Vasu Chetty

Quality selection for Dynamic Adaptive Streaming over HTTP with Scalable Video Coding

Video streaming on the Internet is increasingly using Dynamic Adaptive Streaming over HTTP (DASH), which allows a client to dynamically adjust its video quality by choosing the appropriate quality level for each segment based on the current download rate. In this paper we examine the impact of Scalable Video Coding (SVC) on the clients quality selection policy. Given a variable download rate, when should the client try to maximize the current segments video quality, and when should it instead play it safe and ensure a minimum level of quality for future segments? We use a combination of analysis, dynamic programming, and simulation to show that a client should use a diagonal quality selection policy, which combines prefetching with backfilling to balance both of these concerns. We also illustrate the conditions that affect the slope of the diagonal policy.

Dynamical structure function identifiability conditions enabling signal structure reconstruction

Networks of controlled dynamical systems exhibit a variety of interconnection patterns that could be interpreted as the structure of the system. One such interpretation of system structure is a systems signal structure, characterized as the open-loop causal dependencies among manifest variables and represented by its dynamical structure function. Although this notion of structure is among the weakest available, previous work has shown that if no a priori structural information is known about the system, not even the Boolean structure of the dynamical structure function is identifiable. Consequently, one method previously suggested for obtaining the necessary a priori structural information is to leverage knowledge about target specificity of the controlled inputs. This work extends these results to demonstrate precisely the a priori structural information that is both necessary and sufficient to reconstruct the network from input-output data. This extension is important because it significantly broadens the applicability of the identifiability conditions, enabling the design of network reconstruction experiments that were previously impossible due to practical constraints on the types of actuation mechanisms available to the engineer or scientist. The work is motivated by the proteomics problem of reconstructing the Per-Arnt-Sim Kinase pathway used in the metabolism of sugars.

Network semantics of dynamical systems

Dynamical systems enjoy a rich variety of mathematical representations, from interconnections of convolution operators or rational functions of a complex variable to systems of (possibly stochastic) differential or differential-algebraic equations. Although many of these representations can describe the same behavior, i.e. represent the same constraints on manifest variables, each one may characterize a different notion of system structure. This paper introduces a method for interpreting the semantics of different representations of a network system by exploring the set of realizations consistent with each. We then focus on signal structure, extending its definition, and demonstrate that its semantics differ from other network representations in important and useful ways. In particular, the information cost for identifying a systems signal structure from data can be considerably less than that needed for identifying a systems subsystem structure.

On the necessity of full-state measurement for state-space network reconstruction

Network reconstruction is an important research topic in many different applications, including biochemical reactions, critical infrastructures, social media, and wireless mesh networks. This paper shows that, for a certain important class of systems, all the states in a system must be measured in order to ensure correct reconstruction of the network. Furthermore, we show that this result is strongly necessary, in that if only one state is not measured, the structure of the recovered network could be arbitrarily different from the structure of the actual network. Finally, we note that our results motivate the need for dynamical structure functions, a partial structure system representation that reveals important structural information about the system but requires much less a priori information (than knowledge of full state measurements) for reconstruction from data.

Vulnerability analysis for distributed and coordinated destabilization attacks

This paper focuses on how the vulnerability of an LTI system to destabilizing attacks can be posed as its robustness to external disturbances. First, we extend existing work on single link attack models to a more generalized attack model that allows for multiple link attacks. This is done by extending the partial structure representation of dynamical structure functions to include external perturbations. Given the new model, we then discuss how to determine the vulnerability of the system for both coordinated and distributed destabilizing attacks on a system. Finally, we develop a separability result for vulnerability in feedback systems that will be useful in determining secure architectures for structured controller design.

Robust signal-structure reconstruction

This paper focuses on the reconstruction of the signal structure of a system in the presence of noise and nonlinearities. Previous results on polynomial time reconstruction in this area were restricted to systems where target specificity was part of the inherent structure, [5]. This work extends these results to all reconstructible systems and proposes a faster reconstruction algorithm along with an improved model selection procedure. Finally, a simulation study then details the performance of this new algorithm on reconstructible systems.

A case study of a systematic attack design method for critical infrastructure cyber-physical systems

As cyber-physical systems continue to become more prevalent in critical infrastructures, security of these systems becomes paramount. Unlike purely cyber systems, cyber-physical systems allow cyber attackers to induce physical consequences. The purpose of this paper is to design a general attack methodology for cyber-physical systems and illustrate it using a case study of the Sevier River System in Central Utah (United States). By understanding such attacks, future work can then focus on designing systems that are robust against them.

Farming as feedback control

This paper surveys a brief history of agriculture, demonstrating how advances in genetics, equipment, and management practices have resulted in the remarkable productivity experienced by todays agriculture industrial complex. We then show that progress in each of these areas is, in part, the result of new solutions to a feedback-control problem, whether it be for selective breeding, exploiting new sensor and actuation technology on a tractor or harvester, or using advanced crop and weather models to make better decisions about irrigation or pesticides. The paper concludes with an invitation for the controls community to explore the varied and important feedback-control problems in this area, emphasizing that sustainably feeding the exponentially growing global population, without harming the environment, will demand creative, careful thinking for years to come.

Passive reconstruction of non-target-specific discrete-time LTI systems

Much of the existing literature on the reconstruction of a systems dynamical structure functions has focused on learning the structure of a system using experiments in which each measured state must be perturbed independently. This work develops a reconstruction procedure that does not require multiple targeted experiments, instead determining the structure of the network when inputs are drawn from a Gaussian distribution and are active simultaneously. Although similar reconstruction procedures exist in the literature, this algorithm removes the restriction of target specificity, which states that each input must independently affect a measured state in the system. This allows for the reconstruction procedure to be applied to a larger number of networks that were previously not reconstructible because of their inherent structure. Furthermore, this is the first reconstruction procedure on the dynamical structure function to operate in the time-domain, rather than the frequency domain, in order to avoid the overhead and inaccuracies that could be introduced through transformations.

How good is bad weather

Accurately identifying key parameters in complex systems demands sufficient excitation, so that the resulting data will be informative enough to reveal hidden parameter values. In many situations, however, users choose inputs that attempt to optimize the system response, not necessarily those that yield more informative data. This leads to the classic tradeoff between exploitation and exploration in learning problems. Farmers face a similar issue. Although they would like to identify key soil parameters affecting the growth of their crops, market pressures force them to manage their product to maximize yield, resulting in less informative data. This suggests that weather, and bad weather in particular, may play a critically important role in creating informative data for crop systems by driving them into low-yield regimes that no farmer would otherwise choose to explore. This paper investigates these issues using a standard computational model for corn and real weather data. Two model-based measures characterizing any years weather pattern are introduced. The first measure characterizes how well a particular years weather pattern produces corn, according to the model. The second measure characterizes how well a particular years weather pattern distinguishes the way different soil types affect corn growth. We then use these measures to show that, from the perspective of corn, bad weather can indeed be very good for distinguishing soil type.