Search

Single channel blind source separation (SCBSS) refers to separate multiple sources from a mixed signal collected by a single sensor. The existing methods for SCBSS mainly focus on separating two sources and have weak generalization performance. To address these problems, an algorithm is proposed in this paper to separate multiple sources from a mixture by designing a parallel dual generative adversarial Network (PDualGAN) that can build the relationship between a mixture and the corresponding multiple sources to realize one-to-multiple cross-domain mapping. This algorithm can be applied to any mixed model such as linear instantaneous mixed model and convolutional mixed model. Besides, one-to-multiple datasets are created which including the mixtures and corresponding sources for this study. The experiment was carried out on four different datasets and tested with signals mixed in different proportions. Experimental results show that the proposed algorithm can achieve high performance in peak signal-to-noise ratio (PSNR) and correlation, which outperforms state-of-the-art algorithms.

In a recent study of auditory evoked potential (AEP) based brain-computer interface (BCI), it was shown that, with an encoder-decoder framework, it is possible to translate human neural activity to speech (T-CAS). However, current encoder-decoder-based methods achieve T-CAS often with a two-step method where the information is passed between the encoder and decoder with a shared dimension reduction vector, which may result in a loss of information. A potential approach to this problem is to design an end-to-end method by using a dual generative adversarial network (DualGAN) without dimension reduction of passing information, but it cannot realize one-to-one signal-to-signal translation (see Fig.1 (a) and (b)). In this paper, we propose an end-to-end model to translate human neural activity to speech directly, create a new electroencephalogram (EEG) datasets for participants with good attention by design a device to detect participants' attention, and introduce a dual-dual generative adversarial network (Dual-DualGAN) (see Fig. 1 (c) and (d)) to address an end-to-end translation of human neural activity to speech (ET-CAS) problem by group labelling EEG signals and speech signals, inserting a transition domain to realize cross-domain mapping. In the transition domain, the transition signals are cascaded by the corresponding EEG and speech signals in a certain proportion, which can build bridges for EEG and speech signals without corresponding features, and realize one-to-one cross-domain EEG-to-speech translation. The proposed method can translate word-length and sentence-length sequences of neural activity to speech. Experimental evaluation has been conducted to show that the proposed method significantly outperforms state-of-the-art methods on both words and sentences of auditory stimulus., Comment: 12 pages, 13 figures

A single channel electromyography blind recognition model based on watermarking is proposed in this paper. Single Channel Independent Component Analysis is adopted to avoid complicated circuit connection and the unreliability of hardware and reduce the noise which accompanied with surface Electromyography (sEMG) signals. Embedded watermarking is applied to solve the problem of blind source separation disorder. A self adaptive neural network and some eigenvectors are applied in sEMG features classification. From the classification results, hand gestures can be recognized. In consideration of time-scale synchronization attack, the host sEMG signals are transformed into wavelet domain and the synchronization codes are embedded. The experiment results show that the model proposed in this paper is penetrable against most common signal processing, and can recognize the hand gesture accurately.

Chemical conversion treatment is a basic method to improve the corrosion resistance of magnesium and its alloys. In the method of conventional conversion, however, the solution contains hexavalent chromium compounds which are harmful to the environment. The replacement surface treatment has been strongly emphasized in recent years. A chemical conversion treatment solution containing phytic was developed in this study for magnesium alloys. The surface conversion coating was characterized by SEM and EDS. The factors of influence to the corrosion resistance of conversion coating were discussed through hydrogen evolution method and polarization potential. The corrosion resistance of phytic acid conversion coating was comparable with the chromate containing. The experimental results indicate that the phytic acid conversion coating had better corrosion resistance when magnesium alloy was treated with a solution containing 0.5% to 1% phytic acid at temperature 25 to 60 °C and pH value at the range of 3∼5 for 30 to 60 min.

By use of hand gesture recognition technology, traditional remote manufacturing networks can be operated by disabled people who lose the grasp ability. The surface electromyography (sEMG) network inference is critically important for revealing fundamental hand gesture processes, investigating sEMG functions, and understanding their relations. However, current hand gesture recognition methods using sEMG do not sufficiently consider the temporal behavior of this type of data and lack the capability to capture the complex nonlinear system dynamics. It is therefore imperative to have good methods to explore a more suitable choice, which can avoid the problems mentioned above. This paper presents a rapid prototyping manufacturing network with sEMG using recurrent hidden Markov models (RHMMs) and a particle swarm optimization (PSO) approach. It also provides high-level modeling, programming methods, and running results for this remote manufacturing network in terms of software and hardware. PSO is used to train the RHMM and determine the model parameters. Remote manufacturing is accomplished between the network monitor center and a number of sensor nodes by use of a sEMG sensor circuit and Winsock monitor sensors. Furthermore, it has such advantages as network wireless, low cost, and strong interaction and expansibility, and is easy to maintain and promote. © The Author(s) 2011 Reprints and permissions: sagepub.co.uk/ journalsPermissions.nav.

Traditional hand gesture identification systems require designers to use multiple serial interfaces or multiplexer as tools to record surface Electromyography (sEMG). These ways lead to complicated circuit connection and unreliability of hardware. It is therefore imperative to have good methods to explore a more suitable design choice, which can avoid the problems mentioned above as more as possible. This paper presents a hand gesture identification system of sEMG based on Single Channel Independent Component Analysis (SCICA) which can avoid the problems mentioned above and reduce the noise which accompanied with sEMG signal. It also provides high-level modeling, programming methods and running results for this system in terms of software and hardware. More specifically, real time communication and identification are accomplished accurately between the hand and identification system by taking advantage of USB interface and VC++ s2008. This system also can manage the historical sEMG signals by using database. Furthermore, it has such advantages as low cost and easy to maintain and promote.