The increasing demand for high-capacity and energy-efficient wireless communications has made extreme massive Multiple Input Multiple Output (MIMO) a critical technology for 6G networks. Hybrid beamforming is essential to trade-off spectral efficiency and power consumption, yet existing solutions are typically array-specific. Additionally, most work focuses on single-polarized antenna systems, whereas dual-polarized arrays are widely used in practical deployments. This paper proposes a universal shifted codebook set-based hybrid beamforming framework for dual-polarized massive MIMO. The key lies in decoupling the array architecture impact from the hybrid beamforming optimization, transferring it into a designed shifted codebook set. This enables seamless adaptation across different hybrid array architectures without modifying core algorithm modules. Based on this framework, we develop two efficient algorithms: shifted-codebook orthogonal matching pursuit and shifted-codebook greedy grid-of-beam for joint and two-stage hybrid beamforming, respectively. Simulations using 3GPP dual-polarized channels show that our algorithms outperform the state-of-the-art solution, offering higher spectral efficiency and improved energy efficiency while maintaining flexibility for product evolution and MIMO hardware upgrades.

Beamspace beamforming, which is advantageous in reducing channel estimation efforts and the subsequent precoding complexity, has great potential for massive Multiple Input multiple Output (MIMO) systems with extremely large antenna arrays. Sub-6 GHz will still play an essential role for a wide coverage and a reliable connectivity in future wireless communications. It is thus of prime interest to investigate efficient beamspace beamforming technologies for Frequency Division Duplex (FDD) massive MIMO. Standard beamspace beamforming schemes can be realized by either downlink grid-of-beam or uplink sounding, where beams are selected from the Discrete Fourier Transformation (DFT) codebook or calculated by Eigen-Value Decomposition (EVD) from the measured channel covariance matrix. However, in FDD when channel reciprocity does not hold, it is difficult to design efficient beamspace beamforming algorithms that not only delivers high performance but also has low complexity and low induced overhead. This paper proposes a novel Krylov-beamspace beamforming scheme for FDD massive MIMO. Taking advantages of angular reciprocity in FDD, we develop a generalized algorithm to compensate the uplink channel covariance matrix for the downlink use, which is more robust against scattered channels and imperfect reciprocity than the conventional counterpart. The beamspace beamforming matrix is designed by the derived beam signature vector and the compensated channel covariance matrix, resulting in the so called “beam”-Krylov subspace. Two concepts utilizing limited downlink feedback and/or uplink sounding are proposed, which can provide enough flexibilities in implementing the new scheme. The proposed algorithm does not need any complicated EVD calculations and even outperforms the baseline method, which is low-complexity/low-overhead and quite applicable for FDD massive MIMO.

The cell-free massive Multiple Input Multiple Output (MIMO) architecture is considered for millimeter Wave (mmWave) bands to overcome propagation challenges at higher frequencies such as significant pathloss, easy blockage, and shadowing. Instead of using a single and co-located extreme large antenna array, multiple large arrays with a smaller dimension in a hybrid architecture can be efficiently deployed over multiple Access Points (APs) in a distributed way. We propose a distributed hybrid beamforming concept for mmWave cell-free massive MIMO systems, which supports multiantenna or even hybrid array based User Equipments (UEs) and multi-stream transmissions. Analog beamforming at both APs and UEs can be obtained in a joint and multi-phase manner to maximize the total effective array gain in the cell-free system. We also develop an interference aware digital precoding method to deal with multi-user/multi-stream transmissions, achieving a high multiplexing gain. Simulation results show a good potential of using the distributed hybrid beamforming in cell-free systems.

Extractive text summarization aims at extracting the most representative sentences from a given document as its summary.To extract a good summary from a long text document, sentence embedding plays an important role.Recent studies have leveraged graph neural networks to capture the inter-sentential relationship (e.g., the discourse graph) to learn contextual sentence embedding.However, those approaches neither consider multiple types of inter-sentential relationships (e.g., semantic similarity & natural connection), nor model intra-sentential relationships (e.g, semantic & syntactic relationship among words).To address these problems, we propose a novel Multiplex Graph Convolutional Network (Multi-GCN) to jointly model different types of relationships among sentences and words.Based on Multi-GCN, we propose a Multiplex Graph Summarization (Multi-GraS) model for extractive text summarization.Finally, we evaluate the proposed models on the CNN/DailyMail benchmark dataset to demonstrate the effectiveness of our method.

