In Natural Language Understanding, intent detection and slot filling have been widely used to understand user queries. However, current methods tend to rely on single words and sentences to understand complex semantic concepts, and can only consider local information within the sentence. Therefore, they usually cannot capture long-distance dependencies well and are prone to problems where complex intentions in sentences are difficult to recognize. In order to solve the problem of long-distance dependency of the model, this paper uses ConceptNet as an external knowledge source and introduces its extensive semantic information into the multi-intent detection and slot filling model. Specifically, for a certain sentence, based on confidence scores and semantic relationships, the most relevant conceptual knowledge is selected to equip the sentence, and a concept context map with rich information is constructed. Then, the multi-head graph attention mechanism is used to strengthen context correlation and improve the semantic understanding ability of the model. The experimental results indicate that the model has significantly improved performance compared to other models on the MixATIS and MixSNIPS multi-intent datasets.

A user wearing a Head-Mounted Display (HMD) is likely to feel isolated when sharing virtual reality (VR) experiences with Non-HMD users in the same physical space due to not being able to see the real space outside the virtual world. This research proposes a method for an HMD user to recognize the Non-HMD users' gaze and attention via a projector attached to the HMD. In the proposed approach, the projected HMD user's view is filtered darker than default, and when Non-HMD users point controllers towards the projected view, the filter is removed from a circular area for both HMD and Non-HMD users indicating which region the Non-HMD users are viewing. We conducted two user studies showing that the Non-HMD users' gaze can be recognized with the proposed method, and investigated the preferred range for the alpha value and the size of the area for removing the filter for the HMD user.

Thanks to the development of the 6th generation mobile network that makes it possible for us to move towards an intelligent ubiquitous information society, among which some novel technologies represented by cell-free network has also attracted widespread academic attention. Cell-free network has brought distinguished gains to the network capacity with its strong ability against inter-cell interference. Unfortunately, further improvement demands more base stations (BSs) to be settled, which incurs steep cost increase. To address this issue, reconfigurable intelligent surface (RIS) with low cost and power consumption is introduced in this paper to replace some of the trivial BSs in the system, then, a RIS-aided cell-free network paradigm is formulated. Our objective is to solve the weighted sum-rate (WSR) maximization problem by jointly optimizing the beamforming design at BSs and the phase shift of RISs. Due to the non-convexity of the formulated problem, this paper investigates a joint optimizing scheme based on block coordinate descent (BCD) method. Subsequently, on account of the majority of the precious work reposed perfect channel state information (CSI) setup for the ultimate performance, this paper also extends the proposed algorithm to the case wherein CSI is imperfect by utilizing successive convex approximation (SCA). Finally, simulation results demonstrate that the proposed scheme shows great performance and robustness in perfect CSI scenario as well as the imperfect ones.

In the cellular system, the Worst Case User (WCU), whose distances to three nearest BSs are the similar, usually achieves the lowest performance. Improving user performance, especially the WCU, is a big problem for both network designers and operators. This paper works on the WCU in terms of coverage probability analysis by the stochastic geometry tool and data rate optimization with the transmission power constraint by the reinforcement learning technique under the Stretched Pathloss Model (SPLM). In analysis, only fast fading from the WCU to the serving Base Stations (BSs) is taken into the analysis to derive the lower bound coverage probability. Furthermore, the paper assumes that the Coordinated Multi-Point (CoMP) technique is only employed for the WCU to enhance its downlink signal and avoid the explosion of Intercell Interference (ICI). Through the analysis and simulation, the paper states that to improve the WCU performance under bad wireless environments, an increase in transmission power can be a possible solution. However, in good environments, the deployment of advanced techniques such as Joint Transmission (JT), Joint Scheduling (JS), and reinforcement learning is an suitable solution.

The purpose of this study was to develop a practical functional electrical stimulation (FES) controller for joint movements restoration based on an optimal control technique by cascading a linear model predictive control (MPC) and a nonlinear transformation. The cascading configuration was aimed to obtain an FES controller that is able to deal with a nonlinear system. The nonlinear transformation was utilized to transform the linear solution of linear MPC to become a nonlinear solution in form of optimized electrical stimulation intensity. Four different types of nonlinear functions were used to realize the nonlinear transformation. A simple parameter estimation to determine the value of the nonlinear transformation parameter was also developed. The tracking control capability of the proposed controller along with the parameter estimation was examined in controlling the 1-DOF wrist joint movement through computer simulation. The proposed controller was also compared with a fuzzy FES controller. The proposed MPC-FES controller with estimated parameter value worked properly and had a better control accuracy than the fuzzy controller. The parameter estimation was suggested to be useful and effective in practical FES control applications to reduce the time-consuming of determining the parameter value of the proposed controller.

