Search

The main causes of damage to industrial machinery are aging, corrosion, and the wear of parts, which affect the accuracy of machinery and product precision. Identifying problems early and predicting the life cycle of a machine for early maintenance can avoid costly plant failures. Compared with other sensing and monitoring instruments, sound sensors are inexpensive, portable, and have less computational data. This paper proposed a machine tool life cycle model with noise reduction. The life cycle model uses Mel-Frequency Cepstral Coefficients (MFCC) to extract audio features. A Deep Neural Network (DNN) is used to understand the relationship between audio features and life cycle, and then determine the audio signal corresponding to the aging degree. The noise reduction model simulates the actual environment by adding noise and extracts features by Power Normalized Cepstral Coefficients (PNCC), and designs Mask as the DNN's learning target to eliminate the effect of noise. The effect of the denoising model is improved by 6.8% under Short-Time Objective Intelligibility (STOI). There is a 3.9% improvement under Perceptual Evaluation of Speech Quality (PESQ). The life cycle model accuracy before denoising is 76%. After adding the noise reduction system, the accuracy of the life cycle model is increased to 80%. [ABSTRACT FROM AUTHOR]

In the industrial Internet of Things, the network time protocol (NTP) can be used for time synchronization, allowing machines to run in sync so that machines can take critical actions within 1 ms. However, the commonly used NTP mechanism does not take into account that the network packet travel time over a link is time-varying, which causes the NTP to make incorrect synchronization decisions. Therefore, this paper proposed a low-cost modification to NTP with clock skew compensation and adaptive clock adjustment, so that the clock difference between the NTP client and NTP server can be controlled within 1 ms in the wired network environment. The adaptive clock adjustment skips the clock offset calculation when the NTP packet run trip time (RTT) exceeds a certain threshold. The clock skew compensation addresses the inherent issue that different clocks (or oscillators) naturally drift away from each other. Both adaptive clock adjustment and clock skew compensation are environment dependent and device dependent. The measurement result in our experimental environment shows that the when the RTT threshold is set at 1.7 ms, the best synchronization accuracy is achieved. [ABSTRACT FROM AUTHOR]

This article provides a novel method of reversible data hiding (RDH) by adaptively adjusting the embedding level to achieve minimal distortion and to attain the embedded data confidentiality via data reversion key generation. The peak point queue (PPQ) is used to adaptively determine the embedding level and achieve superior quality in the marked image. In respect to data reversion key for the double protection, our approach, named Athena, makes better utilization of the overhead information and allows users to exchange the key based on a public key infrastructure (PKI). Our experimental results show that the method is capable of providing a better quality image for a range of different test images of various sizes. [ABSTRACT FROM AUTHOR]

Designing a real-time video-streaming system involves the challenges of, network congestion and bandwidth fluctuations, lowering the video quality. In this paper, a realtime video-streaming system is designed and implemented for scalable video coding (SVC) with an adaptive control scheme, containing macro- and micro-control, to resolve the bandwidth fluctuation and network congestion. Under macro-control, the server allocates appropriated video-streaming of SVC for heterogeneous clients based on terminal capacity and network conditions. Under micro-control, the server changes the video frame rate based on video bit rates and network bandwidth. The streaming system with the adaptive control scheme is implemented to improve the peak signal noise ratio (PSNR) by 4.01 % to 10.89 % of the Foreman and Stefan videos. [ABSTRACT FROM AUTHOR]