Feature decoupled knowledge distillation enabled lightweight image transmission through multimode fibers.

Multimode fibers (MMF) show tremendous potential in transmitting high-capacity spatial information. However, the quality of multimode transmission is quite sensitive to inherent scattering characteristics of MMF and almost inevitable external perturbations. Previous research has shown that deep learning may break through this limitation, while deep neural networks are intricately designed with huge computational complexity. In this study, we propose a novel feature decoupled knowledge distillation (KD) framework for lightweight image transmission through MMF. In this framework, the frequency-principle-inspired feature decoupled module significantly improves image transmission quality and the lightweight student model can reach the performance of the sophisticated teacher model through KD. This work represents the first effort, to the best of our knowledge, that successfully applies a KD-based framework for image transmission through scattering media. Experimental results demonstrate that even with up to 93.4% reduction in model computational complexity, we can still achieve averaged Structure Similarity Index Measure (SSIM) of 0.76, 0.85, and 0.90 in Fashion-MNIST, EMNIST, and MNIST images respectively, which are very close to the performance of cumbersome teacher models. This work dramatically reduces the complexity of high-fidelity image transmission through MMF and holds broad prospects for applications in resource-constrained environments and hardware implementations.

Signal recovery in optical wireless communication using photonic convolutional processor.

Deep neural networks (DNNs) have been applied to recover signals in optical communication systems and have shown competence of mitigating linear and nonlinear distortions. However, as the data throughput increases, the heavy computational cost of DNNs impedes them from rapid and power-efficient processing. In this paper, we propose an optical communication signal recovery technology based on a photonic convolutional processor, which is realized by dispersion delay unit and wavelength division multiplexing. Based on the photonic convolutional processor, we implement an optoelectronic convolutional neural network (OECNN) for signal post-equalization and experimentally demonstrate on 16QAM and 32QAM of an optical wireless communication system. With system parameters optimization, we verify that the OECNN can achieve accurate signal recovery where the bit error ratio (BER) is below the 7% forward error correction threshold of 3.8×10-3 at 2Gbps. With adding the OECNN-based nonlinear compensation, compared with only linear compensation, we improve the quality (Q) factor by 3.35 dB at 16QAM and 3.30 dB at 32QAM, which is comparable to that of an electronic neural network. This work proves that the photonic implementation of DNN is promising to provide a fast and power-efficient solution for optical communication signal processing.

3.76-Gbps yellow-light visible light communication system over 1.2 m free space transmission utilizing a Si-substrate LED and a cascaded pre-equalizer network.

Recently, visible light communication (VLC) has emerged as a promising communication method in 6G. To achieve 6G high-speed transmission, wavelength division multiplexing (WDM) based VLC systems are a highly promising candidate. However, the "yellow and green gap" greatly limits the yellow light efficiency of InGaN-based LEDs and also restricts the transmission rate of yellow LEDs. In addition, pre-equalization and post-equalization also have an important impact on high-speed communication. In this paper, we propose to employ a vertical InGaN-based Si-substrate yellow LED with bit-power loading discrete multitone (DMT) modulation and a novel cascaded pre-equalizer network to achieve a high-speed yellow-light VLC system. The proposed cascaded pre-equalizer network is based on a digital Zobel network and a partial nonlinear pre-equalizer (DZNPN). The microscopic time-domain transient response of the high-speed and large-amplitude signal is also investigated to show a severe impairment. Utilizing the DZNPN cascaded pre-equalizer network based on the third-order Volterra series, a record-breaking data rate of 3.764Gbps over 1.2 m free space and 3.808Gbps over 0.7 m are experimentally demonstrated under the hard decision-forward error correction (HD-FEC) threshold of 3.8 × 10-3. The rate can be improved from 2.818Gbps to 3.764Gbps with 650Mbaud compared to the un-preprocessed signal. This is the highest data rate ever reported for yellow-light VLC systems based on a single LED to the best of our knowledge.

Transfer Learning Strategy in Neural Network Application for Underwater Visible Light Communication System.

Post-equalization using neural network (NN) is a promising technique that models and offsets the nonlinear distortion in visible light communication (VLC) channels, which is recognized as an essential component in the incoming 6G era. NN post-equalizer is good at modeling complex channel effects without previously knowing the law of physics during the transmission. However, the trained NN might be weak in generalization, and thus consumes considerable computation in retraining new models for different channel conditions. In this paper, we studied transfer learning strategy, growing DNN models from a well-trained 'stem model' instead of exhaustively training multiple models from randomly initialized states. It extracts the main feature of the channel first whose signal power balances the signal-to-noise ratio and the nonlinearity, and later focuses on the detailed difference in other channel conditions. Compared with the exhaustive training strategy, stem-originated DNN models achieve 64% of the working range with five times the training efficiency at most or more than 95% of the working range with 150% higher efficiency. This finding is beneficial to improving the feasibility of DNN application in real-world UVLC systems.

