Image correction and rectangling are valuable tasks in practical photography systems such as smartphones. Recent remarkable advancements in deep learning have undeniably brought about substantial performance improvements in these fields. Nevertheless, existing methods mainly rely on task-specific architectures. This significantly restricts their generalization ability and effective application across a wide range of different tasks. In this paper, we introduce the Unified Rectification Framework (UniRect), a comprehensive approach that addresses these practical tasks from a consistent distortion rectification perspective. Our approach incorporates various task-specific inverse problems into a general distortion model by simulating different types of lenses. To handle diverse distortions, UniRect adopts one task-agnostic rectification framework with a dual-component structure: a Deformation Module, which utilizes a novel Residual Progressive Thin-Plate Spline (RP-TPS) model to address complex geometric deformations, and a subsequent Restoration Module, which employs Residual Mamba Blocks (RMBs) to counteract the degradation caused by the deformation process and enhance the fidelity of the output image. Moreover, a Sparse Mixture-of-Experts (SMoEs) structure is designed to circumvent heavy task competition in multi-task learning due to varying distortions. Extensive experiments demonstrate that our models have achieved state-of-the-art performance compared with other up-to-date methods.

Entering the 5G era, the mobile network operators (MNO) are facing greater challenges in providing services cost effectively than any other previous generations. The potential solutions to this are lying on the emerging trend of deep convergence of information technology (IT), communication technology (CT) and data technology (DT). In particular, the O-RAN technology, the representation of such ICDT convergence and proposed by the O-RAN ALLIANCE in 2018, is transforming Radio Access Networks towards a new paradigm featuring openness, cloudification and intelligence. O-RAN has gained huge attention from both industry and academia since its inception. In this paper, we presented the recent endeavors from China Mobile, including our deployment scenarios, various test results from open fronthaul, cloud platform to the intelligent controller. Our rich and comprehensive tests have demonstrated the viability and superiority of current O-RAN technologies. Furthermore, we also provide our deep thinking on the O-RAN future evolution in order to better serve the emerging applications such as Metaverse, cloud extended-reality (XR), extensive enterprise private 5G verticals and so on.

Network slicing (NS) can enable diverse communication services for vertical industries. In radio access network slicing, differential service level agreement (SLA) guaranteeing is an essential resource management task. Benefit from powerful data analysis capabilities, deep learning (DL) is suitable for intelligent resource management under the cases of complex constraints and time-varying states. Thus, DL has been used to manage resources for NS recently. However, the training of these DL-assisted methods is time-consuming and it is difficult to keep high SLA satisfaction rates dynamically. To address this problem, we propose a hierarchical intelligent NS resource configuration method via organically integrating NS preconfiguration models based on deep neural networks and NS reconfiguration models using multiarm bandits. A factory automation system is established to evaluate our proposed methods on different industrial services. Simulation and experimental results demonstrate that our proposed methods outperform benchmarks comprehensively.

Integrated AI and Communication have been identified as crucial elements for IMT2030 by ITU-R. The integration of communication and computing in the RAN will facilitate deploying AI technology closer to mobile devices and enable AI computing collaboration with reduced latency. This paper primarily investigates the collaboration of AI inference computation between mobile devices and base stations, with a specific focus on the joint optimization of communication and computing resources in a multi-user environment. A communication and computing integrated scheduling algorithm is proposed to ensure optimal system performance in terms of inference accuracy and computation latency. Simulation results validate the effectiveness of the algorithm, demonstrating superior performance compared to the baseline algorithm.

Ultra-dense network deployment has been proposed as a key technique for achieving capacity goals in the fifth-generation (5G) mobile communication system. However, the deployment of smaller cells inevitably leads to more frequent handovers, thus making mobility management more challenging and reducing the capacity gains offered by the dense network deployment. In order to fully reap the gains for mobile users in such a network environment, we propose an intelligent dual connectivity mechanism for mobility management through deep learning-based mobility prediction. We first use LSTM (Long Short Term Memory) algorithm, one of deep learning algorithms, to learn every user equipment's (UE's) mobility pattern from its historical trajectories and predict its movement trends in the future. Based on the corresponding prediction results, the network will judge whether a handover is required for the UE. For the handover case, a dual connection will be established for the related UE. Thus, the UE can get the radio signal from two base stations in the handover process. Simulation results verify that the proposed intelligent dual connectivity mechanism can significantly improve the quality of service of mobile users in the handover process while guaranteeing the network energy efficiency.

