Conventional meta‐atom designs rely heavily on researchers’ prior knowledge and trial‐and‐error searches using full‐wave simulations, resulting in time‐consuming and inefficient processes. Inverse design methods based on optimization algorithms, such as evolutionary algorithms, and topological optimizations, have been introduced to design metamaterials. However, none of these algorithms are general enough to fulfill multi‐objective tasks. Recently, deep learning methods represented by generative adversarial networks (GANs) have been applied to inverse design of metamaterials, which can directly generate high‐degree‐of‐freedom meta‐atoms based on S‐parameters requirements. However, the adversarial training process of GANs makes the network unstable and results in high modeling costs. This paper proposes a novel metamaterial inverse design method based on the diffusion probability theory. By learning the Markov process that transforms the original structure into a Gaussian distribution, the proposed method can gradually remove the noise starting from the Gaussian distribution and generate new high‐degree‐of‐freedom meta‐atoms that meet S‐parameters conditions, which avoids the model instability introduced by the adversarial training process of GANs and ensures more accurate and high‐quality generation results. Experiments have proven that our method is superior to representative methods of GANs in terms of model convergence speed, generation accuracy, and quality.

In autonomous driving, cars rely on light detection and ranging (lidar) to navigate the surroundings, but interference from the environment makes it difficult to retrieve useful information. To address this problem, this paper develops a noise reduction method to filter lidar point clouds (i.e., an adaptive radius outlier removal filter based on principal component analysis). We believe this method can outperform existing clustering algorithms when applied to point cloud images captured at a large distance from the lidar. Compared to traditional methods, the proposed method has higher precision and recall with an F-score up to 0.876 and complexity reduced by at least 50%.

The demand for high speed data transmission has increased rapidly, leading to advanced optical communication techniques. In the past few years, multiple equalizers based on neural network (NN) have been proposed to recover signal from nonlinear distortions. However, previous experiments mainly focused on achieving low bit error rate (BER) on certain dataset with an offline-trained NN, neglecting the generalization ability of NN-based equalizer when the properties of optical link change. The development of efficient online training scheme is urgently needed. In this article, we've proposed an adaptive online training scheme for NN-based equalizer. Our scheme can fine-tune NN parameters with only received signals rather than original symbols as reference. Once the system enters online stage, no labeled training sequence needs to be transmitted. By introducing data augmentation and virtual adversarial training, the convergence speed has been accelerated by 4.5 times, compared with decision-directed self-training. The proposed adaptive NN-based equalizer is called “AdaNN”. Its BER has been evaluated under two scenarios: a 56 Gb/s PAM4-modulated VCSEL-MMF optical link (100-m), and a 32 Gbaud 16QAM-modulated Nyquist-WDM system (960-km SSMF). In our experiments, with the help of AdaNN, BER values can be quickly stabilized below 1e-3 after receiving 10 ⁵ unlabeled symbols. AdaNN shows great performance improvement compared with non-adaptive NN and conventional MLSE.

Fiber Kerr nonlinearity limits the achievable transmission distance and channel capacity of high speed optical communication systems. In this paper, we study the performance of C-band single carrier 400G and 800G 16-ary quadrature amplitude modulation (16QAM) or 64QAM Nyquist wavelength division multiplexed (WDM) systems with emphasis on fiber nonlinearity and its mitigation. The nonlinear mitigation methods of digital back-propagation (DBP) and coupled-equations digital back-propagation (CE-DBP) are discussed for single channel and multi-channel systems, respectively. The DBP algorithm is a general approach suitable for compensating intra-channel nonlinearity. The CE-DBP is an extension of the DBP, which can deal with both intra- and inter-channel nonlinearity by solving the coupled multi-channel Manakov equations. Based on the first order perturbation theory, we also analyze the application of Kalman filter (KF) to nonlinear compensation, which utilizes correlation between symbols to compensate for Kerr nonlinearity. With numerical simulations, the performance of three nonlinearity equalization schemes of CE-DBP, DBP+KF, and CE-DBP+KF are compared. For nonlinearity mitigation in single carrier 400G and 800G WDM systems, the DBP+KF is more efficient than the CE-DBP considering 5 coupled channels. The CE-DBP+KF method has slightly better performance than the DBP+KF. However, the CE-DBP is quite complex due to multi-channel Manakov equation calculation. Furthermore, for WDM operation, joint receiving multi-channels demands high bandwidth of coherent detection. In contrast, the DBP+KF only need information of the target channel. We thus conclude that the DBP+KF is an appropriate approach to mitigate fiber nonlinearity for single carrier 400G and 800G Nyquist-WDM systems. With the combination of DBP and KF, the transmission distance extensions are achieved as 12%, 13%, 23% and 28% for 400G_64QAM, 800G_64QAM, 400G_16QAM and 800G_16QAM WDM systems, respectively.

In coherent optical orthogonal frequency-division-multiplexing systems, the laser phase noise which induces common phase error (CPE) and intercarrier interference (ICI) is a major impairment to the system's performance. In this letter, we propose an orthogonal basis expansion-based phase noise suppression algorithm which can mitigate CPE as well as ICI effectively. The proposal can significantly reduce the bit-error-rate floor caused by ICI.

Short-distance optical interconnection among servers in data centers has attracted the attentions of a multitude of researchers. The method to make the optical link with high transmission capacity and low cost becomes increasingly crucial. Considering these requirements, we propose a new equalization method for time-domain interference elimination in this letter, employed in a band-limited 850-nm vertical cavity surface-emitting laser (VCSEL) link. The method is named simplified maximum likelihood sequence estimation (MLSE), which reduces about 87.8% computational complexity in digital signal processing compared with the conventional MLSE. Based on the proposed simplified-MLSE and combined with the feed-forward equalizer (FFE), a 60-Gb/s non-return-to-zero (NRZ)-modulated VCSEL link could transmit via 56-m OM4 multimode fiber (MMF), and the bit error rate (BER) is still lower than the 7% hard-decision forward error correction (FEC) threshold. The 3-dB bandwidth of the employed 850-nm VCSEL in this transmission system is only 18 GHz, fulfilling the principle of cost-efficient optical interconnects and showing the enormous potential of our proposal in data centers.