Long video delivery over wireless links whose SNR (Signal-to-Noise Ratio) falls below 5 dB often hits a cliff: conventional codecs combined with FEC remain decodable, but the decoded frames collapse into blocky mosaics that are semantically unusable. This paper presents AdaGenCom (Adaptive Generative Communication), a multi-modal semantic communication framework that preserves meaning when pixel recovery becomes infeasible. AdaGenCom decomposes a source video into (i) structural semantics as sparse visual anchors (motion-aware keyframes and optional flow cues), (ii) textual semantics as compact scene descriptions, and (iii) conditional semantics as lightweight channel-state metadata. Each stream is transmitted with learned joint source-channel coding to enable graceful degradation. At the receiver, a pre-trained video diffusion model (LTX-Video) reconstructs the full sequence while a Channel-Aware Parameter Mapping (CAPM) mechanism adapts diffusion sampling—denoising steps, guidance scale, and sampling stochasticity—to the estimated channel quality without retraining. On 400+ frame sequences from UVG, AdaGenCom maintains CLIP similarity above 0.96 at SNRs where H.264/H.265 outputs become unrecognizable (CLIP below 0.80). These results indicate that generative priors with channel-aware inference provide a practical path for long-video delivery under extreme wireless conditions.

In response to the rapid growth of global video traffic and the limitations of traditional wireless transmission systems, we propose a novel dual-stage vector quantization system, VQ-DeepVSC, tailored to enhance video transmission over wireless channels. In the first stage, we design the adaptive key-frame extractor and interpolator, deployed respectively at the transmitter and receiver, which intelligently select key frames to minimize inter-frame redundancy and mitigate thecliff-effectunder challenging channel conditions. In the second stage, we propose the semantic vector quantization encoder and decoder, placed respectively at the transmitter and receiver, which efficiently compress key frames using advanced indexing and spatial normalization modules to reduce redundancy. Additionally, we propose adjustable index selection and recovery modules, enhancing compression efficiency and enabling flexible compression ratio adjustment. Compared to the existing methods based on joint source-channel coding (JSCC) framework, the proposed method exhibits superior compatibility with current digital communication systems. Experimental results demonstrate that VQ-DeepVSC achieves substantial improvements in both Multi-Scale Structural Similarity (MS-SSIM) and Learned Perceptual Image Patch Similarity (LPIPS) metrics than the H.265 standard, particularly under low channel signal-to-noise ratio (SNR) or multi-path channels, highlighting the significantly enhanced transmission capabilities of our approach.

Interpretable deep-learning models have received widespread attention in the field of image recognition. However, owing to the coexistence of medical-image categories and the challenge of identifying subtle decision-making regions, many proposed interpretable deep-learning models suffer from insufficient accuracy and interpretability in diagnosing images of medical diseases. Therefore, this study proposed a feature-driven inference network (FeaInfNet) that incorporates a feature-based network reasoning structure. Specifically, local feature masks (LFM) were developed to extract feature vectors, thereby providing global information for these vectors and enhancing the expressive ability of FeaInfNet. Second, FeaInfNet compares the similarity of the feature vector corresponding to each subregion image patch with the disease and normal prototype templates that may appear in the region. It then combines the comparison of each subregion when making the final diagnosis. This strategy simulates the diagnosis process of doctors, making the model interpretable during the reasoning process, while avoiding misleading results caused by the participation of normal areas during reasoning. Finally, we proposed adaptive dynamic masks (Adaptive-DM) to interpret feature vectors and prototypes into human-understandable image patches to provide an accurate visual interpretation. Extensive experiments on multiple publicly available medical datasets, including RSNA, iChallenge-PM, COVID-19, ChinaCXRSet, MontgomerySet, and CBIS-DDSM, demonstrated that our method achieves state-of-the-art classification accuracy and interpretability compared with baseline methods in the diagnosis of medical images. Additional ablation studies were performed to verify the effectiveness of each component.

Current methodologies in distributed source coding have predominantly investigated decoder-focused strategies, emphasizing the alignment and exploitation of side information. This study introduces a paradigm shift by presenting an encoder-centric algorithm that conducts proactive optimization in the frequency domain. This shift is motivated by the current deep learning models' tendency to passively extract high-frequency elements, such as contours and content in the spatial domain at the encoder side, without considering the frequency characteristics of these spatial components. Unlike current trends, the proposed scheme actively selects the essential frequency components directly in the frequency domain by introducing an adaptive self-learning filter, enabling the encoder to discern and retain critical frequency components effectively and precisely. Furthermore, we align the side information in the spatial domain before feature extraction and implement an affine transformation-based alignment strategy to utilize the side information better. By leveraging the shared frequency domain components of the image pairs, the proposed algorithm adeptly learns affine coefficients to accomplish precise spatial alignment. This dual strategy of proactive encoder optimization and decoder alignment via affine transformations is highly efficient, outperforming existing state-of-the-art methods in distributed source coding when tested across two diverse datasets by an average of 0.5 dB in PSNR.

