Hierarchical Network Architecture Research Articles

A Deep Gaussian Process-Based Flight Trajectory Prediction Approach and Its Application on Conflict Detection

In this work, a deep Gaussian process (DGP) based framework is proposed to improve the accuracy of predicting flight trajectory in air traffic research, which is further applied to implement a probabilistic conflict detection algorithm. The Gaussian distribution is applied to serve as the probabilistic representation for illustrating the transition patterns of the flight trajectory, based on which a stochastic process is generated to build the temporal correlations among flight positions, i.e., Gaussian process (GP). Furthermore, to deal with the flight maneuverability of performing controller’s instructions, a hierarchical neural network architecture is proposed to improve the modeling representation for nonlinear features. Thanks to the intrinsic mechanism of the GP regression, the DGP model has the ability of predicting both the deterministic nominal flight trajectory (NFT) and its confidence interval (CI), denoting by the mean and standard deviation of the prediction sequence, respectively. The CI subjects to a Gaussian distribution, which lays the data foundation of the probabilistic conflict detection. Experimental results on real data show that the proposed trajectory prediction approach achieves higher prediction accuracy compared to other baselines. Moreover, the conflict detection approach is also validated by a obtaining lower false alarm and more prewarning time.

MFC-CNN: An automatic grading scheme for light stress levels of lettuce (Lactuca sativa L.) leaves

The identification and control of light stress is the key to the high-yield and high-quality production of leafy vegetables in a controlled environment. As a widely grown vegetable in the plant factory, lettuce is responsive to light intensity. Strong or weak light will seriously affect its yield and quality. These differences can be articulated by images and then classified by fine-grained classification methods. Deep learning is commonly used in crop image classification due to its fast and convenient advantages, but conventional fine-grained recognition approaches are still extremely challenging in dealing with this kind of inter-species classification. To address this issue, this work takes lettuce as the research object and established a set of leaf images for lettuce light stress grading. The leaves were divided into four categories depending on the changes in shoot fresh weight. Then, a hierarchical fusion convolutional neural network architecture (MFC-CNN) based on multi-scale input was constructed to grade the light stress. We firstly separate leaf patches from complete leaf and construct four-scale input to expand local leaf vein and texture information. The multi-scale dataset is fed into the main network at different depths according to the characteristics of CNN in feature extraction. Finally, the network is tested through comparative experiments. The results show that the proposed model has obvious advantages in light stress dataset and generalized dataset.

Pkg2Vec: Hierarchical package embedding for code authorship attribution

Authorship attribution of software is the task of identifying the author of a given piece of code. Code attribution is of importance in multiple scenarios, ranging from software plagiarism to cybersecurity. In this paper, we introduce authorship attribution of software packages that better reflect real-world scenarios in which code is organized in packages and written by teams. We present a novel approach for software package authorship attribution called Pkg2Vec, based on a hierarchical deep neural network (DNN) architecture, corresponding to the hierarchical nature of software (code) packages. The hierarchical neural network model consists of a token level encoder and an attention mechanism for a function level encoder, together producing package embedding. Beyond package embedding, we use keywords and API calls as resilient features, which reflect the programmer’s intention and style. Pkg2Vec is evaluated on a large dataset of public packages and compared to a number of other source code authorship attribution state-of-the-art algorithms. We find that Pkg2Vec significantly outperforms other approaches, achieving a 13% improvement in accuracy

