Sea-land segmentation in remote sensing imagery is of great significance for dynamic coastline monitoring. However, single-modality remote sensing images often struggle to capture the complex structure of coastlines effectively. To address this challenge, we propose ECPNet, a multimodal sea-land segmentation network that integrates optical and SAR images. The complex and variable nature of coastlines, coupled with dynamic factor disturbances such as waves and tides, further exacerbates the inherent feature misalignment between optical and SAR modalities due to their fundamentally different imaging mechanisms. Moreover, optical images are easily affected by weather conditions, while SAR images are affected by speckle noise, both of which severely affect the confidence of modal features and limit segmentation accuracy. To overcome these challenges, we propose the edge-guided deformable alignment module (EDAM), which uses edge information as a structural prior to guide fine-grained, pixel-level adaptive alignment and boundary enhancement of shallow optical and SAR features. In addition, we introduce the triple confidence-guided fusion module (TCFM), which dynamically evaluates the reliability of features from each source and adaptively adjusts their fusion weights. This mechanism effectively suppresses the adverse effects of clouds, haze, and speckle noise. Experimental results on a dataset based on Sentinel data demonstrate that our method achieves more accurate segmentation and sharper boundary delineation compared to existing approaches, under both clear and cloud-covered conditions. The dataset is publicly available at https://huggingface.co/datasets/cuibinge/Multimodal-Sea-Land

Monitoring glacier changes is essential for understanding global climate dynamics and assessing their environmental impacts. However, accurate detection remains challenging due to seasonal variations, illumination differences, and heterogeneous textures in remote sensing imagery. To address these issues, we propose VPGCD-Net, a transformer-based dual-branch network that achieves robust glacier change detection through visual prompt engineering. The visual prompting branch integrates threshold segmentation and difference calculation, leveraging a visual prompt transformer (VPT) to encode regions of significant change and generate high-level semantic prompts. Meanwhile, the change detection branch adopts ResNet18 as the backbone to extract dual-temporal features, followed by a transformer module for modeling global spatiotemporal dependencies and a feature-wise linear modulation (FiLM) module for adaptive feature modulation to emphasize real change regions. Complementing the method, we introduce the first polar-glacier-focused dataset specifically designed for deep-learning-based glacier change detection in remote sensing. Experimental results demonstrate that VPGCD-Net outperforms existing state-of-the-art methods, achieving superior accuracy even under complex conditions such as shadow interference. The dataset is publicly available at https://huggingface.co/datasets/cuibinge/Glacier-Dataset

To remove the restriction on code length of polar codes, this paper proposes a construction scheme, called stepwise polar codes, which can generate arbitrary-length polar codes. The stepwise polar codes are generated by sub-polar codes with different code lengths. To improve coding performance, sub-polar codes are united by polarization effect priority algorithm, which can reduce the number of incompletely polarized channels. Then, the construction method of the generator matrix of the stepwise polar code is presented. Furthermore, we prove that the proposed scheme has lower decoding complexity than punctured, multi-kernel polar codes. Simulation results show that the proposed method can achieve similar decoding performance compared with the conventional punctured polar codes, rate-compatible punctured polar code, PC-short and asymmetric polar codes (APC) when code length N=48 and 72, respectively.

针对深度强化学习算法在路径规划的过程中出现与所处环境交互信息不精确、回馈稀疏、收敛不稳定等问题,在竞争网络结构的基础上,提出一种基于自调节贪婪策略与奖励设计的竞争深度 Q网络算法.智能体在探索环境时,采用基于自调节贪婪因子的ε-greedy探索方法,由学习算法的收敛程度决定探索率ε的大小,从而合理分配探索与利用的概率.根据人工势场法物理理论塑造一种势场奖励函数,在目标处设置较大的引力势场奖励值,在障碍物附近设置斥力势场奖励值,使智能体能够更快的到达终点.在二维网格环境中进行仿真实验,仿真结果表明,该算法在不同规模地图下都取得了更高的平均奖赏值和更稳定的收敛效果,路径规划成功率提高了 48.04%,验证了算法在路径规划方面的有效性和鲁棒性.同时与 Q-learning算法对比实验表明,所提算法路径规划成功率提高了 28.14%,具有更好的环境探索和路径规划能力.