5G network is anticipated to meet the challenging requirements of mobile traffic in the 2020's, which are characterized by super high data rate, low latency, high mobility, high energy efficiency, and high traffic density. This paper provides an overview of China Mobile's 5G vision and potential solutions. Targeting a paradigm shift to user-centric network operation from the traditional cell-centric operation, 5G radio access network (RAN) design considerations are presented, including RAN restructure, Turbo charged edge, core network (CN) and RAN function repartition, and network slice as a service. Adaptive multiple connections in the user-centric operation is further investigated, where the decoupled downlink and uplink, decoupled control and data, and adaptive multiple connections provide sufficient means to achieve a 5G network with “no more cells.” Software-defined air interface (SDAI) is presented under a unified framework, in which the frame structure, waveform, multiple access, duplex mode, and antenna configuration can be adaptively configured. New paradigm of 5G network featuring user-centric network (UCN) and SDAI is needed to meet the diverse yet extremely stringent requirements across the broad scope of 5G scenarios.

Non-orthogonal multiple access (NOMA) is expected to be a promising multiple access techniques for 5G networks due to its superior spectral efficiency (SE). In this paper, we study the energy efficiency (EE) optimization for the fading multiple-input multiple-output (MIMO) NOMA systems with statistical channel state information (CSI) at the transmitter. The EE optimization problem is formulated to maximize the system EE (defined by ergodic capacity under unit power consumption) under the total transmit power constraint and the minimum rate constraint of the weak user. We propose the near optimal power allocation schemes and also the suboptimal closed-form solutions. Numerical results show that the proposed NOMA schemes significantly outperform the traditional orthogonal multiple access scheme with open loop MIMO transmission in terms of both SE and EE.

Non-orthogonal multiple access (NOMA) is expected to be a promising technique for future wireless networks due to its superior spectral efficiency. In this paper, the sum rate optimization problem for multiple-input multiple-output (MIMO) NOMA systems is studied with the total transmit power constraint and the minimum rate constraint of weak user. We first derive a channel state information (CSI) condition in which MIMO NOMA systems can achieve full rate transmission, i.e. the transmission rate of the weak user equals to the channel capacity of weak user. Based on the CSI condition, we propose an optimal power allocation scheme for MIMO NOMA systems, which can achieve the capacity region of MIMO Broadcast channel as dirty paper coding. A low complexity suboptimal scheme is proposed as well for all CSI channel conditions. Numerical results show that the proposed NOMA schemes significantly outperform the traditional time division based single user MIMO scheme and the multi-user MIMO scheme.

We propose decentralized relay coordination algorithms and corresponding channel state information (CSI) signaling concepts for weighted sum rate maximization, via linear downlink transceiver optimization in multiuser multi-relay MIMO network. Two decentralized algorithms are developed with the local CSI requirements at the base station and the relay nodes. In the first proposed algorithm, the generalized channel inversion precoding is utilized at the base station to avoid the second hop channel sharing requirements among the relay nodes. Further, decentralized transceiver optimization of the relay nodes and users are developed to maximize the weighted sum rate. For the second proposed approach, we provide a novel structure of the precoding matrix of the base station. Then decentralized joint BS, relay and user transceiver design is developed to further improve the performance. Moreover, CSI signaling and decentralized processing in TDD system are analyzed. Simulation results show the effectiveness of the decentralized relay coordination compared to the existing algorithms.