Learning allocentric representations of space for navigation

The hippocampus of the mammalian brain supports spatial navigation by building cognitive maps of the environments in which the animal explores. Currently, there is little neurocomputational work investigating the encoding and decoding mechanisms of hippocampal neural representations in large-scale environments. We propose a biologically-inspired hierarchical neural network architecture to learn the transformation of egocentric sensorimotor inputs into allocentric spatial representation for navigation. The hierarchical network is composed of two parallel subnetworks mimicking the lateral entorhinal cortex (LEC) and medial entorhinal cortex (MEC), and one convergent subnetwork mimicking the hippocampus. LEC relays time-related visual information and MEC supplies space-related information in the form of multi-resolution grid codes as resulted from integrating movement information. The convergent subnetwork integrates all information from the parallel subnetworks and predicts the position of the agent in the environment. Synaptic weights of the vision-to-place and grid-to-place connections are learned based on the stochastic gradient descent algorithm. Simulations in a large virtual maze demonstrate that hippocampal place units in the model form multiple and irregularly-spaced place fields, similar to those observed in neurobiological experiments. The model is able to accurately decode the positions of the agent from the learned spatial representations. Moreover, the model is capable of adaptation to degraded visual inputs, and therefore is robust against perturbations. When the motion inputs are deprived, the model meets with localization difficulty, suffering from less accuracy in position predictions.

CFSRE: Context-aware based on frame-semantics for distantly supervised relation extraction

In relation extraction with distant supervision, noise labels are a bottleneck problem that hinders the performance of training models. Existing neural models solved this problem using attention mechanisms or multi-instance/multi-label learning to select sentences that are likely to express the relations. They mainly focused on the structural information of entity pairs and ignored semantic information of relation-clue words. Furthermore, these models do not consider the semantic scenario information of an entity pair (As the same entity pair may express different relations in different semantic scenarios). To bridge the gap, this paper proposes a novel model for relation extraction based on frame-semantics, namely Context-aware based on Frame-semantics for Distantly Supervised Relation Extraction (CFSRE). The model combines a hierarchical neural network architecture with FrameNet semantic knowledge base to solve the noise labels and the semantic representation of entity pairs. More specifically, a simple and effective instance selection method is used to select informative positive instances for model training. In addition, we propose a novel sentence representation method by combining sentence context representation with frame semantic representation of entities. We find that the joint representation leads to a better performance because it can obtain more comprehensive semantic representation of text instances. Then, a hierarchical-attention mechanism has been designed to select the most informative features for relation extraction. Finally, a softmax classifier is used to test the performance of the model. Experiments were conducted on two widely used benchmark datasets. The results demonstrate that the proposed method is significantly better than state-of-the-art methods for distantly supervised relation extraction.

Super-resolution visualization of subwavelength defects via deep learning-enhanced ultrasonic beamforming: A proof-of-principle study

Detecting small, subwavelength defect has known to be a challenging task mainly due to the diffraction limit, according to which the minimum resolvable size is in the order of the wavelength of a propagating wave. In this proof-of-concept study, we present a deep learning-enhanced super-resolution ultrasonic beamforming approach that computationally exceeds the diffraction limit and visualizes subwavelength defects. The proposed super-resolution approach is a novel subwavelength beamforming methodology enabled by a hierarchical deep neural network architecture. The first network (the detection network) globally detects defective regions from an ultrasonic beamforming image. Subsequently, the second network (the super-resolution network) locally resolves subwavelength-scale fine details of the detected defects. We validate the proposed approach using two independent datasets: a bulk wave array dataset generated by numerical simulations and guided wave array dataset generated by laboratory experiments. The results demonstrate that our deep learning super-resolution ultrasonic beamforming approach not only enables visualization of fine structural features of subwavelength defects, but also outperforms the existing widely-accepted super-resolution algorithm (time-reversal MUSIC). We also study key factors of the performance of our approach and discuss its applicability and limitations.

Node Deployment of Marine Monitoring Networks: A Multiobjective Optimization Scheme.