Medical triage chatbot is widely used in pre-diagnosis by asking symptom and medical history-related questions. Information collected from patients through an online chatbot system is often incomplete and imprecise, and thus it's essentially hard to achieve precise triaging. In this paper, we propose Multi-relational Hyperbolic Diagnosis Predictor (MHDP) --- a novel multi-relational hyperbolic graph neural network-based approach, to build a disease predictive model. More specifically, in MHDP, we generate a heterogeneous graph consisting of symptoms, patients, and diagnoses nodes, and then derive node representations by aggregating neighborhood information recursively in the hyperbolic space. Experiments conducted on two real-world datasets demonstrate that the proposed MHDP approach surpasses state-of-the-art baselines.

Question answering (QA) has achieved promising progress recently. However, answering a question in real-world scenarios like the medical domain is still challenging, due to the requirement of external knowledge and the insufficient quantity of high-quality training data. In the light of these challenges, we study the task of generating medical QA pairs in this paper. With the insight that each medical question can be considered as a sample from the latent distribution of questions given answers, we propose an automated medical QA pair generation framework, consisting of an unsupervised key phrase detector that explores unstructured material for validity, and a generator that involves a multi-pass decoder to integrate structural knowledge for diversity. A series of experiments have been conducted on a real-world dataset collected from the National Medical Licensing Examination of China. Both automatic evaluation and human annotation demonstrate the effectiveness of the proposed method. Further investigation shows that, by incorporating the generated QA pairs for training, significant improvement in terms of accuracy can be achieved for the examination QA system. 1

Human tackle reading comprehension not only based on the given context itself but often rely on the commonsense beyond.To empower the machine with commonsense reasoning, in this paper, we propose a Commonsense Evidence Generation and Injection framework in reading comprehension, named CEGI.The framework injects two kinds of auxiliary commonsense evidence into comprehensive reading to equip the machine with the ability of rational thinking.Specifically, we build two evidence generators: one aims to generate textual evidence via a language model; the other aims to extract factual evidence (automatically aligned text-triples) from a commonsense knowledge graph after graph completion.Those evidences incorporate contextual commonsense and serve as the additional inputs to the reasoning model.Thereafter, we propose a deep contextual encoder to extract semantic relationships among the paragraph, question, option, and evidence.Finally, we employ a capsule network to extract different linguistic units (word and phrase) from the relations, and dynamically predict the optimal option based on the extracted units.Experiments on the Cos-mosQA dataset demonstrate that the proposed CEGI model outperforms the current state-ofthe-art approaches and achieves the highest accuracy (83.6%) on the leaderboard.

This paper presents a novel framework, MGNER, for Multi-Grained Named Entity Recognition where multiple entities or entity mentions in a sentence could be nonoverlapping or totally nested.Different from traditional approaches regarding NER as a sequential labeling task and annotate entities consecutively, MGNER detects and recognizes entities on multiple granularities: it is able to recognize named entities without explicitly assuming non-overlapping or totally nested structures.MGNER consists of a Detector that examines all possible word segments and a Classifier that categorizes entities.In addition, contextual information and a self-attention mechanism are utilized throughout the framework to improve the NER performance.Experimental results show that MGNER outperforms current state-of-the-art baselines up to 4.4% in terms of the F1 score among nested/non-overlapping NER tasks.* Work was done when the author Yaliang Li was at Tencent America.

Targeting massive Multiple Input Multiple Output (MIMO) systems, we consider an efficient multi- panel array architecture and propose hybrid beamforming techniques for millimeter wave frequencies. Taking advantages of the multi-panel array, hybrid two-stage processing including coarse and fine beamforming can be carried out in a distributed fashion. Panel-coordinated analog beamforming as well as decoupling only requires channels measured at each panel itself. Non- linear precoding is then applied on a per-panel basis. It is shown that the proposed schemes relax the antenna dimensionality constraint and exhibit a good performance even in an overloaded case, which are practically feasible in both the single-/multi-cell scenarios.

For massive multiple input multiple output systems at millimeter wave (mmWave) bands, we consider an efficient hybrid array architecture, namely, overlapped subarray (OSA), and develop a Unified Low Rank Sparse (ULoRaS) recovery algorithm for hybrid beamforming in downlink multiuser scenarios. The ULoRaS scheme takes advantage of the transmit-receive coordinated beamforming procedure to achieve large array gains. It has no dimensionality constraint and can be applied to the generalized OSA architecture, including both the fully connected and the widely discussed non-OSA cases. It is shown that the proposed ULoRaS algorithm for the novel OSA design is a good compromise of the performance and the required hardware complexity.

