Semantic communication has emerged as a promising paradigm in next-generation communication systems, leveraging advanced artificial intelligence (AI) models to extract and transmit semantic representations for efficient information exchange. However, the reliability of received information is often compromised by unpredictable semantic noise, such as corruptions or distortions in the transmitted representations. Traditional methods typically rely on adversarial training with artificially injected noise to improve robustness. Yet, these approaches suffer from limited adaptability to varying noise conditions and incur considerable computational overhead during training. To address these challenges, this paper introduces Semantic communication with High-and-Low Frequency Decomposition (Se-HiLo), a novel noise-resilient scheme designed for image transmission. Se-HiLo integrates a Finite Scalar Quantization (FSQ) module that enhances robustness by constraining encoded representations within predefined discrete spaces, thereby eliminating the need for adversarial training. While FSQ strengthens resistance to noise, it inherently limits representational expressiveness. To mitigate this trade-off, Se-HiLo further incorporates a transformer-based high-and-low frequency decomposition module that separates image representations into distinct frequency components and encodes them into independent FSQ spaces, thus preserving semantic diversity and expressiveness. Extensive experiments validate that Se-HiLo significantly improves noise robustness and maintains accurate semantic communication across a wide range of noise environments.

The high flexibility and mobility enable unmanned aerial vehicles (UAVs) playing a pivotal role in gathering/relaying data in emergency communication. However, communication resources are scarce in emergency communication scenarios, while Semantic Communication (SC) holds promise for overcoming current communication bottlenecks by reducing data transmission volumes. In this paper, we investigate joint resource optimization and trajectory optimization in UAV-assisted emergency semantic communication networks to further enhance communication efficiency. Firstly, considering the importance of data freshness in emergency communication scenarios, we propose a semantic metric, termed semantic transmission efficiency, which integrates both semantic similarity and semantic Age of Information (AoI). Then, we jointly optimize user bandwidth, transmission power, and UAV flight trajectories with the objective of maximizing the long-term semantic transmission efficiency. To solve the joint optimization problem, we propose the Diffusion-Deep Deterministic Policy Gradient (Diffusion-DDPG) algorithm. By leveraging the capability of generative diffusion model to capture data distributions, the algorithm effectively identifies optimal solutions to complex optimization problems, thereby enhancing the exploration ability of agent. Comparative experiments demonstrate that Diffusion-DDPG effectively balances semantic similarity and semantic AoI within semantic transmission efficiency. Particularly in resource-constrained scenarios, it shows superior convergence speeds.

Grouping vehicles into platoons is a promising cooperative driving scenario to enhance the traffic safety and capacity of future vehicular networks. However, fast changing channel conditions in multi-platoon vehicular networks cause tremendous uncertainty for resource allocation. In addition, the unprecedented proliferation of various emerging vehicle-to-infrastructure (V2I) applications may result in some service demands with conflicting quality of experience. In this paper, we formulate a multi-objective resource allocation problem, which maximizes the transmission success ratio of intra-platoon communications and the mean opinion score (MOS) of V2I communication links. To efficiently solve this multi-objective optimization problem, we resort to a deep reinforcement learning (DRL) framework. Specifically, we divide it into a set of scalar optimization subproblems based on the weighted sum approach and model each one as a partially observable stochastic game (P-OSG), where each platoon acts as an agent and the actions taken by all platoons correspond to the resource allocation solution. We further propose a contribution-based dual-clip proximal policy optimization (CD-PPO) algorithm to deal with each subproblem, which is a DRL algorithm based on the actor-critic framework. The network parameters of all subproblems are then optimized collaboratively by using the proposed training algorithm and the neighborhood parameter transfer strategy. The desired Pareto front is obtained when all subproblems are solved. Simulation results reveal that the proposed algorithm can outperform other algorithms in terms of the MOS and transmission success ratio.

