Lance Fiondella

Mobi-Sync: Efficient Time Synchronization for Mobile Underwater Sensor Networks

Time synchronization is an important requirement for many services provided by distributed networks. A lot of time synchronization protocols have been proposed for terrestrial Wireless Sensor Networks (WSNs). However, none of them can be directly applied to Underwater Sensor Networks (UWSNs). A synchronization algorithm for UWSNs must consider additional factors such as long propagation delays from the use of acoustic communication and sensor node mobility. These unique challenges make the accuracy of synchronization procedures for UWSNs even more critical. Time synchronization solutions specifically designed for UWSNs are needed to satisfy these new requirements. This paper proposes Mobi-Sync, a novel time synchronization scheme for mobile underwater sensor networks. Mobi-Sync distinguishes itself from previous approaches for terrestrial WSN by considering spatial correlation among the mobility patterns of neighboring UWSNs nodes. This enables Mobi-Sync to accurately estimate the long dynamic propagation delays. Simulation results show that Mobi-Sync outperforms existing schemes in both accuracy and energy efficiency.

Efficient Software Reliability Analysis With Correlated Component Failures

Correlated component failures (COCOF) may impact the reliability of a software application, and hence these types of failures must be explicitly incorporated into reliability analysis. The influence of COCOF on application reliability must be analyzed within the context of the application architecture. Contemporary reliability analysis approaches that incorporate COCOF, however, cannot scale to even moderate-sized software applications. This paper presents an efficient, scalable approach to analyze the reliability of a component-based software system, considering COCOF within the context of its architecture. The effectiveness of the approach is illustrated through two experimental studies. The results indicate that the approach is simple and efficient, and hence can be applied to large systems to identify correlations that impede system reliability.

A universal generating function-based multi-state system performance model subject to correlated failures

Multi-state system (MSS) reliability modeling is a paradigm that allows both systems and components to exhibit more than two performance levels. While several researchers have introduced correlation or dependence into MSS models to assess its negative influence on performance and associated measures, these methods exhibit complexity that is exponential or worse in the worst case. To overcome this limitation, this paper proposes an extension to the discrete universal generating function approach for MSS to allow correlation between the elements comprising a multi-state component. We subsequently generalize to the continuous case and allow failures to follow any life distribution. The approach possesses an analytical form and therefore enables efficient performance and reliability assessment as well as sensitivity analysis on the impact of correlation. This sensitivity analysis can be applied to a wide range of measures including performance, reliability, the density function, hazard rate, mean time to failure, availability, and mean residual life. The approach is illustrated through a series of examples, demonstrating the efficiency of the approach to assess performance and reliability as well as to conduct sensitivity analysis. The results indicate that the approach can identify the impact of correlation on performance, reliability, and the many measures of interest.

Discrete and continuous reliability models for systems with identically distributed correlated components

Many engineers and researchers base their reliability models on the assumption that components of a system fail in a statistically independent manner. This assumption is often violated in practice because environmental and system specific factors contribute to correlated failures, which can lower the reliability of a fault tolerant system. A simple method to quantify the impact of correlation on system reliability is needed to encourage models explicitly incorporating correlated failures. Previous approaches to model correlation are limited to systems consisting of two or three components or assume that the majority of the subsets of component failures are statistically independent. This paper proposes a method to model the reliability of systems with correlated identical components, where components possess the same reliability and also exhibit a common failure correlation parameter. Both discrete and continuous models are proposed. The method is demonstrated through a series of examples, including derivations of analytical expressions for several common structures such as k-out-of-n: good and parallel systems. The continuous models consider the role of correlation on reliability and metrics, including mean time to failure, availability, and mean residual life. These examples illustrate that the method captures the impact of component correlation on system reliability and related metrics.

RELIABILITY AND SENSITIVITY ANALYSIS OF COHERENT SYSTEMS WITH NEGATIVELY CORRELATED COMPONENT FAILURES

Modeling correlated component failures poses a unique challenge for reliability researchers because it requires ingenuity to devise an approach free from the assumption that components fail in a statistically independent manner. Several studies have addressed this problem with models that introduce additional parameters to describe the correlated failure of components. However, these earlier techniques often require the correlations to be positive and almost always introduce an exponential number of correlation parameters. These restrictions limit the scalability of existing approaches for conducting sensitivity analysis on the correlation parameters, which could identify correlation reductions that would improve system reliability. This paper presents a technique for reliability and sensitivity analysis that requires only a quadratic number of correlation parameters, encompassing systems with both negative and positive component correlations. Unlike previous research, the proposed approach places no unnecessary restrictions on a systems correlation parameters. A series of examples illustrates the flexibility of the approach. The results quantitatively confirm that negative component correlation assists fault-tolerant systems to attain levels of reliability even higher than systems of statistically independent redundant components. Thus, the techniques introduced here offer a methodology to concisely measure the utility of negative component correlations on system reliability improvement.

