Skip to main content

Über dieses Buch

The book constitutes the refereed proceedings of the 13th EAI International Conference on Communications and Networking, held in October 2018 in Chengdu, China. The 71 papers presented were carefully selected from 114 submissions. The papers are organized in topical sections on wireless communications and networking, next generation WLAN, big data networks, cloud communications and networking, ad hoc and sensor networks, satellite and space communications and networking, optical communications and networking, information and coding theory, multimedia communications and smart networking, green communications and computing, signal processing for communications, network and information security, machine-to-machine and IoT, communication QoS, reliability and modeling, cognitive radio and networks, smart internet of things modeling, pattern recognition and image signal processing, digital audio and video signal processing, antenna and microwave communications, radar imaging and target recognition, and video coding and image signal processing.



Main Track


Joint QoS-Aware Downlink and Resource Allocation for Throughput Maximization in Narrow-Band IoT with NOMA

Narrow-Band Internet of Things (NB-IoT) is 3GPPs cellular technology designed for Low-Power Wide Area Network (LPWN) and it is a promising approach that NB-IoT combines with NOMA which is designed for accommodating more devices in the 5G era. Previous works mainly focus on uplink channel resource allocation to achieve connectivity maximization in NB-IoT with NOMA; however, few articles consider NB-IoT downside issues and downlink resource allocation problem to achieve maximum system throughput has not been studied in NB-IoT with NOMA. Thus, in this paper to provide a reliable and seamless service for NB-IoT users (NUs) and maximizing network downlink throughput, we propose a resource allocation algorithm for joint equipment QoS requirements and resource allocation fairness. In this scheme, we design algorithm to implement the mapping between NUs and subchannels for suboptimal system throughput. Then we convert the power allocation problem of the NUs on the same subchannel into a DC problem and we design algorithm to solve it to get suboptimal solution. Numerical results show that the proposed scheme achieves a better performance compared with exiting schemes in terms of the system throughput.

Wei Chen, Heli Zhang, Hong Ji, Xi Li

MIMO-UFMC Transceiver Schemes for Millimeter Wave Wireless Communications

This paper provides results on the use of UFMC modulation scheme in MIMO wireless links operating at mmWave frequencies. First of all, full mathematical details on the processing needed to realize a MIMO-UFMC transceiver at mmWave, taking into account also the hybrid analog/digital nature of the beamformers, are given. Then, we propose several reception structures, considering also the case of continuous packet transmission with no guard intervals among the packets. In particular, an adaptive low complexity MMSE receiver is proposed that is shown to achieve very satisfactory performance. A channel independent transmit beamformer is also considered, as as to avoid the need for channel state information at the transmitter. Numerical results show that the proposed transceiver schemes are effective, as well as that the continuous packet transmission scheme, despite increased interference, attains the highest values of system throughput.

Stefano Buzzi, Carmen D’Andrea, Dejian Li, Shulan Feng

A Novel Mixed-Variable Fireworks Optimization Algorithm for Path and Time Sequence Optimization in WRSNs

To prolong the lifespan of the network, the auxiliary charging equipment is introduced into the traditional Wireless Sensor Networks (WSNs), known as Wireless Rechargeable Sensor Networks (WRSNs). Different from existing researches, in this paper, a periodic charging and data collecting model in WRSNs is proposed to keep the network working perpetually and improve data collection ratio. Meanwhile, the Wireless Charging Vehicle (WCV) has more working patterns, charging, waiting, and collecting data when staying at the sensor nodes. Then, the simultaneous optimization for the traveling path and time sequence is formulated to be a mixed-variable optimization problem. A novel Mixed-Variable Fireworks Optimization Algorithm (MVFOA) is proposed to solve it. A large number of experiments show the feasibility of the MVFOA, and MVFOA is superior to the Greedy Algorithm.

Chengkai Xia, Zhenchun Wei, Zengwei Lyu, Liangliang Wang, Fei Liu, Lin Feng

Aggregating Multidimensional Wireless Link Information for Device-Free Localization

Device-free localization (DFL) is an emerging and promising technique, which can realize target localization without the requirement of attaching any wireless devices to targets. By analyzing the shadowing loss caused by targets on wireless links, we can estimate the target locations. However, for existing DFL approaches, a large number of wireless links is required to guarantee a certain localization precision, which may lead to high hardware cost. In this paper, we propose a novel multi-target device-free localization method with multidimensional wireless link information (MDMI). Unlike previous works that measure RSS only on a single transmission power level, MDMI collects RSS measurements from multiple transmission power levels to enrich the measurement information. Furthermore, the compressive sensing (CS) theory is applied by exploiting the inherent spatial sparsity of DFL. We model the DFL problem as a joint sparse recovery problem and adopt the multiple sparse Bayesian learning (M-SBL) algorithm to reconstruct the sparse vectors of different transmission power levels. Numerical simulation results demonstrate the outstanding performance of the proposed method.

