Journal of Communications and Information Networks
Print ISSN 2096-1081 CN 10-1375/TN
Editor-in-Chief: Wei Zhang
This paper studies the fundamental limit of semantic communications over the discrete memoryless channel. We consider the scenario to send a semantic source consisting of an observation state and its corresponding semantic state, both of which are recovered at the receiver. To derive the performance limitation, we adopt the semantic rate-distortion function (SRDF) to study the relationship among the minimum compression rate, observation distortion, semantic distortion, and channel capacity. For the case with unknown semantic source distribution, while only a set of the source samples is available, we propose a neural-network-based method by leveraging the generative networks to learn the semantic source distribution. Furthermore, for a special case where the semantic state is a deterministic function of the observation, we design a cascade neural network to estimate the SRDF.For the case with perfectly known semantic source distribution,we propose a general Blahut-Arimoto(BA)algorithm to effectively compute the SRDF.Finally,experimental results validate our proposed algorithms for the scenarios with ideal Gaussian semantic source and some practical datasets.
Deep learning enables real-time resource allocation for ultra-reliable and low-latency communications (URLLC), one of the major use cases in the next-generation cellular networks. Yet the high training complexity and weak generalization ability of neural networks impede the practical use of the learning-based methods in dynamic wireless environments. To overcome these obstacles, we propose a parameter generation network(PGN)to efficiently learn bandwidth and power allocation policies in URLLC. The PGN consists of two types of fully-connected neural networks (FNNs). One is a policy network, which is used to learn a resource allocation policy or a Lagrangian multiplier function. The other type of FNNs are hypernetworks, which are designed to learn the weight matrices and bias vectors of the policy network. Only the hypernetworks require training.Using the well-trained hypernetworks,the policy network is generated through forward propagation in the test phase. By introducing a simple data processing, the hypernetworks can well learn the weight matrices and bias vectors by inputting their indices, resulting in low training cost. Simulation results demonstrate that the learned bandwidth and power allocation policies by the PGNs perform very close to a numerical algorithm.Moreover,the PGNs can be well generalized to the number of users and wireless channels, and are with significantly lower memory costs,fewer training samples,and shorter training time than the traditional learning-based methods.
Low-earth orbit (LEO) satellite networks ignite global wireless connectivity. However, signal outages and co-channel interference limit the coverage in traditional LEO satellite networks where a user is served by a single satellite. This paper explores the possibility of satellite cooperation in the downlink transmissions.Using tools from stochastic geometry, we model and analyze the downlink coverage of a typical user with satellite cooperation under Nakagami fading channels. Moreover, we derive the joint distance distribution of cooperative LEO satellites to the typical user.Our model incorporates fading channels, cooperation among several satellites, satellites’ density and altitude, and co-channel interference. Extensive Monte Carlo simulations are performed to validate analytical results. Simulation and numerical results suggest that coverage with LEO satellites cooperation considerably exceeds coverage without cooperation. Moreover,there are optimal satellite density and satellite altitude that maximize the coverage probability, which gives valuable network design insights.
In order to increase the capacity of future satellite communication systems, faster-than-Nyquist (FTN) signaling is increasingly considered. Existing methods for compensating for the high power amplifier (HPA) nonlinearity require perfect knowledge of the HPA model. To address this issue, we analyze the FTN symbol distribution and propose a complex-valued deep neural network (CVDNN) aided compensation scheme for the HPA nonlinearity,which does not require perfect knowledge of the HPA model and can learn the HPA nonlinearity during the training process. A model-driven network for nonlinearity compensation is proposed to further enhance the performance.Furthermore,two training sets based on the FTN symbol distribution are designed for training the network.Extensive simulations show that the Gaussian distribution is a good approximation of the FTN symbol distribution. The proposed model-driven network trained by employing a Gaussian distribution to approximate an FTN signaling can achieve a performance gain of 0.5 dB compared with existing methods without HPA’s parameters at the receiver. The proposed neural network is also applicable for non-linear compensation in other systems,including orthogonal frequency-division multiplexing(OFDM).
