Semantic communication (SemCom) has emerged as a promising paradigm for achieving unprecedented communication efficiency in sixth-generation (6G) networks by leveraging artificial intelligence (AI) to extract and transmit the underlying meanings of source data. However, deploying SemCom over digital systems presents new challenges, particularly in ensuring robustness against transmission errors that may distort semantically critical content. To address this issue, this paper proposes a novel framework, termed generative feature imputing, which comprises three key techniques. First, we introduce a spatial-error-concentration packetization strategy that spatially concentrates feature distortions by encoding feature elements based on their channel mappings—a property crucial for both the effectiveness and reduced complexity of the subsequent techniques. Second, building on this strategy, we propose a generative feature imputing method that utilizes a diffusion model to efficiently reconstruct missing features caused by packet losses. Finally, we develop a semantic-aware power allocation scheme that enables unequal error protection by allocating transmission power according to the semantic importance of each packet. Experimental results demonstrate that the proposed framework outperforms conventional approaches, such as Deep Joint Source-Channel Coding (DJSCC) and JPEG2000, under block fading conditions, achieving higher semantic accuracy and lower Learned Perceptual Image Patch Similarity (LPIPS) scores.

Semantic communications (SemComs) have emerged as a promising paradigm for joint data and task-oriented transmissions, combining the demands for both the bit-accurate delivery and end-to-end (E2E) distortion minimization. However, current joint source-channel coding (JSCC) in SemComs is not compatible with the existing communication systems and cannot adapt to the variations of the sources or the channels, while separate source-channel coding (SSCC) is suboptimal in the finite blocklength regime. To address these issues, we propose an adaptive source-channel coding (ASCC) scheme for SemComs over parallel Gaussian channels, where the deep neural network (DNN)-based semantic source coding and conventional digital channel coding are separately deployed and adaptively designed. To enable efficient adaptation between the source and channel coding, we first approximate the E2E data and semantic distortions as functions of source coding rate and bit error ratio (BER) via logistic regression, where BER is further modeled as functions of signal-to-noise ratio (SNR) and channel coding rate. Then, we formulate the weighted sum E2E distortion minimization problem for joint source-channel coding rate and power allocation over parallel channels, which is solved by the successive convex approximation. Finally, simulation results demonstrate that the proposed ASCC scheme outperforms typical deep JSCC and SSCC schemes for both the single- and parallel-channel scenarios while maintaining full compatibility with practical digital systems.

This paper considers a multi-user semantic and data communication (MU-SemDaCom) system, where a base station (BS) simultaneously serves users with different semantic and data tasks through a downlink multi-user multiple-input single-output (MU-MISO) channel. The coexistence of heterogeneous communication tasks, diverse channel conditions, and the requirements for digital compatibility poses significant challenges to the efficient design of MU-SemDaCom systems. To address these issues, we propose a multi-user adaptive source-channel coding (MU-ASCC) framework that adaptively optimizes deep neural network (DNN)-based source coding, digital channel coding, and superposition broadcasting according to the channel conditions. First, we employ a data-regression method to approximate the end-to-end (E2E) semantic and data distortions, for which no closed-form expressions exist due to the complex coupling between DNN-based source coding and channel codes. The obtained logistic formulas decompose the E2E distortion as the addition of the source and channel distortion terms, in which the logistic parameter variations are task-dependent and jointly determined by both the DNN and channel parameters. Then, based on the derived formulas, we formulate a weighted-sum E2E distortion minimization problem that jointly optimizes the source-channel coding rates, power allocation, and beamforming vectors for both the data and semantic users. Finally, an alternating optimization (AO) framework is developed, where the adaptive rate optimization is solved using the subgradient descent method, while the joint power and beamforming is addressed via the uplink-downlink duality (UDD) technique. Simulation results demonstrate that, compared with the conventional separate source-channel coding (SSCC) and deep joint source-channel coding (DJSCC) schemes that are designed for a single task, the proposed MU-ASCC scheme achieves simultaneous improvements in both the data recovery and semantic task performance.

Federated learning (FL) has emerged as a widely adopted paradigm for enabling edge learning with distributed data while ensuring data privacy. However, the traditional FL with deep neural networks trained via backpropagation can hardly meet the low-latency learning requirements in the sixth generation (6G) mobile networks. This challenge mainly arises from the high-dimensional model parameters to be transmitted and the numerous rounds of communication required for convergence due to the inherent randomness of the training process. To address this issue, we adopt the state-of-the-art principle of maximal coding rate reduction to learn linear discriminative features and extend the resultant white-box neural network into FL, yielding the novel framework of Low-Latency Federated Learning (LoLaFL) via forward-only propagation. LoLaFL enables layer-wise transmissions and aggregation with significantly fewer communication rounds, thereby considerably reducing latency. Additionally, we propose twononlinearaggregation schemes for LoLaFL. The first scheme is based on the proof that the optimal NN parameter aggregation in LoLaFL should be harmonic-mean-like. The second scheme further exploits the low-rank structures of the features and transmits the low-rank-approximated covariance matrices of features to achieve additional latency reduction. Theoretic analysis and experiments are conducted to evaluate the performance of LoLaFL. In comparison with traditional FL, the two nonlinear aggregation schemes for LoLaFL can achieve reductions in latency of over 87% and 97%, respectively, while maintaining comparable accuracies.

