Search

The poly(m-phenylene isophthalamide) (PMIA) paper has attracted extensive interests due to its ultrahigh mechanical properties as an ideal protective material for anti-impact damage applications. In the pursuit of additional properties, composites based on the PMIA matrix and various fillers are widely explored. However, additional improvements are frequently obtained at the expense of mechanical properties because of the serious interfacial compatibility brought by different components. In this study, a self-reinforced doping strategy is proposed by combining microscale PMIA fibers as the fillers and nanoscale PMIA fibers as the matrix to form a micronano paper. Without the limitation of the interfacial compatibility issues, the nanofibers are tightly aligned and adhered to the microfibers, enabling the in situ generation of hydrogen bonds at the interfaces. A compact interfacial structure is thus constructed with reduced porosity on the surface. It indicates that the microfibers have a positive impact on the improvement of mechanical properties. In our optimized sample with 5 wt % microfibers, the elastic modulus, tensile strength, and elongation are 1530 MPa, 24.8 MPa, and 5.3%, respectively, which are 142, 49.4, and 65% higher than those of the pristine nano-PMIA paper. In addition, the insulating performance is also improved, facilitating its further application extended to broad fields.

Recently, electroencephalographic (EEG) emotion recognition attract attention in the field of human-computer interaction (HCI). However, most of the existing EEG emotion datasets primarily consist of data from normal human subjects. To enhance diversity, this study aims to collect EEG signals from 30 hearing-impaired subjects while they watch video clips displaying six different emotions (happiness, inspiration, neutral, anger, fear, and sadness). The frequency domain feature matrix of EEG signals, which comprise power spectral density (PSD) and differential entropy (DE), were up-sampled using cubic spline interpolation to capture the correlation among different channels. To select emotion representation information from both global and localized brain regions, a novel method called Shifted EEG Channel Transformer (SECT) was proposed. The SECT method consists of two layers: the first layer utilizes the traditional channel Transformer (CT) structure to process information from global brain regions, while the second layer acquires localized information from centrally symmetrical and reorganized brain regions by shifted channel Transformer (S-CT). We conducted a subject-dependent experiment, and the accuracy of the PSD and DE features reached 82.51% and 84.76%, respectively, for the six kinds of emotion classification. Moreover, subject-independent experiments were conducted on a public dataset, yielding accuracies of 85.43% (3-classification, SEED), 66.83% (2-classification on Valence, DEAP), and 65.31% (2-classification on Arouse, DEAP), respectively.

Recently, rapid progress has been made in the application of organic-inorganic halide perovskites in electronic devices, such as memristors and artificial synaptic devices. Organic-inorganic halide perovskite is considered as a promising candidate for the next generation of computing devices due to its ion migration property and advantages in manufacturing. In this work, a two-dimensional (2D)-3D organic-inorganic hybrid perovskite memristor was studied, using the stacking structure of indium tin oxide (ITO)/FA1−yMAyPbI3−xClx/(PEA)2PbI4/Au. The results show that this new type of memristor has novel resistance switching characteristics, such as scanning-rate-dependent current switching property, good current-voltage (I–V) curve repeatability, and ultralow energy consumption. A defect-modulated electron tunneling mechanism is demonstrated using the p-i-n junction model, and it is proven that the conductance state of the memristive device is determined by the defect concentration in the perovskite film near the electrode sides. In addition to the good memristive properties, this 2D-3D perovskite memristor can also function well as an artificial synapse, and its internal defect movement can faithfully simulate the inflow and extrusion of Ca2+in biological synapses. Moreover, this perovskite-based artificial synapse has ultra-low power consumption due to the switchable p-i-n structure in organic-inorganic halide perovskites. Our finding highlights the immense application potential of the 2D-3D perovskite memristor in the future neuromorphic computing system.