Due to the gradual scarcity of spectrum resources, orbital angular momentum (OAM) technology has been proposed and developed continuously to broaden channel capacity. To solve this problem, some ultra-wideband reflective phase metasurface antennas working in millimeter band are designed to generate high purity vortex waves which carry OAM. Based on the Pancharatnam-Berry (PB) phase concept, the unit cell is composed of a metasurface, dielectric, metal grounding layer. Through the optimization design of the unit structure parameters,the reflected wave efficiency can be as high as 95% when covering 2π rotating phase, which realizes the basic requirements of PB phase concept and the relative bandwidth of 116%. Then the metasurface arrays are arranged according to vortex wave generation formula and phase compensation principle.In the 25 GHz to 35 GHz frequency wide band,integer(l=±1,±2,±3) decimal(l=±1.5)and high-mode(l=±8)OAM vortex beams are generated, respectively. The OAM purity analysis shows that the antennas can generate millimeter wave OAM beams with high purity in a wide band range, and with a maximum gain of up to 23.6 dBi.
This paper introduces an innovative approach to address the trajectory optimization challenge for cellular-connected unmanned aerial vehicles (UAVs) operating in three-dimensional(3D)space. In most cases, optimizing UAV trajectories necessitates ensuring reliable network connectivity. However, achieving dependable connectivity in 3D space poses a significant challenge due to terrestrial base stations primarily designed for ground users. Additionally, UAVs possess network information only for the areas they have visited, with global network information being inaccessible. To address this issue, we propose a collaborative approach in which multiple UAVs create a global model of outage probability using federated learning, enabling more precise and effective trajectory design. Building upon the constructed global information, we conduct the trajectory design. Initially, we introduce A-star(A*)algorithm for trajectory design in small-scale scenarios. Nevertheless, recognizing the limitations of A* algorithm in large-scale scenarios, we further introduce improved rapidly-exploring random trees (RRTs) algorithm for weighted path optimization. Simulation results are provided to validate the effectiveness of the proposed algorithms.
This paper analyses the performance of fullduplex(FD)dual-hop vehicular cooperative network with decode-and-forward (DF) protocol. At FD relay nodes, we examine the effects of non-linear hybrid power time splitting (PTS) based energy harvesting (EH). All three nodes—source (S), relay (R), and destination (D)—are assumed to be moving vehicles. The expressions for the system’s outage probability (OP) over double (cascaded) Rayleigh fading channels are derived. We also analyse the impact of residual self-interference (RSI) caused at FD relay on system’s performance. We compare the performance of system with two relay selection techniques, namely,maximum channel gain-based(Max-G)relay selection and minimum RSI-based(Min-SI)relay selection. This paper considers the joint effect of time splitting ratio and self-interference cancellation (SIC) level to find the optimum EH duration. Additionally, the effect of time splitting ratio and average signal-to-noise ratio(SNR)on outage and throughput performance of the system are also investigated in this paper. The derived expressions are validated through Monte Carlo simulations.
This paper introduces a deep learning(DL) algorithm for estimating doubly-selective fading channel and detecting signals in orthogonal frequency division multiplexing (OFDM) communication systems affected by hardware impairments (HIs). In practice, hardware imperfections are present at the transceivers, which are modeled as direct current(DC) offset, carrier frequency offset (CFO), and in-phase and quadrature-phase (IQ) imbalance at the transmitter and the receiver in OFDM system. In HIs, the explicit system model could not be mathematically derived, which limits the performance of conventional least square (LS) or minimum mean square error (MMSE) estimators. Thus, we consider time–frequency response of a channel as a 2D image, and unknown values of the channel response are derived using known values at the pilot locations with DL-based image super-resolution,and image restoration techniques. Further, a deep neural network (DNN) is designed to fit the mapping between the received signal and transmit symbols, where the number of outputs equals to the size of the modulation order. Results show that there are no significant effects of HIs on channel estimation and signal detection in the proposed DL-assisted algorithm. The proposed DL-assisted detection improves the OFDM performance as compared to the conventional LS/MMSE under severe HIs.
