In this paper, a fixed-point quantization noise estimator aiming at Digital Signal Processing (DSP) algorithms is presented. The estimator enables significant reduction in the computation time required to perform comp...
详细信息
ISBN:
(纸本)9781424464708
In this paper, a fixed-point quantization noise estimator aiming at Digital Signal Processing (DSP) algorithms is presented. The estimator enables significant reduction in the computation time required to perform complex word-length optimizations while providing a high accuracy. Affine Arithmetic (AA) is used to provide a Signal-to-Quantization Noise-Ratio (SQNR) estimation for differentiable non-linear algorithms with and without feedbacks. The estimation is based on the parameterization of the statistical properties of the noise at the output of fixed-point algorithms. This parameterization allows to relate the fixed-point formats of the signals with the output noise distribution by means of fast matrix operations. Thus, a fast estimation is achieved and the word-length optimization computation time is significantly reduced. The estimator is tested using a subset of non-linear algorithms such as vector operations, power computation IIR filter, adaptive filters, and channel equalizers. The Word-length optimization computation time is boosted by three orders of magnitude while keeping the average estimation error down to 6% for most cases.
Within the current threat landscape of cyberattacks, adversaries use a wide array of techniques to gain a persistent and covert presence on systems. Rootkits are a popular form of attack vector because they are a spec...
详细信息
ISBN:
(纸本)9781450386975
Within the current threat landscape of cyberattacks, adversaries use a wide array of techniques to gain a persistent and covert presence on systems. Rootkits are a popular form of attack vector because they are a specialized form of malware that use stealth and administrative privilege to operate without detection. They can also use sophisticated active and passive measures to actually cripple normal defensive software. Zero-day rootkits are particularly dangerous in advanced persistent threat (APT) scenarios because typical signature-based methods of malware scanning are ineffective at detecting them. In this paper, we consider a more robust technique of dynamic malware detection that allows use of side-channel system properties as process-indicative data, allowing observation outside of a rootkit's ability to influence. In particular, we extend prior work that has used CPU power measurements as a means to detect rootkit execution. In this work we use a Data Acquisition (DAQ) system to collect data from an embedded analog power sensor to record low-frequency digital readings from three channels of a PC power supply. Of particular focus, we use a data-driven nonlinear phase space algorithm (NLPSA) to analyze power readings and perform supervised learning to discern infected versus non-infected states. Our initial case study using Windows 10 and Ubuntu rootkits shows that NLPSA can achieve perfect accuracy in distinguishing rootkit execution and normal system operation, confirming other similar studies where side-channel data are in view.
One of the most important elements for the rehabilitation process regards to the correct evaluation of the biomechanical and the electrophysiological responses. This evaluation must be done during the therapy. In gene...
详细信息
ISBN:
(纸本)9781424441242
One of the most important elements for the rehabilitation process regards to the correct evaluation of the biomechanical and the electrophysiological responses. This evaluation must be done during the therapy. In general, the improvements achieved by the treatment are slightly and difficult to be distinguished. This is a difficult task when the changes in the signals obtained by the bio-amplifiers (EMG, electro-goniometry, etc) are evaluated by a wired system because the patient cannot interact with its environment freely. The present work tackles the design, construction and implementation of a platform to carry out biomechanical analysis for upper and lower limbs. The included variables in the biomechanical system are the angular position, linear acceleration, electromyography signals and force executed by the limbs. The designed scheme considers the wireless monitoring of relevant signals;such variables allow us to analyze the effectiveness achieved by the therapy. Processing and data exhibition are carry out in a personal computer. Two application examples regarding the biomechanical wrist evaluation and the EMG correlation are presented. nonlinear algorithms to analyze the information obtained in the system are used to evaluate the biomechanical responses produced in different patients.
Tire-road friction is the most important characteristic defining the planar dynamics of wheeled vehicles. It has consequences on the drivability, stability and tuning of the active vehicle dynamics control systems. Th...
详细信息
Tire-road friction is the most important characteristic defining the planar dynamics of wheeled vehicles. It has consequences on the drivability, stability and tuning of the active vehicle dynamics control systems. This paper proposes two online friction estimation methods designed for the adaptation of vehicle dynamics control algorithms. The problem is framed as a classification problem where inertial measurements are used to discriminate between high and low friction regimes. The first method merges a recursive least-squares (RLS) algorithm with a heuristic bistable logic to classify the friction condition and promptly react to its changes. The second method runs a classification algorithm on the slip-acceleration characteristic. Both methods simultaneously account for the longitudinal and lateral dynamics and are tested on experimental data.
暂无评论