The imbalance between limited training samples and extreme high spectral dimensions is a challenge for hyperspectral image classification. To significantly improve the classification of hyperspectral images in the case of small samples, we proposed a novel hyperspectral image classification method based on semantic filtering and ensemble learning. Semantic filtering synergistically combines the efficiency of the recursive filter and the effectiveness of the recent edge detector for scale-aware edge-preserving filtering, which can efficiently extract subjectively meaningful structures from natural scenes containing multiple-scale objects. Ensemble learning is used to fuse multi-scale features and to enhance the information-rich features. Experiments on three data sets prove that both semantic information and ensemble learning can improve classification accuracy, and their combination obtained the state-of-the-art performance.

The precise segmentation of multimodal MRI images is the primary stage of tumor diagnosis and treatment. Current segmentation strategies often underutilize multiscale features, which can easily lead to loss of contextual information, reduction of low-level features and noise interference. To overcome these issues, a 3D multiscale local cross-channel residual denoising network (MLRD-Net) for an MRI-based brain tumor segmentation algorithm is proposed in this paper. Specifically, we employ encoder-decoder structure to connect local and global features, and enhance the receptive field of the network. Random slice operation has been conducted to enhance robustness. Then, residual blocks with pre-activation operation are developed in down-sampling stage, which effectively improves signal propagation along the network and alleviates network overfitting. Finally, the local cross-channel denoising mechanism is established to eliminate unimportant features without dimensionality reduction. Our proposal was evaluated in Brain Tumor Segmentation 2020 dataset (BraTS 2020), obtaining significantly improved results with mean Dice Similarity Coefficient metric of 0.91, 0.79, and 0.73 for the complete, tumor core, and enhancing tumor regions respectively. Besides, we conduct further practice on BraTS 2019, with the mean Dice Similarity Coefficient metric of 0.89, 0.80, and 0.75. Massive experiments demonstrate that our method is powerful and reliable. It increases little model complexity while achieving very competitive performance.

Polar code has been selected as the control channel coding scheme for the 5th generation mobile communication technology (5G), and the performance of short polar codes is receiving intensive attention. However, belief propagation (BP) algorithm suffers from performance loss under short codes. In order to improve decoding performance, this paper proposes a BP flip decoding algorithm based on convolutional neural network (CNN-BP flip). Cyclic redundancy check (CRC) is used to check the results after BP decoding, and extracts the left messages in the factor graph of the last iteration that have not passed the CRC. The extracted messages are mapped into an image as the input of the convolutional neural network (CNN), the network model is adopted to classify the flipping bits. Compared with the critical set-based BP flip (CS-BP flip) algorithm, the CNN-BP flip algorithm achieves 0.23 dB gain at the number of flip attempts T1=5 and the block error rate (BLER)=10-2 for polar code with N=64.

Due to the phenomenon of mixed pixels in low-resolution remote sensing images, the green tide spectral features with low Enteromorpha coverage are not obvious. Super-resolution technology based on deep learning can supplement more detailed information for subsequent semantic segmentation tasks. In this paper, a novel green tide extraction method for MODIS images based on super-resolution and a deep semantic segmentation network was proposed. Inspired by the idea of transfer learning, a super-resolution model (i.e., WDSR) is first pre-trained with high spatial resolution GF1-WFV images, and then the representations learned in the GF1-WFV image domain are transferred to the MODIS image domain. The improvement of remote sensing image resolution enables us to better distinguish the green tide patches from the surrounding seawater. As a result, a deep semantic segmentation network (SRSe-Net) suitable for large-scale green tide information extraction is proposed. The SRSe-Net introduced the dense connection mechanism on the basis of U-Net and replaces the convolution operations with dense blocks, which effectively obtained the detailed green tide boundary information by strengthening the propagation and reusing features. In addition, the SRSe-Net reducs the pooling layer and adds a bridge module in the final stage of the encoder. The experimental results show that a SRSe-Net can obtain more accurate segmentation results with fewer network parameters.