Accuracy and diversity represent two critical quantifiable performance metrics in the generation of natural and semantically accurate captions. While efforts are made to enhance one of them, the other suffers due to the inherent conflicting and complex relationship between them. In this study, we demonstrate that the suboptimal accuracy levels derived from human annotations are unsuitable for machine-generated captions. To boost diversity while maintaining high accuracy, we propose an innovative variational transformer (VaT) framework. By integrating "invisible information prior (IIP)" and "auto-selectable Gaussian mixture model (AGMM)," we enable its encoder to learn precise linguistic information and object relationships in various scenes, thus ensuring high accuracy. By incorporating the "range-median reward (RMR)" baseline into it, we preserve a wider range of candidates with higher rewards during the reinforcement-learning-based training process, thereby guaranteeing outstanding diversity. Experimental results indicate that our method achieves simultaneous improvements in accuracy and diversity by up to 1.1% and 4.8%, respectively, over the state-of-the-art. Furthermore, our approach demonstrates its performance that is the closest to human annotations in semantic retrieval, with its score of 50.3 versus the human score of 50.6. Thus, the method can be readily put into industrial use.

With the widespread application of deep learning technology in medical image analysis, the effective explanation of model predictions and improvement of diagnostic accuracy have become urgent problems that need to be solved. Attribution methods have become key tools to help doctors better understand the diagnostic basis of models, and are used to explain and localize diseases in medical images. However, previous methods suffer from inaccurate and incomplete localization problems for fundus diseases with complex and diverse structures. To solve these problems, we propose a weakly supervised interpretable fundus disease localization method called hierarchical salient patch identification (HSPI) that can achieve interpretable disease localization using only image-level labels and a neural network classifier (NNC). First, we propose salient patch identification (SPI), which divides the image into several patches and optimizes consistency loss to identify which patch in the input image is most important for the network’s prediction, in order to locate the disease. Second, we propose a hierarchical identification strategy to force SPI to analyze the importance of different areas to neural network classifier’s prediction to comprehensively locate disease areas. Conditional peak focusing is then introduced to ensure that the mask vector can accurately locate the disease area. Finally, we propose patch selection based on multi-sized intersections to filter out incorrectly or additionally identified non-disease regions. We conduct disease localization experiments on fundus image datasets and achieve the best performance on multiple evaluation metrics compared to previous interpretable attribution methods. Additional ablation studies are conducted to verify the effectiveness of each method.

Traditional TCP has been the dominant protocol for internet traffic after years of development in both academia and industry. However, the emergence of live video streaming applications and the increasing demand for low-latency video transmission, particularly in the context of sports and game live streaming, has posed a challenge to traditional TCP congestion control algorithms. As a representative TCP algorithm, BBR has been widely used in industry. However, BBR tends to inject more packets than the actual bottleneck in the dynamic network resulting in high latency. Because the maximum bandwidth in the past period is used to calculate the congestion window (CWND). To address this issue, we develop a recursive least squares (RLS) model to predict future bandwidth based on past bandwidth samples and update BBR's CWND periodically. To overcome the difficulty of modifying the kernel congestion control algorithm, we use extended Berkeley Packet Filter (eBPF) technology to rewrite TCP BBR and use eBPF MAP to exchange data between the kernel space and the data space. Experiments show that our algorithm can effectively reduce latency for live streaming in the dynamic network. Compared with CUBIC, our algorithm can achieve 76.9% average latency reduction with 2.3% average throughput loss and bring 42.4% average latency reduction with 1.9% average throughput loss compared with BBR.

In this paper, we propose a puncturing-based joint source channel coding scheme with a hierarchical down-sampling layer (PJSCC). The proposed hierarchical down-sampling layer fully exploits both frequency and spatial priors. Moreover, to achieve adaptive compression ratio control, PJSCC utilizes a shared puncturing table as a global prior shared between the sender and receiver. This puncturing table plays a vital role in selectively pruning or padding symbols, and the adaptation of the compression ratio is achieved through the manual configuration of hyperparameters to adjust the puncturing rate. Experimental results show that the proposed scheme achieves superior reconstruction performance across several classic datasets.

