The joint detection uses Kalman filtering (KF) to estimate the prior probability of control outputs to assist channel decoding. In this paper, we regard the joint detection as maximum a posteriori (MAP) decoding and derive the lower and upper bounds based on the pairwise error probability considering system interference, quantization interval, and weight distribution. We first derive the limiting bounds as the signal-to-noise ratio (SNR) goes to infinity and the system interference goes to zero. Then, we construct an infinite-state Markov chain to describe the consecutive packet losses of the control systems to derive the MAP bounds. Finally, the MAP bounds are approximated as the bounds of the transition probability from the state with no packet loss to the state with consecutive single packet loss. The simulation results show that the MAP performance of $\left(64,16\right)$ polar code and 16-bit CRC coincides with the limiting upper bound as the SNR increases and has $3.0$dB performance gain compared with the normal approximation of the finite block rate at block error rate $10^{-3}$.

This paper proposes a unified analytical framework for reconfigurable intelligent surface (RIS)-assisted two-way full-duplex (TW-FD) communication systems in 6G Internet of Things (IoT) scenarios. The proposed framework specifically addresses RISs with N reflective elements, facilitating efficient bidirectional communication. A novel unified moment-based analytical approach is developed, accommodating diverse fading models including Rayleigh, Nakagami-n, Weibull, Nakagami-m, and κ-μ, thereby significantly enhancing the versatility and practicality for complex 6G IoT environments. To comprehensively validate the applicability of our analysis, both independently identically distributed (i.i.d.) and independently non-identically distributed (i.n.i.d.) fading channel scenarios are investigated. By employing the Edgeworth expansion method, we derive analytical expressions for the probability density function (PDF) and cumulative distribution function (CDF) of the end-to-end signal-to-interference-plus-noise ratio (SINR). Additionally, closed-form expressions for probability, average symbol error rate (SER) for various modulation schemes, and end-to-end ergodic rate are provided. Monte Carlo simulations demonstrate the accuracy and robustness of the theoretical models proposed. The results presented in this work not only contribute substantially to the analytical methodologies for RIS-assisted communication but also offer practical guidance for the design and optimization of future 6G IoT systems.

As envisioned in the sixth-generation (6G) networks, connectivity density of users and/or machines is expected to surge in future Internet of Everything. However, most current unsourced multiple access schemes experience a swift decline in performance as the number of active users increases due to higher level of interference, limited symbol representation space and available spectrum resources. To tackle these, we propose in this paper a generalized finite-field unsourced multiple access (FFUMA) with either explicit or implicit pilots. The first several information bits in each user message are used to select the spreading pattern, and the remaining information bits are encoded and spread over the finite field. For the explicitly-piloted transmission, the first few bits are mapped to a low-correlated or compressed-sensing pilot which is then transmitted through the channel. At the receiver, a two-stage algorithm consisting of pilot detection and finite-field data decoding is proposed. For the implicitly-piloted transmission, the first few bits are not physically transmitted, but they are in fact delivered to the receiver once the spreading pattern is correctly recognized. We establish the algorithm for blindly detecting the active finite-field spreading patterns and set forth the joint iterative data detection and decoding over the finite field. Error analyses of the proposed finite-field UMA, including error event decomposition and extrinsic information transfer (EXIT) analysis, are provided to evaluate the performance. The advantages of the proposed FFUMA have been demonstrated by the simulation results in terms of error rates and energy-efficiency.

Real-time speech communication over wireless networks remains challenging, as conventional channel protection mechanisms cannot effectively counter packet loss under stringent bandwidth and latency constraints. Semantic communication has emerged as a promising paradigm for enhancing the robustness of speech transmission by means of joint source channel coding (JSCC). However, its cross-layer design hinders practical deployment due to the incompatibility with existing digital communication systems. To address this, we perform JSCC over the network layer to combat packet loss and support real deployment. Inspired by the generative latent modeling, we propose Glaris, a generative latent-prior-based resilient speech semantic communication framework that performs resilient transform coding in the generative latent space. Generative latent priors enable high-quality packet loss concealment (PLC) at the receiver side, well-balancing semantic consistency and reconstruction fidelity. Additionally, an integrated error resilience mechanism is designed to mitigate the error propagation and improve the effectiveness of PLC. Compared with traditional packet-level forward error correction (FEC) strategies, our new method achieves enhanced robustness over dynamic wireless networks while reducing redundancy overhead significantly. Experimental results on the LibriSpeech dataset demonstrate that Glaris consistently outperforms existing error-resilient codecs, achieving JSCC-level robustness while maintaining seamless compatibility with existing systems, and it also strikes a favorable balance between transmission efficiency and error resilience.

Representing and rendering dynamic scenes with complex motions remains challenging in computer vision and graphics. Recent dynamic view synthesis methods achieve high-quality rendering but often produce physically implausible motions. We introduce NeHaD, a neural deformation field for dynamic Gaussian Splatting governed by Hamiltonian mechanics. Our key observation is that existing methods using MLPs to predict deformation fields introduce inevitable biases, resulting in unnatural dynamics. By incorporating physics priors, we achieve robust and realistic dynamic scene rendering. Hamiltonian mechanics provides an ideal framework for modeling Gaussian deformation fields due to their shared phase-space structure, where primitives evolve along energy-conserving trajectories. We employ Hamiltonian neural networks to implicitly learn underlying physical laws governing deformation. Meanwhile, we introduce Boltzmann equilibrium decomposition, an energy-aware mechanism that adaptively separates static and dynamic Gaussians based on their spatial-temporal energy states for flexible rendering. To handle real-world dissipation, we employ second-order symplectic integration and local rigidity regularization as physics-informed constraints for robust dynamics modeling. Additionally, we extend NeHaD to adaptive streaming through scale-aware mipmapping and progressive optimization. Extensive experiments demonstrate that NeHaD achieves physically plausible results with a rendering quality-efficiency trade-off. To our knowledge, this is the first exploration leveraging Hamiltonian mechanics for neural Gaussian deformation, enabling physically realistic dynamic scene rendering with streaming capabilities.