Digital twin technology can be combined with wire-less virtual reality (VR) to provide workers with an immersive virtual industrial manufacturing environment. In this paper, we consider a scenario of utilizing computing resources at the edge servers to render the virtual representation of physical industrial scenes for workers. In order to provide an immersive experience, the rendered virtual representation in the form of high-quality video needs to be delivered with low latency. The quality of delivered contents is associated with rendering resources and content bitrate, and high-quality contents result in large rendering and transmission delay. The joint resource allocation problem has been formulated, allocating limited bandwidth and edge computing resources, and determining the bitrate of content to achieve low latency and high content quality, in which two objectives are conflicting. To solve this problem, we employ a deep reinforcement learning (DRL) based multi-objective optimization approach which decomposes the original problem into a series of subproblems. Instead of training the DRL model for each subproblem with equal number of episodes, an adaptive subproblem model training scheme has been proposed which terminates the training of subproblem model early based on the model convergence to improve the overall training efficiency. Simulation results verify the effectiveness of the proposed resource allocation strategy in terms of content quality and delay.

Panoramic or 360-degree virtual reality videos have high resolution, frame rate, and visual quality that demand efficient coding. Although a user watching a 360-degree video can switch viewing angles, only a portion of the video in the user’s Field of View (FoV) is displayed at any time. In this paper, we propose an FoV-based coding scheme for 360-degree videos, which allocates more bits to tiles of the predicted FoV area than other tiles. Taking possible FoV prediction error into account, the proposed scheme aims to minimize the expected weighted distortion of the FoV region, where different weights are given to tiles at different locations representing the influence of projection from spherical domain to the 2D plane. Accordingly, an adaptive tile-level quantization parameter (QP) selection scheme is derived. Simulation results demonstrate the effectiveness of the proposed scheme.

With the rapid development of deep learning, more and more images and videos are used for machine analysis. The amount of images and video content consumed by machines has exceeded that of humans. However, the traditional image and video coding schemes are designed for human vision system, where information that is vital to machine vision, e.g., boundary of a salient object, may not be preserved during compression. In this paper, based on high efficiency video coding (HEVC) intra coding, we propose an image compression scheme for machines using boundary-enhanced saliency. Using image classification as an example task, Grad-CAM, a deep learning visualization method, is used to interpret classification results to generate a pixel-level saliency map for each image. Object segmentation and edge detection are then performed to generate boundary map of the salient object. With boundary-enhanced saliency map, we derive a coding tree unit (CTU)-level QP adjustment scheme, where more bits are allocated to salient regions of image concerning machine vision. Experimental results show that, compared with HEVC, our proposed scheme could achieve up to 29.94% and 31.53% bitrate saving with the same TOP 1 and TOP5 accuracy performance in image classification, respectively.

360° video streaming provides an immersive viewing experience, but demands high bitrate transmission with low latency. To accommodate the fluctuating network condition and different user requirements, 360° videos are streamed with different quality versions, which makes the bandwidth more demanding. Edge servers with limited resources can be utilized to cache and transcode video contents near users to support 360° video streaming. The edge caching strategy and the transcoding strategy are correlated, as the cached contents limit the transcoding operations, while transcoding extends the use of cached contents at the cost of delay and provides new contents for possible caching. In this article, we consider the problem of collaborative edge transcoding and caching in an edge cluster for tile-based 360° video streaming, which aims to jointly allocate storage and computing resources to reduce quality mismatch level, delay, and transmission cost. The formulated problem is complicated considering dynamic viewers’ Field of View (FoV), video popularity, different quality versions, future user requests, and possible FoV prediction errors. To solve this problem, the process of collaborative transcoding and caching is modeled as a Markov decision process (MDP). Then, a model-free deep reinforcement learning approach, the deep deterministic policy gradient (DDPG), is used to obtain the caching replacement and computing power allocation strategy. Since the action space for caching replacement and power allocation is large, an FoV-guided scheme has been designed to speed the training of the DDPG agent. The simulation results with real viewing traces show that the proposed scheme can increase the cache hit ratio, reduce transmission cost, and improve users’ viewing experience effectively.