The increasing demands for real-time marine monitoring call for the wide deployment of Marine Monitoring Networks (MMNs). The low-rate underwater communications over a long distance, long propagation delay of underwater acoustic channel, and high deployment costs of marine sensors in a large-scale three-dimensional space bring great challenges in the network deployment and management of MMN. In this paper, we first propose a multitier, hierarchical network architecture of MMN with the support of edge computing (HMMN-EC) to enable efficient monitoring services in a harsh marine environment, taking into consideration the salient features of marine communications. Specifically, HMMN-EC is composed of three subnetworks, i.e., underwater acoustic subnetwork, the sea-surface wireless subnetwork, and the air wireless subnetwork, with a diversity of network nodes with different capabilities. We then jointly investigate the deployment diverse network nodes with various constraints in different subnetworks of HMMN-EC. To this end, we formulate a Multiobjective Optimization (MO) problem to minimize the network deployment cost while achieving the maximal network lifetime, subject to the limited energy of different marine nodes and the complex deployment environment. To solve the formulated problem, we present an Ant-Colony-based Efficient Topology Optimization (AC-ETO) algorithm to find the optimal locations of nodes in different subnetworks of MMN in a large-scale deployment. The time complexity of the proposed algorithm is also analyzed. Finally, extensive simulations are carried out to validate the superior performance of the proposed algorithm compared with some existing solutions.

Joint Channel Allocation and Resource Management for Stochastic Computation Offloading in MEC

To accommodate ever-increasing computational workloads while satisfying the requirements of delay-sensitive tasks, mobile edge computing (MEC) is proposed to offload the tasks to nearby edge servers. Nevertheless, it can introduce new technical issues in terms of transmission and computation overheads affected by the underlying offloading decisions. In this paper, we investigate the computation offloading problem in a hierarchical network architecture, where tasks can be offloaded to nearby micro-BS and further forwarded to macro-BS equipped with an MEC server. Specifically, we propose a scheme of joint channel allocation and resource management, named JCRM, to make offloading decisions and maximize the long-term network utility with considering stochastic task arrival/dispatch and dynamic changes in available resources. As the formulated utility maximization problem is a mixed-integer non-linear stochastic programming problem that is directly intractable, we, therefore, leverage the Lyapunov optimization technique to decouple the original problem into three separate sub-problems. Based on the solutions to those sub-problems, our proposed scheme can make optimal offloading-downloading decisions with maximizing the overall task offloading rate. Finally, we verify the long-term network stability and near-optimal performance of JCRM via both theoretical analysis and extensive simulations.

On recommendation-aware content caching for 6G: An artificial intelligence and optimization empowered paradigm

Recommendation-aware Content Caching (RCC) at the edge enables a significant reduction of the network latency and the backhaul load, thereby invigorating ubiquitous latency-sensitive innovative services. However, the effectiveness of RCC strategies is highly dependent on explicit information as regards subscribers’ content request patterns, the sophisticated caching placement policy, and the personalized recommendation tactics. In this article, we investigate how the potentials of Artificial Intelligence (AI) and optimization techniques can be harnessed to address those core issues and facilitate the full implementation of RCC for the upcoming intelligent 6G era. Towards this end, we first elaborate on the hierarchical RCC network architecture. Then, the devised AI and optimization empowered paradigm is introduced, whereas AI and optimization techniques are leveraged to predict the users’ content preferences in real-time situations with the assistance of their historical behavior data and determine the cache pushing and recommendation decision, respectively. Through extensive case studies, we validate the effectiveness of AI-based predictors in estimating users’ content preference and the superiority of optimized RCC policies over the conventional benchmarks. At last, we shed light on the opportunities and challenges in the future.

Predicting Age with Deep Neural Networks from Polysomnograms.

The aim of this study was to design a new deep learning framework for end-to-end processing of polysomnograms. This framework can be trained to analyze whole-night polysomnograms without the limitations of and bias towards clinical scoring guidelines. We validated the framework by predicting the age of subjects. We designed a hierarchical attention network architecture, which can be pre-trained to predict labels based on 5-minute epochs of data and fine-tuned to predict based on whole-night polysomnography recordings. The model was trained on 511 recordings from the Cleveland Family study and tested on 146 test subjects aged between 6 to 88 years. The proposed network achieved a mean absolute error of 7.36 years and a correlation to true age of 0.857. Sleep can be analyzed using our end-to-end deep learning framework, which we expect can generalize to learning other subject-specific labels such as sleep disorders. The difference in the predicted and chronological age is further proposed as an estimate of biological age.

