Departmental Papers (ESE)

Pricing Strategies for User-Provided Connectivity Services

M. H. Afrasiabi et al. — Tue, 17 Jan 2012 14:39:11 PST

User-provided connectivity (UPC) services offer a possible alternative, orcomplement, to existing infrastructure-based connectivity. A userallows other users to occasionally connect through its "home base" inexchange for reciprocation, or possibly compensation. This service modelexhibits strong positive and negative externalities. A large user basemakes the service more attractive, as it offers more connectivity options toroaming users, but it also implies a greater volume of (roaming) trafficpassing through a user's home base, which can increase congestion. Theseinteractions make it difficult to predict the eventual success of such aservice offering, and in particular how to effectively price it. This paperinvestigates a two-price policy where the first price is an introductory price that expires onceservice adoption reaches a certain level. The paper uses a simplifiedanalytical model to investigate pricing strategies under this policy, and their sensitivity to changesin system parameters. The insight and practicalguidelines this yields are validated numerically under morerealistic conditions.

Assessing IPv6 Through Web Access - A Measurement Study and Its Findings

Mehdi Nikkhah et al. — Tue, 08 Nov 2011 12:08:29 PST

Transitioning an infrastructure the size of the Internet is no small feat. We are in the midst of such a transition, \ie from IPv4 to IPv6. IPv6 was standardized 15~years ago, but until recently there were few incentives to adopt it. The allocation of the last large block of IPv4 addresses changed that, and migrating to an IPv6 Internet has become more urgent. This migration is, however, still rife with uncertainties and challenges. The goal of this paper is to provide insight into this transition, and possibly make it smoother. The focus is on the ``network,'' and the paper reports on extensive measurements that compare and contrast IPv6 and IPv4. Two important hypotheses, denoted as H1 and H2, were identified and validated. H1 argues that the IPv6 and IPv4 data planes now perform by and large comparably. In contrast, H2 points to routing differences as the primary culprit behind occurrences of poorer IPv6 performance. In other words, promoting IPv6 and IPv4 peering parity is probably the single most effective step towards equal IPv6 and IPv4 performance

Autonomous Legged Hill and Stairwell Ascent

Aaron M. Johnson et al. — Tue, 08 Nov 2011 12:08:13 PST

This paper documents near-autonomous negotiation of synthetic and natural climbing terrain by a rugged legged robot, achieved through sequential composition of appropriate perceptually triggered locomotion primitives. The first, simple composition achieves autonomous uphill climbs in unstructured outdoor terrain while avoiding surrounding obstacles such as trees and bushes. The second, slightly more complex composition achieves autonomous stairwell climbing in a variety of different buildings. In both cases, the intrinsic motor competence of the legged platform requires only small amounts of sensory information to yield near-complete autonomy. Both of these behaviors were developed using X-RHex, a new revision of RHex that is a laboratory on legs, allowing a style of rapid development of sensorimotor tasks with a convenience near to that of conducting experiments on a lab bench. Applications of this work include urban search and rescue as well as reconnaissance operations in which robust yet simple-to-implement autonomy allows a robot access to difficult environments with little burden to a human operator.

Exploring User-Provided Connectivity - A Simple Model

M. H. Afrasiabi et al. — Wed, 17 Aug 2011 16:33:54 PDT

The advent of cheap and ubiquitous wireless access has intro- duced a number of new connectivity paradigms. This paper investigates one of them, user-provided connectivity or UPC. In contrast to tradi- tional infrastructure-based connectivity, e.g., connectivity through the up-front build-out of expensive base-stations, UPC realizes connectivity organically as users join and expand its coverage. The low(er) deployment cost this aords is one of its main attractions. Conversely, the disadvan- tages of connectivity sharing and a high barrier-to-entry from low initial penetration create strong disincentives to its adoption. The paper's con- tributions are in formulating and solving a simple model that captures key aspects of UPC adoption, and in articulating guidelines to make it successful. For analytical tractability, the model is arguably simplistic, but the robustness of its ndings is demonstrated numerically across a wide range of more general (and more realistic) congurations.

Large Negative Magnetoresistance and Strong Localization in Highly Disordered Electrospun Pregraphitic Carbon Nanofiber

Yu Wang et al. — Wed, 17 Aug 2011 16:33:40 PDT

A highly disordered pregraphitic carbon nanofiber with the product of its quasi-Fermi wave vector and mean free path close to 1 was fabricated using electrospinning technique. Strong localization made the conductivity vary with temperature as ln σ ∝ T^−1/2 from 300 to 5 K, suggesting variable range hopping as the conductivity mechanism, and resulted in a large negative magnetoresistance from 300 K down to 1.9 K that can still be quantitatively described using weak localization and electron interaction models.