Wireless virtualization has become a key concept in future cellular networks which can provide multiple virtualized wireless networks for different mobile virtual network operators (MVNOs) over the same physical infrastructure. Resource allocation is a main challenging issue in wireless virtualization for which auction approaches have been widely used. However, for most existing auction-based allocation schemes, the objective is to maximize the social welfare (i.e., the sum of all valuations of winning bidders) due to its simplicity. While in reality, MVNOs are more interested in maximizing their own revenues (i.e., received payments from auction winners). However, the revenue-optimal auction problem is much more complex since the payment price is unknown before calculation. In this paper, we aim to design a revenue-optimal auction mechanism for resource allocation in wireless virtualization. Considering the complexity, deep learning techniques are applied. Specifically, we construct a multi-layer feed-forward neural network based on the analysis of optimal auction design. The neural network adopts users’ bids as the input and the allocation rule and conditional payment rule for the users as the output. The proposed auction mechanism possesses several desirable properties, e.g., individual rationality, incentive compatibility and budget constraint. Finally, simulation results demonstrate the effectiveness of the proposed scheme. Comparing with second-price auction and optimization-based schemes, the proposed scheme can increase the revenue by 10 and 30 percent on average, for single MVNO and multi-MVNO cases, respectively.

Scalable video coding (SVC) has received much attention for video transmission over wireless due to its flexibility. However, most previous work only considered SVC video streaming from a single base station (BS). At present, the densification of BSs enables a user equipment (UE) to connect to multiple BSs in ultra-dense networks (UDNs). In this paper, we consider the problem of SVC video streaming in a UDN, which allows different layers of a video block to be downloaded from different BSs. An optimization problem is formulated aiming to maximize the quality of experience (QoE) of users by selecting the optimal connection strategy and optimal number of video layers. Considering the complexity, to efficiently solve the problem in a distributed manner, the problem of choosing connection strategy is formulated as a multi-agent multi-armed bandit (MA-MAB) problem with only few information exchange. Each user can adapt its connection strategy in a distributed self-learning system. To obtain the optimal arm for the MA-MAB problem, we propose a multi-user arm decision algorithm. To avoid large computation and handover costs, we adopt the same connection strategy for the entire video sequence. Then for each video block, with the given connection strategy, the number of video layers is adjusted adaptively according to dynamic network conditions. Finally, based on the above designs, we provide the SVC-based video downloading scheme to obtain an approximate optimal solution to the original optimization problem. Extensive simulations and comparisons show the feasibility and superiority of the proposed scheme.

In a PoW-based blockchain network, miners participate in a block-discovery racing game for financial incentives. As the total computing power becomes overwhelming, miners join in the mining pool which combines the scattered computing power to win a stable profit. Miners in the same mining pool work as a team and once they successfully mine a valid block, mining pool plays a role in distributing the payoff to miners according to its reward sharing mechanism. Specifically, two main reward sharing strategies: Pay-Per-Share (PPS) and Pay-Per-Last-NShare (PPLNS) are considered. In the mining system model, a miner can choose to join a pool and adapt the selection for improving the expected reward. And we formulate the dynamic pool selection problem as an evolutionary game. We consider the required hash rate, network delay and reward sharing strategy as the main factors which affect the choice of miners. Evolutionary stable equilibrium (ESS) is considered to be the solution, and we conduct theoretical analysis on the existence and stability of the ESS for a case of two mining pools. A low complexity distributed algorithm is proposed for miners to reach the equilibrium. Numerical results show the evolution of miners and demonstrate the theoretical findings of our study.

The mobile blockchain has been recognized as an emerging solution to address the security and privacy issues in a mobile application system. The mining process in mobile blockchain requires high computing resources which could overwhelm that which mobile devices can offer. In this case, mobile edge computing servers (MESs) can be involved to offer computing services to miners in mobile blockchain. Note that the resources of MESs are also limited; MESs could further request resources from the cloud computing server (CCS). Accordingly, the issue of hierarchical computing resource allocation arises. In this paper, we first consider a simple case with single-seller multiple buyers and a hierarchical single-seller multibuyer combinatorial auction model is proposed to solve this problem, based on which efficient and truthful frameworks are provided. We then extend the model to consider multiple CCSPs and propose a hierarchical multiple-seller multiple-buyer combinatorial auction model. For both models, the winner determination problems are formulated and computationally tractable algorithms are proposed. Also, pricing schemes are proposed to ensure the property of incentive compatibility and individual rationality. Finally, we evaluate the proposed schemes via simulations.

