Compared to the industrial manufacturing process from the decomposition of ammonium nitrate (NH4NO3) that inevitably involves intermediate products (of hydrogen and synthetic ammonia), the direct synthesis route for N2O uses N2 and O2, which is very simple and does not involve any intermediate products. For N2O synthesis, challenges still remain in the direct synthesis from N2 and O2, in particular, with the high selectivity. In terms of the rate-determining step of activation of the inert N2 for direct synthesis of N2O, plasma catalysis is more effective to load electric energies on the N2 molecules than thermal catalysis. Here we report a direct synthesis of N2O from N2 and O2 by plasma catalysis using a cycled storage-discharge (CSD) mode in a dielectric barrier discharge (DBD) reactor with supported catalysts of MnO x /Al2O3. For such a CSD mode, it has two sequential stages in each cycle of the storage of surface-oxygen species only on the catalyst surface generated by oxygen plasma (with a postpurge of nitrogen), followed by the discharge of nitrogen plasma, compared to the normal mode using a mixture of two gases. Under such a CSD mode, it has 6.1 times the selectivity, as high as 91%, of N2O (that far exceeds the literature) compared to the normal mode. This exceptional high selectivity of N2O may be attributed to the outstanding capability of N2 plasma to generate highly reactive nitrogen species that allows efficiently proceeding with surface oxygen on MnO x /Al2O3 catalysts, as well as no extra oxygen species in the gaseous phase (from oxygen plasma of the mixture of two gases) that dismisses the highly reactive nitrogen species (as reductants).

To solve the problems of existing wavelet-deep learning image watermarking methods lack comprehensive robustness or have high computational complexity, this paper proposes RWWA, a robust watermarking framework that integrates wavelet-transform convolution (WTConv) with attention mechanism unit (AMU). The WTConv expands the receptive field and preserves spatial-frequency details while maintaining parameter efficiency. The AMU enhances robust features through channel-spatial attention mechanisms, while the multi-scale feature extraction (MSFE) modules in the decoder integrate with WTConv to optimize watermark extraction accuracy. Experiments show RWWA outperforms HiDDeN with 93.02

To address the imbalance between robustness and imperceptibility in digital image watermarking caused by error correction coding, this paper proposes a robust multi-technique fusion watermarking method that integrates spline interpolation, polar code encoding, Discrete Wavelet Transform (DWT), Discrete Cosine Transform (DCT), Speeded-Up Robust Features (SURF), and Linear Particle Swarm Optimization (LPSO). The proposed method first employs spline interpolation to expand the capacity of cover image, followed by embedding the polar code encoded watermark into the block-DCT domain of the DWT low frequency sub-band. The key embedding parameters are optimized via LPSO under diverse attack scenarios. Additionally, this paper uses the SURF algorithm to detect the feature information of cover and watermarked images and applies it to geometric correction. Experimental results demonstrate that the proposed method achieves outstanding robustness against both single attacks (NC > 0.9325) and combined attacks (BER < 0.0195) while maintaining high imperceptibility (PSNR > 41 dB). The proposed approach addresses the critical robustness-imperceptibility balance in watermarking, while proving the efficacy of technical fusion as a strategy for advancing image security.

The key to single-channel time-domain speech separation lies in encoding the mixed speech into latent feature representations by encoder and obtaining the accurate estimation of the target speaker masks by the separation network. Despite the advanced separation network contribute to separate target speech, but due to the limitation of the time-domain encoder-decoder framework, these separation models commonly improve the separation performance by setting a small convolution kernel size of encoder to increase the length of the coded sequence, which will result in increased computational complexity and training costs for the model. Therefore, in this paper, we propose an efficient time-domain speech separation model using short-sequence encoder-decoder framework (ESEDNet). In this model, we construct a novel encoder-decoder framework to accommodate short encoded sequences, where the encoder consists of multiple convolution and downsampling operations to reduce length of high-resolution sequence, while the decoder utilizes the encoded features to reconstruct the fine-detailed speech sequence of the target speaker. Since the output sequence of the encoder is shorter, when combined with our proposed multi-temporal resolution Transformer separation network (MTRFormer), ESEDNet can efficiently obtains separation masks for the short encoded feature sequence. Experiments show that compared with previous state-of-the-art (SOTA) methods, ESEDNet is more efficient in terms of computational complexity, training speed and GPU memory usage, while maintaining competitive separation performance.