High-Reliability Computing For The Smarter Planet

Heather Quinn et al. — Thu, 23 Jun 2011 11:07:22 PDT

As computer automation continues to increase in our society, the need for greater radiation reliability is necessary. Already critical infrastructure is failing too frequently. In this paper, we will introduce the Cross-Layer Reliability concept for designing more reliable computer systems.

Multiplicative Updates for Nonnegative Quadratic Programming

Fei Sha et al. — Wed, 15 Jun 2011 13:42:12 PDT

Many problems in neural computation and statistical learning involve optimizations with nonnegativity constraints. In this article, we study convex problems in quadratic programming where the optimization is confined to an axis-aligned region in the nonnegative orthant. For these problems, we derive multiplicative updates that improve the value of the objective function at each iteration and converge monotonically to the global minimum. The updates have a simple closed form and do not involve any heuristics or free parameters that must be tuned to ensure convergence. Despite their simplicity, they differ strikingly in form from other multiplicative updates used in machine learning.We provide complete proofs of convergence for these updates and describe their application to problems in signal processing and pattern recognition.

Experimental Validation of Strong Directional Selectivity in Nonsymmetric Metallic Gratings with a Subwavelength Slit

Semih Cakmakyapan et al. — Wed, 30 Mar 2011 15:08:06 PDT

Strong directional selectivity is theoretically predicted and experimentally validated at the microwave frequencies in the beaming regime for a single subwavelength slit in nonsymmetric metallic gratings with double-side corrugations. The operation regime can be realized at a fixed angle of incidence when the surface-plasmon assisted transmission is significant within a narrow range of observation angles, if illuminating one of the grating interfaces, and tends to vanish for all observation angles, if illuminating the opposite interface. The studied effect is connected with asymmetry (nonreciprocity) in the beaming that occurs if the surface plasmon properties are substantially different for the two interfaces being well isolated from each other.

Electromagnetic tunneling through a single-negative slab paired with a double-positive bilayer

Giuseppe Castaldi et al. — Wed, 23 Mar 2011 15:57:27 PDT

We show that resonant tunneling of electromagnetic fields can occur through a three-layer structure composed of a single-negative (i.e., either negative permittivity or negative permeability) slab paired with a bilayer made of double-positive (i.e., positive permittivity and permeability) media. In particular, one of the two double-positive media can be chosen arbitrarily (even vacuum), while the other may exhibit extreme (either near-zero or very high) permittivity and permeability values. Our results on this counterintuitive tunneling phenomenon also demonstrate the possibility of synthesizing double-positive slabs that effectively exhibit single-negative-like wave-impedance properties within a moderately wide frequency range.

On the Feasibility and Efficacy of Protection Routing in IP Networks

Kin Wah (Eric) Kwong et al. — Wed, 09 Mar 2011 14:49:37 PST

With network components increasingly reliable, routing is playing an ever greater role in determining network reliability. This has spurred much activity in improving routing stability and reaction to failures, and rekindled interest in centralized routing solutions, at least within a single routing domain. Centralizing decisions eliminates uncertainty and many inconsistencies, and offers added flexibility in computing routes that meet different criteria. However, it also introduces new challenges; especially in reacting to failures where centralization can increase latency. This paper leverages the flexibility afforded by centralized routing to address these challenges. Specifically, we explore when and how standby backup forwarding options can be activated, while waiting for an update from the centralized server after the failure of an individual component (link or node). We provide analytical insight into the feasibility of such backups as a function of network structure, and quantify their computational complexity. We also develop an efficient heuristic reconciling protectability and performance, and demonstrate its effectiveness in a broad range of scenarios. The results should facilitate deployments of centralized routing solutions.

Shaping Light Beams in the Nanometer Scale: A Yagi-Uda Nanoantenna in the Optical Domain

Jingjing Li et al. — Thu, 27 Jan 2011 14:40:05 PST

AYagi-Uda-like optical nanoantenna concept using resonant core-shell plasmonic particles as its “reflectors” and “directors” is studied numerically. Such particles when placed near an optical dipole source in a certain arrangement may exhibit large induced dipole moments, resulting in shaping the far-field radiation pattern, analogous to the far field of classical Yagi-Uda antennas in the microwave regime. The variation of the ratio of radii in concentric core-shell nanostructure is used to tailor the phase of the polarizabilities of the particles and, consequently, the antenna’s far-field pattern. The idea of a nanospectrum analyzer is also briefly proposed for molecular spectroscopy.