The interface engineering is always an important way to modulate the behavior of the ferroelectric tunnel memristor (FTM), which directly affects its biological synaptic properties. Here, to investigate the effect of interface on bio-synapse performance, FTMs with the structure of Pt/ BaTiO3/La0.67Sr0.33MnO3were studied, of which the interfaces of the device can be controlled through tailoring the termination of the SrTiO3substrate and the unit-cell of BaTiO3thin film by its growth mode. FTMs with different interfaces present opposite resistive switching behaviors due to the contrary polarization directions and the different band alignments in the BaTiO3film. More importantly, the synaptic learning properties of FTMs can also be tuned by tailoring the interface. FTMs with different interface terminations could modulate different properties of long-term potentiation (LTP), long-term depression (LTD), spike-timing-dependent plasticity (STDP), and paired-pulse facilitation (PPF). Based on the synaptic behaviors of these two interface-engineered FTMs, the artificial neural network (ANN) system could be constructed to complete a handwritten digital image recognition process, and the accuracies of both are close to 90%. Our results provide a useful reference for tuning the functionalities of memristors through nanoscale interface engineering.

Machine learning is regarded as an attractive solution for soft failure management in optical networks; however, the performance of trained models working on unseen data is of growing concern, due to scarce historical data and high training costs. Hence, the issues of reducing training data and improving generalization have received considerable attention. In this paper, a soft failure detection (SFD) and identification scheme is proposed based on digital residual spectrum that leverages auto-encoder (AE) and support vector machine. The digital residual spectrum acquired by the coherent receiver is computed by subtracting the averaged spectrum of multiple normal digital spectra from the digital spectrum. The scheme features: (i) high generalization, i.e., a model trained for a specific transmission setup performs well in other setups with different fiber lengths; (ii) low cost, i.e., the digital residual spectrum is easily obtained from a coherent receiver without additional hardware; and (iii) easy training, i.e., only normal samples are needed to train the SFD model (less pressure on the collection of rare soft-failure data). Using the model trained for any specific setup, we demonstrated an area under the curve and identification accuracy above 99.24% and 96.45%, respectively, for five experimental setups.

To mitigate nonlinear distortions in multi-user analog radio over fiber (A-RoF) system, a nonlinear equalizer (NLE) which combines transfer learning (TL) and waveform regression-based artificial neural network (ANN) is proposed and validated experimentally in this research. Unlike traditional ANNs, waveform regression ANNs can be regarded as waveform training ANNs and avoid complex-value training. In this paper, we first analyze the modulation distortions that appear in multi-user RoF systems and then propose a waveform regression ANN-based nonlinear equalizer (ANN-NLE). Subsequently, we use TL to reduce training costs and increase the compatibility of the ANN-NLE. Finally, an A-RoF system with multicore fiber is adopted to verify the proposed approach. According to the simulation and experimental results, the bit error ratio (BER) of a multi-user signal is decreased to the threshold of 3.8 × 10−3, and the improved optical receiver sensitivity is greater than 1.5 dB. This proposed method could realize nonlinear equalization and display good compatibility with different systems.

In this paper, we investigated the effects of error-feedback noise shaping (EFNS) on a low-resolution Kramers-Kronig (KK) detection system for the first time. Both 16-ary quadrature amplitude modulation (16-QAM) and 64-ary quadrature amplitude modulation (64-QAM) were considered in the experiment. The carrier tone was added on the transmitter side using the virtual carrier-assisted single sideband (VCA-SSB) method to meet the minimum phase (MP) condition. Under the MP condition, KK detection only needs the intensity of the optical signal to be detected by one photodiode (PD) to retrieve the phase information. However, the virtual carrier tone increases the carrier-to-signal power ratio (CSPR), which results in a high peak-to-average power ratio (PAPR) at the transmitter, causing an increase in quantization noise. To mitigate quantization noise in the low-resolution KK detection system, EFNS was applied. The application of EFNS increased the optimal CSPR of the KK detection system and improved receiver sensitivity. Using a 4-bit DAC and the EFNS, 80 km transmission was achieved in the 60-Gbit/s 64-QAM KK detection system.