To address the limitations in feature extraction and cross-modal fusion in audio-visual speech enhancement, a multistage deep fusion method was proposed for low signal-to-noise ratio (SNR) conditions. The method consisted of an audio-visual encoding network, a fusion network, and an auditory decoding network. A multi-branch collaborative unit (MCU) was introduced in the auditory encoder, along with an audio-visual attention fusion module (AVAFM) between each visual and auditory layer. A fusion weighting block (FWB) was also designed to optimize and dynamically weight features at each stage. Experiments on TMSV and LGRID datasets showed that the proposed method significantly improved PESQ and STOI scores under various low-SNR conditions. Compared to audio-only enhancement, average gains of 38.95% in PESQ and 33.92% in STOI were achieved at -5 dB, -2 dB, and 1 dB. These results demonstrate the method’s strong denoising ability and the effectiveness of visual information.

Pointing at the vexed question of blind recognition in the convolutional code class, this paper proposes a convolutional code blind identification method via convolutional neural networks (CNNs). First, this algorithm uses the traditional method to generate different convolutional codes, and the feature extraction algorithm adopts the theorem of Euclid’s algorithm. Then, the input signal is loaded to the CNN; next, the feature is extracted by convolutional kernel. Finally, the Softmax activation function is applied to full‐connection layer network. After the input signals pass through the above layers, the system classifies the signals. The research results indicate that the presented algorithm has improved the recognition performance of code length and rate. For different convolutional codes with parameters of (5, 7), (15, 17), (23, 35), (53, 75), and (133, 171) and similar convolutional codes with parameters of (3, 1, 6), (3, 1, 7), (2, 1, 7), (2, 1, 6), and (2, 1, 5), the recognition rate of parameter classification can reach 100% at signal‐to‐noise ratio (SNR) of 3 dB.

As a selective state space model, Mamba exhibits outstanding performance and efficiency in sequence modeling tasks. Therefore, in this letter, we use Mamba as the fundamental network component to construct a novel speech separation model, Speech Conv-Mamba. Specifically, this model embeds Mamba within a U-shaped convolutional network to build the encoder and decoder network for high-dimensional representation and waveform reconstruction of speech signals. Additionally, we stack multiple temporal dilated convolutions and Mamba to create the separation network for separation task. Our comparative experiments on the GRID2Mix and Libri2Mix datasets demonstrate that the proposed model Speech Conv-Mamba, which achieves 98% and 89% of SepFormer's separation accuracy on two datasets using only 9% (2.4 M) of its model size, provides much less computational complexity and training cost.

Abstract This article discusses the ‘power-to-X’ (P2X) concept, highlighting the integral role of non-thermal plasma (NTP) in P2X for the eco-friendly production of chemicals and valuable fuels. NTP with unique thermally non-equilibrium characteristics, enables exotic reactions to occur under ambient conditions. This review summarizes the plasma-based P2X systems, including plasma discharges, reactor configurations, catalytic or non-catalytic processes, and modeling techniques. Especially, the potential of NTP to directly convert stable molecules including CO2, CH4 and air/N2 is critically examined. Additionally, we further present and discuss hybrid technologies that integrate NTP with photocatalysis, electrocatalysis, and biocatalysis, broadening its applications in P2X. It concludes by identifying key challenges, such as high energy consumption, and calls for the outlook in plasma catalysis and complex reaction systems to generate valuable products efficiently and sustainably, and achieve the industrial viability of the proposed plasma P2X strategy.

The burgeoning necessity to discover new methodologies for the synthesis of long-chain hydrocarbons and oxygenates, independent of traditional reliance on high-temperature, high-pressure, and fossil fuel-based carbon, is increasingly urgent. In this context, we introduce a nonthermal plasma-based strategy for the initiation and propagation of long-chain carbon growth from biogas constituents (CO2 and CH4). Utilizing a plasma reactor operating at atmospheric room temperature, our approach facilitates hydrocarbon chain growth up to C40 in the solid state (including oxygenated products), predominantly when CH4 exceeds CO2 in the feedstock. This synthesis is driven by the hydrogenation of CO2 and/or amalgamation of CHx radicals. Global plasma chemistry modeling underscores the pivotal role of electron temperature and CHx radical genesis, contingent upon varying CO2/CH4 ratios in the plasma system. Concomitant with long-chain hydrocarbon production, the system also yields gaseous products, primarily syngas (H2 and CO), as well as liquid-phase alcohols and acids. Our finding demonstrates the feasibility of atmospheric room-temperature synthesis of long-chain hydrocarbons, with the potential for tuning the chain length based on the feed gas composition.