Dongping Yu, Yan Guo, Ning Li, Sixing Yang

Software Defined Industrial Network: Architecture and Edge Offloading Strategy

The integration of the internet and the traditional manufacturing industry has identified the “Industrial Internet of Things” (IIoT) as a popular research topic. However, traditional industrial networks continue to face challenges of resource management and limited raw data storage and computation capacity. In this paper, we propose a Software Defined Industrial Network (SDIN) architecture to address the existing drawbacks in IIoT such as resource utilization, data processing and system compatibility. The architecture is developed based on the Software Defined Network (SDN) architecture, combining hierarchical cloud and edge computing technologies. Based on the SDIN architecture, a novel centralized computation offloading strategy in industrial application is proposed. The simulation results confirm that the SDIN architecture is feasible and effective in the application of edge computing.

Fangmin Xu, Huanyu Ye, Shaohua Cui, Chenglin Zhao, Haipeng Yao

Multi-agent Deep Reinforcement Learning Based Adaptive User Association in Heterogeneous Networks

Nowadays, lots of technical challenges emerge focusing on user association in ever-increasingly complicated 5G heterogeneous networks. With distributed multiple attribute decision making (MADM) algorithm, users tend to maximize their utilities selfishly for lack of cooperation, leading to congestion. Therefore, it is efficient to apply artificial intelligence to deal with these emerging problems, which enables users to learn with incomplete environment information. In this paper, we propose an adaptive user association approach based on multi-agent deep reinforcement learning (RL), considering various user equipment types and femtocell access mechanisms. It aims to achieve a desirable trade-off between Quality of Experience (QoE) and load balancing. We formulate user association as a Markov Decision Process. And a deep RL approach, semi-distributed deep Q-network (DQN), is exploited to get the optimal strategy. Individual reward is defined as a function of transmission rate and base station load, which are adaptively balanced by a designed weight. Simulation results reveal that DQN with adaptive weight achieves the highest average reward compared with DQN with fixed weight and MADM, which indicates it obtains the best trade-off between QoE and load balancing. Compared with MADM, our approach improves by $${4\%\sim 11\%}$$ , $${32\%\sim 40\%}$$ , $${99\%}$$ in terms of QoE, load balancing and blocking probability, respectively. Furthermore, semi-distributed framework reduces computational complexity.

Weiwen Yi, Xing Zhang, Wenbo Wang, Jing Li

A Novel Double Modulation Technique with High Spectrum Efficiency for TDCS

The modulation techniques in traditional transform domain communication system (TDCS) exist some drawbacks, such as low transmission rate and low spectrum efficiency. We propose a novel double modulation technique with high spectrum efficiency for TDCS in this paper. First, we divide the basis function averagely into several orthogonal modules, and conduct the CSK modulation. Then, the double modulation signal waveform can be obtained by employing bipolar modulation for different module combination. Furthermore, we propose two demodulation schemes for the proposed modulation technique, namely the cyclic shift keying (CSK)-bipolar and bipolar-CSK demodulation. We also derive the mathematical expressions of their bit error rate (BER) performance. Simulation results show that for different signal-to-noise ratio (SNR), the two demodulation schemes can both achieve reliable performance, satisfactory anti-interference capabilities and effectively improve spectrum efficiency. In addition, it can be verified that CSK-bipolar demodulation can achieve the same BER with less SNR compared with bipolar-CSK demodulation.

Bo Zheng, Heng-Yang Zhang, Le Sun, Hua-Xin Wu, Wei-Lun Liu

Quality of Experience Prediction of HTTP Video Streaming in Mobile Network with Random Forest

As video is witnessing a rapid growth in mobile networks, it is crucial for network service operators to understand if and how Quality of Service (QoS) metrics affect user engagement and how to optimize users’ Quality of Experience (QoE). Our aim in this paper is to infer the QoE from the observable QoS metrics using machine learning techniques. For this purpose, Random Forest is applied to predict three objective QoE metrics, i.e., rebuffering frequency, mean bitrate and bitrate switch frequency, with the initial information of each video session. In our simulation, QoE of four different video streamings are analyzed with eight different system loads. Results show that sufficient prediction accuracy can be achieved for all QoE metrics with the attributes we adopted, especially with low and middle system loads. In terms of type of streamings, the prediction of all metrics for static users performs better than mobile users. Feature selection is also implemented under the highest load to examine the effect of different attributes on each QoE metric and the correlation among attributes.