In this paper, we propose a hierarchical Stackelberg game based resource allocation algorithm (HGRAA) to jointly allocate the wireless and computational resources of a mobile edge computing (MEC) system. The proposed HGRAA is composed of two levels: the lower-level evolutionary game (LEG) minimizes the cost of mobile terminals (MTs), and the upper-level exact potential game (UEPG) maximizes the utility of MEC servers. At the lower-level, the MTs are divided into delay-sensitive MTs (DSMTs) and non-delay-sensitive MTs (NDSMTs) according to their different quality of service (QoS) requirements. The competition among DSMTs and NDSMTs in different service areas to share the limited available wireless and computational resources is formulated as a dynamic evolutionary game. The dynamic replicator is applied to obtain the evolutionary equilibrium so as to minimize the costs imposed on MTs. At the upper level, the exact potential game is formulated to solve the resource sharing problem among MEC servers and the resource sharing problem is transferred to nonlinear complementarity. The existence of Nash equilibrium (NE) is proved and is obtained through the Karush-Kuhn-Tucker (KKT) condition. Simulations illustrate that substantial performance improvements such as average utility and the resource utilization of MEC servers can be achieved by applying the proposed HGRAA. Moreover, the cost of MTs is significantly lower than other existing algorithms with the increasing size of input data, and the QoS requirements of different kinds of MTs are well guaranteed in terms of average delay and transmission data rate.

Pedestrian detection is an essential but challenging task in computer vision, especially in crowded scenes due to heavy intra-class occlusion. In human visual system, head information can be used to locate pedestrian in a crowd because it is more stable and less likely to be occluded. Inspired by this clue, we propose a dual-task detector which detects head and human body simultaneously. Concretely, we estimate human body candidates from head regions with statistical head-body ratio. A head-body alignment map is proposed to perform relational learning between human bodies and heads based on their inherent correlation. We leverage the head information as a strict detection criterion to suppress common false positives of pedestrian detection via a novel pull-push loss. We validate the effectiveness of the proposed method on the CrowdHuman and CityPersons benchmarks. Experimental results demonstrate that the proposed method achieves impressive performance in detecting heavy-occluded pedestrians with little additional computation cost.

Well annotated dataset is crucial to the training of object detectors. However, the production of finely annotated datasets for object detection tasks is extremely labor-intensive, therefore, cloud sourcing is often used to create datasets, which leads to these datasets tending to contain incorrect annotations such as inaccurate localization bounding boxes. In this study, we highlight a problem of object detection with noisy bounding box annotations and show that these noisy annotations are harmful to the performance of deep neural networks. To solve this problem, we further propose a framework to allow the network to modify the noisy datasets by alternating refinement. The experimental results demonstrate that our proposed framework can significantly alleviate the influences of noise on model performance.

The current approaches to semantic role classification usually first define a representation vector for a candidate role and feed the vector into a deep neural network to perform classification. The representation vector contains some lexicalization features like word embeddings, lemmar embeddings. From linguistics, the semantic role frame of a sentence is a joint structure with strong dependencies between arguments which is not considered in current deep SRL systems. Therefore, this paper proposes a global deep reranking model to exploit these strong dependencies. The evaluation experiments on the CoNLL 2009 shared tasks show that our system can outperforms a strong local system significantly that does not consider role dependency relations.

Approximate Simultaneous Diagonalization (ASD) is a problem to find a common similarity transformation which approximately diagonalizes a given square-matrix tuple. Many data science problems have been reduced into ASD through ingenious modelling. For ASD, the so-called Jacobi-like methods have been extensively used. However, the methods have no guarantee to suppress the magnitude of off-diagonal entries of the transformed tuple even if the given tuple has an exact common diagonalizer, i.e., the given tuple is simultaneously diagonalizable. In this paper, to establish an alternative powerful strategy for ASD, we present a novel two-step strategy, called Approximate-Then-Diagonalize-Simultaneously (ATDS) algorithm. The ATDS algorithm decomposes ASD into (Step 1) finding a simultaneously diagonalizable tuple near the given one; and (Step 2) finding a common similarity transformation which diagonalizes exactly the tuple obtained in Step 1. The proposed approach to Step 1 is realized by solving a Structured Low-Rank Approximation (SLRA) with Cadzow's algorithm. In Step 2, by exploiting the idea in the constructive proof regarding the conditions for the exact simultaneous diagonalizability, we obtain an exact common diagonalizer of the obtained tuple in Step 1 as a solution for the original ASD. Unlike the Jacobi-like methods, the ATDS algorithm has a guarantee to find an exact common diagonalizer if the given tuple happens to be simultaneously diagonalizable. Numerical experiments show that the ATDS algorithm achieves better performance than the Jacobi-like methods.