As a decentralized ledger to record all transaction information, blockchain can be applied to address the security and privacy issues in mobile application system. We term the blockchain applied to mobile applications as mobile blockchain. The mining process in mobile blockchain requires high computing capacity and energy which could overwhelm that mobile devices can offer. In this case, mobile edge computing servers (MESs) can be involved to offer computation services to miners in mobile blockchain. Note that the resources of MESs are also limited, MESs could further request resources from the cloud computing server (CCS). Accordingly, in this paper, both mobile edge computing and cloud computing are considered to support the mobile blockchain applications which makes the problem a hierarchical one. Naturally, the issue of hierarchical resource allocation arises. And a hierarchical combinatorial auction model is proposed to solve this problem, based on which an efficient and truthful framework is provided. Specifically, we formulate winner determination problems (WDPs) for mobile edge computing service providers and cloud computing service provider, and computationally tractable algorithms to address both problems are proposed. Finally, numerical analysis shows the effectiveness of the proposed scheme.

User association is indispensable for the operation of wireless network and has critical impacts on system performance. For most existing work, user associations are typically coupled, which require a user equipment (UE) to associate with the same base station (BS) in uplink (UL) and downlink (DL). However, wireless networks are becoming heterogeneous and densifying, which generates intrinsic distinctions (transmission power, data traffic and backhaul capacity etc.) between UL and DL. Accordingly, coupled association may no longer be optimal. In this work, we explore decoupled user association in full-duplex ultra-dense networks (UDNs), which allows a UE to associate with different BSs in UL and DL respectively. Furthermore, to fully exploit the benefits of UDNs, multiple association, referring to associating a UE with multiple BSs, is jointly adopted in UL and DL. Considering the dynamic and complicated association process, an evolutionary game (EG) is formulated, where UEs are players, and their strategies are association selections in UL/DL. Particularly, evolutionary equilibrium is viewed as the stable solution to the formulated problem. Moreover, an EG-based algorithm with low complexity is proposed for decoupled multiple association. Numerical results validate the convergence of the proposed algorithm for strategy adoption. Besides, the impacts of information exchange delay and learning rate are investigated for providing a better association decision.

User association is a crucial factor that affects the performance of wireless networks. In current cellular networks, user association is typically coupled, which means an user equipment (UE) must associate with the same base station (BS) in uplink (UL) and downlink (DL). For single-tier wireless networks, such mechanism is simple and effective. However, in heterogeneous ultra-dense networks (UDNs), there are distinct differences in transmission power, data traffic and channel quality etc., for which coupled association could restrict the performance of system. To cope with it, the concept of decoupled UL-DL (DUDe) association has been introduced recently, which enables a UE to associate with different BSs in UL and DL. In this paper, we investigate decoupled UL-DL user association in UDNs. Considering the existence of asymmetric information (i.e., channel gains and intercell interferences), which can be seen as the private information for UE, we propose a contract-theoretic user association approach. Particularly, we model the decoupled association process as a monopoly labor market, where BSs act as employers and offer contracts to employees (i.e., UEs). The contract items cover the available associated bandwidths, transmitted powers and corresponding prices. Then BS broadcasts these drafted contract information, and UE selects to sign the optimal contract by considering her own demands. Numerical results show significant superiorities of DUDe than coupled UL-DL association in perspective of nodes utilities and social surplus, and compared with the existing user association methods, contract-theoretic approach shows a certain improvement in performance.

The new trends in network denisification, heterogeneity, and the introduction of new techniques (e.g., full-duplex) introduce new challenges for user association. For most existing user association schemes, the uplink (UL) and downlink (DL) access are coupled. That is, a user equipment (UE) is associated with the same BS for UL and DL transmissions. However, in ultra-dense heterogeneous networks (UDNs), due to the large disparities among base stations in different tiers and among uplink and downlink, the coupled UL-DL user association will limit the system performance. In this paper, we propose a novel concept of decoupled multiple association for user association in UDNs, which allows a UE to be associated with multiple base stations (BSs) for UL and DL in a decoupled manner. Furthermore, the context information of UEs is considered when making association decisions. Specifically, a decoupled multiple association matching game is formulated and a context-aware swap matching algorithm is proposed. The proposed scheme could attain higher data rates and could satisfy the quality of service (QoS) requirements of different UEs. Additionally, it could overcome the back-haul limitation of individual BSs. We compare the proposed scheme with three other association schemes, and the simulation results show significant performance gains of our proposed scheme in UDNs.