Yue Yu, Yu Liu, Yumei Wang

Performance of Linearly Modulated SIMO High Mobility Systems with Channel Estimation Errors

This paper studies the error performance of linearly modulated single-input multiple-output (SIMO) high mobility communication systems with channel estimation errors. Channel estimation errors are unavoidable in high mobility systems, due to the rapid time-varying fading of the channel caused by severe Doppler effects, and this might have non-negligible adverse impacts on system performance. However, in high mobility communications, rapid time-varying fading channels induce Doppler diversity which can be exploited to improve system performance. Based on the statistical attributes of minimum mean square error (MMSE) channel estimation, a new optimum diversity receiver for MASK, MPSK and MQAM SIMO high mobility systems with channel estimation errors is proposed. The exact analytical error probability expressions of MPSK, MASK, and MQAM of the SIMO diversity receiver are identified and expressed as a unified expression. It quantifies the impacts of both Doppler diversity and channel estimation errors. The result is expressed as an explicit function of the channel temporal correlation, pilot and data signal-to-noise ratios (SNRs). Simulations results are used to validated analytical results. Simulation results show that MPSK, MASK, and MQAM systems have the same Doppler diversity order even though they differ in symbol error rates(SERs). Moreover, simulation results show that MQAM systems achieve better spectral efficiency than its MPSK and MASK counterparts.

Mahamuda Alhaji Mahamadu, Zheng Ma

Minimum Cost Offloading Decision Strategy for Collaborative Task Execution of Platooning Assisted by MEC

In this paper, we study the offloading decision of collaborative task execution between platoon and MEC (Mobile Edge Computing) server. The mobile application is represented by a series of fine-grained tasks that form a linear topology, each of which is either executed on a local vehicle, offloaded to other members of the platoon, or offloaded to a MEC server. The objective of the design is to minimize the cost of task offloading and meet the deadline of tasks execution. We transform the cost minimized task decision problem into the shortest path problem, which is limited by the deadline of the tasks on a directed acyclic graph. The classical LARAC algorithm is used to solve the problem approximately. Numerical analysis shows that the scheduling method of the tasks decision can be well applied to the platoon scenario and execute the task in cooperation with the MEC server. In addition, compared with different execution models, the optimal offloading decision for collaborative task execution can significantly reduce the cost of task execution and meet lower deadlines.

Taiping Cui, Xiayan Fan, Chunyan Cao, Qianbin Chen

Cluster-Based Caching Strategy with Limited Storage in Ultra Dense Networks

Ultra dense network (UDN) is considered as one of the key techniques to boost the network capacity in 5G. In order to reduce the huge backhaul cost and end-to-end transmission delay, caching the popular content at the edge of UDNs is an inspiring approach. Considering that the storage capacity of a single small base station (SBS) in UDNs is usually limited, SBSs cooperation to store respective file fragments is an interesting approach that needs further investigation. In this paper, we propose a cluster-based caching strategy (CBCS) for limited storage SBSs in UDNs. A novel clustering scheme based on SBSs’s load capacity and location is designed with consideration on files fragments and SBSs cooperation. We target the minimum average download delay under the constraint of the number of SBSs in a cluster. The simulation results show that the proposed algorithm could achieve a better hit ratio and has a lower average download delay.

Chengjia Hu, Xi Li, Hong Ji, Heli Zhang

Image Retrieval Research Based on Significant Regions

Deep Convolution neural networks (CNN) has achieved great success in the field of image recognition. But in the image retrieval task, the global CNN features ignore local detail description for paying too much attention to semantic information of images. So the MAP of image retrieval remains to be improved. Aiming at this problem, this paper proposes a local CNN feature extraction algorithm based on image understanding, which includes three steps: significant regions extraction, significant regions description and pool coding. This method overcomes the semantic gap problem in traditional local characteristic and improves the retrieval effect of global CNN features. Then, we apply this local CNN feature in the image retrieval task, including the same category retrieval task by feature fusion strategy and the instance retrieval task by re-ranking strategy. The experimental results show that this method has achieved good performance on the Caltech 101 and Caltech 256 classification datasets, and competitive results on the Oxford 5k and Paris 6k instance retrieval datasets.