Semantic communications (SemCom) is considered one of the key technologies for next-generation communications. However, most SemCom systems utilize Deep Learning (DL) based joint source-channel coding (JSCC), which are incompatible with existing digital communication systems. In this paper, we propose a novel digital SemCom system based on variable product quantization (VPQ-SemCom), which harnesses multiple lightweight codebooks to represent images and dynamically optimize bitrates according to the entropy of semantic features to adapt to transmission scenarios with multiple bandwidths and SNRs. Specifically, product quantization (PQ), which can represent semantic features with several lightweight codebooks, is introduced to provide powerful representation capacities of semantic features. Furthermore, a rate adaption module, which can flexibly adjust feature length based on the entropy of semantic features, is proposed to integrate with PQ to improve rate-distortion performance. The experimental results demonstrate that VPQ-SemCom shows 32.4% improvement at high SNRs and 62.2% improvement at SNR = 2dB in Learned Perceptual Image Patch Similarity (LPIPS) compared to current state-of-the-art vector quantization (VQ) based digital SemCom systems.

We propose “MaskDSC”, a novel system designed to facilitate robust visual data transmission over unreliable wireless channels. MaskDSC effectively balances compression efficiency and transmission resilience by leveraging contextual modeling within the semantic latent space, complemented by unequal error protection mechanism at the physical layer, ensuring compatibility with existing digital communication systems. The novelty of our approach lies in a dual-functional masked Transformer architecture that exploits causal-order contextual dependencies among visual tokens. This architecture not only enhances compression efficiency through improved contextual entropy modeling but also provides robust error concealment capabilities to address diverse transmission error patterns inherent in volatile wireless channels. Our experimental evaluations conducted on image datasets demonstrate that MaskDSC outperforms state-of-the-art transmission systems, especially in terms of efficiency and resilience under dynamic wireless channel conditions.

Deep autoencoder-based joint source-channel coding (JSCC) has gained significant attention for end-to-end image semantic communication systems. However, existing methods typically optimize a uniform bandwidth-distortion trade-off over the entire image, potentially leading to the loss of crucial details and inconsistent content for tasks with diverse regions of interest. In this paper, we propose a flexible fine-grained bandwidth allocation method for deep JSCC that enables highly efficient, task-scalable image transmission across various semantic communication scenarios using a single codec. Our method optimizes the bandwidth-distortion trade-off by constraining image distortion through a 2D pixel-wise quality map. Guided by the pixel-wise quality map, we introduce a novel conditional affine transformation that generates dedicated semantic feature maps tailored to specific tasks. Additionally, we introduce a semantic guidance network to automatically generate task-aware quality maps via backpropagation without additional retraining. This approach leverages a pretrained variable-length neural JSCC codec and adjusts the transmission quality on a fine-grained level, eliminating the need to train separate models for different tasks. Experimental results demonstrate the effectiveness of our bandwidth allocation method, enhancing task-specific performance in various goal-oriented image communication scenarios without additional training.

End-to-end visual communication systems typically optimize a trade-off between channel bandwidth costs and signal-level distortion metrics. However, under challenging physical conditions, this traditional coding and transmission paradigm often results in unrealistic reconstructions with perceptible blurring and aliasing artifacts, despite the inclusion of perceptual or adversarial losses for optimizing. This issue primarily stems from the receiver’s limited knowledge about the underlying data manifold and the use of deterministic decoding mechanisms. To address these limitations, this paper introducesDiffCom, a novel end-to-endgenerative communicationparadigm that utilizes off-the-shelf generative priors and probabilistic diffusion models for decoding, thereby improving perceptual quality without heavily relying on bandwidth costs and received signal quality. Unlike traditional systems that rely on deterministic decoders optimized solely for distortion metrics, ourDiffComleverages raw channel-received signal as a fine-grained condition to guide stochastic posterior sampling. Our approach ensures that reconstructions remain on the manifold of real data with a novel confirming constraint, enhancing the robustness and reliability of the generated outcomes. Furthermore,DiffComincorporates a blind posterior sampling technique to address scenarios with unknown forward transmission characteristics. Extensive experimental validations demonstrate thatDiffComnot only produces realistic reconstructions with details faithful to the original data but also achieves superior robustness against diverse wireless transmission degradations. Collectively, these advancements establishDiffComas a new benchmark in designing generative communication systems that offer enhanced robustness and generalization superiorities.

In this paper, we propose “SoundSpring”, a cutting-edge error-resilient audio transceiver that marries the robustness benefits of joint source-channel coding (JSCC) while also being compatible with current digital communication systems. Unlike recent deep JSCC transceivers, which learn to directly map audio signals to analog channel-input symbols via neural networks, our SoundSpring adopts the layered architecture that delineates audio compression from digital coded transmission, but it sufficiently exploits the impressive in-context predictive capabilities of large language (foundation) models. Integrated with the casual-order mask learning strategy, our single model operates on the latent feature domain and serve dual-functionalities: as efficient audio compressors at the transmitter and as effective mechanisms for packet loss concealment at the receiver. By jointly optimizing towards both audio compression efficiency and transmission error resiliency, we show that mask-learned language models are indeed powerful contextual predictors, and our dual-functional compression and concealment framework offers fresh perspectives on the application of foundation language models in audio communication. Through extensive experimental evaluations, we establish that SoundSpring apparently outperforms contemporary audio transmission systems in terms of signal fidelity metrics and perceptual quality scores. These new findings not only advocate for the practical deployment of SoundSpring in learning-based audio communication systems but also inspire the development of future audio semantic transceivers.

The forthcoming sixth-generation (6G) wireless system brings forth the need to meet increasingly stringent performance criteria, necessitating refinement and advanced design of channel coding and modulation techniques. Within this context, polar codes stand out as a competitive candidate due to their favorable attributes. This study delves into probabilistically-shaped polar-coded modulation (PS-PCM) for its potential to enhance spectral efficiency and error-correction performance, offering a pragmatic design geared toward standardization. Specifically, we present a coding chain for PS-PCM that ensures backward compatibility with legacy 5G polar codes. We also propose, for the first time, a channel-independent method for constructing PS-PCM that eliminates the need for on-the-fly computations, which is an imperative attribute for practical system implementation. The core of this code construction method is the utilization of a surrogate channel to allocate rates to each component code while considering the varying probability distributions and protection levels of label bits. The assigned component code rates can be seamlessly integrated into the modulation and coding scheme (MCS) table to facilitate rapid construction for both base stations and user equipment. Furthermore, we devise flexible rate matching methods to allow fine granularity adjustment of rates and blocklengths for PS-PCM. Simulation results validate the efficacy of our proposed approaches, which also provide insights and benchmarks for the capability of PS-PCM to meet requirements for enhanced spectral efficiency, higher reliability, and more flexibility.

