arXiv subjects

Yi Zhong

Publications and source records attributed to Yi Zhong.

At least 91 records · Page 5Linked to original sources

arXiv2020

Effect of Spatial and Temporal Traffic Statistics on the Performance of Wireless Networks

The traffic in wireless networks has become diverse and fluctuating both spatially and temporally due to the emergence of new wireless applications and the complexity of scenarios. The purpose of this paper is to quantitatively analyze the impact of the wireless traffic, which fluctuates both spatially and temporally, on the performance of the wireless networks. Specially, we propose to combine the tools from stochastic geometry and queueing theory to model the spatial and temporal fluctuation of traffic, which to our best knowledge has seldom been evaluated analytically. We derive the spatial and temporal statistics, the total arrival rate, the stability of queues and the delay of users by considering two different spatial properties of traffic, i.e., the uniformly and non-uniformly distributed cases. The numerical results indicate that although the fluctuation of traffic (reflected by the variance of total arrival rate) when the users are clustered is much fiercer than that when the users are uniformly distributed, the unstable probability is smaller. Our work provides a useful reference for the design of wireless networks when the complex spatio-temporal fluctuation of the traffic is considered.

cs.NI↗

arXiv2020

Spatio-temporal Modeling for Massive and Sporadic Access

The vision for smart city imperiously appeals to the implementation of Internet-of-Things (IoT), some features of which, such as massive access and bursty short packet transmissions, require new methods to enable the cellular system to seamlessly support its integration. Rigorous theoretical analysis is indispensable to obtain constructive insight for the networking design of massive access. In this paper, we propose and define the notion of massive and sporadic access (MSA) to quantitatively describe the massive access of IoT devices. We evaluate the temporal correlation of interference and successful transmission events, and verify that such correlation is negligible in the scenario of MSA. In view of this, in order to resolve the difficulty in any precise spatio-temporal analysis where complex interactions persist among the queues, we propose an approximation that all nodes are moving so fast that their locations are independent at different time slots. Furthermore, we compare the original static network and the equivalent network with high mobility to demonstrate the effectiveness of the proposed approximation approach. The proposed approach is promising for providing a convenient and general solution to evaluate and design the IoT network with massive and sporadic access.

cs.IT↗

arXiv2020

Using Deep Convolutional Neural Networks to Diagnose COVID-19 From Chest X-Ray Images

The COVID-19 epidemic has become a major safety and health threat worldwide. Imaging diagnosis is one of the most effective ways to screen COVID-19. This project utilizes several open-source or public datasets to present an open-source dataset of COVID-19 CXRs, named COVID-19-CXR-Dataset, and introduces a deep convolutional neural network model. The model validates on 740 test images and achieves 87.3% accuracy, 89.67 % precision, and 84.46% recall, and correctly classifies 98 out of 100 COVID-19 x-ray images in test set with more than 81% prediction probability under the condition of 95% confidence interval. This project may serve as a reference for other researchers aiming to advance the development of deep learning applications in medical imaging.

eess.IV↗

arXiv2020

Triple Memory Networks: a Brain-Inspired Method for Continual Learning

Continual acquisition of novel experience without interfering previously learned knowledge, i.e. continual learning, is critical for artificial neural networks, but limited by catastrophic forgetting. A neural network adjusts its parameters when learning a new task, but then fails to conduct the old tasks well. By contrast, the brain has a powerful ability to continually learn new experience without catastrophic interference. The underlying neural mechanisms possibly attribute to the interplay of hippocampus-dependent memory system and neocortex-dependent memory system, mediated by prefrontal cortex. Specifically, the two memory systems develop specialized mechanisms to consolidate information as more specific forms and more generalized forms, respectively, and complement the two forms of information in the interplay. Inspired by such brain strategy, we propose a novel approach named triple memory networks (TMNs) for continual learning. TMNs model the interplay of hippocampus, prefrontal cortex and sensory cortex (a neocortex region) as a triple-network architecture of generative adversarial networks (GAN). The input information is encoded as specific representation of the data distributions in a generator, or generalized knowledge of solving tasks in a discriminator and a classifier, with implementing appropriate brain-inspired algorithms to alleviate catastrophic forgetting in each module. Particularly, the generator replays generated data of the learned tasks to the discriminator and the classifier, both of which are implemented with a weight consolidation regularizer to complement the lost information in generation process. TMNs achieve new state-of-the-art performance on a variety of class-incremental learning benchmarks on MNIST, SVHN, CIFAR-10 and ImageNet-50, comparing with strong baseline methods.

