Derived from the regular perturbation treatment of the nonlinear Schrodinger equation, a machine learning-based scheme to mitigate the intra-channel optical fiber nonlinearity is proposed. Referred to as the perturbation theory-aided (PA) learned digital back-propagation (LDBP), the proposed scheme constructs a deep neural network (DNN) in a way similar to the split-step Fourier method: linear and nonlinear operations alternate. Inspired by the perturbation analysis, the intra-channel cross-phase modulation term is conveniently represented by matrix operations in the DNN. The introduction of this term in each nonlinear operation considerably improves the performance, as well as enables the flexibility of PA-LDBP by adjusting the numbers of spans per step. The proposed scheme is evaluated by numerical simulations of a single carrier optical fiber communication system operating at 32 Gbaud with 64-quadrature amplitude modulation and 20*80 km transmission distance. The results show that the proposed scheme achieves approximately 3.5 dB, 1.8 dB, 1.4 dB, and 0.5 dB performance gain in terms of Q2 factor over the linear compensation, when the numbers of spans per step are 1, 2, 4, and 10, respectively. Two methods are proposed to reduce the complexity of PALDBP, i.e., pruning the number of perturbation coefficients and chromatic dispersion compensation in the frequency domain for multi-span per step cases. Investigation of the performance and complexity suggests that PA-LDBP attains improved performance gains with reduced complexity when compared to LDBP in the cases of 4 and 10 spans per step.
This letter investigates the reconfigurable intelligent surface (RIS)-assisted multiple-input single-output (MISO) wireless system, where both half-duplex (HD) and full-duplex (FD) operating modes are considered together, for the first time in the literature. The goal is to maximize the rate by optimizing the RIS phase shifts. A novel deep reinforcement learning (DRL) algorithm is proposed to solve the formulated non-convex optimization problem. The complexity analysis and Monte Carlo simulations illustrate that the proposed DRL algorithm significantly improves the rate compared to the non-optimized scenario in both HD and FD operating modes using a single parameter setting. Besides, it significantly reduces the computational complexity of the downlink HD MISO system and improves the achievable rate with a reduced number of steps per episode compared to the conventional DRL algorithm.
Reconfigurable intelligent surface (RIS) is considered as a revolutionary technology for future wireless communication networks. In this letter, we consider the acquisition of the time-varying cascaded channels, which is a challenging task due to the massive number of passive RIS elements and the small channel coherence time. To reduce the pilot overhead, a deep learning-based channel extrapolation is implemented over both antenna and time domains. We divide the neural network into two parts, i.e., the time-domain and the antenna-domain extrapolation networks, where the neural ordinary differential equations (ODE) are utilized. In the former, ODE accurately describes the dynamics of the RIS channels and improves the recurrent neural network's performance of time series reconstruction. In the latter, ODE is resorted to modify the relations among different data layers in a feedforward neural network. We cascade the two networks and jointly train them. Simulation results show that the proposed scheme can effectively extrapolate the cascaded RIS channels in high mobility scenario.
In a time-varying massive multiple-input multipleoutput (MIMO) system, the acquisition of the downlink channel state information at the base station (BS) is a very challenging task due to the prohibitively high overheads associated with downlink training and uplink feedback. In this paper, we consider the hybrid precoding structure at BS and examine the antennatime domain channel extrapolation. We design a latent ordinary differential equation (ODE)-based network under the variational auto-encoder (VAE) framework to learn the mapping function from the partial uplink channels to the full downlink ones at the BS side. Specifically, the gated recurrent unit is adopted for the encoder and the fully-connected neural network is used for the decoder. The end-to-end learning is utilized to optimize the network parameters. Simulation results show that the designed network can efficiently infer the full downlink channels from the partial uplink ones, which can significantly reduce the channel training overhead.
In this paper, we propose a deep reinforcement learning (DRL) approach for solving the optimisation problem of the network's sum-rate in device-to-device (D2D) communications supported by an intelligent reflecting surface (IRS). The IRS is deployed to mitigate the interference and enhance the signal between the D2D transmitter and the associated D2D receiver. Our objective is to jointly optimise the transmit power at the D2D transmitter and the phase shift matrix at the IRS to maximise the network sum-rate. We formulate a Markov decision process and then propose the proximal policy optimisation for solving the maximisation game. Simulation results show impressive performance in terms of the achievable rate and processing time.