Recent advancements in neural image codecs (NICs) are of significant compression performance, but limited attention has been paid to their error resilience. These resulting NICs tend to be sensitive to packet losses, which are prevalent in real-time communications. In this paper, we investigate how to elevate the resilience ability of NICs to combat packet losses. We propose ResiComp, a pioneering neural image compression framework with feature-domain packet loss concealment (PLC). Motivated by the inherent consistency between generation and compression, we advocate merging the tasks of entropy modeling and PLC into a unified framework focused on latent space context modeling. To this end, we take inspiration from the impressive generative capabilities of large language models (LLMs), particularly the recent advances of masked visual token modeling (MVTM). In specific, ResiComp develops a bi-directional masked Transformer to model the contextual dependencies among latents with dual-functionality: 1) iteratively acts as a conditional entropy model to boost compression efficiency; 2) operates latent PLC to improve resilience. During training, we integrate MVTM to mirror the effects of packet loss, enabling a dual-functional Transformer to restore the masked latents by predicting their missing values and conditional probability mass functions. Our ResiComp jointly optimizes compression efficiency and loss resilience. Moreover, ResiComp provides flexible coding modes, allowing for explicitly adjusting the efficiency-resilience trade-off in response to varying Internet or wireless network conditions. Extensive experiments demonstrate that ResiComp can significantly enhance the NIC’s resilience against packet losses, while exhibits a worthy trade-off between compression efficiency and packet loss resilience. Additionally, packet-level simulations, conducted using diverse network models based on real traces, demonstrate that ResiComp exhibits much better robustness to fluctuating network conditions compared to redundancy-based approaches like VTM + FEC.

Qihang Ai, Jiafan Li, Jincheng Dai, Jianwu Zhou, Lemao Liu, Haiyun Jiang, Shuming Shi. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Deep learning (DL)-based channel state information (CSI) feedback holds substantial promise for boosting spectrum efficiency in massive MIMO systems. However, prevailing studies often treat compressed CSI bits uniformly, assuming their accurate transmission over noisy channels. Such assumptions falter when confronted with bandwidth limitations or low signal-to-noise ratios (SNRs), significantly impairing CSI reconstruction quality. In this paper, we introduce a novel soft variational approach to implement deep joint source-channel coding for CSI feedback within noisy environments, conceptualizing the system as an end-to-end rate-distortion (RD) optimization framework. Specifically, our model utilizes a nonlinear transform to extract the latent representation of CSI and employs an entropy model as a prior to guide the variable-rate transmission of latent features across noisy channels. Experimental results demonstrate that our approach achieves substantial improvements over benchmark schemes, saving 40% in feedback bandwidth under normalized mean squared error (NMSE) and ensuring the robustness to varying wireless channels.

Toward next-generation multiple access, we propose a multiuser transmission framework via non-binary factor graphs defined over Galois field GF(q). The non-orthogonal multiuser transmission is realized by applying the GF(q) multiplication and superposition. In coordination with the high-order modulation, the proposed GF(q) codebook improves the cardinality of the superimposed constellation and reduces interference at each resource element. The proposed structure also characters for flexibility by spatially coupling the base graph when applying to different user scales. We have theoretically proved that both the system capacity and the error performance are improved due to the GF(q) operations. We utilize the simulated annealing algorithm to optimize both the base factor graph and the spatially-coupled graph, therefore short girths are eliminated or mitigated. The detection of the proposed scheme can be performed through iterative belief-propagation (BP) based on symbol-level log-domain probability vectors derived on the GF(q) factor graph. The numerical results in terms of upper bounds of maximum likelihood (ML) detection and symbol error rates through BP detection are both provided, which demonstrate the superiority of the GF(q) factor graphs to their binary counterparts. The performance can be further improved by increasing the field order q and/or by spatially coupling the non-binary factor graph.

This paper conducts the first analysis of the bit error probabilityPbof polar codes under successive cancellation (SC) decoding, with a focus on information bits. We introduce the concept of component bit error probabilityPb(i), which corresponds to thei-th stage of SC decoding, andPbis computed as the sum of allPb(i) over the information set. To facilitate the evaluation ofPb(i) andPb, we propose an analytical approach that leverages the algebraic structure of a particular type of polar codes termed polar subcodes. This approach allows us to derive closed-form approximations for the upper bounds on bothPb(i) andPb. Our analysis applies to both systematic and non-systematic polar codes, where we further investigate the homogeneity and introduce the input-output weight distribution of polar subcodes to obtain more concise expressions for the proposed approximate upper bounds. Experimental results are also presented to validate the effectiveness of these approximations.

The scope of the 6G wireless communication system is envisioned to expand beyond delivering data to humans and towards connecting machines that constantly upload computation-intensive status updates to obtain real-time situational awareness. Under dynamic environments, the amount of useful information contained in status updates degrades over time, which could be measured based on the concept of age of information. In this paper, we develop an analytical framework to investigate the temporal value of status updates, in terms of the peak age of information. Given the temporal dynamics of observed physical process, the procedure of transmission and computing is modeled as tandem queues for both parallel processing and series processing modes at the edge server. The obtained closed-form expressions explicitly characterize the coupling among information generation, transmission, and usage, which can be exploited as performance metrics for task-oriented resource optimization. The accuracy of our analysis is verified with simulation results. Based on the theoretical analysis, we formulate an optimization problem to simultaneously minimize the age of status updates and energy consumption for multiple devices. Numerical results reveal that the computation and transmission time could be traded off to obtain timely status updates at low energy cost.