Poly (m-phenylene isophthalamide)-based loose nanofiltration membranes were fabricated by electrospinning and subsequential electrostatic layer by-layer self-assembly methods, and the polyelectrolyte multilayers (PEMs) generated by polyacrylic acid (PAA) and polyethyleneimine (PEI) were adopted to contribute the surface functional layer of nanofiber membranes. The incidences of spraying parameter and nano-SiO2content on the morphology, chemical constitution and dyes separation performance of PEMs functional layer was investigated, respectively. The results showed that the construction of the PEMs functional layer could obviously enhance the Congo red (CR) filtration efficiency by increasing the rejection from 4.9% to 99.8% and improve the surface hydrophilicity by reducing the water contact angle (WCA) from 90.1° to 51.8°. Nano-SiO2is introduced as a filler to regulate the assembly density of PEMs functional layers, the prepared PEMs/PMIA/SiO2membrane achieved a relatively stable permeation flux (22.9 L·m−2·h−1), and maintained high separation efficiency of CR/NaCl solution, revealing the excellent dye desalination performance, which exhibited a potential for the dye/salt wastewater treatment. In particular, we offered a new method to prepare loose nanofiltration (LNF) membranes by electrospinning and electrostatic self-assembly technique with accurate separation of inorganic salts and dye molecules.

ABSTRACTLong non-coding RNA (lncRNA) BBOX1 antisense RNA 1 (BBOX1-AS1) was reported to participate in ovarian cancer, while its role in other ovarian disorders is unclear. We speculated that BBOX1-AS1 could interact with microRNA(miR)-146b, which is involved in premature ovarian failure (POF). This study was therefore carried out to explore its role in POF. In this study, 60 patients with POF and 60 controls were enrolled. The expression of BBOX1-AS1 and miR-146b were analyzed by RT-qPCRs. The direct interaction between miR-146b and the wild type BBOX1-AS1 (BBOX1-AS1-WT) or mutant BBOX1-AS1 (BBOX1-AS1-mut) was explored with RNA-RNA pulldown assay. Subcellular location of BBOX1-AS1 in COV434 granulosa cells was detected by subcellular fractionation. The role of BBOX1-AS1 and miR-146b in the apoptosis of COV434 cells was evaluated by cell apoptosis assay. Overexpression assay was applied to explore the relationship between BBOX1-AS1 and miR-146b. We found that the expression levels of BBOX1-AS1 were increased, while the expression levels of miR-146b were decreased in POF patients. BBOX1-AS1-WT, but not BBOX1-AS1-mut, directly interacted with miR-146b. BBOX1-AS1 was detected in both nucleus and cytoplasm, while they did not affect the expression of each other. BBOX1-AS1 suppressed the role of miR-146b in cell apoptosis. Therefore, BBOX1-AS1 may increase the apoptosis of granulosa cells in POF by sponging miR-146b.

Biomaterial-based memristors (bio-memristors) are often adopted to emulate biological synapse functions and applied to construct neural computing networks in brain-inspired chip systems. However, the randomness of conductive filament formation in bio-memristors inhibits their switching performance by causing the dispersion of the device-switching parameters. In this case, a facile porous silk fibroin (p-SF) memristor was obtained through a protein surface reconstruction strategy, in which the size of the hole can be adjusted by the density of hybrid nanoseeds. The porous SF memristors exhibit greatly enhanced electrical characteristics, including uniform I–Vcycles, centralized distribution of the switching voltages, and both high and low resistances, compared to devices without pores. The results of three-dimensional (3D) simulations based on classical density functional theory (cDFT) suggest that the reconstructed pores in the SF layers guide the formation and fracture of Ag filaments under an electric field and enhance the overall conductivity by separating Ag+ion and electron diffusion pathways. Ag+ions are predicted to preferentially diffuse through pores, whereas electrons diffuse through the SF network. Interestingly, the device conductance can be bidirectionally modulated gradually by positive and negative voltages, can faithfully simulate short-term and long-term plasticity, and can even realize the triplet-spike-timing-dependent plasticity (triplet-STDP) rule, which can be used for pattern recognition in biological systems. The simulation results reveal that a memristor network of this type has an accuracy of ∼95.78% in memory learning and the capability of pattern learning. This work provides a facile technology route to improve the performance of bionic-material memristors.