Massive Millimeter Wave (mmWave) Multiple Input Multiple Output (MIMO) systems utilize hybrid beamforming techniques to alleviate the implementation complexity of combining a large number of antennas. To solve the optimization problem in the hybrid design supporting multiple users and multiple data streams per user, we first propose a coordinated Radio Frequency (RF) beamforming technique based on the Generalized Low Rank Approximation of Matrices (GLRAM) approach and then develop an efficient Modified GLRAM (MGLRAM) algorithm. The proposed scheme only requires the information of the composite channel, instead of the complete physical channel matrix which is assumed to be known in the existing literature. It takes advantage of the coordination between the base station and users to achieve a maximal array gain and has no dimensionality constraint. The multiplexing gain is then exploited by applying the Block Diagonalization (BD) technique. It is shown that our proposed scheme is a practical and competing solution, which can be easily applied to both the Time Division Duplex (TDD) and Frequency Division Duplex (FDD) systems.

Given a health-related question (such as "I have a bad stomach ache. What should I do?"), a medical self-diagnosis Android inquires further information from the user, diagnoses the disease, and ultimately recommend best solutions. One practical challenge to build such an Android is to ask correct questions and obtain most relevant information, in order to correctly pinpoint the most likely causes of health conditions. In this paper, we tackle this challenge, named "relevant symptom question generation": Given a limited set of patient described symptoms in the initial question (e.g., "stomach ache"), what are the most critical symptoms to further ask the patient, in order to correctly diagnose their potential problems? We propose an augmented long short-term memory (LSTM) framework, where the network architecture can naturally incorporate the inputs from embedding vectors of patient described symptoms and an initial disease hypothesis given by a predictive model. Then the proposed framework generates the most important symptom questions. The generation process essentially models the conditional probability to observe a new and undisclosed symptom, given a set of symptoms from a patient as well as an initial disease hypothesis. Experimental results show that the proposed model obtains improvements over alternative methods by over 30% (both precision and mean ordinal distance).

As fast development in communication field, sub-6 GHz spectrum band becomes ultra crowd. Even though Carrier Aggregation (CA) and higher order modulation used, available spectrum is still the bottleneck of communication speed and network latency. Nowadays, more and more companies and institutes are paying more attention to over 6GHz communication system research. Recently released news indicate frequency points are mainly on 12GHz, 15GHz, 28GHz, 38GHz even 70GHz, and accompanied working bandwidth is about 400MHz to 1GHz. As working band goes higher, antenna conceptions will also be updated. A CPW-Fed monopole antenna working in KU-band with more than 20% relative bandwidth is hereby introduced, which might be a good candidate for future 5G cellular networks and also satellite based DBS reception system. Proposed antenna has small profile lower than λg/4 with dual-polarization, which is thus suitable for base station and mobile device conceptions.

As one of the key LTE-A technologies, coordinated multi-point transmission (CoMP) has been extensively investigated. Among various CoMP schemes, coherent joint transmission scheme (CJT) shows the largest CoMP gains in both cell average performance and cell edge performance. However, for pursuit of the upper bound of CoMP gains, centralized scheduling was adopted in those proposed CJT schemes. With CoMP cluster size increasing, those proposed CoMP schemes face multiple severe challenges, such as the unacceptable computational complexity and more strict latency requirements. In this paper, a novel CJT scheme is proposed, which is based on distributed scheduling and joint transmit precoder design. The complexity of the new CJT scheme is theoretically analyzed and compared with that of centralized scheduling CJT schemes. Moreover, comprehensive performance evaluations of the new CJT scheme are performed by system level simulations in various transmission scenarios. Taking the performance of non-CoMP scheme as baseline, the new CJT scheme can achieve over 35% cell average gain and over 32% cell edge gain, although there is around 10% cell edge CoMP gain loss than that of centralized scheduling CJT scheme.

Coordinated multi-point transmission (CoMP) has been regarded as one of the key LTE-A technologies. Various CoMP schemes has been extensively investigated,and coherent joint transmission scheme (CJT) has been demonstrated to extract the largest CoMP gains in both cell average performance and cell edge performance. In CJT scheme, multiple coordinated cells constitute a supper cell adopting centralized multi-user scheduling and resource allocation. However, with CoMP cluster size increasing, traditional CJT scheme faces multiple severe challenges, such as the unacceptable computational complexity and more strict latency requirements. In this paper, Distributed scheduling scheme is proposed for downlink CJT to alleviate the computational complexity and other system strict requirements. The complexity of the proposed scheme is analyzed. Comprehensive performance evaluations of the proposed scheme are performed by system level simulations in various transmission scenarios. Taking the performance of single cell as the baseline, the proposed scheme can achieve over 35% cell average gain and over 32% cell edge gain with low complexity.