This paper proposes an adaptive source-channel coding (ASCC) scheme for point-to-point digital semantic communications over parallel Gaussian channels, where the deep neural network (DNN)-based semantic source coding and conventional digital channel coding are separately deployed and adaptively designed. To enable efficient adaptation between the source and channel coding, we first approximate the E2E data and semantic distortions as functions of source coding rate and bit error ratio (BER) via logistic regression, where BER is further modeled as functions of signal-to-noise ratio (SNR) and channel coding rate. Then, we formulate the weighted sum E2E distortion minimization problem for joint source-channel coding rate and power allocation over parallel channels, which is solved by the successive convex approximation. Finally, simulation results demonstrate that the proposed ASCC scheme outperforms typical separate and deep joint source-channel coding schemes while maintaining full compatibility with practical digital systems.

Semantic communications (SemCom) have emerged as a new paradigm for supporting sixth-generation applications, where semantic features of data are transmitted using artificial intelligence algorithms to attain high communication efficiencies. Most existing SemCom techniques utilize deep neural networks (DNNs) to implement analog source-channel mappings, which are incompatible with existing digital communication architectures. To address this issue, this paper proposes a novel framework of digital deep joint source-channel coding (D2-JSCC) targeting image transmission in SemCom. The framework features digital source and channel codings that are jointly optimized to reduce the end-to-end (E2E) distortion. First, deep source coding with an adaptive prior model is designed to encode semantic features according to their distributions. Second, channel coding is employed to protect encoded features against channel distortion. To facilitate their joint design, the E2E distortion is characterized as a function of the source and channel rates via the analysis of the Bayesian model and Lipschitz assumption on the DNNs. Then to minimize the E2E distortion, a two-step algorithm is proposed to control the source-channel rates for a given channel signal-to-noise ratio. Simulation results reveal that the proposed framework outperforms classic deep JSCC and mitigates the cliff and leveling-off effects, which commonly exist for separation-based approaches.

Semantic communications (SemCom) have emerged as a new paradigm for supporting sixth-generation applications with high communication efficiencies. Most existing SemCom techniques utilize deep neural networks (DNNs) to implement analog source-channel mappings, which are incompatible with existing digital communication architectures. To address this issue, this paper proposes a novel framework of digital deep joint source-channel coding ($\mathrm{D}^{2}$-JSCC) targeting image transmission in SemCom. The framework features digital source and channel codings that are jointly optimized to reduce the end-to-end (E2E) distortion. First, deep source coding with an adaptive density model is designed to efficiently extract and encode semantic features according to their different distributions. Second, channel coding is employed to protect encoded features against channel distortion. To facilitate their joint design, the E2E distortion is characterized as a function of the source and channel rates. Then to minimize the E2E distortion, we propose an efficient two-step algorithm to find the optimal trade-off between the source and channel rates for a given channel signal-to-noise ratio (SNR). Via experiments on simulating the $\mathbf{D}^{2}$-JSCC with different channel codes and real datasets, the proposed framework is observed to outperform the classic deep JSCC and separation-based approaches.

Semantic communications are expected to accomplish various semantic tasks with relatively less spectrum resource by exploiting the semantic feature of source data. To simultaneously serve both the data transmission and semantic tasks, joint data compression and semantic analysis has become a pivotal issue in semantic communications. This article proposes a deep separate source-channel coding (DSSCC) framework for the joint task and data-oriented semantic communications (JTD-SCs) and utilizes the variational autoencoder approach to solve the rate-distortion problem with semantic distortion. First, by analyzing the Bayesian model of the DSSCC framework, we derive a novel rate-distortion optimization problem via the Bayesian inference approach for general data distributions and semantic tasks. Next, for a typical application of joint image transmission and classification, we combine the variational autoencoder approach with a forward adaption scheme to effectively extract image features and adaptively learn the density information of the obtained features. Finally, an iterative training algorithm is proposed to tackle the overfitting issue of deep learning models. Simulation results reveal that the proposed scheme achieves better coding gain as well as data recovery and classification performance in most scenarios, compared to the classical compression schemes and the emerging deep joint source-channel schemes.

This paper proposes a deep separate source-channel coding (DSSCC) scheme for the semantic-aware image transmission, where image is lossily compressed and transmitted to receiver for recovery and processing certain semantic tasks. To improve the compression efficiency, the forward adaption (FA) method is incorporated into the DSSCC scheme to capture the density information of compressed features as side information. For a typical application of image classification task, we derive a novel rate-distortion optimization problem by analyzing the Bayesian model of the FA-based DSSCC framework. Then, a variational autoencoder approach is proposed to effectively compress image for semantic-aware transmission by minimizing the proposed rate-distortion problem. Simulation results reveal that the proposed FA-based DSSCC scheme achieves better image recovery and classification performance in most scenarios, compared to the classical compression schemes and the emerging deep joint source-channel schemes.

