Search

Formal verification is used to establish the compliance of software and hardware systems with important classes of requirements. System compliance with functional requirements is frequently analyzed using techniques such as model checking, and theorem proving. In addition, a technique called quantitative verification supports the analysis of the reliability, performance, and other quality-of-service (QoS) properties of systems that exhibit stochastic behavior. In this paper, we extend the applicability of quantitative verification to the common scenario when the probabilities of transition between some or all states of the Markov models analyzed by the technique are unknown, but observations of these transitions are available. To this end, we introduce a theoretical framework, and a tool chain that establish confidence intervals for the QoS properties of a software system modelled as a Markov chain with uncertain transition probabilities. We use two case studies from different application domains to assess the effectiveness of the new quantitative verification technique. Our experiments show that disregarding the above source of uncertainty may significantly affect the accuracy of the verification results, leading to wrong decisions, and low-quality software systems. [ABSTRACT FROM PUBLISHER]

We developed a multitrack average amplitude profile method to study adjacent track interference (ATI) in perpendicular magnetic recording. We have integrated a ATI test module into our spin stand. The consequence of demagnetization field on the thermal decay is characterized by signal-to-noise ratio (SNR) decay. The correlation between ATI and SNR decay suggests that the ATI is thermally activated decay. [ABSTRACT FROM PUBLISHER]

We developed a new magnetic test to characterize the track overlapping effect in shingled magnetic recording. Media noise power and media noise track profile as a function of track trimming are directly measured on a spinstand tester. Pseudo-jitter and on-track noise depth are introduced as new recording parameters. Their effects on shingled track density are discussed. [ABSTRACT FROM AUTHOR]

Spiral projection imaging (SPI) is a 3D, spiral based magnetic resonance imaging (MRI) acquisition scheme that allows for self-navigated motion estimation of all six degrees-of-freedom. The trajectory, a set of spiral planes, is enhanced to accommodate motion tracking by adding orthogonal planes. Rigid-body motion tracking is accomplished by comparing the overlapping data and deducing the motion that is consistent with the comparisons. The accuracy of the proposed method is quantified for simulated data and for data collected using both a phantom and a volunteer. These tests were repeated to measure the effect of off-resonance blurring, coil sensitivity, gradient warping, undersampling, and nonrigid motion (e.g., neck). The artifacts of off-resonance, coils sensitivity, and gradient warping impose an unnotable effect on the accuracy of motion estimation. The worst mean accuracy is 0.15^\circ and 0.20 mm for the phantom while the worst mean accuracy is 0.48^\circ and 0.34 mm when imaging a brain, indicating that the nonrigid component in human subjects slightly degrades accuracy. When applied to in vivo motion, the proposed technique considerably reduces motion artifact. [ABSTRACT FROM AUTHOR]