This work does the statistical quality-of-service (QoS) analysis of a block-fading device-to-device (D2D) link in a multi-tier cellular network that consists of a macro-BS (BSMC) and a micro-BS (BSmC) which both operate in full-duplex (FD) mode. For the D2D link under consideration, we first formulate the mode selection problem-whereby D2D pair could either communicate directly, or, through the BSmC, or, through the BSMC-as a ternary hypothesis testing problem. Next, to compute the effective capacity (EC) for the given D2D link, we assume that the channel state information (CSI) is not available at the transmit D2D node, and hence, it transmits at a fixed rate r with a fixed power. This allows us to model the D2D link as a Markov system with six-states. We consider both overlay and underlay modes for the D2D link. Moreover, to improve the throughput of the D2D link, we assume that the D2D pair utilizes two special automatic repeat request (ARQ) schemes, i.e., Hybrid-ARQ (HARQ) and truncated HARQ. Furthermore, we consider two distinct queue models at the transmit D2D node, based upon how it responds to the decoding failure at the receive D2D node. Eventually, we provide closed-form expressions for the EC for both HARQ-enabled D2D link and truncated HARQ-enabled D2D link, under both queue models. Noting that the EC looks like a quasi-concave function of r, we further maximize the EC by searching for an optimal rate via the gradient-descent method. Simulation results provide us the following insights: i) EC decreases with an increase in the QoS exponent, ii) EC of the D2D link improves when HARQ is employed, iii) EC increases with an increase in the quality of self-interference cancellation techniques used at BSmC and BSMC in FD mode.
Channel estimation is challenging for the reconfigurable intelligence surface (RIS) assisted millimeter wave (mmWave) communications. Since the number of coefficients of the cascaded channels in such systems is closely dependent on the product of the number of base station antennas and the number of RIS elements, the pilot overhead would be prohibitively high. In this letter, we propose a cascaded channel estimation framework for an RIS assisted mmWave multiple-input multiple-output system, where the wideband effect on transmission model is considered. Then, we transform the wideband channel estimation into a parameter recovery problem and use a few pilot symbols to detect the channel parameters by the Newtonized orthogonal matching pursuit algorithm. Moreover, the Cramer-Rao lower bound on the channel estimation is introduced. Numerical results show the effectiveness of the proposed channel estimation scheme.
Reconfigurable intelligent surface (RIS) is considered as a revolutionary technology for future wireless communication networks. In this letter, we consider the acquisition of the cascaded channels, which is a challenging task due to the massive number of passive RIS elements. To reduce the pilot overhead, we adopt the element-grouping strategy, where each element in one group shares the same reflection coefficient and is assumed to have the same channel condition. We analyze the channel interference caused by the element-grouping strategy and further design two deep learning based networks. The first one aims to refine the partial channels by eliminating the interference, while the second one tries to extrapolate the full channels from the refined partial channels. We cascade the two networks and jointly train them. Simulation results show that the proposed scheme provides significant gain compared to the conventional element-grouping method without interference elimination.
The novel concept of simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) is investigated, where incident signals can be transmitted and reflected to users located at different sides of the surface. In particular, the fundamental coverage range of STAR-RIS aided two-user communication networks is studied. A sum coverage range maximization problem is formulated for both non-orthogonal multiple access (NOMA) and orthogonal multiple access (OMA), where the resource allocation at the access point and the transmission and reflection coefficients at the STAR-RIS are jointly optimized to satisfy the communication requirements of users. For NOMA, we transform the non-convex decoding order constraint into a linear constraint and the resulting problem is convex, which can be optimally solved. For OMA, we first show that the optimization problem for given time/frequency resource allocation is convex. Then, we employ the one dimensional search-based algorithm to obtain the optimal solution. Numerical results reveal that the coverage can be significantly extended by the STAR-RIS compared with conventional RISs.
The steering dynamics of re-configurable intelligent surfaces (RIS) have hoisted them to the front row of technologies that can be exploited to solve skip-zones in wireless communication systems. They can enable a programmable wireless environment, turning it into a partially deterministic space that plays an active role in determining how wireless signals propagate. However, RIS-based communication systems' practical implementation may face challenges such as noise generated by the RIS structure. Besides, the transmitted signal may face a double-fading effect over the two portions of the channel. This article tackles this double-fading problem in near-terrestrial free-space optical (nT-FSO) communication systems using a RIS module based upon liquid-crystal (LC) on silicon (LCoS). A doped LC layer can directly amplify a light when placed in an external field. Leveraging on this capacity of a doped LC, we mitigate the double-attenuation faced by the transmitted signal. We first revisit the nT-FSO power loss scenario, then discuss the direct-light amplification, and consider the system performance. Results show that at 51 degrees of the incoming light incidence angle, the proposed LCoS design has minimal RIS depth, implying less LC's material. The performance results show that the number of bit per unit bandwidth is upper-bounded and grows with the ratio of the sub-links distances. Finally, we present and discuss open issues to enable new research opportunities towards the use of RIS and amplifying-RIS in nT-FSO systems.