Current point cloud compression methods based on deep learning cannot guarantee that the reconstructed points are constrained to the surface, resulting in low reconstruction quality at low bitrates. Hence, this paper proposes an efficient deep learning-based point cloud geometry compression algorithm. Specifically, by introducing a two-dimensional plane at the decoder, the reconstructed local patch is constrained within a manifold, preserving sufficient surface features. This strategy ensures the decoder can reconstruct high-quality point clouds even at low bitrates. Moreover, we use the anchor features obtained by the neural network to compress the local features at the encoder. The experimental results show that, under the condition of the same restoration quality, the proposed method improves the point-to-plane PSNR by more than 2dB compared to the state-of-the-art methods. The code is available at https://github.com/zbaoye/SurfPCC.

Despite the remarkable accomplishments of deep neural networks in computer vision tasks, the inherent opacity of their operations remains a pressing concern. Attribution methods generating visual explanatory maps representing the importance of image pixels for model classification are popular for explaining neural network decisions. However, the small and diverse decision regions in fine-grained or medical images limit the precision and comprehensiveness of the existing attribution methods when explaining decisions made for such a data type. This paper introduces a novel attribution method called hierarchical dynamic masks (HDM) to overcome these concerns to generate saliency maps with high recognition reliability and localization capability. Specifically, we suggest dynamic masks (DM), which enable multiple small-sized benchmark mask vectors to learn the image's critical information roughly through an optimization method. The benchmark mask vectors guide the learning of the large-sized combination mask vectors so that their overlay mask accurately learns detailed pixel importance information. Additionally, we construct the HDM by hierarchically concatenating DM modules. These DM modules search and combine the regions of interest in the remaining neural network classification decisions within the masked image in a learning-based way. Since HDM forces DM to perform importance analysis in different areas, it makes the fused saliency map more comprehensive. The experiments reveal that the proposed method outperforms existing approaches significantly regarding recognition credibility and positioning ability when qualitatively and quantitatively tested on CUB-200-2011 and iChallenge-PM datasets.

The quest for enhancing the interpretability of neural networks has become a prominent focus in recent research endeavors. Prototype-based neural networks have emerged as a promising avenue for imbuing models with interpretability by gauging the similarity between image components and category prototypes to inform decision-making. However, these networks face challenges as they share similarity activations during both the inference and explanation processes, creating a tradeoff between accuracy and interpretability. To address this issue and ensure that a network achieves high accuracy and robust interpretability in the classification process, this article introduces a groundbreaking prototype-based neural network termed the “Decoupling Prototypical Network” (DProtoNet). This novel architecture comprises encoder, inference, and interpretation modules. In the encoder module, we introduce decoupling feature masks to facilitate the generation of feature vectors and prototypes, enhancing the generalization capabilities of the model. The inference module leverages these feature vectors and prototypes to make predictions based on similarity comparisons, thereby preserving an interpretable inference structure. Meanwhile, the interpretation module advances the field by presenting a novel approach: a “multiple dynamic masks decoder” that replaces conventional upsampling similarity activations. This decoder operates by perturbing images with mask vectors of varying sizes and learning saliency maps through consistent activation. This methodology offers a precise and innovative means of interpreting prototype-based networks. DProtoNet effectively separates the inference and explanation components within prototype-based networks. By eliminating the constraints imposed by shared similarity activations during the inference and explanation phases, our approach concurrently elevates accuracy and interpretability. Experimental evaluations on diverse public natural datasets, including CUB-200-2011, Stanford Cars, and medical datasets like RSNA and iChallenge-PM, corroborate the substantial enhancements achieved by our method compared to previous state-of-the-art approaches. Furthermore, ablation studies are conducted to provide additional evidence of the effectiveness of our proposed components.

In intelligent reflecting surface (IRS) assisted communication, beam search is usually time-consuming as the multiple-input multiple-output (MIMO) of IRS is usually very large. The hierarchical codebook is a widely accepted method for reducing the complexity of searching time. The performance of this method strongly depends on the design scheme of beamforming of different beamwidths. In this paper, a non-constant phase difference (NCPD) beamforming algorithm is proposed. To implement the NCPD algorithm, we first model the phase shift of IRS as a continuous function and then determine the parameters of the continuous function through the analysis of its array factor. Then, we propose a hierarchical codebook and two beam training schemes, namely the joint searching (JS) scheme and direction-wise searching (DWS) scheme by using the NCPD algorithm which can flexibly change the width, direction, and shape of the beam formed by the IRS array. Numerical results show that the NCPD algorithm is more accurate with smaller side lobes, and also more stable on IRS of different sizes compared to other wide beam algorithms. The misalignment rate of the beam formed by the NCPD method is significantly reduced. The time complexity of the NCPD algorithm is constant, thus making it more suitable for solving the beamforming design problem with practically large IRS.