Class imbalance refers to the instance where the number of training samples for the majority classes is far more than that of the minority classes (relative imbalance), and the quality of training samples for the minority classes is inferior to that of the majority classes (absolute imbalance), which are further complicated by other imbalance factors, e.g., data overlapping. Video background subtraction aims to classify each pixel into two classes: foreground and background. This paper first reveals that background subtraction is a class imbalance problem, where the foreground and background are the minority and majority classes, respectively. By exploring spatial and temporal correlation inherent in video data, we present an imbalance compensation framework for background subtraction, which consists of two sequential modules, imbalance-compensated bilayer modeling, and imbalance-compensated Bayesian classification. In the first module, spatio-temporal oversampling (SOS) and selective downsampling (SDS) are proposed to compensate the imbalance at data level. SOS attempts to synthesize representative samples appended to the minority sample set, while SDS selectively deletes a number of majority samples in data overlapping areas. The rebalanced samples are then used to learn a bilayer model. In the second module, novel cost functions are proposed to compensate the effect of class imbalance at algorithm level. The cost functions are based on imbalance measurement, and used to construct the prior term in the Bayesian classification scheme. Experiments are conducted on public databases to demonstrate the effectiveness of the proposed method.

In this paper, H.264/AVC primary and redundant slices interlaced multiple description video coding (PRSI-MDVC) is studied due to its high coding efficiency and ease of constructing multiple descriptions by interleaving primary and redundant slices, where the problem of optimal description generation in the rate-distortion sense is addressed. Optimal description generation requires minimization of end-to-end distortion consisting of both source coding distortion determined by quality of primary slices and channel distortion associated with redundant slice coding, subject to a rate constraint. The relevant existing works on the description generation mainly focus on the estimation of channel distortion to determine the amount of inserted redundancy (quality of redundant slices) but ignore the source distortion estimation for the quality of primary slices, thus a comprehensive end-to-end rate-distortion optimization is still unavailable. In this paper, we attack the minimization of the end-to-end distortion by fully exploring temporal coding dependency. Specifically, on one hand, a most recently developed source distortion temporal propagation model is employed to determine coding options of primary slices in the PRSI-MDVC. On the other hand, channel distortion estimation is mainly concerned with the mismatch error estimation when primary slices are lost. Unlike the existing channel distortion estimation approach under the asymptotic fine quantization assumption which is not valid in most practical cases (e.g., at low or medium coding rates), we develop a novel and more feasible estimation scheme, based on which coding parameters of the redundant slices can be better determined. Simulation results show the effectiveness of the proposed frame-level rate-distortion optimized description generation scheme compared with the relevant approaches.

Multiple description coding (MDC) is an effective technique to combat transmission loss over unreliable lossy networks. Network coding which allows coding at the intermediate nodes in the network promises to increase throughput of the whole network and better utilizes network resources. To provide both robustness and efficiency for image multicast over wireless ad hoc networks, a scheme based on joint multiple description coding (MDC) and network coding is proposed in this paper. Multiple description lattice vector quantization (MDLVQ) is employed to encode an image at the source node, and different descriptions are transmitted on an individually or mixed base. Linear network coding is used to mix selected packets at the intermediate nodes provided that corresponding receivers can decode the mixed packet. At the destination nodes, received original packets and mixed packets can result in a reproduction with certain quality. Experimental results validate that the proposed scheme can benefit the image transmission with better reconstructed image quality, lower failure rate and less energy consumptions.

In this paper, we propose a novel 2-D layered multiple description coding (2DL-MDC) for error-resilient video transmission over unreliable networks. The proposed 2DL-MDC scheme allocates multiple description sub-bitstreams of a 2-D scalable bitstream to two network paths with unequal loss rates. We formulate the 2-D scalable rate-distortion problem and derive the expected distortion for the proposed scheme. To minimize the end-to-end distortion given the total rate budget and packet loss probabilities, we need to optimally allocate source and channel rates for each hierarchical sublayer of the scalable bitstream. The conventional Lagrangian multiplier method can be utilized to solve this problem but with overwhelming computational complexity. Therefore, we consider the use of the genetic algorithm to solve the rate-distortion optimization problem. The simulation results verify that the proposed method is able to achieve significant performance gain as opposed to the conventional equal rate allocation method.