.4.2 --- Varying Amount of the Allocated Bandwidth --- p.48 / Chapter 3.4.3 --- Varying Number of Sequences --- p.50 / Chapter 3.5 --- Conclusion --- p.53 / Chapter 3.6 --- Appendix: Alternative Implementation of MPEG Video Aggre- gation --- p.53 / Chapter 3.6.1 --- Profile Approach --- p.54 / Chapter 3.6.2 --- Bit-Plane Approach --- p.54 / Chapter 4 --- A Control-Theoretic Study of Video Traffic Adaptation --- p.58 / Chapter 4.1 --- Review of Previous Adaptation Schemes --- p.60 / Chapter 4.1.1 --- A Generic Model for Adaptation Scheme --- p.60 / Chapter 4.1.2 --- Objectives of Adaptation Controller --- p.61 / Chapter 4.2 --- Motivation for Control-Theoretic Study --- p.64 / Chapter 4.3 --- Linear Feedback Controller Model --- p.64 / Chapter 4.3.1 --- Encoder Model --- p.65 / Chapter 4.3.2 --- Adaptation Controller Model --- p.69 / Chapter 4.4 --- Analysis --- p.72 / Chapter 4.4.1 --- Stability --- p.73 / Chapter 4.4.2 --- Robustness against Coding-mode Switching --- p.83 / Chapter 4.4.3 --- Unit-Step Responses and Unit-Sample Responses --- p.84 / Chapter 4.5 --- Implementation --- p.91 / Chapter 4.6 --- Experimental Results --- p.95 / Chapter 4.6.1 --- Overall Performance of the Adaptation Scheme --- p.97 / Chapter 4.6.2 --- Weak-Control verus Strong-Control --- p.99 / Chapter 4.6.3 --- Varying Amount of Reserved Bandwidth --- p.101 / Chapter 4.7 --- Conclusion --- p.103 / Chapter 4.8 --- Appendix I: Further Research --- p.103 / Chapter 4.9 --- Appendix II: Review of Previous Adaptation Schemes --- p.106 / Chapter 4.9.1 --- Watanabe. et. al.'s Scheme --- p.106 / Chapter 4.9.2 --- MPEG's Scheme --- p.107 / Chapter 4.9.3 --- Lee et.al.'s Modification --- p.109 / Chapter 4.9.4 --- Chen's Adaptation Scheme --- p.110 / Chapter 5 --- Conclusion --- p.116 / Bibliography --- p.118

by Shum Kam Hong. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1992. / Includes bibliographical references (leaves 125-131). / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Background --- p.1 / Chapter 1.2 --- Outline of the thesis --- p.6 / Chapter 1.3 --- Current Art in Packet Switching --- p.9 / Chapter 2 --- Management of Control Information --- p.17 / Chapter 2.1 --- Inter-node Exchange of Link Congestion Status --- p.21 / Chapter 2.2 --- Consistency of Control Information --- p.24 / Chapter 2.3 --- Alternate Format of Control Information --- p.26 / Chapter 3 --- Traffic Flow Control --- p.29 / Chapter 3.1 --- Control of Traffic Influx into the Network --- p.29 / Chapter 3.2 --- Control of Traffic Loading from the Node --- p.30 / Chapter 3.3 --- Flow Control for Connection Oriented Traffic --- p.32 / Chapter 3.4 --- Judgement of Link Status --- p.38 / Chapter 3.5 --- Starvation-free and Deadlock-free --- p.42 / Chapter 4 --- Call Set-up Algorithm Traffic Modelling --- p.47 / Chapter 4.1 --- Basic Algorithm --- p.47 / Chapter 4.2 --- Minimization of Bandwidth Overhead --- p.48 / Chapter 4.3 --- Two-way Transmission --- p.51 / Chapter 4.4 --- Traffic Modelling --- p.52 / Chapter 4.4.1 --- Aggregate Traffic Models --- p.53 / Chapter 4.4.2 --- Traffic Burstiness --- p.57 / Chapter 5 --- Parameters Tuning and Analysis --- p.76 / Chapter 5.1 --- Scheme I : Scout Pumping --- p.76 / Chapter 5.2 --- Scheme II : Speed-up Scout Pumping --- p.85 / Chapter 5.3 --- Blocking Probability --- p.90 / Chapter 5.4 --- Scout Stream Collision --- p.92 / Chapter 6 --- Simulation Modelling & Performance Evaluation --- p.96 / Chapter 6.1 --- The Network Simulator --- p.96 / Chapter 6.1.1 --- Simulation Event Scheduling --- p.97 / Chapter 6.1.2 --- Input Traffic Regulation --- p.100 / Chapter 6.1.3 --- Actual Offered Load --- p.101 / Chapter 6.1.4 --- Static and Dynamic Parameters --- p.103 / Chapter 6.2 --- Simulation Results --- p.107 / Chapter 7 --- Conclusions --- p.123 / Chapter A --- List of Symbols --- p.132