cs.LG↗

arXiv2020

An Actor-Critic-Based UAV-BSs Deployment Method for Dynamic Environments

In this paper, the real-time deployment of unmanned aerial vehicles (UAVs) as flying base stations (BSs) for optimizing the throughput of mobile users is investigated for UAV networks. This problem is formulated as a time-varying mixed-integer non-convex programming (MINP) problem, which is challenging to find an optimal solution in a short time with conventional optimization techniques. Hence, we propose an actor-critic-based (AC-based) deep reinforcement learning (DRL) method to find near-optimal UAV positions at every moment. In the proposed method, the process searching for the solution iteratively at a particular moment is modeled as a Markov decision process (MDP). To handle infinite state and action spaces and improve the robustness of the decision process, two powerful neural networks (NNs) are configured to evaluate the UAV position adjustments and make decisions, respectively. Compared with the heuristic algorithm, sequential least-squares programming and fixed UAVs methods, simulation results have shown that the proposed method outperforms these three benchmarks in terms of the throughput at every moment in UAV networks.

cs.NI↗

arXiv2019

Edge Caching with Real-Time Guarantees

In recent years, optimization of the success transmission probability in wireless cache-enabled networks has been studied extensively. However, few works have concerned about the real-time performance of the cache-enabled networks. In this paper, we investigate the performance of the cache-enabled with real-time guarantees by adopting the age of information (AoI) as the metric to characterize the timeliness of the delivered information. We establish a spatial-temporal model by utilizing the stochastic geometry and queuing theory which captures both the temporal traffic dynamics and the interferers' geographic distribution. Under the random caching framework, we achieve the closed-form expression of AoI by adopting the maximum average received power criterion for the user association. Finally, we formulate a convex optimization problem for the minimization of the Peak AoI (PAoI) and obtain the optimal caching probabilities by utilizing the Karush-Kuhn-Tucker (KKT) conditions. Numerical results demonstrate that the random caching strategy is a better choice rather than the most popular content and uniform caching strategy to improve the real-time performance for the cached file as well as maintaining the file diversity at a relatively high level.

cs.IT↗

arXiv2019

Graph-augmented Convolutional Networks on Drug-Drug Interactions Prediction

We propose an end-to-end model to predict drug-drug interactions (DDIs) by employing graph-augmented convolutional networks. And this is implemented by combining graph CNN with an attentive pooling network to extract structural relations between drug pairs and make DDI predictions. The experiment results suggest a desirable performance achieving ROC at 0.988, F1-score at 0.956, and AUPR at 0.986. Besides, the model can tell how the two DDI drugs interact structurally by varying colored atoms. And this may be helpful for drug design during drug discovery.

cs.LG↗

arXiv2019

A New Small-World IoT Routing Mechanism based on Cayley Graphs

An increasing number of low-power Internet of Things (IoT) devices will be widely deployed in the near future. Considering the short-range communication of low-power devices, multi-hop transmissions will become an important transmission mechanism in IoT networks. It is crucial for lowpower devices to transmit data over long distances via multihop in a low-delay and reliable way. Small-world characteristics of networks indicate that the network has an advantage of a small Average Shortest-path Length (ASL) and a high Average Clustering Coefficient (ACC). In this paper, a new IoT routing mechanism considering small-world characteristics is proposed to reduce the delay and improve the reliability. The ASL and ACC are derived for performance analysis of small-world characteristics in IoT networks based on Cayley graphs. Besides, the reliability and delay models are proposed for Small-World IoT based on Cayley grapHs (SWITCH). Simulation results demonstrate that SWITCH has lower delay and better reliability than that of conventional Nearest Neighboring Routing (NNR). Moreover, the maximum delay of SWITCH is reduced by 50.6% compared with that by NNR.

cs.NI↗

arXiv2019

SBSGAN: Suppression of Inter-Domain Background Shift for Person Re-Identification

Cross-domain person re-identification (re-ID) is challenging due to the bias between training and testing domains. We observe that if backgrounds in the training and testing datasets are very different, it dramatically introduces difficulties to extract robust pedestrian features, and thus compromises the cross-domain person re-ID performance. In this paper, we formulate such problems as a background shift problem. A Suppression of Background Shift Generative Adversarial Network (SBSGAN) is proposed to generate images with suppressed backgrounds. Unlike simply removing backgrounds using binary masks, SBSGAN allows the generator to decide whether pixels should be preserved or suppressed to reduce segmentation errors caused by noisy foreground masks. Additionally, we take ID-related cues, such as vehicles and companions into consideration. With high-quality generated images, a Densely Associated 2-Stream (DA-2S) network is introduced with Inter Stream Densely Connection (ISDC) modules to strengthen the complementarity of the generated data and ID-related cues. The experiments show that the proposed method achieves competitive performance on three re-ID datasets, ie., Market-1501, DukeMTMC-reID, and CUHK03, under the cross-domain person re-ID scenario.