Currently, audio-visual speech separation methods utilize the speaker's audio and visual correlation information to help separate the speech of the target speaker. However, these methods commonly use the approach of feature concatenation with linear mapping to obtain the fused audio-visual features, which prompts us to conduct a deeper exploration for audio-visual fusion. Therefore, in this paper, according to the speaker's mouth landmark movements during speech, we propose a novel time-domain single-channel audio-visual speech separation method: audio-visual fusion with temporal convolution attention network for speech separation model (AVTCA). In this method, we design temporal convolution attention network (TCANet) based on the attention mechanism to model the contextual relationships between audio and visual sequences, and use TCANet as the basic unit to construct sequence learning and fusion network. In the whole deep separation framework, we first use cross attention to focus on the cross-correlation information of the audio and visual sequences, and then we use the TCANet to fuse the audio-visual feature sequences with temporal dependencies and cross-correlations. Afterwards, the fused audio-visual features sequences will be used as input to the separation network to predict mask and separate the source of each speaker. Finally, this paper conducts comparative experiments on Vox2, GRID, LRS2 and TCD-TIMIT datasets, indicating that AVTCA outperforms other state-of-the-art (SOTA) separation methods. Furthermore, it exhibits greater efficiency in computational performance and model size.

Concentrating on the joint recognition problem of modulation and space–time coding, algorithm for blind identification of modulation and space–time coding based upon fourth-order cyclic cumulant and feature parameter extraction in correlation matrix is proposed in this paper. The algorithm first uses the fourth-order cyclic cumulant to identify different modulation modes (4PSK, 8PSK, 16QAM, 32QAM, 64QAM). Different space–time block codes (Alamouti, OSTBC3, OSTBC4, NOSTBC2, and NOSTBC4) are identified by extracting feature parameters in the correlation matrix. The five recognized modulation modes and five space–time codes are sent to convolutional neural network, then 25 combination modes of modulation and space–time coding are recognized by using the Dropout layer, Gaussian noise layer, Flatten layer, dense layer, and fully connected layer successively by applying the Softmax activation function. The research results indicate that the emanated algorithm may effectively identify 5 modulation modes and 5 space–time codes, and can identify 25 combination modes of modulation and space–time coding. Under the condition that SNR is greater than or equal to 0 dB, the joint recognition rate of modulation and space–time coding can reach 100%.

With the escalation of global climate change, reduction of carbon emissions and achieving carbon neutrality have gradually become significant concerns. Conversion of CO2 into valuable products is regarded as a viable solution to address these challenges. In comparison to other catalytic technologies, non-thermal plasma (NTP) offers diverse reaction pathways for CO2 conversion under mild process conditions and can ensure selective production of value-added chemicals and fuels when combined with catalytic materials. However, further research is needed to translate plasma-based approaches to the industrial scale. This article focuses on three crucial characteristics of CO2 conversion in NTP: energy efficiency, conversion rates, and selectivity. We overview recent research advances, outline challenges for future technological advances, and propose potential directions for future research.

With the emergence of more advanced separation networks, significant progress has been made in time-domain speech separation methods. These methods typically use a temporal encoder–decoder structure to encode speech feature sequences, thereby accomplishing the separation task. However, due to the limitation of traditional encoder–decoder structure, the separation performance decreases sharply when the encoded sequence is short, and when encoded sequence is sufficiently long, the separation performance improves, but which leads to an increase in computational complexity and training cost. Therefore, this paper compresses and reconstructs the speech feature sequence through a multi-layer convolution structure, and proposes a multi-layer encoder–decoder time-domain speech separation model (MLED). In this model, our encoder–decoder structure can compress speech sequence to a short length while ensuring the separation performance does not decrease. And combined with our multi-scale temporal attention (MSTA) separation network, MLED achieves efficient and precise separation of short encoded sequences. Therefore, compared to previous advanced time-domain separation methods, our experiments show that MLED achieves competitive separation performance with smaller model size, lower computational complexity, and training cost.

In order to reconstruct the sparse check matrix of LDPC code,a sparse check matrix reconstruction algorithm for LDPC code at high BER was proposed based on modified LBP decoding.Firstly,some bits were selected randomly from the codeword matrix to construct the codeword analysis matrix,and Gaussian elimination on it was performed to find the dual space.Secondly,by determining whether the pairwise space vectors were sparse or not,it improved the effi-ciency of the subsequent suspected check vectors determination.Finally,in the case of insufficient received codes,the known check vectors were combined with the modified LBP decoding method to correct the wrong codes,so as to speed up the reconstruction of the sparse check matrix of LDPC code and improve the reconstruction performance.The simula-tion results show that the reconstruction rate of sparse check matrix of(648,324)LDPC codes in IEEE 802.11n protocol is improved by 52.16% compared with the existing algorithms,and can reach 92.28% at high BER of 0.004 5.

This paper proposes to model and simulate the radio frequency interference (RFI), especially the narrowband RFI and the wideband RFI, for the high-frequency over-the-horizon (OTH) radar. Based on the theories of random process and linear filtering, the proposed models uses the white Gaussian noise as the input of a linear filter, so that the output is a random process with small bandwidth. In the filter design, seven kinds of response functions are proposed for the convolution method, and four sets of coefficients are proposed for the auto-regressive and moving-average method. Besides, extended approaches are also provided to increase the diversity of the RFI simulation. Numerical experiments demonstrate that the proposed methods can simulate most range-Doppler maps of real RFI.

