This paper shows the results related to a binary pattern matching (BPM) task executed on an Analog In-memory Computing (AIMC) unit based on an embedded Phase-Change Memory (ePCM), both designed in a 90-nm CMOS STMicro...
详细信息
This paper shows the results related to a binary pattern matching (BPM) task executed on an Analog In-memory Computing (AIMC) unit based on an embedded Phase-Change Memory (ePCM), both designed in a 90-nm CMOS STMicroelectronics technology. The Hit Rate in pattern recognition is characterized and modeled in different scenarios, in order to evaluate the influence of cells Conductance Time Drift (CTD) in a real application. In particular, two PCM multilevel programming algorithms and different cells conductances are considered, and their effects on CTD are experimentally observed. Results suggest that the adoption of a SET staircase (SSC) sequence implies a lower CTD on PCM cells with respect to a RESET Staircase (RSC) sequence, as well as an increased Hit Rate, even with lower levels of employed conductance.
Malware databases have been unintentionally collecting garbage (incomplete malware) together with malware through the Internet. This paper focuses on finding garbage (incomplete malware) from large malware datasets us...
详细信息
ISBN:
(纸本)9781728151731
Malware databases have been unintentionally collecting garbage (incomplete malware) together with malware through the Internet. This paper focuses on finding garbage (incomplete malware) from large malware datasets using binary pattern matching and speed up the matching by using nested clustering as a preprocessing. To verify the effectiveness of our method, we conduct experiments on various malware datasets. The results show that our method works efficiently while maintaining high accuracy.
The speech signal "conveys information on different time scales from short (20-40 ms) time scale or segmental, associated to phonological and phonetic information to long (150-250 ms) time scale or supra segmenta...
详细信息
The speech signal "conveys information on different time scales from short (20-40 ms) time scale or segmental, associated to phonological and phonetic information to long (150-250 ms) time scale or supra segmental, associated to syllabic and prosodic information. Linguistic and neurocognitive studies recognize the phonological classes at segmental level as the essential and invariant representations used in speech temporal organization. In the context of speech processing, a deep neural network (DNN) is an effective computational method to infer the probability of individual phonological classes from a short segment of speech signal. A vector of all phonological class probabilities is referred to as phonological posterior. There are only very few classes comprising a short term speech signal;hence, the phonological posterior is a sparse vector. Although the phonological posteriors are "estimated at segmental level, we claim that they convey supra segmental information. Specifically, we demonstrate that phonological posteriors are indicative of syllabic and prosodic events. Building on findings from converging linguistic evidence on the gestural model of Articulatory Phonology as well as the neural basis of speech perception, we hypothesize that phonological posteriors convey properties of linguistic classes at multiple time scales, and this information is embedded in their support (index) of active coefficients. To verify this hypothesis, we obtain a binary representation of phonological posteriors at the segmental level which is referred to as first-order sparsity structure;the high order structures are obtained by the concatenation of first-order binary vectors. It is then confirmed that the classification of supra-segmental linguistic events, the problem known as linguistic parsing, can be achieved with high accuracy using a simple binary pattern matching of first-order or high-order structures. (C) 2016 Elsevier B.V. All rights reserved.
Pressure ulcer is a prevalent complication for bed-bound patients who are not able to shift their body weights over time. Continuous monitoring of patient's postures in the bed can be helpful for caregivers in ord...
详细信息
ISBN:
(纸本)9781479927616
Pressure ulcer is a prevalent complication for bed-bound patients who are not able to shift their body weights over time. Continuous monitoring of patient's postures in the bed can be helpful for caregivers in order to keep track of patient's movements and quality of their repositioning during a day. This information allows hospitals to plan an effective repositioning schedule for each patient. In this paper, a high speed and robust posture classification algorithm is proposed that can be employed in any pervasive patient's monitoring system. First, a whole-body pressure image is recorded using a commercial pressure mat system. Image enhancement is then applied to the raw pressure images and a binary signature for each different posture is constructed. Finally, using a binary pattern matching technique, a given posture can be classified to one of the known posture classes. Our extensive experiments show that the proposed algorithm is able to predict in-bed postures with more than 97% average accuracy.
暂无评论