In this paper, for an energy-constrained two-stage selective decode-and-forward (DF) relaying network which employs a Best-Select form of cooperation, we focus on maximizing the minimum node lifetime when only the mean channel state information (CSI) toward the destination is available at each relay node. If only the mean CSI can be exploited, an optimal relay selection strategy for lifetime maximization in an energy-constrained network is not readily available. In this paper, a simple but effective relay-selection strategy is proposed by exploiting only the mean CSI and the residual energy information (REI) at relay nodes. For any given transmit power at each node, the proposed strategy can maximize the minimum node lifetime while achieving fairly stable end-to-end outage performance. A decentralized implementation with minimal overhead is also designed, so that only the local mean CSI and REI is required at each relay node. Both asymptotic performance analyses and realistic simulations verify the advantages of the proposed relay-selection strategy.

A new time-domain transmit beamforming algorithm is proposed for cancelling inter-channel-interference (ICI) due to Doppler frequency shift under high speed train communication scenario. Simulation results show that by employing the algorithm a high speed train communication system is capable of providing continuous 100Mbps data rate for passengers at a speed of 450km/h. This would guarantee continuous data-intensive services for today's high speed train passengers.

Orthogonal frequency-division multiplexing (OFDM) and multiple-input-multiple-output (MIMO) are two key technologies adopted in future wireless communication systems, such as LTE/LTE-A. There are many papers addressed the resource allocation problems for OFDM and MIMO systems independently. However few discussions targets on the joint frequency-spatial resource allocation problem, particularly on eigenmode scheduling level. In this paper, we try to propose novel resource allocation schemes on eigenmode selection level, with the joint consideration of precoding problem in spatial domain, eigenmode scheduling and power allocation cross spatial and frequency domains.

For co-channel deployment of Macrocell and indoor Femtocell, in order to mitigate downlink (DL) interference toward user equipments (UEs) of Macrocell, DL power control (PC) at Femto has been considered as the primary inter-cell interference cancelation (ICIC) strategy in 3GPP standardization. Further, currently, the 3GPP-agreed baseline for backhaul is that there is no direct X2/S1 interface between Macrocell and Femtocell, which means the real-time Macro-Femto coordination does not exist. Thus, autonomous triggering of Femto DL ICIC and autonomous PC at Femto have been recommended in 3GPP standardization. In this paper, to enhance the feasibility and performance of Femto's autonomous PC, we firstly design a strategy for Femto to properly determine the self-configured threshold used to autonomously trigger DL ICIC. Then, more importantly, to improve the autonomous parameter-estimation approaches used for implementing autonomous PC at Femto, an effective strategy is designed for Femto to autonomously estimate whether there exists at least one indoor victim Macro UE (MUE). Simulations show that the proposed strategies work well and provide obviously enhanced ICIC benefit for the focused autonomous Femto PC scheme.

In this paper, we investigate the problem of energy saving (ES) for wireless access networks, especially for the inter-eNB scenario 2 which is a typical non-overlapped scenario specified by 3rd generation partnership project (3GPP). Due to the uneven distribution of the traffic load both over time and over cells, energy efficiency can be improved by switching off some eNBs or reducing their transmission power according to the actual traffic demand. Considering this characteristic of wireless networks, we propose a flexible and efficient solution based on the load information of each cell and the idea of multi-hop load release. The proposed solution is flexible to support different heterogeneous network architecture and different optimization cases with less signaling overhead. Moreover, the gradual multi-hop load release mechanism can efficiently release the load to the whole network and then maximize the total energy efficiency. Simulation results demonstrate that the ES gain of the proposed solution can reach about 25% with appropriate values of different thresholds for event triggering.

The inherent high peak-to-average power ratio (PAPR) of multicarrier transmission, such as OFDM or DMT, is a major drawback that can significantly worsen the power efficiency and performance of these systems. As a consequence, a number of PAPR reduction techniques have been proposed in the literature each offering a different combination of system performance, complexity and bandwidth efficiency. However, most of the technique require high computational complexity which is undesirable. In this paper, we propose a unique PAPR reduction method using time domain signal modification which does not require any additional IFFT operations. It will be shown that our proposed technique can achieve significant PAPR reduction with low complexity and latency.