Jie Xu, Shuwei Sheng, Yuhao Cai, Yin Bian, Du Xu

Partial Systematic Polar Coding

Due to having a better performance of bit error rate (BER), systematic polar codes have been potentially applied in digital data transmission. In the systematic polar coding, source bits are transmitted transparently. In this paper, we propose a scheme of novel partial systematic polar coding in which the encoded codeword is only composed of partial source bits with respect to the encoded word of systematic polar codes. To effectively reduce the resource consumption of the systematic encoder/decoder under all-zero frozen bits, the partial systematic polar codes are introduced subsequently. Then the simulation results in terms of core $$ F = \left[ {\begin{array}{*{20}l} 1 \hfill & 0 \hfill \\ 1 \hfill & 1 \hfill \\ \end{array} } \right] $$ are provided to demonstrate the aforementioned analysis with negligible difference of BER performance.

Hongxu Jin, Rongke Liu

An Adaptive Code Rate Control of Polar Codes in Time-Varying Gaussian Channel

Under a benchmark of bit error rate (BER) in data transmission, a just perfect trade-off between maximizing code rate (CR) and reliable communication presents a significant coordinated challenge in the time-varying additive white Gaussian noise (T-AWGN) channel. In this paper, based on the guidance of a tight bound as coding parameters of polar code rate $$ R $$ , block length $$ N $$ with the capacity $$ I(W) $$ in channel $$ W $$ of $$ N \ge \beta /(I(W) - R)^{\mu } $$ , a criteria of effectively adjusting the size of the parameter $$ \mu $$ will achieve a better trade-off between the CR and the reliability, where $$ \beta $$ depends only on block error probability. In the circumstance of a round-clock traffic light (RTL) simulation, numerical results show that this scheme has a good preference for the guaranteed reliability for the wireless communication.

Hongxu Jin, Bofeng Jiang

The Quaternion-Fourier Transform and Applications

It is well-known that the Fourier transforms plays a critical role in image processing and the corresponding applications, such as enhancement, restoration and compression. For filtering of gray scale images, the Fourier transform in $$\mathbb {R}^2$$ is an important tool which converts the image from spatial domain to frequency domain, then by applying filtering mask filtering is done. To filter color images, a new approach is implemented recently which uses hypercomplex numbers (called Quaternions) to represent color images and uses Quaternion-Fourier transform for filtering. The quaternion Fourier transform has been widely employed in the colour image processing. The use of quaternions allow the analysis of color images as vector fields, rather than as color separated components. In this paper we mainly focus on the theoretical part of the Quaternion Fourier transform: the real Paley-Wiener theorems for the Quaternion-Fourier transform on $$\mathbb {R}^2$$ for Quaternion-valued Schwartz functions and $$L^p$$ -functions, which generalizes the recent results of real Paley-Wiener theorems for scalar- and quaternion-valued $$L^2$$ -functions.

Shanshan Li, Jinsong Leng, Minggang Fei

An Optimized Algorithm on Multi-view Transform for Gait Recognition

Gait is one of the common used biometric features for human recognition, however, for some view angles, it is difficult to exact distinctive features, which leads to hindrance for gait recognition. Considering the challenge, this paper proposes an optimized multi-view gait recognition algorithm, which creates a Multi-view Transform Model (VTM) by adopting Singular Value Decomposition (SVD) on Gait Energy Image (GEI). To achieve the goal above, we first get the Gait Energy Image (GEI) from the gait silhouette data. After that, SVD is used to build the VTM, which can convert the gait view-angles to $$ 90^\circ $$ to get more distinctive features. Then, considering the image matrix is so large after SVD in practice, Principal Component Analysis (PCA) is used in our experiments, which helps to reduce redundancy. Finally, we measure the Euclidean distance between gallery GEI and transformed GEI for recognition. The experimental result shows that our proposal can significantly increase the richness of multi-view gait features, especially for angles offset to $$ 90^\circ $$ .