Unsourced multiple access (UMA) is a promising technique for the Internet-of-Things systems with a large number of users. In this paper, we explore the good algebraic properties of finite fields and propose a non-binary coding scheme for the UMA. The scheme is based on the slotted structure and T-fold ALOHA, and each active user randomly selects one slot to transmit its message. Finite-field operations are performed for both the channel coding and the multiple access. Compared to binary-field operations, the finite-field sparse spreading and multiuser superposition lead to improved diversity and less interference on each channel use. In addition, the non-binary LDPC code boosts the performance under short/moderate code length. We derive the joint belief-propagation decoding algorithm for the concatenated finite-field factor graph, which is then verified by the extrinsic information transfer (EXIT) analysis. The numerical results in terms of energy-efficiency performance are provided, which clearly demonstrate the superiority to the existing schemes in the region with large user numbers. The proposed scheme has the potential to carry out larger-scale user connections, and the energy-efficiency performance can be further improved by optimizing the coding parameters.

In this paper, we propose a new class of high-efficiency semantic coded transmission methods to realize end-to-end speech transmission over wireless channels. We name the whole system as Deep Speech Semantic Transmission (DSST). Specifically, we introduce a nonlinear transform to map the speech source to semantic latent space and feed semantic features into source-channel encoder to generate the channel-input sequence. Guided by the variational modeling idea, we set an entropy model on the latent space to estimate the importance diversity among semantic feature embeddings. Accordingly, these semantic features of different importance can be reasonably allocated with different coding rates, which maximizes the system coding gain. Furthermore, we introduce a channel signal-to-noise ratio (SNR) adaptation mechanism such that a single model can be applied over various channel states. The end-to-end optimization of our model leads to a flexible rate-distortion (RD) tradeoff, supporting an adaptive rate wireless speech semantic transmission. Experimental results verify that our DSST system clearly outperforms current engineered speech transmission systems on both objective and subjective metrics. Compared with existing neural speech semantic transmission methods, our model saves up to 75% of channel bandwidth costs when achieving the same quality. Audio samples are available at https://ximoo123.github.io/DSST.

In this paper, we aim to redesign the vision Transformer (ViT) as a new backbone to realize semantic image transmission, termed wireless image transmission transformer (WITT). Previous works build upon convolutional neural networks (CNNs), which are inefficient in capturing global dependencies, resulting in degraded end-to-end transmission performance especially for high-resolution images. To tackle this, the proposed WITT employs Swin Transformers as a more capable backbone to extract long-range information. Different from ViTs in image classification tasks, WITT is highly optimized for image transmission while considering the effect of the wireless channel. Specifically, we propose a spatial modulation module to scale the latent representations according to channel state information, which enhances the ability of a single model to deal with various channel conditions. As a result, extensive experiments verify that our WITT attains better performance for different image resolutions, distortion metrics, and channel conditions. The code is available at https://github.com/KeYang8/WITT.

Humans tend to focus on only a few regions of interest (ROI) rather than perceiving the entire scene. This insight is also useful for machine tasks. Built upon the properties of ROI, in this paper, we propose a learned image transmission framework toward machine tasks, which ensures both the image reconstruction quality and the task accuracy. The whole system is optimized under a tripartite RDA tradeoff across the channel bandwidth cost (rate, R), the signal reconstruction quality (distortion, D), and the machine task performance (accuracy, A). According to the image content complexity distribution and the specific task, we incorporate both the entropy model and the ROI map to guide the source-channel coding rate allocation. As a result, we obtain the system coding gain. During this process, we develop two types of real-time ROI generation methods, suitable for high and low bandwidth cost regions, respectively. Experimental results show that our approach vastly outperforms state-of-the-art engineered image transmission methods and emerging image transmission methods. Moreover, we conduct an extensive ablation study to demonstrate the importance of individual components in our method, by which we expect to facilitate future research on this novel approach for machine-type semantic communications.

Learned image transmission (LIT) has shown promising progress in recent years to boost the end-to-end transmission performance in semantic communications. To further enhance the system efficiency, in this paper, we propose a novel LIT framework built on multiple-input multiple-output (MIMO) fading channels. In particular, the proposed framework supports concurrent transmission of multiple streams, which can maximize the multiplexing gain in end-to-end semantic communication systems. By jointly considering the entropy distribution of the image semantic features and the wireless MIMO channel states, we design a spatial multiplexing mechanism that can adaptively realize coding rate allocation and stream mapping. As a result, source content and channel environment will be seamlessly coupled, which maximizes the coding gain. Moreover, the proposed LIT model is versatile: a single model can support various transmission rates. The whole model is optimized under the constraint of transmission rate-distortion (RD) tradeoff. Experimental results verify that our scheme substantially increases the throughput of semantic communication systems, and outperforms traditional MIMO communication systems under realistic fading channels.

Guessing random additive noise decoding (GRAND) is a recently proposed decoding algorithm which can achieve the error performance of maximum likelihood (ML) decoding. However, GRAND and its variants are only suitable for some short codes with high code rates and have large average query numbers. To mitigate these problems, we propose a successive cancellation list (SCL)-GRAND decoding algorithm for the cyclic redundancy check concatenated polar (CRC-polar) codes. The proposed decoder first divides the received sequence into two subblocks. Then SCL is used to decode the upper subblock and output several candidates into the candidate list. For each candidate, GRAND is used to decode the lower subblock and finally choose the most-likely codeword as the decoded result. Since the SCL is integrated into the SCL-GRAND algorithm, this algorithm can achieve lower complexity and better flexibility than the original GRAND.

In this paper we propose a novel non-orthogonal multiple access (NOMA) scheme via non-binary factor graphs by introducing operations over Galois field GF(q). By utilizing sparse GF(q) spreading, the proposed scheme benefits from an improved diversity of the superimposed symbols and accordingly a lower probability of constellation overlapping. Meanwhile, the coordination with high-order modulation helps to reduce the interference at the receiver, and therefore a better error performance can be expected. Iterative belief-propagation (BP) detection based on log-domain probabilities is derived for the GF(q) NOMA. The upper bounds of maximum likelihood (ML) detection and simulation results in terms of symbol error rate are provided, which demonstrate the superiority of the proposed scheme to its binary counterparts.