Locally repairable codes (LRCs) are a type of new erasure codes designed for modern distributed storage systems (DSSs). In order to obtain ternary LRCs of distance 6, firstly, we propose constructions with disjoint repair groups and construct several families of LRCs with 1 ≤ r ≤ 6, where codes with 3 ≤ r ≤ 6 are obtained through a search algorithm. Then, we propose a new method to extend the length of codes without changing the distance. By employing the methods such as expansion and deletion, we obtain more LRCs from a known LRC. The resulting LRCs are optimal or near optimal in terms of the Cadambe-Mazumdar (C-M) bound.

Pilot contamination is addressed in massive multiple-input multiple-output (MIMO) uplink. The main ideas of pilot decontamination are twofold: One is to design transmission timing of pilot sequences such that the pilot transmission periods in different cells do not fully overlap with each other, as considered in previous works. The other is joint channel and data estimation via approximate message-passing (AMP) for bilinear inference. The convergence property of conventional AMP is bad in bilinear inference problems, so that adaptive damping was required to help conventional AMP converge. The main contribution of this paper is a modification of the update rules in conventional AMP to improve the convergence property of AMP. Numerical simulations show that the proposed AMP outperforms conventional AMP in terms of estimation performance when adaptive damping is not used. Furthermore, it achieves better performance than state-of-the-art methods based on subspace estimation when the power difference between cells is small.

In this paper, we reformulate a sensitivity analysis method for function-expansion-based topology optimization method without using gray area. In the conventional approach based on function expansion method, permittivity distribution contains gray materials, which are intermediate materials between core and cladding ones, so as to let the permittivity differentiable with respect to design variables. Since this approach using gray area dose not express material boundary exactly, it is not desirable to apply this approach to design problems of strongly guiding waveguide devices, especially for plasmonic waveguides. In this study, we present function-expansion-method-based topology optimization without gray area. In this approach, use of gray area can be avoided by replacing the area integral of the derivative of the matrix with the line integral taking into acount the rate of boundary deviation with respect to design variables. We verify the validity of our approach through applying it to design problems of a T-branching power splitter and a mode order converter.

The joint representations of knowledge graph have become an important approach to improve the quality of knowledge graph, which is beneficial to machine learning, data mining, and artificial intelligence applications. However, the previous work suffers severely from the noise in text when modeling the text information. To overcome this problem, this paper mines the high-quality reference sentences of the entities in the knowledge graph, to enhance the representation ability of the entities. A novel framework for joint representation learning of knowledge graphs and text information based on reference sentence noise-reduction is proposed, which embeds the entity, the relations, and the words into a unified vector space. The proposed framework consists of knowledge graph representation learning module, textual relation representation learning module, and textual entity representation learning module. Experiments on entity prediction, relation prediction, and triple classification tasks are conducted, results show that the proposed framework can significantly improve the performance of mining and fusing the text information. Especially, compared with the state-of-the-art method[15], the proposed framework improves the metric of H@10 by 5.08% and 3.93% in entity prediction task and relation prediction task, respectively, and improves the metric of accuracy by 5.08% in triple classification task.

In this letter, we adopt two multi-carrier relay selections, i.e., bulk and per-subcarrier (PS), to the multi-hop decode-and-forward relaying orthogonal frequency-division multiplexing with index modulation (OFDM-IM) system. Particularly, in the form of average outage probability (AOP), the influence of joint selection and non-joint selection acting on the last two hops on the system is analyzed. The closed-form expressions of AOPs and the asymptotic AOPs expressions at high signal-to-noise ratio are given and verified by numerical simulations. The results show that both bulk and PS can achieve full diversity order and that PS can provide additional power gain compared to bulk when JS is used. The theoretical analyses in this letter provide an insight into the combination of OFDM-IM and cooperative communication.