Salient Object Detection by Fusing Local and Global Contexts

Benefiting from the powerful discriminative feature learning capability of convolutional neural networks (CNNs), deep learning techniques have achieved remarkable performance improvement for the task of salient object detection (SOD) in recent years. However, most existing deep SOD models do not fully exploit informative contextual features, which often leads to suboptimal detection performance in the presence of a cluttered background. This paper presents a context-aware attention module that detects salient objects by simultaneously constructing connections between each image pixel and its local and global contextual pixels. Specifically, each pixel and its neighbors bidirectionally exchange semantic information by computing their correlation coefficients, and this process aggregates contextual attention features both locally and globally. In addition, an attention-guided hierarchical network architecture is designed to capture fine-grained spatial details by transmitting contextual information from deeper to shallower network layers in a top-down manner. Extensive experiments on six public SOD datasets show that our proposed model demonstrates superior SOD performance against most of the current state-of-the-art models under different evaluation metrics.

Hierarchical Offloading for Delay-Constrained Applications in Fog RAN

By computation offloading, fog radio access network (RAN) can support delay-constrained applications, since such a hierarchical network architecture enables parallel computing at local computing layer (LCL), edge computing layer (ECL), and cloud computing layer (CCL). Comparing to processing applications exclusively at the LCL, the assistance of processing at the ECL and the CCL can reduce computation latency, while introduce extra communication latency. These two latency are not only influenced by the hierarchical offloading decision but also by the available communication and computational resources at multiple computing layers. In this regard, this paper aims to jointly provision communication and computational resources for a powerful hierarchical offloading mechanism in fog RAN to complete delay-constrained applications with high-efficiency resource utilization. Specifically, we formulate a deadline-constrained energy minimization problem, which is followed by feasibility analysis, effective solution, and application extension. Through theoretical analyses and simulation results, we provide some insights on hierarchical offloading conditions and demonstrate advantages of our proposed algorithms in terms of the minimum affordable latency, total energy consumption, and scalability.

FDIPP: False Data Injection Prevention Protocol for Smart Grid Distribution Systems.

In this paper, a false data injection prevention protocol (FDIPP) for smart grid distribution systems is proposed. The protocol is designed to work over a novel hierarchical communication network architecture that matches the distribution system hierarchy and its vast number of entities. The proposed protocol guarantees both system and data integrity via preventing packet injection, duplication, alteration, and rogue node access. Therefore, it prevents service disruption or damaging power network assets due to drawing the wrong conclusions about the current operating status of the power grid. Moreover, the impact of the FDIPP protocol on communication network performance is studied using intensive computer simulations. The simulation study shows that the proposed communication architecture is scalable and meets the packet delay requirements of inter-substation communication as mandated by IEC 61850-90-1 with a minimal packet loss while the security overhead of FDIPP is taken into account.

Joint Optimization of MapReduce Scheduling and Network Policy in Hierarchical Data Centers