Recently a number of low-density parity-check (LDPC) decoding algorithms based on deep learning have been proposed in the literature. However, most of the work has been targeted for additive white Gaussian noise (AWGN) channels. For more practical scenarios, in this paper we investigate the neural-network based min-sum (MS) decoding for protograph LDPC codes in fading channels. Since the wireless channel is complex and varying, accurate channel state information (CSI) cannot be always available at the receiver. We classify the scenarios into three cases with perfect CSIs, imperfect CSIs, and no CSIs. By assigning learnable weights on the edges in the iterative decoding, the proposed neural decoder compensates for the performance loss caused by the error/lack of CSIs. The trajectory-based extrinsic information transfer (T-EXIT) chart is employed as a theoretical tool to select the proper training dataset for the neural network and the proper channel initialization scheme for the receiver. Numerical results in terms of block error rates are provided, which agree with the T-EXIT analysis. It can be seen that the proposed neural MS decoder clearly outperforms the traditional MS algorithm in Rayleigh fading channels. Meanwhile, the proposed decoder shows a good compatibility so that it can be applied to the cases with different accuracy of CSIs.

We propose a novel neural waveform compression method to catalyze emerging speech semantic communications. By introducing nonlinear transform and variational modeling, we effectively capture the dependencies within speech frames and estimate the probabilistic distribution of the speech feature more accurately, giving rise to better compression performance. In particular, the speech signals are analyzed and synthesized by a pair of nonlinear transforms, yielding latent features. An entropy model with hyperprior is built to capture the probabilistic distribution of latent features, followed by quantization and entropy coding. The proposed waveform codec can be optimized flexibly towards arbitrary rate, and the other appealing feature is that it can be easily optimized for any differentiable loss function, including perceptual loss used in semantic communications. To further improve the speech quality, we incorporate residual coding to mitigate the degradation arising from quantization distortion at the latent space. Results indicate that achieving the same perceptual quality score, the proposed method saves up to 27% coding rate than widely used adaptive multi-rate wideband (AMR-WB) codec as well as emerging neural waveform coding methods.

How to efficiently transmit a special video over wireless channels? While the established systems work by combining H.26x video coding and 5G LDPC channel coding, its end-to-end transmission efficiency is still far away from the extreme for video sources in a specific domain. In this paper, we seek to design a special semantic communication system tailored for transmitting video calling streams over the wireless channels. Inspired by the recent progress in talking-head animation, we propose a talking- head semantic transmission (THST) system, which can efficiently transmit motion keypoint representation as compact semantic information to drive the free-view talk-heading synthesis at the receiver. Since the motion semantic key points are correlated, our THST system learns a nonlinear analysis transform to map the key points across multiple frames into latent space, then transmits the latent hyper semantic representation to the receiver via deep joint source-channel coding. Our system incorporates a latent prior to estimate the importance diversity on the semantic key points, accordingly, we realize variable rate joint source-channel coding to obtain system level coding gain. Extensive experimental validation shows that our THST system outperforms engineered competing systems on benchmark datasets. Moreover, due to the system level joint source and channel design, our method provides much more robust performance over noisy channels with only 33% bandwidth cost versus the current talking-head compression combined with 5G LDPC coded transmission systems.

The emerging field semantic communication is driving the research of end-to-end data transmission. By utilizing the powerful representation ability of deep learning models, learned data transmission schemes have exhibited superior performance than the established source and channel coding methods. While, so far, research efforts mainly concentrated on architecture and model improvements toward a static target domain. Despite their successes, such learned models are still suboptimal due to the limitations in model capacity and imperfect optimization and generalization, particularly when the testing data distribution or channel response is different from that adopted for model training, as is likely to be the case in real-world. To tackle this, in this paper, we propose a novel online learned joint source and channel coding approach that leverages the deep learning model’s overfitting property. Specifically, we update the off-the-shelf pre-trained models after deployment in a lightweight online fashion to adapt to the distribution shifts in source data and environment domain. We take the overfitting concept to the extreme, proposing a series of implementation-friendly methods to adapt the codec model or representations to an individual data or channel state instance, which can further lead to substantial gains in terms of the end-to-end rate-distortion performance. Accordingly, the streaming ingredients include both the semantic representations of source data and the online updated decoder model parameters. The system design is formulated as a joint optimization problem whose goal is to minimize the loss function, a tripartite trade-off among the data stream bandwidth cost, model stream bandwidth cost, and end-to-end distortion. The proposed methods enable the communication-efficient adaptation for all parameters in the network without sacrificing decoding speed. Extensive experiments, including user study, on continually changing target source data and wireless channel environments, demonstrate the effectiveness and efficiency of our approach, on which we outperform existing state-of-the-art engineered transmission scheme (VVC combined with 5G LDPC coded transmission).

In this paper, we design a new class of high-efficiency deep joint source-channel coding methods to achieve end-to-end video transmission over wireless channels. The proposed methods exploit nonlinear transform and conditional coding architecture to adaptively extract semantic features across video frames, and transmit semantic feature domain representations over wireless channels via deep joint source-channel coding. Our framework is collected under the name deep video semantic transmission (DVST). In particular, benefiting from the strong temporal prior provided by the feature domain context, the learned nonlinear transform function becomes temporally adaptive, resulting in a richer and more accurate entropy model guiding the transmission of current frame. Accordingly, a novel rate adaptive transmission mechanism is developed to customize deep joint source-channel coding for video sources. It learns to allocate the limited channel bandwidth within and among video frames to maximize the overall transmission performance. The whole DVST design is formulated as an optimization problem whose goal is to minimize the end-to-end transmission rate-distortion performance under perceptual quality metrics or machine vision task performance metrics. Across standard video source test sequences and various communication scenarios, experiments show that our DVST can generally surpass traditional wireless video coded transmission schemes. The proposed DVST framework can well support future semantic communications due to its video content-aware and machine vision task integration abilities.