Wireless networked control systems (WNCSs) are control systems whose components are connected through wireless networks. In WNCSs, a controlled object (CO) could become unstable due to bursty packet losses in addition to random packet losses and round-trip delays on wireless networks. In this paper, to reduce these network-induced effects, we propose a new design for multihop TDMA-based WNCSs with two-disjoint-path switching, where two disjoint paths are established between a controller and a CO, and they are switched if bursty packet losses are detected. In this system, we face the following two difficulties: (i) link scheduling in TDMA should be done in such a way that two paths can be switched without rescheduling, taking into account of the constraint of control systems. (ii) the conventional cross-layer design method of control systems is not directly applicable because round-trip delays may vary according to the path being used. Therefore, to overcome the difficulties raised by the two-path approach, we reformulate link scheduling in multihop TDMA and cross-layer design for control systems. Simulation results confirm that the proposed WNCS achieves better performance in terms of the 2-norm of CO's states.

To estimate angle, velocity, and range information of multiple targets jointly in FMCW MIMO radar, two-dimensional (2D) MUSIC with matched filtering and FFT algorithm is proposed. By reformulating the received FMCW signal of the colocated MIMO radar, we exploit 2D MUSIC to estimate the angle and Doppler frequency of multiple targets. Then by using a matched filter together with the estimated angle and Doppler frequency and FFT operation, the range of the target is estimated. To effectively estimate the parameters of multiple targets with large distance differences, we also propose a successive interference cancellation method that uses the orthogonal projection. That is, rather than estimating the multiple target parameters simultaneously using 2D MUSIC, we estimate the target parameters sequentially, in which the parameters of the target having strongest reflected power are estimated first and then, their effect on the received signal is canceled out by using the orthogonal projection. Simulations verify the performance of the proposed algorithm.

Mobile video traffic is expected to increase explosively because of the proliferating number of Wi-Fi terminals. An overloaded multiple-input multiple-output (MIMO) technique allows the receiver to implement smaller number of antennas than the transmitter in exchange for degradation in video quality and a large amount of computational complexity for postcoding at the receiver side. This paper proposes a novel linear precoder for high-quality video streaming in overloaded multiuser MIMO systems, which protects visually significant portions of a video stream. A low complexity postcoder is also proposed, which detects some of data symbols by linear detection and the others by a prevoting vector cancellation (PVC) approach. It is shown from simulation results that the combination use of the proposed precoder and postcoder achieves higher-quality video streaming to multiple users in a wider range of signal-to-noise ratio (SNR) than a conventional unequal error protection scheme. The proposed precoder attains 40dB in peak signal-to-noise ratio even in poor channel conditions such as the SNR of 12dB. In addition, due to the stepwise acquisition of data symbols by means of linear detection and PVC, the proposed postcoder reduces the number of complex additions by 76% and that of multiplications by 64% compared to the conventional PVC.

Distributed compressive video sensing (DCVS) has received considerable attention due to its potential in source-limited communication, e.g., wireless video sensor networks (WVSNs). Multi-hypothesis (MH) prediction, which treats the target block as a linear combination of hypotheses, is a state-of-the-art technique in DCVS. The common approach is under the supposition that blocks that are dissimilar from the target block are given lower weights than blocks that are more similar. This assumption can yield acceptable reconstruction quality, but it is not suitable for scenarios with more details. In this paper, based on the joint sparsity model (JSM), the authors present a Tikhonov-regularized MH prediction scheme in which the most similar block provides the similar common portion and the others blocks provide respective unique portions, differing from the common supposition. Specifically, a new scheme for generating hypotheses and a Euclidean distance-based metric for the regularized term are proposed. Compared with several state-of-the-art algorithms, the authors show the effectiveness of the proposed scheme when there are a limited number of hypotheses.

This paper addresses the association problem of tracking closely spaced targets in group or formation. In the Labeled Multi-Bernoulli Filter (LMB), the weight of a hypothesis is directly affected by the distance between prediction and measurement. This may generate false associations when dealing with the closely spaced multiple targets. Thus we consider utilizing structure information among the group or formation. Since, the relative position relation of the targets in group or formation varies slightly within a short time, the targets are considered as nodes of a topological structure. Then the position relation among the targets is modeled as a hypergraph. The hypergraph matching method is used to resolve the association matrix. At last, with the structure prior information introduced, the new joint cost matrix is re-derived to generate hypotheses, and the filtering recursion is implemented in a Gaussian mixture way. The simulation results show that the proposed algorithm can effectively deal with group targets and is superior to the LMB filter in tracking precision and accuracy.