The concept of small cells such as femtocell has been regarded as a promising solution to achieve green communication and deal with the coverage hole and growth of traffic within macrocells. However, the unpredictable deployment pattern of such nodes makes static network planing impractical and the overlaying macrocell and femtocell tiers may suffer significant cross-tier interference from each other. In this paper, we analyze the cross-tier interference scenarios and propose corresponding semi-static resource allocation schemes to achieve the best trade off between spectrum efficiency and interference cancellation by applying hybrid shared and partitioned spectrum usage according to the location based measurement of the femtocell. Extensive system simulation is performed to demonstrate that the proposed scheme can effectively avoid spectrum collision and significantly improve the overall network capacity.

Under the co-channel deployment of indoor Home-eNB (HeNB) with closed-subscriber-group (CSG) access mode and Macro-eNB (MeNB), it is very important to mitigate downlink (DL) interference toward Macro user equipments (UEs) which are located in the vicinity of a HeNB. For this scenario, based on discussions in 3GPP standardization, power setting of HeNB is recommended to be the primary DL inter-cell interference cancelation (ICIC) solution, which limits the transmit power of HeNB so that the nearby Macro UE (MUE) will not be interfered severely. Because acceptable/good control channel (CCH) performance is a pre-condition for the data channel performance, in this paper, with the focus on DL ICIC for CCH, we propose an enhanced HeNB power setting scheme. In particular, the proposed scheme considers providing effective DL CCH interference mitigation for MUE while maintaining acceptable/good DL CCH performance for Home UE (HUE). Based on simulation results, even if there is no MeNB-HeNB coordination, the proposed scheme can work well and provide better performance than existing schemes.

In order to achieve higher bandwidth and higher throughput, the carrier aggregation (CA) technique has been adopted by the specifications of 3GPP Release 10 for Long Term Evolution (LTE) advanced systems, where the user equipment (UE) may operate over up to 5 component carriers (CCs). On the other hand, 3GPP specifications also define a mechanism named discontinuous reception (DRX) in order to save the energy consumption of UE operation, where a UE is allowed to stop monitoring Physical Downlink Control CHannel (PDCCH) during some period of the operation time. Obviously, the duration and the frequency of these non-monitoring periods are important parameters which significantly impact the energy saving (ES) efficiency and the system performance. The conventional UE-specific DRX scheme is not efficient since all CCs are applied with the same DRX operation. Moreover, multiple services with different delay requirements may operate simultaneously for the same UE in the realistic communication systems. Hence, the mechanism of applying identical DRX configuration over all CCs is not optimal and will cause large energy waste. In this paper, we propose a new efficient DRX scheme for LTE-A systems with multiple services, where some important parameters are further optimized according to the characteristics of multi-CC and multi-service systems. Simulation results demonstrate that the proposed DRX scheme can improve the energy saving efficiency significantly compared with the existing no-DRX scheme and the conventional UE-specific DRX scheme, even with 10% retransmission probability.

In order to overcome the indoor coverage problem for residential users, the deployment of home base stations (also referred as home-eNB) has been regarded as a promising solution. However, the unpredictable deployment pattern of home-eNBs makes centralized network planning impractical and requires distributed spectrum planning to achieve effective interference management. In this paper, by enabling adjacent home-eNBs to share the measured pathloss information, we present a cognitive-based distributed interference management strategy in which each home-eNB selects component carriers based on the estimated mutual interference impact. In particular, practical estimations of mutual interference impact are derived firstly for single-antenna systems. More importantly, the methodology is further extended to multi-antenna systems via analyzing post-processing signal-to-interference ratio (SIR). System-level simulation results demonstrate that the presented scheme can effectively avoid spectrum collision and significantly improve network throughput for home-eNB systems with either single-antenna or multi-antenna configuration.

The concept of small cells such as femtocells has been regarded as a promising solution to overcome the indoor coverage problem and deal with the growth of traffic within macrocells. However, the unpredictable deployment pattern of such cells makes centralized network planning impractical and requires more efficient interference management and distributed spectrum planning schemes to realize the benefit of femtocells. In this paper, we propose a cognitive-based interference management solution for LTE-Advanced (LTE-A) femtocells by sharing measured pathloss information among neighbors and selecting component carriers according to the estimated mutual interference. System-level simulation results demonstrate that the proposed scheme can effectively avoid spectrum collision and significantly improve overall network capacity.