Lingyun Chi, Cheng Dai, Jingren Yan, Xingang Liu

A Novel Real-Time EEG Based Eye State Recognition System

With the development of brain-computer interface (BCI) technology, fast and accurate analysis of Electroencephalography (EEG) signals becomes possible and has attracted a lot of attention. One of the emerging applications is eye state recognition based on EEG signals. A few schemes like the K* algorithm have been proposed which can achieve high accuracy. Unfortunately, they are generally complex and hence too slow to be used in a real-time BCI framework such as an instance-based learner. In this paper, we develop a novel effective and efficient EEG based eye state recognition system. The proposed system consists of four parts: EEG signal preprocessing, feature extraction, feature selection and classification. First, we use the ‘sym8’ wavelet to decompose the original EEG signal and select the 5th floor decomposition, which is subsequently de-noised by the heuristic SURE threshold method. Then, we propose a novel feature extraction method by utilizing the information accumulation algorithm based on wavelet transform. By using the CfsSubsetEval evaluator based on the BestFirst search method for feature selection, we identify the optimal features, i.e., optimal scalp electrode positions with high correlations to eye states. Finally, we adopt Random Forest as the classifier. Experiment results show that the accuracy of the overall EEG eye state recognition system can reach 99.8% and the minimum number of training samples can be kept small.

Zijia Zhou, Pan Li, Jianqi Liu, Weikuo Dong

Tracking Performance of Improved Convex Combination Adaptive Filter Based on Maximum Correntropy Criterion

A convex combination adaptive filter based on maximum correntropy criterion (CMCC) was widely used to solve the contradiction between the step size and the misadjustment in impulsive interference. However, one of the major drawbacks of the CMCC is its poor tracking ability. In order to solve this problem, this paper proposes an improved convex combination based on the maximum correntropy criterion (ICMCC), and investigates its estimation performance for system identification in the presence of non-Gaussian noise. The proposed ICMCC algorithm implements the combination of arbitrary number of maximum correntropy criterion (MCC) based adaptive filters with different adaption steps. Each MCC filter in the ICMCC is capable of tracking a specific change speed, such that the combined filter can track a variety of the change speed of weight vectors. In terms of normalized mean square deviation (NMSD) and tracking speed, the proposed algorithm shows good performance in the system identification for four non-Gaussian noise scenarios.

Wenjing Wu, Zhonghua Liang, Qianwen Luo, Wei Li

Module Selection Algorithm Based on WSS/SSS-Hybrid AoD Node in Dynamic Elastic Optical Networks

Driven by the emerging applications based on Internet, optical backbone networks need to improve their transmission capabilities while ensuring high reliability, flexibility, and scalability. Elastic optical networks and space-division multiplexing optical networks are seen as the potential solutions. In order to implement these technologies, innovative nodes are required to provide flexibility, reliability, and scalability for the optical networks. Architecture on Demand (AoD) node is a new type of elastic optical node structure proposed in the recent years and can dynamically provide a customizable structure according to the exchange and processing requirements of the network traffic. Spectrum Selector Switches (SSS) is one of the key modules but has not been widely used because of its excessive cost. To solve the problem of how to select the Wavelength Selective Switch (WSS)/SSS coexistence in the current network, we propose a pre-built algorithm for the modules in the AoD nodes. Simulation results show that the proposed algorithm performs better than the benchmarks in different network scenarios and provides a solution to the gradual upgrade of AoD nodes.

Ziqin Li, Xiaosong Yu, Shimulin Xie, Yan Wang, Yuhui Wang, Yongli Zhao, Jie Zhang

Joint Optimization of Energy Efficiency and Interference for Green WLANs

In the past years, the issues of energy efficiency and interference are becoming increasingly serious in wireless local area network (WLAN) since lots of access points (AP) are deployed densely to provide high-speed users access. However, current works focus on solving the two issues separately and the influence of each other is rarely considered. To address these problems, we propose a joint optimization scheme of energy efficiency and interference to reduce energy consumption and interference together without sacrificing users’ traffic demands. Firstly, based on energy consumption measurement of AP and network interference analysis, we establish energy efficiency and interference models respectively. Then, the weighting method is introduced to build the joint optimization to quantify the effects of user-AP association, AP switch, AP transmit power and AP channel on energy consumption and interference. Lastly, we formulate the joint optimization as an Mixed Integer Non-Linear Programming (MINLP) problem. Since the MINLP problem is NP-hard, we proposed an Joint Optimization of Energy Efficiency and Interference (JOEI) algorithm based on greedy method to simplify its computational complexity. The evaluation results show that the proposed algorithm can effectively reduce the network energy consumption while improve the capacity of WLANs.