Nanoscaled Piezoelectric Aluminum Nitride Contour- Mode Resonant Sensors

Gianluca Piazza et al. — Thu, 27 Jan 2011 14:39:54 PST

This paper reports on a new class of nanoscaled piezoelectric aluminum nitride contour-mode resonant sensors (CMR-S) that have been developed for the gravimetric detection of volatile organic chemicals (VOC). The use of the CMR-S and its scaling for the making of large arrays of detectors is justified in terms of the superior sensitivity and limit of detection (LOD ~ zg/μm²) that this technology attains with respect to any other available acoustic device. The choice of a novel functionalization layer based on ss-DNA is introduced as an effective way to selectively detect multiple VOCs without altering the electromechanical characteristics of the resonator. Experimental results confirming the advantages of scaling the device dimensions and its frequency of operation in terms of improved LOD and measurement speed are presented. Furthermore, preliminary data showing selective detection of dymethyl-methylphosphonate (DMMP) and dinitroluene (DNT) (with LOD in the order of ppt) are reported.

Use of a Single Multiplexed CMOS Oscillator as Direct Frequency Read-Out for an Array of Eight AlN Contour-Mode NEMS Resonant Sensors

Matteo Rinaldi et al. — Thu, 27 Jan 2011 14:39:44 PST

This paper reports on the first demonstration of a single multiplexed CMOS oscillator circuit employed as direct frequency readout for an array of 8 nanoscaled aluminum nitride Contour-Mode Resonant Sensors (CMR-S). In this first prototype 8 thin-film (250 nm) AlN CMR-S operating at 186 MHz were fabricated on the same chip and simultaneously wire-bonded to a Pierce-like oscillator circuit (fabricated in the ON Semiconductor 0.5 µm CMOS process) by means of 8 CMOS transmission gates addressed via a 3 bit on-chip decoder. The 8 CMR-S were simultaneously exposed to different concentrations of methanol (0.1–1% of the saturated vapor pressure) and their response was monitored in a time-multiplexed mode. Frequency shifts of 300 Hz corresponding to changes of mass per unit area of 7 ag/µm² were experimentally detected. Values of phase noise derived Allan deviation as low as 0.9 Hz were measured. Such Allan deviation translates in an estimated limit of detection of 21 zg/µm².

High Frequency Piezoelectric Resonant Nanochannel for Bio-Sensing Applications in Liquid Environment

Chiara Zuniga et al. — Thu, 27 Jan 2011 14:39:33 PST

This paper reports on the first demonstration of a 457 MHz AlN Piezolectric Resonant Nanochannel (PRN) for biosensing applications in liquid environment. A novel process consisting of 7 lithographic steps was developed to fabricate the PRN. The new resonant device shows an unchanged value of the electromechanical coupling, k_t² (about 0.8 %), whether the channel is filled with air or water and a quality factor, Q, in liquid of approximately 170. The value of k_t² and Q are respectively about 2.7 and 2 times the ones recorded for conventional laterally vibrating AlN Contour Mode Resonant Sensors (CMR-Ss) submerged in water. Overall, these results translate in a ~ 5 fold enhancement in the figure of merit (k_t² - Q product) of the resonant device when operated in liquid and simultaneously permit the efficient delivery of ultra-low concentrations of fluid samples directly on the surface of the sensor.

Validating Agent Based Social Systems Models

Gnana K. Bharathy et al. — Tue, 18 Jan 2011 09:24:55 PST

Validating social systems is not a trivial task. The paper outlines some of our past efforts in validating models of social systems with cognitively detailed agents. It also presents some of the challenges faced by us. A social system built primarily of cognitively detailed agents can provide multiple levels of correspondence, both at observable and abstract aggregated levels. Such a system can also pose several challenges including large feature spaces, issues in information elicitation with database, experts and news feeds, counterfactuals, fragmented theoretical base, and limited funding for validation. Our own approach to validity assessment is to consider the entire life cycle and assess the validity under four broad dimensions of methodological validity, internal validity, external validity and qualitative, causal and narrative validity. In the past, we have employed a triangulation of multiple validation techniques, including face validation as well as formal validation tests including correspondence testing.

Optimal Robust Multihop Routing for Wireless Networks of Mobile Micro Autonomous Systems

Jonathan Fink et al. — Tue, 18 Jan 2011 09:24:46 PST

This paper develops algorithms to ensure that agents of a mobile micro autonomous system (MMAS) maintain integrity of communication flows as they move to accomplish their task. Due to inherent uncertainties in estimation of wireless channels, we advocate a stochastic approach whereby achievable communication rates of point-to-point links are regarded as random variables with known means and variances. To achieve reliable end-to-end communication flows, terminals route their traffic through various alternative paths to reduce the effect of uncertainty in individual link rates. The proposed algorithms are optimal and robust in that routes are obtained as solutions of optimization problems subject to constraints on minimum required rates and maximum acceptable variances. Algorithms are tested in an event-based simulator that uses an accurate data-driven model of radio communications to model both the structure of code running independently on multiple robots as well as the transmission of messages via a real radio. Simulation results corroborate that rates of end-to-end flows are maintained at target levels despite variations in the rates of individual links.