cs.CV↗

arXiv2019

Power-Consumption Outage Challenge in Next-Generation Cellular Networks

The conventional outage in wireless communication systems is caused by the deterioration of the wireless communication link, i.e., the received signal power is less than the minimum received signal power. Is there a possibility that the outage occurs in wireless communication systems with a good channel state? Based on both communication and heat transfer theories, a power-consumption outage in the wireless communication between millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) base stations (BSs) and smartphones has been modeled and analyzed. Moreover, the total transmission time model with respect to the number of power-consumption outages is derived for mmWave massive MIMO communication systems. Simulation results indicate that the total transmission time is extended by the power-consumption outage, which deteriorates the average transmission rate of mmWave massive MIMO BSs.

cs.NI↗

arXiv2019

Gravitational resonances in mimetic thick branes

In this work, we investigate gravitational resonances in both single and double mimetic thick branes, which can provide a new way to detect the extra dimension. For the single brane model, we apply the relative probability proposed in [Phys. Rev. D. 80 (2009) 065019]. For the double brane model, we investigate the resonances quasi-localized on the double brane, on the sub-branes and between the sub-branes, respectively. To investigate the resonances quasi-localized on the double brane, we introduce two different definitions of the relative probability and find that the corresponding mass spectra of gravitational resonances are almost the same. For the gravitational resonances quasi-localized on sub-branes and between the sub-branes, the influence of the distance between the two sub-branes and the thickness of the sub-branes are analyzed and new features are found in both cases.

gr-qc↗

arXiv2019

Energy Efficient Optimization of Wireless-powered 5G Full Duplex Cellular Networks: A Mean Field Game Approach

This paper studies the power allocation of an ultra-dense cellular network consisting of multiple full-duplex (FD) base stations (BSs) serving a large number of half-duplex (HD) user equipments (UEs) located in a wide geographical area. Each BS consists of a baseband unit (BBU) that is responsible for signal processing and base station control, and a radio remote unit (RRU) that corresponds to a radio transceiver remotely built closer to the UEs. We consider a wireless-powered cellular network in which the BBU can periodically charge the RRU. We model the energy efficiency and coverage optimization problem for this network as a mean field game. We consider the weighted energy efficiency of the BS as the main performance metrics and evaluate the optimal strategy that can be adopted by the FD BSs in an ultra-dense network setting. Based on the interference and network energy efficiency models of the mean field game theory, Nash equilibrium of our proposed game is derived. Utilizing solutions of Hamilton-Jacobi-Bellman (HJB) and Fokker-Planck-Kolmogorov (FPK) equations, a new power transmission strategy is developed for the BSs to optimize the energy efficiency of 5G cellular networks with full duplex transmissions. Simulation results indicate that the proposed strategy not only improves the energy efficiency but also ensures the average network coverage probability converges to a stable level.

cs.NI↗

arXiv2019

A Novel JT-CoMP Scheme in 5G Fractal Small Cell Networks

To satisfy the requirement of the fifth generation (5G) mobile communications that offers an ultra high data rate of 100Mbps to 1Gbps anytime and anywhere, the coordinated multipoint (CoMP) technique is proposed to mitigate inter-cell interference to improve the coverage of high data rate services, cell-edge throughput, and system capacity. However, the joint transmission (JT) CoMP technique is difficult to be applied in practice due to the critical time synchronization for multiple coordination links and the bottleneck of backhaul capacity and radio resource at each small cell base stations (SBSs). Moreover, since the coordination SBSs in the conditional scheme are entirely separate from each other, different time of arrivals at the user cause the severe time synchronization problem. The anisotropic propagation environment in the urban scenario makes the implementation condition even worse. To tackle these issues, we propose a novel JT-CoMP scheme with the anisotropic path loss model to minimize the network backhaul traffic subject to the constraints on the radio resource and the differences in time of arrivals. Simulation results demonstrate that the proposed distance-resource-limited CoMP scheme can obtain the maximum achievable rate with the minimum network backhaul traffic, compared with existing schemes.

cs.NI↗

arXiv2018

The Stochastic Geometry Analyses of Cellular Networks with α-Stable Self-Similarity