Hybrid frequency domain aided temporal convolutional neural network with low network complexity utilized in UVLC system.

Deep neural network has been used to compensate the nonlinear distortion in the field of underwater visible light communication (UVLC) system. Considering the tradeoff between the equalization performance and the network complexity is the priority in practical applications. In this paper, we propose a novel hybrid frequency domain aided temporal convolutional neural network (TFCNN) with attention scheme as the post-equalizer in CAP modulated UVLC system. Experiments illustrate that the proposed TFCNN can achieve better equalization performance and remain the bit error rate (BER) below the 7% hard-decision forward error correction (HD-FEC) limit of 3.8×10-3 when other equalizers loss effectiveness under serious distortion condition. Compared with the standard deep neural network, TFCNN shows 76.4% network parameters complexity reduction.

Adaptive deep-learning equalizer based on constellation partitioning scheme with reduced computational complexity in UVLC system.

Visible light communication (VLC) system has emerged as a promising solution for high-speed underwater data transmission. To tackle with the linear and nonlinear impairments, deep learning inspired equalization is introduced into VLC. Despite their success in accuracy, deep learning approaches often come with high computational budget. In this paper, we propose an adaptive deep-learning equalizer based on complex-valued neural network and constellation partitioning scheme for 64 QAM-CAP modulated underwater VLC (UVLC) system. Inspired by the fact that symbols modulated at different levels experience various extent of nonlinear distortion, we adaptively partition the received symbols in constellation and design compact equalization networks for specific regions to reduce computation consumption. Experiments demonstrate that the partitioned equalizer can achieve the bit error rate below the 7% hard-decision forward error correction (HD-FEC) limit of 3.8 × 10-3 at 2.85 Gbps similar to the standard complex-valued network, yet with 56.1% total computational complexity reduction. This work paves the path for online data processing in high speed UVLC system.

Highly Efficient and Stable Self-Powered Mixed Tin-Lead Perovskite Photodetector Used in Remote Wearable Health Monitoring Technology.

Realization of remote wearable health monitoring (RWHM) technology for the flexible photodiodes is highly desirable in remote-sensing healthcare systems used in space stations, oceans, and forecasting warning, which demands high external quantum efficiency (EQE) and detectivity in NIR region. Traditional inorganic photodetectors (PDs) are mechanically rigid and expensive while the widely reported solution-processed mixed tin-lead (MSP) perovskite photodetectors (PPDs) exhibit a trade-off between EQE and detectivity in the NIR region. Herein, a novel functional passivating antioxidant (FPA) strategy has been introduced for the first time to simultaneously improve crystallization, restrain Sn2+ oxidization, and reduce defects in MSP perovskite films by multiple interactions between thiophene-2-carbohydrazide (TAH) molecules and cations/anions in MSP perovskite. The resultant solution-processed rigid mixed Sn-Pb PPD simultaneously achieves high EQE (75.4% at 840 nm), detectivity (1.8 × 1012 Jones at 840 nm), ultrafast response time (trise /tfall = 94 ns/97 ns), and improved stability. This work also highlights the demonstration of the first flexible photodiode using MSP perovskite and FPA strategy with remarkably high EQE (75% at 840 nm), and operational stability. Most importantly, the RWHM is implemented for the first time in the PIN MSP perovskite photodiodes to remotely monitor the heart rate of humans at rest and after-run conditions.

[Overexpression of a fusion protein of 4-coumaroyl-CoA ligase and polyketide synthase for raspberry ketone production in Chlamydomonas reinhardtii].

Raspberry ketones have important therapeutic properties such as anti-influenza and prevention of diabetes. In order to obtain raspberry ketone from Chlamydomonas reinhardtii, two enzymes catalyzing the last two steps of raspberry ketone synthesis, i.e. 4-coumaryl-CoA ligase (4CL) and polyketide synthase (PKS1), were fused using a glycine-serine-glycine (GSG) tripeptide linker to construct an expression vector pChla-4CL-PKS1. The fusion gene 4CL-PKS1 driven by a PSAD promoter was transformed into a wild-type (CC125) and a cell wall-deficient C. reinhardtii (CC425) by electroporation. The results showed the recombinant C. reinhardtii strain CC125 and CC425 with 4CL-PKS1 produced raspberry ketone at a level of 6.7 µg/g (fresh weight) and 5.9 µg/g (fresh weight), respectively, both were higher than that of the native raspberry ketone producing plants (2-4 µg/g).