As large-scale data analytic becomes norm in various industries, using MapReduce frameworks to analyze ever-increasing volumes of data will keep growing. In turn, this trend drives up the intention to move MapReduce into multi-tenant clouds. However, the application performance of MapReduce can be significantly affected by the time-varying network bandwidth in a shared cluster. Although many recent studies improve MapReduce performance by dynamic scheduling to reduce the shuffle traffic, most of them do not consider the impact by widely existing hierarchical network architectures in data centers. In this paper, we propose and design a Hierarchical topology (Hit) aware MapReduce scheduler to minimize overall data traffic cost and hence to reduce job execution time. We first formulate the problem as a Topology Aware Assignment (TAA) optimization problem while considering dynamic computing and communication resources in the cloud with hierarchical network architecture. We further develop a synergistic strategy to solve the TAA problem by using the stable matching theory, which ensures the preference of both individual tasks and hosting machines. Finally, we implement the proposed scheduler as a pluggable module on Hadoop YARN and evaluate its performance by testbed experiments and simulations. The testbed experimental results show Hit-scheduler can improve job completion time by 28% and 11% compared to Capacity Scheduler and Probabilistic Network-Aware scheduler, respectively. Our simulations further demonstrate that Hit-scheduler can reduce the traffic cost by 38% at most and the average shuffle flow traffic time by 32% compared to Capacity scheduler. In this manuscript, we have extended Hit-scheduler to a decentralized heuristic scheme to perform the policy-aware allocation in data center environments. Many existing centralized approximation approaches are too complex and infeasible to implement over a data center, which typically include large amounts of servers, containers, switches and traffic flows. In the extension, we have designed a decentralized heuristic scheme to perform the Policy-Aware Task (PAT) allocation by using existing centralize algorithm to approximately maximize the total gained utility. Finally, the simulation based experimental results show that the proposed PAT policy reduces the communication cost by 33.6% compared with the default scheduler in data centers.

Blockchain-Based Hierarchical Trust Networking for JointCloud

The Internet of Things (IoT) is gradually becoming mature and has already entered our daily life, which interconnects more machines and makes communication more convenient and more intelligent. Massive IoT devices produce innumerable data which need to be analyzed in joint cloud computation (JointCloud) with diversified services. However, due to the weak security of IoT devices, the existing JointCloud architecture hardly provides a secure trusted trade environment for users, which affects severely the application in the IoT network. In this article, we propose a hierarchical trust networking architecture based on permissioned blockchain to implement JointCloud (HTJC). The proposed Hyperledger fabric-based architecture has a better performance than those based on Ethereum in latency. By introducing the credit bonus-penalty strategy (CBPS), HTJC can solve the trust problem and provide users with a secure trusted trade environment. The availability of the proposed architecture is evaluated and compared to the existing models. The numerical results show that the HTJC can defend distributed denial-of-service (DDoS) attacks and provide users with a trusted and effective trade platform.

Synthesis of hierarchical Mn3O4 nanowires on reduced graphene oxide nanoarchitecture as effective pseudocapacitive electrodes for capacitive desalination application

We present a hierarchical network architecture fabrication of pseudocapacitive Mn3O4 nanowires immobilized on reduced graphene oxide (Mn3O4/RGO) hybrids for effective capacitive deionization (CDI). A hydrothermal synthesis process was employed for constructing Mn3O4/RGO nanoarchitecture with hierarchical pores to ease the interaction with salt ions. Physico-chemical analysis verified the Mn3O4 nanowires with few microns sized length and diameter of 20–30 nm were evenly immobilized on the surfaces of the RGO. With this tailored nanoarchitecture, the Mn3O4/RGO based electrode shows ideal pseudocapacitive behavior with a higher capacitance of (Cs) of 437 F g−1, the energy density of 41.12 Wh kg−1 and power density of 400 W kg−1 at 1 A g−1 in 1 M NaCl solution. Benefiting from the fascinating electrochemical features, the Mn3O4/RGO nanoarchitecture constructed CDI electrode exhibited high electrosorption capacity (SAC) of 34.5 mg g−1 at 1.2 V with a high salt adsorption rate (ASAR) of 1.15 mg g−1 min−1 in 1000 mg L−1 NaCl solution. The much improved SAC, ASAR, and recyclability could be attributed to the distinctive pseudocapacitive nanoarchitecture, which improves the sodiation/desodiation. The present investigation indicates that the Mn3O4/RGO nanoarchitecture is a capable CDI electrode material for desalination application.

Hierarchical Information-Centric Networking Framework

In information-centric networking, one content publishing operation involves a great number of content routers, and different name prefixes pollute forwarding tables and result in explosion of forwarding entries. Moreover, flooding is employed to achieve information-centric communications, and this flooding incurs huge costs. Taking these issues into account, this paper proposes a hierarchical information-centric networking framework and aims to reduce the content communication costs and latency. This framework proposes the hierarchical information-centric networking architecture and name structure to achieve forwarding aggregation and avoid forwarding table pollution. This framework is evaluated, and the data show that it reduces the content communication cost and latency by nearly 49.2% and 9.5%, respectively.