To understand the spatial deployment of base stations (BSs) is the first step to analyze the performance of cellular networks and further design efficient networking protocols. Poisson point process (PPP), which has been widely adopted to characterize the deployment of BSs and established the reputation to give tractable results in the stochastic geometry analyses, usually assumes a static BS deployment density in homogeneous PPP (HPPP) models or delicately designed location-dependent density functions in in-homogeneous PPP (IPPP) models. However, the simultaneous existence of attractiveness and repulsiveness among BSs practically deployed in a large-scale area defies such an assumption, and the $α$-stable distribution, one kind of heavy-tailed distributions, has recently demonstrated superior accuracy to statistically model the varying BS density in different areas. In this paper, we start with these new findings and investigate the intrinsic feature (i.e., the spatial self-similarity) embedded in the BSs. Afterwards, we refer to a generalized PPP setup with $α$-stable distributed density and theoretically derive the related coverage probability. In particular, we give an upper bound of the derived coverage probability for high signal-to-interference-plus-noise ratio (SINR) thresholds and show the monotonically decreasing property of this bound with respect to the variance of BS density. Besides, we prove that our model could reduce to the single-tier HPPP for some special cases, and demonstrate the superior accuracy of the $α$-stable model to approach the real environment.

cs.IT↗

arXiv2018

Age of Information in Poisson Networks

The age of information (AoI) has been extensively studied in recent years. However, few works have focused on the analysis of AoI in large wireless networks. In this work, we analyze this metric to characterize timeliness in a Poisson bipolar network, in which spatial distribution, fading, and interference are taken into consideration. We further study the effect of deadline constraint on the AoI. We derive upper and lower bounds for the cumulative distribution function (cdf) of the average age in network. Numerical results show that the inclusion of deadline constraint can substantially improve the performance of the system.

cs.IT↗

arXiv2018

How Much of Wireless Rate Can Smartphones Support in 5G Networks?

Due to the higher wireless transmission rates in the fifth generation (5G) cellular networks, higher computation overhead is incurred in smartphones, which can cause the wireless transmission rates to be limited by the computation capability of wireless terminals. In this case, is there a maximum receiving rate for smartphones to maintain stable wireless communications in 5G cellular networks? The main objective of this article is to investigate the maximum receiving rate of smartphones and its influence on 5G cellular networks. Based on Landauer's principle and the safe temperature bound on the smartphone surface, a maximum receiving rate of the smartphone is proposed for 5G cellular networks. Moreover, the impact of the maximum receiving rate of smartphones on the link adaptive transmission schemes has been investigated. Numerical analyses imply that the maximum receiving rate of smartphones cannot always catch up with the downlink rates of future 5G cellular networks. Therefore, the link adaptive transmission scheme for future 5G cellular networks has to take the maximum receiving rate of smartphones into account.

cs.NI↗

arXiv2018

Base Station Switch-off with Mutual Repulsion in 5G Massive MIMO Networks

When small cells are densely deployed in the fifth generation (5G) cellular networks, switching off a part of base stations (BSs) is a practical approach for saving energy consumption considering the variation of traffic load. The small cell network with the massive multi-input multi-output (massive MIMO) system is analyzed in this paper due to the dense deployment and low power consumption. Based on the BS switch-off strategy with distance constraints, the energy and coverage efficiency are investigated to illustrate the performance of the BS switch-off strategy. Simulation results indicate that the energy efficiency and coverage efficiency of the proposed strategy are better than the random strategy. The energy efficiency increases with the BS intensity and the minimal distance, and a maximum coverage efficiency can be achieved with the increase of the BS intensity and the minimum distance. In this case, the optimal BS switch-off strategy can be designed under this work in the actual scene.

cs.NI↗

arXiv2018

Inflation in mimetic f(G) gravity

Mimetic gravity is analysed in the framework of some extensions of General Relativity, where a function of the Gauss-Bonnet invariant in four dimensions is considered. By assuming the so-called mimetic condition, the conformal degree of freedom is isolated and a pressureless fluid naturally arises. Then, the complete set of field equations for mimetic Gauss-Bonnet gravity is established and some inflationary models are analysed, for which the corresponding gravitational action is reconstructed. The spectral index and tensor-to-scalar ratio are obtained and compared with observational bounds from Planck and BICEP2/Keck array data. The full agreement with above data is achieved for several versions of the mimetic Gauss-Bonnet gravity. Finally, some extensions of Gauss-Bonnet mimetic gravity are considered and the possibility of reproducing inflation is also explored.

gr-qc↗