Typical adversarial-training-based unsupervised domain adaptation (UDA) methods are vulnerable when the source and target datasets are highly complex or exhibit a large discrepancy between their data distributions. Recently, several Lipschitz-constraint-based methods have been explored. The satisfaction of Lipschitz continuity guarantees a remarkable performance on a target domain. However, they lack a mathematical analysis of why a Lipschitz constraint is beneficial to UDA and usually perform poorly on large-scale datasets. In this article, we take the principle of utilizing a Lipschitz constraint further by discussing how it affects the error bound of UDA. A connection between them is built, and an illustration of how Lipschitzness reduces the error bound is presented. A local smooth discrepancy is defined to measure the Lipschitzness of a target distribution in a pointwise way. When constructing a deep end-to-end model, to ensure the effectiveness and stability of UDA, three critical factors are considered in our proposed optimization strategy, i.e., the sample amount of a target domain, dimension, and batchsize of samples. Experimental results demonstrate that our model performs well on several standard benchmarks. Our ablation study shows that the sample amount of a target domain, the dimension, and batchsize of samples, indeed, greatly impact Lipschitz-constraint-based methods' ability to handle large-scale datasets. Code is available at https://github.com/CuthbertCai/SRDA.

This paper proposes a novel scheme for mitigating strong interferences, which is applicable to various wireless scenarios, including full-duplex wireless communications and uncoordinated heterogeneous networks. As strong interferences can saturate the receiver’s analog-to-digital converters (ADC), they need to be mitigated both before and after the ADCs, i.e., via hybrid processing. The key idea of the proposed scheme, namely the Hybrid Interference Mitigation using Analog Prewhitening (HIMAP), is to insert an$M$-input$M$-output analog phase shifter network (PSN) between the receive antennas and the ADCs to spatially prewhiten the interferences, which requires no signal information but only an estimate of the covariance matrix. After interference mitigation by the PSN prewhitener, the preamble can be synchronized, the signal channel response can be estimated, and thus a minimum mean squared error (MMSE) beamformer can be applied in the digital domain to further mitigate the residual interferences. The simulation results verify that the HIMAP scheme can suppress interferences 80dB stronger than the signal by using off-the-shelf phase shifters (PS) of 6-bit resolution.

Satellite communication has the potential to play a key role in many applications of Internet of Things (IoT). In this paper, we consider a satellite-based IoT and investigate the technology that can improve the spectral efficiency. In general, one beam in satellite systems serves one user. To serve multiple users, time division multiplexing or frequency division multiplexing is usually used. In this paper, we propose a novel forward-link multiplexed scheme, by which the signals of different users can be transmitted simultaneously using the same frequency band. Specifically, at the transmitter, we first map each combination of the users' constellation points to a higher-order constellation point, which is referred to as constellation coding, and then transmit such higher-order modulation signals. At the user side, after receiving and detecting the transmitted signal, each user obtain its own signal by the corresponding demapping, which is referred to as constellation decoding. The total system capacity over an additive white Gaussian noise channel is analyzed in this paper. Simulation results demonstrate that the proposed scheme can greatly improve the spectral efficiency.

For frequency-division duplexing (FDD) massive multiple-input multiple-output (MIMO) systems, it is challenging for the base station (BS) to acquire downlink channel state information (CSI) since the number of channel parameters is proportional to the number of BS antennas. To reduce the downlink training and the uplink feedback overhead for orthogonal frequency-division multiplexing (OFDM)-based FDD massive MIMO systems, we propose an effective and practical downlink channel estimation approach that does not require statistical information of the channels. In the proposed method, we parameterize the wideband MIMO channel by a limited number of distinct paths, each characterized by path delay, path angle and path gain. Based on this parametric channel model, we first estimate the reciprocal channel properties, i.e., the number of paths, path delays and path angles at the BS in uplink, and then estimate the non-reciprocal channel properties, i.e., the path gains at the user-side in downlink. A compressive sensing (CS)-based algorithm is proposed to obtain the estimates of the path number and the path delays, and based on these estimates, a low- complexity algorithm is presented to estimate the path angles. Simulation results demonstrate that the proposed method can provide reliable channel estimation for FDD massive MIMO OFDM systems.