Mill load is a key parameter for the safety and optimal control of the grinding process in mineral processing. Grinding sound signal is usually used to detect the mill load indirectly. However, the relationship between grinding sound signal and the mill load is really complicated. In this study, a mill load identification method is proposed by combining the machine learning algorithm with the sound recognition technology. Firstly, a geometric spectral subtraction denoising method based on auto regressive (AR) spectrum estimation is proposed to preprocess the grinding sound signal. Then, the ensemble empirical mode decomposition (EEMD) method is used to decompose the grinding signal. Suitable IMF components are then selected to reconstruct the grinding signal and the box fractal dimension feature is extracted. Finally, an optimized extreme learning machine (ELM) method was proposed to identify the mill load. The simulation results using industrial data show that the proposed method has better overall recognition accuracy compared with other machine learning methods.

Polyimide (PI) has garnered increasing attention in the field of membrane technology owing to its excellent comprehensive performance. In this work, PI nanofibrous membranes with a controllable pore structure were fabricated by the thermal imidization of electrospun polyamide acid (PAA) nanofibrous membranes. Polytetrafluoroethylene (PTFE) and poly(tetrafluoroethylene-hexafluoropropylene) (FEP) particles were introduced into the PAA nanofibrous membrane with the aim of enhancing the membrane's hydrophobicity and regulating its pore structure by electrospraying the PTFE/FEP mixed aqueous dispersion. During the thermal imidization process, melt casting of FEP particles occurred, leading to the adhesion of PI nanofibrous membranes and subsequent regulation of the membrane's pore structure. Meanwhile, due to the imidization temperature being lower than PTFE’s melting temperature, the unmelted PTFE particles would further enhance the membrane’s hydrophobicity. We have successfully provided a novel method for fabricating PI nanofibrous membranes with controllable pore structures and excellent hydrophobicity to expand their applications in membrane separation. Owing to the hydrophobicity and lipophilicity, the obtained PI nanofibrous membranes were applied to oil-water (kerosene/water) separation to characterize the separation performance. The results showed that, following the regulation of the pore structure, the PI nanofibrous membranes exhibited a high oil flux (203.83 L/m2·h) and separation efficiency (＞99.00%) without external pressure.

The analog photonics link (APL) is widely used in microwave photonics. However, in wideband and multi-carrier systems, the third inter-modulation distortion (IMD3) and cross-modulation distortion (XMD) will jointly limit the spurious-free dynamic range (SFDR) of links. In this paper, we experimentally present a linearized wideband and multi-carrier APL, in which the IMD3 and XMD are mitigated simultaneously by using artificial neural networks with transfer learning (TL-ANN). In this experiment, with different artificial neural networks, which are trained with the knowledge obtained from the two- or three-sub-carrier system, the IMD3 and XMD are suppressed by 21.71 dB and 11.11 dB or 22.38 dB and 16.73 dB, and the SFDR is improved by 13.4 dB or 14.3 dB, respectively. Meanwhile, compared with previous studies, this method could reduce the training time and training epochs to 16% and 25%.

Youth violence is a growing concern in Singapore. To address this complex social issue, we surveyed the psychology, social science, and criminology literature to identify a total of 11 intrinsic (familial, individual, school) and 2 extrinsic (peer) factors linked to youth violence, and also their interdependencies. We then developed a complex agent network model where each complex agent is represented by a complex factor network of the 13 factors along with youth violence, coupled to each other through the extrinsic factors to form a complex social network. We simulated the model using as initial conditions the results from a large-scale school-based survey of the factors and random social ties. We find factors in each complex agent evolving with time under the influences from other factors, and the social ties between agents evolving with time as a result of behavioral imitation between agents. We ran a sensitivity analysis on the model, to find that the model is most sensitive to the parameters linking (1) non-intact family, (2) delinquency in general, (3) school disengagement, (4) peer delinquency, and (5) friends in gangto gang involvement. We also ran a series of intervention scenario simulations, and our results show that it is critical to intervene early, and successful interventions work by tipping the balance between competing intrinsic and extrinsic factors. Mental health professionals and school counsellors can then apply this unique insight from the model to design more effective interventions.