J.D. Birdwell

Efficient Implementation of the Chan-Vese Models Without Solving PDEs

Efficient implementation methods are proposed for Chan-Vese models. The proposed methods do not require solutions of PDEs and are therefore fast. The advantages of level set methods, such as automatic handling of topological changes, are preserved. These methods utilize region information to guide the evolution of initial curves. Gaussian smoothing is applied to regularize the evolving curves. These algorithms are able to automatically and efficiently segment objects in complicated images. Experimental results show that the proposed methods work efficiently for images without strong noise. However, they still have initialization problems, as do the Chan-Vese models

The effect of time delays on the stability of load balancing algorithms for parallel computations

A deterministic dynamic nonlinear time-delay system is developed to model load balancing in a cluster of computer nodes used for parallel computations. The model is shown to be self consistent in that the queue lengths cannot go negative and the total number of tasks in all the queues and the network are conserved (i.e., load balancing can neither create nor lose tasks). Further, it is shown that using the proposed load balancing algorithms, the system is stable in the sense of Lyapunov. Experimental results are presented and compared with the predicted results from the analytical model. In particular, simulations of the models are compared with an experimental implementation of the load balancing algorithm on a distributed computing network.

On reliable control system designs with and without feedback reconfigurations

This paper contains an overview of a theoretical framework for the design of reliable multivariable control systems, with special emphasis on actuator failures and necessary actuator redundancy levels. Using a linear model of the system, with Markovian failure probabilities and quadratic performance index, an optimal stochastic control problem is posed and solved. The solution requires the iteration of a set of highly coupled Riccati-like matrix difference equations; if these converge one has a reliable design; if they diverge, the design is unreliable, and the system design cannot be stabilized. In addition, it is shown that the existence of a stabilizing constant feedback gain and the reliability of its implementation is equivalent to the convergence properties of a set of coupled Riccati-like matrix difference equations. In summary, these results can be used for off-line studies relating the open loop dynamics, required performance, actuator mean time to failure, and functional or identical actuator redundancy, with and without feedback gain reconfiguration strategies.

Linear time delay model for studying load balancing instabilities in parallel computations

A linear time-delay system is proposed to model load balancing in a cluster of computer nodes used for parallel computations. The linear model is analysed for stability in terms of the delays in the transfer of information between nodes and the gains in the load balancing algorithm. This model is compared with an experimental implementation of the algorithm on a parallel computer network.

Distributed estimation: constraints on the choice of the local models

The following estimation problem is considered: a coordinator must reconstruct the (global) probability density of a nonlinear random process, conditioned on the noise-corrupted observation history. The coordinator can only access the (local) probability density produced by local processing of the observation history using a (local) model different from the process model. It is shown that if the local model satisfies an algebraic constraint, the coordinator can reconstruct the same conditional density of the state process as the one obtained if the observations were processed using the coordinator (process) model. It is assumed that the random process is a nonlinear stochastic differential equation driven by a Brownian motion, and the observation process is corrupted by additive Brownian motion, which is identically modeled by the coordinator and the local processor. >

Expert Systems Techniques in a Computer-Based Control System Analysis and Design Environment

Abstract We explore the fusion of expert control system analysis and design tools into a prototype computer environment. The use of expert systems technology allows the transfer of recent developments in control system design to users who may not be experts in either theoretical developments or computer technology, in a vehicle which provides considerable design flexibility. The technology does not provide as general a framework as, for example, a command language based environment; however, in specific applications, such as system modeling and linear control systems, it is useful and appropriate. Criteria for such applications is that they have a mature theoretical basis, and that the user interaction influence the sequence of the design process.

Resource-Constrained Load Balancing Controller for a Parallel Database

This brief documents experimental results using a deterministic dynamic nonlinear system for load balancing, previously reported by Tang et al. in a cluster of computer nodes used for parallel computations in the presence of time delays and resource constraints. While previous publications by the authors have provided theoretical analysis of this load-balancing strategy using an idealized model, and have documented experiments using a simulated database, experimental results using a complete database for DNA profiles are documented here. Evaluation of the proposed load-balancing strategy using an actual database was critical because of several characteristics of the database that cannot be accurately captured using either a simulation model or database, including the variation in times required for the database to perform search operations, the time-varying and task-dependent computational load the database imposes upon each node of the parallel computer, and the time-varying network traffic imposed by both the communication of database search requests and results, mixed with the traffic generated by the load-balancing strategy. Although the load-balancing strategy can be represented in a relatively straightforward manner using mathematics, its implementation is by necessity an approximation to its mathematical description. The reported experimental results serve to validate the superiority of using the controller based on the anticipated work loads to a controller based on local work loads, which has been predicted with experiments using a simulated database and documented in prior publications. The experiments demonstrate the efficacy of the load-balancing strategy using an anticipated pattern of work loads and provide support for scalability of the approach.

The effect of time delays in the stability of load balancing algorithms for parallel computations

Deterministic dynamic nonlinear time-delay systems are developed to model load balancing in a cluster of computer nodes used for parallel computations. The model is shown to be self consistent in that the queue lengths cannot go negative and the total number of tasks in all the queues are conserved (i.e., load balancing can neither create nor lose tasks). Further, it is shown that using the proposed load balancing algorithms, the system is stable. Experimental results are presented and compared with the predicted results from the analytical model. In particular, simulations of the models are compared with an experimental implementation of the load balancing algorithm on a parallel computer network.

Preferential Image Segmentation Using Trees of Shapes

A novel preferential image segmentation method is proposed that performs image segmentation and object recognition using mathematical morphologies. The method preferentially segments objects that have intensities and boundaries similar to those of objects in a database of prior images. A tree of shapes is utilized to represent the content distributions in images, and curve matching is applied to compare the boundaries. The algorithm is invariant to contrast change and similarity transformations of translation, rotation and scale. A performance evaluation of the proposed method using a large image dataset is provided. Experimental results show that the proposed approach is promising for applications such as object segmentation and video tracking with cluttered backgrounds.

Load balancing in the presence of random node failure and recovery

In many distributed computing systems that are prone to either induced or spontaneous node failures, the number of available computing resources is dynamically changing in a random fashion. A load-balancing (LB) policy for such systems should therefore be robust, in terms of workload re-allocation and effectiveness in task completion, with respect to the random absence and re-emergence of nodes as well as random delays in the transfer of workloads among nodes. In this paper two LB policies for such computing environments are presented: The first policy takes an initial LB action to preemptively counteract the consequences of random failure and recovery of nodes. The second policy compensates for the occurrence of node failure dynamically by transferring loads only at the actual failure instants. A probabilistic model, based on the concept of regenerative processes, is presented to assess the overall performance of the system under these policies. Optimal performance of both policies is evaluated using analytical, experimental and simulation-based results. The interplay between node-failure/recovery rates and the mean load-transfer delay are highlighted