Experimental Demonstration of a Petabit per Second SDM Network Node

This paper extends a recent demonstration of a hierarchical spatial division network architecture testbed for petabit per second throughput. A spatial cross-connect is implemented to handle 1.08 Pb/s per line side. The spatial cross-connect is based on a low loss 44 × 44 MEMS switch, which enables switching up to 22 spatial channels, each carrying 202-channel, 24.5 GBaud polarization-multiplexed 64-ary quadrature amplitude modulated signals. Transmission is performed using a 22-core homogeneous multicore fiber. The demonstration includes spatial bypass and spatial add and drop scenarios as well as optical bypass and optical add and drop scenarios. The latter uses wavelength cross-connects implemented using conventional wavelength selective switches. A network scenario implementing 1 + 1 spatial multiplex section protection is also demonstrated using three 7-core joint spatial switches to protect 21 spatial channels. Finally, the use of hybrid spatial division multiplexing networks with multi-core and few-mode fibers is addressed, showing that skew between spatial channels may be the most significant obstacle to the implementation of such networks. This work concludes with an analysis of the inter-core skew introduced by the 22-core multicore fiber and its impact on a potential implementation of a hierarchical spatial division network.

Cluster-Based Control Plane Messages Management in Software-Defined Flying Ad-Hoc Network.

Collaboration between multiple Unmanned Aerial Vehicles (UAVs) to establish a Flying Ad-hoc Network (FANET) is a growing trend since future applications claim for more autonomous and rapidly deployable systems. In this context, Software-Defined Networking FANET (SDN-FANET ) separates the control and data plane and provides network programmability, which considers a centralized controller to perform all FANET control functions based on global UAV context information, such as UAV positions, movement trajectories, residual energy, and others. However, control message dissemination in an SDN-FANET with low overhead and high performance is not a trivial task due to FANET particular characteristics, i.e., high mobility, failures in UAV to UAV communication, and short communication range. With this in mind, it is essential to predict UAV information for control message dissemination as well as consider hierarchical network architecture, reducing bandwidth consumption and signaling overhead. In this article, we present a Cluster-bAsed control Plane messages management in sOftware-defined flying ad-hoc NEtwork, called CAPONE. Based on UAV contextual information, the controller can predict UAV information without control message transmission. In addition, CAPONE divides the FANET into groups by computing the number of clusters using the Gap statistics method, which is input for a Fuzzy C-means method to determine the group leader and members. In this way, CAPONE reduces the bandwidth consumption and signaling overhead, while guaranteeing the control message delivering in FANET scenarios. Extensive simulations are used to show the gains of the CAPONE in terms of Packet Delivery Ratio, overhead, and energy compared to existing SDN-FANET architectures.

Air-Ground Heterogeneous Networks for 5G and Beyond via Integrating High and Low Altitude Platforms

UAVs are expected to be an important complementary component for 5G (and beyond) communication systems to achieve the goal of global access to the Internet for all. To fully exploit the benefits of the distinct features of various UAVs, this article proposes a novel hierarchical network architecture enabled by software defined networking, which integrates cross-layer high and low altitude platforms into conventional terrestrial cellular networks to inject additional capacity and expand the coverage for underserved areas in a flexible, seamless, and cost-effective manner. Specifically, we first present a comprehensive comparison and review of different types of UAVs for communication services. Then, we propose an integrated airground heterogeneous network architecture and outline its characteristics and potential advantages. Next, several key enabling techniques for the integrated system are discussed in detail. In addition, we identify the potential application scenarios where the system can further enhance the performance of traditional terrestrial networks, followed by a case study to demonstrate the effectiveness of the proposed architecture. Finally, the discussions on challenges and open research issues are given.