As one novel approach to realize end-to-end wireless image semantic transmission, deep learning-based joint source-channel coding (deep JSCC) method is emerging in both deep learning and communication communities. However, current deep JSCC image transmission systems are typically optimized for traditional distortion metrics such as peak signal-to-noise ratio (PSNR) or multi-scale structural similarity (MS-SSIM). But for low transmission rates, due to the imperfect wireless channel, these distortion metrics lose significance as they favor pixel-wise preservation. To account for human visual perception in semantic communications, it is of great importance to develop new deep JSCC systems optimized beyond traditional PSNR and MS-SSIM metrics. In this paper, we introduce adversarial losses to optimize deep JSCC, which tends to preserve global semantic information and local texture. Our new deep JSCC architecture combines encoder, wireless channel, decoder/generator, and discriminator, which are jointly learned under both perceptual and adversarial losses. Our method yields human visually much more pleasing results than state-of-the-art engineered image coded transmission systems and traditional deep JSCC systems. A user study confirms that achieving perceptually similar end-to-end image transmission quality, the proposed method can save about 50% wireless channel bandwidth costs.

The recent deep learning-based joint source-channel coding (deep JSCC) framework has shown superior performance on end-to-end wireless image transmission without suffering from the “cliff effect”. However, a fundamental limit of current deep JSCC schemes is that the unbalanced regional importance of the source image has not been explicitly taken into account. It evenly distributes the coding rate to every image patch leading to an evident degradation of the overall coding efficiency. To break this fundamental limit, we propose a novel end-to-end wireless image transmission scheme in this paper. Our scheme integrates the deep JSCC architecture and the quadtree-structured regional rate allocation strategy adopted in the HEVC standard, collected under the name “resolution-adaptive deep JSCC (RaDJSCC)”. Our new architecture perceives the content of the transmitted image and adaptively allocates more channel bandwidth to the complex pixel blocks. Results show that for high-resolution images, the proposed RaDJSCC transmission method generally outperforms the emerging analog transmission schemes using deep JSCC and the digital transmission schemes using classical separated source and channel coding, e.g., BPG + LDPC.

We study the problem of deep joint source-channel coding (D-JSCC) for correlated image sources, where each source is transmitted through a noisy independent channel to the common receiver. In particular, we consider a pair of images captured by two cameras with probably overlapping fields of view transmitted over wireless channels and reconstructed in the center node. The challenging problem involves designing a practical code to utilize both source and channel correlations to improve transmission efficiency without additional transmission overhead. To tackle this, we need to consider the common information across two stereo images as well as the differences between two transmission channels. In this case, we propose a deep neural networks solution that includes lightweight edge encoders and a powerful center decoder. Besides, in the decoder, we propose a novel channel state information aware cross attention module to highlight the overlapping fields and leverage the relevance between two noisy feature maps. Our results show the impressive improvement of reconstruction quality in both links by exploiting the noisy representations of the other link. Moreover, the proposed scheme shows competitive results compared to the separated schemes with capacity-achieving channel codes.

In this paper, we propose a new class of distributed joint source-channel coding (DJSCC) methods, namely triple polar codes (T-PC), for transmitting a pair of correlated binary sources over noisy channels. In the T-PC structure, one source is protected by a systematic polar code (SPC), and the other source is encoded into a double polar code (D-PC) word. Following this, we prove the T-PC approaches the corner point of the achievable rate-region of DJSCC. We further propose a distributed joint source-channel decoding algorithm, which involves two components: a cyclic redundancy check (CRC) aided successive cancellation list (CA-SCL) decoding of the SPC and a joint successive cancellation list (J-SCL) decoding of the D-PC. The CA-SCL and J-SCL decoding procedures alternately generate hard-decisions of sources which are iteratively exchanged as the side information and result in superior performance compared with the state-of-the-art polar code based DJSCC scheme.

In this paper, we investigate the joint design and optimization of source-channel polar coding with 2m-ary transmission. A joint framework is proposed which includes a source polar code to compress the redundant source, followed by a set of component polar codes over a 2m-ary modulation to protect the source against errors and achieve increased spectral efficiency. We prove that our scheme suffices to achieve the theoretical limit of source-channel separation theorem. For finite-length cases, a joint decoder that exploits both the residual redundancy and channel characteristics is also derived to further reduce the error rate. Simulation results verify the effectiveness of the scheme.

In this paper, we propose a class of high-efficiency deep joint source-channel coding methods that can closely adapt to the source distribution under the nonlinear transform, it can be collected under the name nonlinear transform source-channel coding (NTSCC). In the considered model, the transmitter first learns a nonlinear analysis transform to map the source data into latent space, then transmits the latent representation to the receiver via deep joint source-channel coding. Our model incorporates the nonlinear transform as a strong prior to effectively extract the source semantic features and provide side information for source-channel coding. Unlike existing conventional deep joint source-channel coding methods, the proposed NTSCC essentially learns both the source latent representation and an entropy model as the prior on the latent representation. Accordingly, novel adaptive rate transmission and hyperprior-aided codec refinement mechanisms are developed to upgrade deep joint source-channel coding. The whole system design is formulated as an optimization problem whose goal is to minimize the end-to-end transmission rate-distortion performance under established perceptual quality metrics. Across test image sources with various resolutions, we find that the proposed NTSCC transmission method generally outperforms both the analog transmission using the standard deep joint source-channel coding and the classical separation-based digital transmission. Notably, the proposed NTSCC method can potentially support future semantic communications due to its content-aware ability and perceptual optimization goal.

In this paper, we propose a block polarization (BP) scheme as a framework to study polar-coded hybrid automatic repeat request (Polar-HARQ) under bit-interleaved polar-coded modulation (BIPCM). The BP scheme of BIPCM can combine multiple independent polar subcodes to form a longer polar code under the block polarization between polar subcodes. The BP scheme of HARQ further carries out block polarization based on the BP scheme of BIPCM, which can combine multiple transmitted BP schemes of BIPCM together to form a longer polar code. When the underlying channels are non-uniform, such as the polarization effect of high-order modulation will lead to non-uniform reliabilities, it is not suitable to use the fixed sequence (e.g., the polarization weight (PW) sequence or Polar sequence in the 5G standard) to construct polar codes. The underlying channels on each polar subcode have uniform reliability by dividing different blocks according to the reliabilities of the underlying channel. Based on this characteristic of BP scheme, we propose a rate-allocation (RA) method to study fast code construction of BP scheme. The key idea of the proposed RA method is finding out the equivalent channel whose average symmetric capacity equals the target transmission rate. The allocated rates are computed according to the symmetric capacities of split channels under the block polarization between polar subcodes. Then the ranking of bit indices within each polar subcode can be obtained by fixed sequence. In this way, the RA code construction is concise and robust for diverse configurations which are the desired features for practical implementation. Simulation results show that the proposed BP scheme under RA construction can achieve almost identical performance as the Gaussian approximation construction with much lower complexity.