Synthesis of Palladium with Different Nanoscale Structures by Sputtering Deposition onto Fiber Templates

Victor M. Pantojas et al. — Tue, 18 Jan 2011 09:24:38 PST

A flexible and versatile method combining sputtering and electrospinning techniques was used to shape different palladium morphological structures with nanoscale features. The samples were prepared by dc-magnetron sputtering onto thermally degradable polymer templates. The sputtering parameters were chosen to deposit the metal under low adatom-mobility conditions. After deposition, the template was removed by heat treatment, thereby forming different palladium morphologies with shapes resembling ribbons and half tubes, amongst others. X-ray diffraction studies demonstrated that they are composed of crystalline palladium or palladium oxide, depending on the heat treatment. The cylindrical walls are composed of 30 nm or smaller crystallites, as measured from transmission electron microscopy images. A mathematical simulation demonstrate that the morphological structures obtained are a consequence of the sputtering line-of-sight deposition process. This fabrication process can be varied to modify three types of structures at the nanoscale level: the external shape, the columnar shape of the walls, and the nano-crystallinity. The external shape can be modified by controlling the deposition time and the fiber template diameter. The columnar shape of the walls and the nano-crystallinity can be modified by changes in the sputtering process parameters. The nanoscale morphologies created have potential uses in sensing and photonic applications.

Spectrum Auction Framework for Access Allocation in Cognitive Radio Networks

Guarav S. Kasbekar et al. — Tue, 18 Jan 2011 09:24:30 PST

In cognitive radio networks, there are two categories of networks on different channels: primary networks, which have high-priority access, and secondary networks, which have low-priority access. We develop an auction-based framework that allows networks to bid for primary and secondary access based on their utilities and traffic demands. The bids are used to solve the access allocation problem, which is that of selecting the primary and secondary networks on each channel either to maximize the auctioneer’s revenue or to maximize the social welfare of the bidding networks, while enforcing incentive compatibility. We first consider the case when the bids of a network depend on which other networks it will share channels with. When there is only one secondary network on each channel, we design an optimal polynomial- time algorithm for the access allocation problem based on reduction to a maximum matching problem in weighted graphs. When there can be two or more secondary networks on a channel, we show that the optimal access allocation problem is NP-complete. Next, we consider the case when the bids of a network are independent of which other networks it will share channels with. We design a polynomial-time dynamic programming algorithm to optimally solve the access allocation problem when the number of possible cardinalities of the set of secondary networks on a channel is upper-bounded. Finally, we design a polynomial-time algorithm that approximates the access allocation problem within a factor of 2 when the above upper bound does not exist.

Cloaked Near-Field Scanning Optical Microscope Tip for Noninvasive Near-Field Imaging

Andrea Alù et al. — Tue, 18 Jan 2011 09:24:21 PST

Near-field imaging is a well-established technique in biomedical measurements, since closer to the detail of interest it is possible to resolve subwavelength details otherwise unresolved by regular lenses. A near-field scanning optical microscope (NSOM) tip may indeed overcome the resolution limits of far-field optics, but its proximity inherently perturbs the measurement. Here, we apply the recent concept of a ‘‘cloaked sensor’’ to an NSOM device in collection mode, showing theoretically how a proper plasmonic cover applied to an NSOM tip may drastically improve its overall measurement capabilities.

Maximum Damage Malware Attack in Mobile Wireless Networks

MHR Khouzani et al. — Tue, 11 Jan 2011 09:12:08 PST

Malware attacks constitute a serious security risk that threatens to slow down the large scale proliferation of wireless applications. As a first step towards thwarting this security threat, we seek to quantify the maximum damage inflicted on the system owing to such outbreaks and identify the most vicious attacks. We represent the propagation of malware in a battery-constrained mobile wireless network by an epidemic model in which the worm can dynamically control the rate at which it kills the infected node and also the transmission range and/or the media scanning rate. At each moment of time, the worm at each node faces the following trade-offs: (i) using larger transmission range and media scanning rate to accelerate its spread at the cost of exhausting the battery and thereby reducing the overall infection propagation rate in the long run or (ii) killing the node to inflict a large cost on the network, however at the expense of loosing the chance of infecting more susceptible nodes at later times. We mathematically formulate the decision problems and utilize Pontryagin Maximum Principle from optimal control theory to quantify the damage that the malware can inflict on the network by deploying optimum decision rules. Next, we establish structural properties of the optimal strategy of the attacker over time. Specifically, we prove that it is optimal for the attacker to defer killing of the infective nodes in the propagation phase for a certain time and then start the slaughter with maximum effort. We also show that in the optimal attack policy, the battery resources are used according to a decreasing function of time, i.e., mostly during the initial phase of the outbreak. Finally, our numerical investigations reveal a framework for identifying intelligent defense strategies that can limit the damage by appropriately selecting network parameters.