Considering the high decoding latency of polar code, an early stopping criterion for belief propagation is presented, which terminates the decoding by monitoring the convergence of codeword estimate (x) over cap. In this paper, Gaussian approximation is used to analyze and select Q bit with low error probability to construct the comparison space. Because the number of bit to be compared is small and only XOR and OR operation is used, the computational complexity is low. Different from other criteria based on (u) over cap, the proposed criterion does not lead to additional latency for it has been completed before calculating (u) over cap. Simulation and FPGA Synthesis results show that compared with G-matrix, Worst Information Bit (WIB) and Frozen Bit Error Rate (FBER), this criterion can effectively save hardware resource. When the maximum iteration number is set to 40, compared with the G-matrix criterion, the average iteration time is increased by 29.98% at 3.5 dB, and the average iteration times are reduced by 39.44% and 27.67% respectively compared with the WIB and FBER schemes.

Polar code has been adopted as the control channel coding scheme for the fifth generation (5G), and the performance of short polar codes is receiving intensive attention. The successive cancellation flipping (SC flipping) algorithm suffers a significant performance loss in short block lengths. To address this issue, we propose a double long short-term memory (DLSTM) neural network to locate the first error bit. To enhance the prediction accuracy of the DLSTM network, all frozen bits are clipped in the output layer. Then, Gaussian approximation is applied to measure the channel reliability and rank the flipping set to choose the least reliable position for multi-bit flipping. To be robust under different codewords, padding and masking strategies aid the network architecture to be compatible with multiple block lengths. Numerical results indicate that the error-correction performance of the proposed algorithm is competitive with that of the CA-SCL algorithm. It has better performance than the machine learning-based multi-bit flipping SC (ML-MSCF) decoder and the dynamic SC flipping (DSCF) decoder for short polar codes.

极化码是一种被严格证明到达信道容量的信道编码方法.虽然串行抵消列表比特翻转(SCLF)译码算法可提高译码性能,但导致较高的译码复杂度.为降低译码复杂度,提出一种分段CRC辅助串行抵消列表比特翻转极化码译码算法.该算法在码字构造过程中,通过添加分段CRC校验,可提前终止翻转译码过程.在中短码长下,可显著降低极化码比特翻转译码复杂度.仿真结果表明,当L=8,Eb N0=1.5 dB时,与SCLF方法翻转2 bit译码算法相比,提出方法的译码复杂度可降低71.9％,同时获得较好的性能增益.

In order to achieve fast decoding and improve the throughput, this paper uses the polar code to encode the spin transfer torque MARAM (STT-MRAM) channel. Based on the Fast-SSC algorithm, a (256, 220) hardware architecture is designed, including the controller, processing element, Kronecker product and memory module. This paper reduces the complexity of data process by splitting the data of nodes, and reduces the memory bandwidth by increasing the reusability of data. The decoder is synthesized on Stratix V 5SGXEA7N2F45C2, the decoding latency is 0.68us, and it can achieve 375 Mbps at 167 MHz.

In this Letter, a new scheduling scheme for belief propagation (BP) is proposed to reduce the decoding latency. All the stages update messages at the same time, it improves the reliability and accelerates the decoding convergence. The authors also remove the last stage of factor graph to reduce computational complexity. Simulations show that the average decoding latency of a (1024,512) polar code is reduced to 29.4 clocks.

In order to strengthen protection of personal safety and important places, this paper designs a pedestrian crossing the cordon detection system to process the image information captured by the camera intelligently in real time. Firstly, Gaussian mixture model is established to detect pedestrians. The detected target image is processed and initialized. Then target information is transferred to the Camshift algorithm for tracking, which solves the problem of manually selecting tracking targets. The Camshift algorithm combined with the Kalman filter which has update and prediction functions can overcome tracking failure due to interference and occlusion especially in the dark environment. On this basis, system can achieve crossing the cordon warning and information reporting to ensure the safety of people and important places. This system can be used in mines, culverts, tunnels and other dangerous and dark areas for safety protection, and has great practical application value.