In this paper, the problem of neural compression based image transmission over wireless channels is studied. Since all procedures are considered over wireless links, the quality of training is affected by wireless factors such as packet errors. In the considered model, compressed data given by the neural source encoder (NSE) are fed into an error-control channel encoder and modulated as discrete symbols sent over a memoryless channel. In the receiving end, the channel decoder and the neural source decoder (NSD) forms an iterative structure to reconstruct the original image. Since all neural compressed data are transmitted over wireless channels, the training of NSD is affected by wireless channel factors such as residual bit errors given by the channel decoder. Meanwhile, during outer-loop iterations, the NSD needs to match the variant of information reliability output by the channel decoder so as to build a global optimal receiver. To this end, a refiner neural network is first attached after the NSD to adjust its output as the format of a priori information sent into the channel decoder. Then, the extrinsic information transfer (EXIT) functions of channel decoder and NSD are derived. At each iteration, the reliability of messages sent into the NSD is explicitly predicted by using the EXIT chart. By this means, the NSD can be trained in a residual bit error aware manner, and we realize a joint learning and iterative decoding framework to ensure the quality of neural image transmission over realistic wireless channels.

Multilevel polar-coded modulation with multistage decoding is a capacity-achieving coded modulation scheme. In this paper, we propose a general formulation for analyzing the performance of multilevel polar-coded modulation and then derive the error probability upper bounds. The analysis explicitly reveals the effect of the modulation scheme on the performance of component polar codes. Based on the derived upper bounds, we also propose two construction methods for multilevel polar-coded modulation. Compared with conventional methods, such as density evolution and Gaussian approximation which involve complicated recursive calculations, the proposed methods have a linear computational complexity. Simulation results also show that the proposed construction methods can achieve comparable performance to existing methods under SC decoding, and even better performance under SC list decoding.

In this paper, layered min-sum (MS) iterative decoding is formulated as a customized neural network following the sequential scheduling of check node (CN) updates. By virtue of the lifting structure of protograph low-density parity-check (LDPC) codes, identical network parameters are shared among all derived edges originating from the same edge in the protograph, which makes the number of learn- able parameters manageable. The proposed neural layered MS decoder can support arbitrary codelengths consequently. Moreover, an iteration-wise greedy training method is proposed to tune the parameters such that it avoids the vanishing gradient problem and accelerates the decoding convergence.

We propose a rate-filling method as a framework to study fast construction of bit-interleaved polar-coded modulation (BIPCM). In particular, the proposed method makes full use of the nesting structure of polar codes. We show that BIPCM can be constructed by appropriately allocating the rates of each polar subcode whose information set is easily extracted by a predetermined index sequence, e.g., the Polar sequence in the 5G standard. By this means, the reliability calculation and ranking operation of bit channels can be discarded, which is a desired feature for low-complex construction. Simulation results show the proposed fast construction method performs independently of the actual channel condition and is robust to diverse modulation and coding schemes in the 5G standard.

In this paper, a novel polar-coded MIMO (PC-MIMO) framework is proposed to enhance the system polarization effect so that the transmission reliability is improved. The key idea is asynchronously transmitting coded bits within one block and then spatially coupling multiple blocks by joint antenna mapping. This consequently leads to the enhancement of the polarization diversity among antenna subchannels under finite block length. Combining with binary polar coding, the gain achieved by the antenna subchannel polarization is finally propagated to bit polarized channels. The proposed asynchronous PC-MIMO is proved to be capacity-achieving under infinite block length and to realize performance gain with respect to synchronous PC-MIMO under finite block length.

Abstract In this study, a coded cooperation scheme using polar codes is explored for high‐reliability and low‐user complexity applications such as Internet of Things. Firstly, the cooperation way among the users is carefully designed based on systematic polar codes and cooperative decoding, without decode‐and‐forward protocol at the cooperative user. Meanwhile repetition aided code construction and decoding algorithm of polar code is proposed for the non‐cooperative case in terms of imperfect inter‐user channel. Performance analysis and numerical simulation demonstrate the bit‐error‐rate advantages of proposed polar coded cooperation in fading channels, compared to the no‐cooperation case and normal coded cooperation schemes. Furthermore, overall rate of proposed polar‐coded cooperation is considered to optimize adaptively to increase throughput efficiency.

The recent development of deep learning methods provides a new approach to optimize the belief propagation (BP) decoding of linear codes.However, the limitation of existing works is that the scale of neural networks increases rapidly with the codelength, thus they can only support short to moderate codelengths.From the point view of practicality, we propose a high-performance neural min-sum (MS) decoding method that makes full use of the lifting structure of protograph low-density parity-check (LDPC) codes.By this means, the size of the parameter array of each layer in the neural decoder only equals the number of edge-types for arbitrary codelengths.In particular, for protograph LDPC codes, the proposed neural MS decoder is constructed in a special way such that identical parameters are shared by a bundle of edges derived from the same edge-type.To reduce the complexity and overcome the vanishing gradient problem in training the proposed neural MS decoder, an iteration-byiteration (i.e., layer-by-layer in neural networks) greedy training method is proposed.With this, the proposed neural MS decoder tends to be optimized with faster convergence, which is aligned with the early termination mechanism widely used in practice.To further enhance the generalization ability of the proposed neural MS decoder, a codelength/rate compatible training method is proposed, which randomly selects samples from a set of codes lifted from the same base code.As a theoretical performance evaluation tool, a trajectory-based extrinsic information transfer (T-EXIT) chart is developed for various decoders.Both T-EXIT and simulation results show that the optimized MS decoding can provide faster convergence and up to 1dB gain compared with the plain MS decoding and its variants with only slightly increased complexity.In addition, it can even outperform the sum-product algorithm for some short codes.