Massive machine-type communications (mMTC) are expected to support a large amount of randomly deployed users for short package transmissions. Noncoherent random access provides an efficient and practical multi-access protocol for mMTC, and also poses new challenges for the receiver design. In this paper, we leverage two well-known methods, i.e., message passing and deep learning, to jointly detect the user activity and the desired data for the noncoherent mMTC. First, by exploiting the exact distribution information of the received signal, a generalized approximate message passing (GAMP)-based algorithm is proposed, which is shown to jointly detect the user activity and the desired data by two modules: inter-user interference elimination and data detection for each user. Inspired by the two-module GAMP-based algorithm, we then propose a model-driven deep learning method, which utilizes the deep neural networks (DNNs) to approximate both the two modules. The loss function for training the DNNs is derived by formulating the two-module detection as an unconstrained optimization problem. Simulation results reveal that the proposed GAMP-based algorithm outperforms the proposed deep learning method when the channel distribution is perfectly known, while it suffers from a significant performance degradation for the case with imperfect channel distribution information.

Massive machine-type communications (mMTC) for the Internet of Things (IoT) are expected to support a large number of devices/users for short packet transmissions with low complexity and low energy consumption. By utilizing the simple while efficient noncoherent energy-based transmission scheme, this work aims to jointly detect the user activity and the desired data for mMTC. First, by exploiting the sparse characteristics of the user activity, approximation message passing (AMP) algorithm is proposed to eliminate the multi-user interference, and a denoiser is designed to minimize the mean-squared error (MSE) of the transmitted signals. Then, maximum${a}$posteriori(MAP) criterion is adopted to approximately detect the user activity and the desired data. By minimizing the symbol error probability of the above two-step algorithm, the power constellation for each user is designed, and it is shown to be asymptotically optimal as the number of the receiver antennas goes to infinity. Finally, simulation results reveal that the proposed noncoherent scheme outperforms the coherent one in the low SNR regime and for short packet transmissions.

Massive machine-type communications (mMTC) are expected to support bunches of low cost devices, which are stochastically active. By utilizing the simple while efficient noncoherent energy-based transmission scheme, this work aims to jointly detect the user activity and the data in mMTC. First, by exploiting the sparse characteristics of the user activity, approximation message passing (AMP) algorithm is proposed to suppress the multi-user interference. Then, maximum a posteriori (MAP) criterion is adopted to approximately detect the user activity and the desired data. By minimizing the symbol error probability of the above two-step algorithm, the power constellations at each user are designed. Finally, the scaling behavior of the considered system is analyzed, and it is shown that to guarantee reliable communications, the number of the receiver antennas per user should vanish in the order of O([1/log(N)]) with N being the number of the total users.

This paper considers a massive single-input multiple-output (SIMO) system, where multiple single-antenna transmitters simultaneously communicate with a receiver equipped with a large number of antennas. Different from the conventional noncoherent transceivers which require a certain level of the statistical information on the channel fading, we propose a joint transceiver design method based on machine learning, requiring a limited number of channel realizations. In the proposed method, the multiple transmitters, the channel, and the receiver are represented with a deep neural network (NN), and an autoencoder is adopted to minimize the end-to-end transmission error probability. Simulation results show that the proposed NN-based transceiver achieves lower transmission error probability in typical scenarios, and is more robust against the channel parameters variation compared with the existing methods.

This article considers the design of the transmitter and receiver in a noncoherent massive single-input and multiple-output (SIMO) system over a multipath channel, representing a typical Internet-of-Things (IoT) scenario that consists of multiple single-antenna transmitters and one receiver with a large number of antennas. In particular, the autoencoders, which consist of multiple independent neural networks (NNs), are adopted at the transmitters and the receiver and are trained jointly, while working separately. To avoid the delicate design for mitigating the intersymbol interference (ISI) caused by multipath channels, the modulation schemes at the transmitters and the demodulation rule at the receiver are learned by the NNs over a limited number of channel samples. Moreover, the relationship between the number of channel samples and the performance of the trained transceiver is analyzed. The simulation results show that the proposed method achieves a lower error probability in comparison with the conventional optimization-based methods under typical channel conditions.

In this paper, we focus on a two-transmitter and two-receiver interference channel (IC), where each transmitter sends a message to the desired receiver. Especially, the full-duplex (FD) amplify-and-forward (AF) protocol is adopted to build up the receiver cooperations. With the considered scheme, the equivalent channel model is analyzed, and the statistics of the accumulated residual interference and noise (ARIN), generated by the imperfect self interference (SI) cancellation and AF scheme, are calculated. Then, the achievable rate regions for both the single-user and joint decoding schemes are characterized by a concave-convex procedure (CCCP). Next, from the achievable rates, one-side cooperation is analyzed to explain its optimality. Simulation results show that the achievable rate regions can be improved by the proposed scheme in certain scenarios.