Far field wireless recharging based on microwave power transfer (MPT) will release mobile terminals from interruption due to finite battery storage. Wireless energy transfer (WET), which integrating MPT with wireless communications, is a promising technology to provide perpetual energy supplies to wireless networks. In this paper, we propose a novel power allocation algorithm for Multiuser MIMO (MU MIMO) systems with simultaneous wireless information and power transfer (SIPT). For the purpose of universal relevance, a K +L+1 node network is investigated, in which K receivers decode information and the other L receivers harvest energy from the signals broadcasted by the base station. The proposed integrated scheme, based on the well-known Block Diagonalization (BD) preceding method, can completely suppress the multiuser interference and perform an outer-bound which is characterized by the boundary of the rate-energy (R-E) region as well.

Energy efficiency is an important issue for the future network since the stemming requirements of the ubiquitous access and the limited battery of the mobile devices. In this paper, we investigate energy efficient transmission for two-way relay network. Theoretically, we derive the optimal transmission data rate, which minimizes the total energy consumption per information bit. The energy efficiency region for two-way relay systems is also investigated to illustrate the effects of relay locations on the energy efficiency of the system, and the optimal relay location is also found to maximize the energy efficiency. Numerical results are presented to demonstrate the tradeoff between energy efficiency and the spectral efficiency.

A new selective transmission mode in a three-node relay system is proposed when interference cancellation (IC) is not applicable on receivers. According to the effective Signal-to-Noise-Ratio (SNR) of each link, the source determines whether to transmit new information in the second slot while the relay node is forwarding the information received during the first slot. By the proposed monitoring mechanism, we can improve the ergodic capacity of the whole system compared to conventional relay protocols without too much feed-back overhead. Moreover, we prove that under careful power control of each node, full spatial diversity and capacity gain will be achieved even when there is no interference cancellation processing at the destination node. Thus we can reduce the burden of the receiver significantly. Two methods with perfect IC are compared as a supplement to explore the flexibility when there is no complexity limit at receivers. Both Decode-and-Forward (DF) and Amplify-and-Forward (AF) modes are analyzed respectively.

The performance of a system, which uses singular value decomposition (SVD) over a multiple-input multiple-output (MIMO) channel, depends on the accuracy of the channel state information (CSI) at the transmitter and the receiver. And although, SVD-based transmission can be done without any feedback by utilizing the channel reciprocity in time-division duplex (TDD) systems, the CSI at the transmitter is always outdated in time-varying channels, especially in the high-speed mobility cases. As in such cases, a typical system employing conventional SVD-based transmission will suffer great degradation in both capacity and BER performance, due to the interference among the sub-channels. So, a more efficient SVD-based transmission strategy against high-speed mobility is proposed in this paper. Compared with other transmission schemes, the novel transmission scheme can achieve significant performance improvement with lower computation complexity. And the simulation results do indicate that the proposed scheme is more robust against the outdated CSI. Besides, the analysis of both signal to interference-plus-noise ratio (SINR) and computation complexity are provided.

In this paper, we investigate the energy efficient relay selection scheme for two-way relay network. Two optimal joint relay selection and power minimization schemes are studied. One approach is to minimize the total transmit power, including the relay power and the transceiver power, subject to the signal to noise ratio (SNR) requirements on the two transceiver. And the other focuses on relay power minimization. Closed-form solutions are obtained and only few parameters need to be broadcast to all relays. Based on the close form solution, a new relay selection criterion is proposed, which simply relies on the relay channel coefficients. Simulation results are presented to show the merits of the proposed scheme.

In this paper, we evaluate the existing transmission schemes in multi-cell environment, such as single-cell transmission with Inter-cell interference (ICI) as well as Collaborative Multi-Point transmission/reception (CoMP) technique in LTE-A, and propose an more effective solution to deal with the inter-cell interference. The proposed transmission structure takes advantage of the cooperation between base stations (BS), and combines both closed-loop and open-loop techniques. With the concept of Effective Channel, the classic Space Frequency Block Coding (SFBC) decode scheme can be used directly in our receiver for single layer transmission. Thus, we prove that the scenario of multiple base stations can employ SFBC conveniently while the performance is guaranteed. Analysis and simulation results show that our method outperforms conventional single cell transmission. Moreover, the proposed transmission scheme can obtain elegant performance enhancement compared with local precoding and global precoding of CoMP in terms of capacity, and doubly reduce the feedback overhead at the same time.