Modeling and learning of brain activity patterns represent a huge challenge to the brain-computer interface (BCI) based on electroencephalography (EEG). Many existing methods estimate the uncorrelated instantaneous demixing of EEG signals to classify multiclass motor imagery (MI). However, the condition of uncorrelation does not hold true in practice, because the brain regions work with partial or complete collaboration. This work proposes a novel method, termed as a common Bayesian network (CBN), to discriminate multiclass MI EEG signals. First, with the constraints of a Gaussian mixture model on every channel, only related channels are selected to construct a normal Bayesian network. Second, the nodes that have both common and varying edges are selected to construct a CBN. Third, the probabilities on common edges are used to learn about the support vector machine for classification. To validate the proposed method, we conduct experiments on two well-known BCI datasets and perform a numerical analysis of the propose algorithm for EEG classification in a multiclass MI BCI. Experimental results show that the proposed CBN method not only has excellent classification performance, but also is highly efficient. Hence, it is suitable for the cases where a system is required to respond within a second.

In multicell multiuser massive multi-input multi-output (MIMO) systems, pilot contamination degrades the uplink (UL) channel estimation performance. To mitigate the effect of pilot contamination, we propose a semiblind channel estimation method that does not require cell cooperation or statistical information of the channels. In the proposed method, we first sequentially estimate the UL data from different users in the target cell. To do that, for each user, we solve a constrained minimization problem to obtain an extracting vector and then use it to extract the desired data source from the observed mixture signal. An efficient algorithm is presented to solve the optimization problem. After the ambiguities in the extracted source are corrected with the aid of the pilot sequence, the estimates of the user UL data can be obtained. Based on the demodulated UL data of all users in the target cell, we finally obtain the least squares (LS) estimate of the channel. The pilot contamination effect is shown to be reduced as the UL data length grows. Simulation results demonstrate that the proposed method significantly outperforms some existing channel estimation methods that do not require cell cooperation or channel statistics.

In orthogonal frequency-division multiplexing (OFDM)-based cognitive radio (CR) systems, the subcarriers already occupied by the primary users cannot be used by the secondary users. This leads to possibly non-contiguous positions of the available subcarriers for the secondary users. The conventional pilot design methods are no longer effective for such systems. In this paper, we propose a new practical pilot design method for OFDM-based CR systems. We first formulate the pilot design as a new optimization problem. Instead of minimizing the mean-square error (MSE) of the least-squares (LS) channel estimator, we minimize an upper bound which is related to this MSE. We then propose an efficient scheme to solve the optimization problem. Specifically, the pilot indices are obtained sequentially by solving a series of one-dimensional optimization problems of significantly lower complexity. The computational complexity of the proposed scheme is low since it only involves real additions. Simulation results show that the pilot index sequences obtained by the proposed method exhibit significantly better performance than those obtained by existing pilot design methods.

In orthogonal frequency division multiplexing (OFDM)-based cognitive radio (CR) systems, the subcarriers used by the licensed users (LUs) have to be deactivated to avoid interference. Thus only part of subcarriers can be used for transmission and the positions of the activated subcarriers may be non-contiguous. The conventional pilot design methods are no longer effective for such systems. In this paper, we propose a new practical pilot design method for OFDM-based CR systems. We first formulate the pilot design as a new optimization problem, where a simple objective function related to the mean-square error (MSE) of the least-squares (LS) channel estimation method is minimized. We then propose an efficient scheme to solve the optimization problem. Specifically, the pilot tones are obtained sequentially by solving some one-dimensional optimization problems. Only real additions are needed in the proposed scheme. The simulation results show that the pilot sequence obtained by the proposed method exhibits better performance than those obtained by existing pilot design methods.

Channel estimation for OFDM systems in rapidly time-varying environments is challenging. In this paper, relying on a basis expansion channel model, we propose a scheme for estimating channel parameters varying within a transmission block. Along with the estimation scheme, we also derive the optimal pilot sequence and optimal placement of pilot tones with respect to the mean square error (MSE) of the channel estimate. It is shown that the optimal pilot sequence consists of some adjacent equipowered and equispaced subsequences that are constrained by certain phase conditions. Simulation results demonstrate the performance of the proposed scheme in rapidly time-varying scenarios.

In orthogonal frequency division multiplexing (OFDM) systems, some subcarriers at the borders of the allocated bandwidth are usually used as guard band. Since these subcarriers, which are often referred to as virtual subcarriers, are not used for transmission, approach of conventional uniformly placed pilot tones is not applicable any more in some situations. Therefore, it is necessary to derive the optimal pilot sequences based on the nonuniform placement of pilot tones. In this paper, we first compute the mean square error (MSE) of the least squares (LS) channel estimate for multiple-input multiple-output (MIMO) OFDM systems. Then, based on nonuniform pilot tone placement, we derive the optimal pilot sequences with respect to this MSE. Simulation results demonstrate the effectiveness of the proposed approach