Bearing fault is the most likely to occur in mechanical fault, and stochastic resonance (SR), as a noise enhanced signal processing tool, can find mechanical faults as early as possible, so as to avoid larger problems. However, most of the existing research methods are based on the first-order Langevin equation. According to the previous studies of many scholars, the weak signal detection ability of the second-order system is better than that of the first-order system, and the coupled system also has better performance due to the addition of the control system. So, in order to detect the fault signal more easily, a second-order coupled tristable stochastic resonance system (SCTSR) based on the adaptive genetic algorithm (AGA) is proposed, it is an improvement on improving the first-order coupled tristable stochastic resonance system (FCTSR). First, based on the fourth-order Runge–Kutta algorithm (F-RK), the performances of monostable, bistable and tristable control systems to SCTSR are compared, it is verified that the monostable system has the best performance as SCTSR’s control system. Secondly, the equivalent potential function of SCTSR is derived, and the influences of each system parameters on it are researched. The output signal-to-noise ratio gain ( SNRG ) is chosen as a measure to verify that SCTSR’s performance is better than that of FCTSR, and the influences of parameters on SNRG are discussed. SCTSR and FCTSR are used to detect low-, high- and multi-frequency cosine signals combined with AGA. The simulation results are compared with the wavelet transform method, which proves the performance superiority of SR, and also prove that SCTSR is easier to detect weak signals and has a stronger de-noising ability. Finally, SCTSR and FCTSR are applied in bearing fault detection under Gaussian white noise and trichotomous noise. The results also prove that SCTSR can get larger peaks and SNRG , and it is easier to detect fault signals. This proves that SCTSR’s performance is superior that of other methods in bearing fault detection, and has better engineering application value.

With the advent of the Industry 5.0 era, the Internet of Things (IoT) devices face unprecedented proliferation, requiring higher communications rates and lower transmission delays. Considering its high spectrum efficiency, the promising filter bank multicarrier (FBMC) technique using offset quadrature amplitude modulation (OQAM) has been applied to Beyond 5G (B5G) industry IoT networks. However, due to the broadcasting nature of wireless channels, the FBMC-OQAM industry IoT network is inevitably vulnerable to adversary attacks from malicious IoT nodes. The FBMC-OQAM industry cognitive radio network (ICRNet) is proposed to ensure security at the physical layer to tackle the above challenge. As a pivotal step of ICRNet, blind modulation recognition (BMR) can detect and recognize the modulation type of malicious signals. The previous works need to accomplish the BMR task of FBMC-OQAM signals in ICRNet nodes. A novel FBMC BMR algorithm is proposed with the transform channel convolution network (TCCNet) rather than a complicated two-dimensional convolution. Firstly, this is achieved by designing a low-complexity binary constellation diagram (BCD) gridding matrix as the input of TCCNet. Then, a transform channel convolution strategy is developed to convert the image-like BCD matrix into a series like data format, accelerating the BMR process while keeping discriminative features. Monte Carlo experimental results demonstrate that the proposed TCCNet obtains a performance gain of 8% and 40% over the traditional inphase/quadrature (I/Q)-based and constellation diagram (CD)-based methods at a signal noise ratio (SNR) of 12 dB, respectively. Moreover, the proposed TCCNet can achieve around 29.682 and 2.356 times faster than existing CD-Alex Network (CD-AlexNet) and I/Q-Convolutional Long Deep Neural Network (I/Q-CLDNN) algorithms, respectively.

Traditional optical spatial modulation techniques all rely on the premise that the system receiving end can obtain accurate channel state information.The optical differential spatial modulation (ODSM) system effectively avoids complex channel estimation, but related research only analyzes the system performance under a single turbulent state, and ignores the effects of factors such as pointing errors and path loss in actual free space optical communication systems.Therefore, the Málaga turbulence channel, which could characterize all atmospheric turbulence states, was used to derive the upper bound expression and diversity order of the bit error rate of the ODSM system under the atmospheric joint effect such as turbulence, pointing error, and path loss.The analysis results show that compared with other optical spatial modulation schemes, the ODSM system avoids complex channel estimation, making the system no longer affected by channel estimation errors, thus possessing higher anti-interference and stability.The bit error rate performance of the ODSM system decreases as the intensity of turbulence and pointing error increase.By optimizing parameters such as the number of optical antennas, the number of photodetectors, and the modulation signal order, the bit error rate performance of the ODSM system can be further improved.