目前,我国教室的管理方式在智能化、专业化、现代化上略显不足,因此,本文从需求入手,以青岛西海岸新区的职业教育与高等教育学校为研究对象,总结并针对当前高校智慧教室建设和管理方面面临的关键问题,研究开发了一套基于物联网应用下的智慧教室管控系统.该系统设计并实现了教室端硬件管控平板,并开发了基于Android的平板端应用与基于MVC框架的教室管理平台,连接校内智慧校园、师生办公系统及校内教务系统,搭建了交互良好的管控平台,解决传统B/S架构下的移动端操作不便问题,将教室管理、人员考勤、信息展示和环境控制等功能融为一体.实现了将教室的多种功能进行集中管控,推进智慧校园建设进一步发展.

In this paper, a puncturing algorithm for mixing 2-kernel and 3-kernel polar codes is presented. The puncturing sequence is generated based on the capacity of channels and the upper bound of minimum block error probability for successive cancellation (SC) decoding. We use the capacity-zero puncturing model, the decoding algorithm of mother codes can still be adopted. An improved greedy algorithm of computing the maximization of the minimum distance is proposed to select the information set. The maximum number of punctured bits is limited to [1,2^n-2) when the length of subcodes M∈ (2^n-2*3,2^n) . Simulation results show that the block error rate based on the mixing kernels is better than that based on 2-kernel.

Rapid development in the field of internet of things (IoT) has increased numbers of applications. Meanwhile, security and privacy threats are also introduced. Various authentication protocols are devised to resist the malicious attacks. Li et al. proposed a remote user authentication protocol using smart cards and they claimed their protocol was secure. However, we find that it cannot resist DoS attack, stolen-verifier attack and replay attack. Then we propose a three-factor remote authentication protocol using smart card based on biometric. The proposed protocol can resist DoS attack effectively by increasing local verification of user identity and password.

The decoding performance of successive cancellation (SC) can be improved by correcting the error bits caused by channel noises. In order to identify and correct those error bits, we propose an improved progressive bit-flipping (IPBF) decoding algorithm. First, a more efficient method of constructing the set of error bit positions (EBPs) is proposed. It can significantly reduce the size of the set of EBPs, especially at medium to high signal to noise ratio (SNR). Second, a pruning technique of reducing the number of error bit-flipping combinations (EBFCs) is presented. In the case of ensuring that the decoding performance of PBF algorithm is not degraded, the decoding complexity is reduced as much as possible with the help of the pruning scheme. From the simulation results, compared to the conventional PBF algorithm, the proposed IPBF algorithm not only significantly reduces the average decoding complexity, but also has a slight improvement in decoding performance. At 2.0dB, the decoding complexity can be reduced by 63.6%, 87.5%, and 97.6% when ω is equal to 2, 3 and 4, respectively.

Polar code has been proven to achieve the symmetric capacity of memoryless channels. However, the successive cancellation decoding algorithm is inherent serial in nature, which will lead to high latency and low throughput. In order to obtain high throughput, we design a deeply pipelined polar decoder and optimize the processing elements and storage structure. We also propose an improved fixed-point nonuniform quantization scheme, and it is close to the floating-point performance. Two-level control strategy is presented to simplify the controller. In addition, we adopt FIFO structure to implement the α_memory and β_memory and propose the 348-stage pipeline decoder.

The age of Internet of things gives rise to more challenges to various secure demands when designing the protocols, such as object identification and tracking, and privacy control. In many of the current protocols, a malicious server may cheat users as if it was a legal server, making it vital to verify the legality of both users and servers with the help of a trusted third-party, such as a registration center. Li et al. proposed an authentication protocol based on dynamic identity for multi-server environment, which is still susceptible to password-guessing attack, eavesdropping attack, masquerade attack, and insider attack etc. Besides, their protocol does not provide the anonymity of users, which is an essential request to protect users’ privacy. In this article, we present an improved authentication protocol, depending on the registration center in multi-server environments to remedy these security flaws. Different from the previous protocols, registration center in our proposed protocol is one of parties in authentication phase to verify the legality of the users and the servers, thus can effectively avoid the server spoofing attack. Our protocol only uses nonce, exclusive-OR operation, and one-way hash function in its implementation. Formal analysis has been performed using the Burrows–Abadi–Needham logic to show its security.