A new polar-coded modulation (PCM) framework with the bit interleaving is proposed to enhance the polarization diversity among the bit polarized subchannels under the finite block length, consequently, the transmission reliability is further improved. The key idea is asynchronously transmitting the coded bits within one block and spatially coupling multiple coded blocks by the joint modulation, and the polarization diversity among the modulation synthesized subchannels under the parallel partition is enhanced. Combining the binary polar coding, this polarization enhancement at the modulation partition stage is then delivered to the final bit polarized subchannels. The capacity-achieving property under the infinite block length and the polarization superiority with respect to state-of-the-art PCM schemes under the finite block length are proved. Finally, the simulation results indicate the performance gain compared to the conventional PCM and 5G LDPC coded modulation schemes.

Code construction is a critical issue for polar coding. The expected construction method is of an accurate estimate of the reliability of bit-channels, but the current methods usually require high computational complexity. In this paper, we concern the input-output weight distribution of each bit-channel and derive its recursive calculation algorithm for systematic coding. The union bound and union-Bhattacharyya bound on the bit error probability are also derived to evaluate the reliability of bit-channels. Furthermore, by calculating the logarithmic form of the union-Bhattacharyya bound, we also propose two novel construction methods named the union-Bhattacharyya bound weight of the bit error probability (UBWB) and the simplified UBWB (SUBWB). Numerical results show that the proposed UBWB/SUBWB construction methods can achieve comparable performance to current methods under successive cancellation (SC) decoding and obtain obvious performance gain under SC list (SCL) decoding.

In this paper, a weighted successive cancellation (WSC) algorithm is proposed to improve the decoding performance of polar codes with the quantized log-likelihood ratio (LLR). The weights used in the WSC are automatically learned by a neural network (NN). A novel NN model and its simplified architecture are build to select the optimal weights of the WSC, and all-zero codewords can train the NN. Besides, we impose the constraints on weights to direct the learning process. The small number of trainable parameters lead to faster learning without performance loss. Simulation results show that the WSC algorithm is valid to various codewords and the trained weights make it outperform SC algorithm with the same quantization precision. Notably, the WSC with 3-bit quantization precision achieves a near floating point performance for short length.

Sparse code multiple access (SCMA) is a promising non-orthogonal multiple access scheme for 5G systems. In this paper, based on the sphere decoding (SD), we design an optimal receiver for SCMA, which provides much lower complexity than the maximum likelihood (ML) detection without sacrifice of the optimal performance. Regarding the coded SCMA system, we propose a list sphere decoding (LSD) method to output each user's bit soft decisions to the corresponding channel decoders. In addition, the existing lattice points in the list are employed to set the original radius so as to further reduce the complexity of LSD. The performance and complexity analyses indicate that the proposed optimal receiver provides notable gain compared to the most popular SCMA receiver equipped with the message passing algorithm.

Polar coded modulation (PCM) is one of the most promising approaches towards high spectral efficiency. It is able to provide excellent performance comparing to the conventional turbo coded modulation schemes. In this paper, we consider the design of the PCM schemes with high modulation order, e.g., 256QAM or 1024QAM etc. Different with the legacy PCM of low modulation order, the coded bits in high-order PCM with all-zero frozen-sequence will demonstrate obvious correlations which lead to catastrophic performance loss. In order to evaluate this loss, we introduce a quantitative metric, named the bit conditional mutual information (BCMI), to analyze the correlation. Then a new PCM scheme based on the constrained frozen-sequence is proposed to eliminate the correlations among the coded bits. Theoretical analysis and simulation results show that the proposed scheme can ensure the performance of PCM and significantly outperform that of the turbo coded modulation schemes.

In this paper, polar coding is combined with the generalized frequency division multiplexing (GFDM) and the channel polarization idea is extended to the GFDM system, which is a flexible multicarrier modulation scheme proposed for future waveform. First, the framework of polar coded GFDM (PC-GFDM) is constructed based on a two-stage channel polarization transform. In the first stage, on the basis of bit-interleaved code modulation, GFDM channel is divided into a set of parallel binary-input GFDM synthesized channels. In the second stage, by using the binary channel polarization, the GFDM synthesized channels are further polarized into a set of bit polarized channels. The polar codes are efficiently constructed by calculating the reliabilities of the two-stage polarized channels. Then, an empirically optimal interleaver is designed for PC-GFDM. Furthermore, regarding the features of GFDM synthesized channels, a simplified block interleaving scheme is proposed to approach the optimal interleaver and reduce the implementation complexity. Compared with the conventional turbo coded GFDM (TC-GFDM), simulation results indicate that our proposed PC-GFDM system can outperform the TC-GFDM system by at least 1dB for different scenarios.

Sparse code multiple access (SCMA) has recently emerged as one of the most favorable multiple access schemes for 5G networks, which allows overloading with a large number of users so as to enable massive connectivity. In this paper, we design a hardware framework of a uplink system for SCMA. First, we propose a unified quantization scheme based on density evolution optimization which is independent of signal-to-noise ratios (SNR). Second, we apply a fast convergence message passing algorithm (FC-MPA) in SCMA multiuser detection, in which the function nodes updating and variable nodes updating are processed synchronously so as to make the FC-MPA converge about 2 times faster than standard MPA. Finally, based on FC- MPA, we design a pipelined decoding structure for SCMA so as to increase throughput. FPGA results demonstrate that the fix- point performance achieve a near floating-point performance for different SNR and the pipelined hardware structure we design is feasible.

In this paper, polar codes are first applied in non-orthogonal multiple access (NOMA) and the channel polarization idea is extended to NOMA, which is a major multiple access technique in 5G systems. The polar coded NOMA (PC-NOMA) scheme is proposed, whereby the NOMA channel is decomposed into a series of binary-input channels under a two-stage channel polarization transform. In the first stage, the NOMA channel is divided into a group of user synthesized channels by using the multi-level coding structure. In the second stage, based on the structure of bit-interleaved code modulation, user synthesized channels are further decomposed into binary polarized channels. Then, a joint successive cancellation decoding scheme is given to construct the multiuser receiver of PC-NOMA. Finally, a low complexity search algorithm is proposed to schedule the NOMA decoding order which improves the error performance by enhanced polarization among user synthesized channels. The block error ratio performances over additive white Gaussian noise channels indicate that the proposed PC-NOMA obviously outperforms the turbo coded NOMA scheme due to the advantages of the two-stage polarization.