Suave: Swarm underwater autonomous vehicle localization

Swarms of autonomous underwater vehicles (AUVs) forming mobile underwater networks often operate in moving currents, which introduce severe turbulence that interferes with coordinated and stealthy navigation of fleet. Therefore, individual AUV must adjust their heading whenever needed to ensure it can reach a pre-determined destination. To achieve accurate navigation, AUVs must maintain precise knowledge of their locations. This paper develops the “Suave” (Swarm underwater autonomous vehicle localization) algorithm to localize swarms of AUVs operating in rough waters. The purpose of Suave is to ensure that all AUVs arrive at their destinations by preserving localization throughout the entire mission. Suave lowers the probability that an AUV swarm is detected by reducing the number of occasions that vehicles must surface to obtain accurate location information from external sources such as satellites. The Suave algorithm also achieves better energy conservation through improved control of localization reference messages. Simulations show Suave significantly improves localization accuracy, lowers energy consumption, and the probability of swarm detection.

System Performance and Reliability Modeling of a Stochastic-Flow Production Network: A Confidence-Based Approach

Production network performance and reliability are essential to satisfy customer orders in a timely manner. This paper proposes a statistical method for a production system to satisfy customer demand with a desired level of confidence, referred to as yield confidence, while simultaneously considering system reliability, defined as the probability that the amount of input can be processed based on the capacities of the individual workstations. The approach models a production system as a stochastic-flow production network, characterized by a discrete time Markov chain (DTMC), where one or more rework actions are possible. This model quantifies the probability that raw input is transformed into a finished product, which is subsequently used to calculate the amount of raw input needed to satisfy demand with a user-specified level of yield confidence. A pair of case studies, taken from the tile and circuit board industries, illustrates the assessment techniques as well as methods to identify workstation level enhancements that can improve network performance and reliability most significantly. Our results indicate that improving the reliability of workstations can enhance yield confidence because a lower volume of raw input can produce the desired volume of output, thereby minimizing the load placed on the production network.

Software reliability with architectural uncertainties

Architecture-based software reliability analysis can provide early identification of critical components which can then be targeted for cost-effective reliability improvement of the application. However, an important challenge in conducting this analysis early in the life cycle is that it is nearly impossible to estimate the architectural and component parameters with certainty. The issue of estimating software application reliability in the presence of uncertain component reliabilities has been addressed in the previous research. In this paper we consider the estimation of software reliability in the presence of architectural uncertainties. We present a methodology to estimate the confidence levels in the architectural parameters using limited testing or simulation data based on the theory of confidence intervals of the multinomial distribution. The sensitivity of the system reliability to uncertain architectural parameters can then be quantified by varying these parameters within their confidence intervals. The illustration of the methodology using a case study indicates that the impact of the uncertainty in a given architectural parameter on the overall application reliability is determined by the inherent branching behavior of the application and the component reliabilities.

Optimal redundancy allocation to maximize multi-state computer network reliability subject to correlated failures

Abstract Modern society depends on the stability of computer networks. One way to achieve this goal is to determine the optimal redundancy allocation such that system reliability is maximized. Redundancy requires that each edge in computer networks possess several binary-state physical lines allocated in parallel. A computer network implementing redundancy allocation is called a multi-state computer network (MSCN), since each edge can exhibit multiple states with a probability distribution according to the number of binary-state physical lines that are operational. However, past research often fails to consider the possibility of correlated failures. This study applies a correlated binomial distribution to characterize the state distribution of each edge within a network and a redundancy optimization approach integrating simulated annealing (SA), minimal paths, and correlated binomial distribution is proposed. The approach is applied to four practical computer networks to demonstrate the computational efficiency of the proposed SA relative to several popular soft computing algorithms.

Quantifying the Impact of Correlated Failures on Stochastic Flow Network Reliability

This paper develops two techniques to analyse the performance of a stochastic-flow network (SFN) model, considering correlated failures. The first approach utilizes a correlated binomial distribution to characterize the failure behavior of the physical lines and routers internal to the individual edges and nodes in the network. The second employs a simulation technique, which can characterize correlated failures between every pair of physical lines and routers in the different edges and nodes comprising the network. Both approaches quantify the probability that a given amount of data can be sent from a source to a sink through this network. This probability that the network satisfies a specified level of demand is referred to as the SFN reliability. The techniques are demonstrated in the context of two case studies, including the Taiwan Academic Network, the backbone of the national computer network connecting all educational institutions in Taiwan. Experimental results demonstrate that correlation can produce a significantly negative impact on reliability, especially when there is a high level of network demand. The proposed approaches, thus, capture the influence of correlation on SFN reliability, offering methods to quantify the utility of reducing correlation.