Zhenzhen Han, Chuan Xu, Guofeng Zhao, Rongtong An, Xinheng Wang, Jihua Zhou

User Assisted Dynamic RAN Notification Area Configuration Scheme for 5G Inactive UEs

The new radio resource control inactive state has become the main status of user equipments (UEs) in 5G networks, because of its low power consumption and energy saving features. To deal with the massive signaling overhead in 5G networks, in this paper, we introduce a UE assisted dynamic RAN notification area (RNA) configuration scheme to effectively reducing the paging and the RNA update overhead of inactive UEs. Especially, UEs are divided into two categories, namely, the speed-priority type and the rate-priority type based on their communication rate, mobility, as well as the location. Accordingly, we further extensively investigate the dynamic RNA configuration update process in both the theoretical and the practical manner. The performance of proposed schemes is evaluated via simulations and the results demonstrate the effectiveness and the efficiency in achieving the design goals, which could achieve a considerable performance improvement with respect to schemes in literatures.

Chunyan Cao, Xiaoge Huang, Xiayan Fan, Qianbin Chen

Transmission Capacity Analysis of Distributed Scheduling in LTE-V2V Mode 4 Communication

LTE-V2X sidelink/PC5 communication aimed at supporting device-to-device (D2D) communications in vehicular scenario has been developed as an appropriate technology by 3GPP. Particularly, mode 4 operating without cellular coverage permits vehicles autonomously to select resources and has the potential to achieve an efficient and reliable transmission for vehicle safety applications. However, there is very little research conducted on theoretical understanding of the characteristics and performance of mode 4. In this work, we propose a tractable mathematical analysis to evaluate the performance of LTE-V2V in mode 4. Specifically, we assume that vehicles driving on 1-D abstract lane follow a Poisson Point Process (PPP). By means of probability model, we analyze the event that vehicles randomly select the same resource inducing collision, and investigate the failure probability of transmission. Also, the distance between adjacent vehicles is log-normally distributed and the transmission outage probability under a fixed threshold is given. Furthermore, we derive the expression of transmission capacity. To this end, numerical results verify that the transmission capacity of mode 4 can be improved to a certain extent with the increasing of density of vehicles.

Jie Lv, Xinxin He, Jianfeng Li, Huan Wang, Tao Luo

A Time-slot Based Coordination Mechanism Between WiFi and IEEE 802.15.4

Both WiFi and IEEE 802.15.4 are wide-spread wireless communication technologies utilized particularly in indoor environments such as home, offices and buildings. Since these wireless networks are normally operating in the license-free Industrial Scientific Medical (ISM) frequency band and share the same wireless medium, where no coordination mechanism is available to guarantee communications, unavoidably it leads to interference among them. In order to address this problem, this paper proposes a time-slot based coordination mechanism between WiFi and IEEE 802.15.4, which is achieved by introducing Access Suppression Notification (ASN) frame into IEEE 802.15.4. The static scheduling algorithm is designed and the experiments show that proposed coordination mechanism demonstrates an overall improvement in both IEEE 802.15.4 packet loss ratio and packet transmission rate.

Xiao Wang, Kun Yang

Modeling a Datacenter State Through a Novel Weight Corrected AHP Algorithm

Analytic Hierarchy Process (AHP) is an effective algorithm for determining the weight of each module of a model. It is generally used in the process of multi-indicator decision making. But, when using AHP for evaluation, it is inevitable to introduce the evaluator’s subjectivity. In this paper, an algorithm based on Bayes’ formula is proposed for correcting the weights determined by the analytic hierarchy process. This algorithm can reduce the subjectivity of the evaluator introduced during the evaluation process. At the same time, the common operational indicators of a data center are summarized and classified. I chose some relatively important indicators and established an evaluation model for the operational status of the data center. The weight of the modules of the established model is corrected using this improved algorithm.

Weiliang Tan, Yuqing Lan, Daliang Fang

Research on Semantic Role Labeling Method

Semantic role labeling task is a way of shallow semantic analysis. Its research results are of great significance for promoting Machine Translation [1], Question Answering [2], Human Robot Interaction [3] and other application systems. The goal of semantic role labeling is to recover the predicate-argument structure of a sentence, based on the sentences entered and the predicates specified in the sentence. Then mark the relationship between the predicate and the argument, such as time, place, the agent, the victim, and so on. This paper introduces the main research directions of semantic role labeling and the research status at home and abroad in recent years. And summarized a large number of research results based on statistical machine learning and deep neural networks. The main purpose is to analyze the method of semantic role labeling and its current status. Summarize the development trend of the future semantic role labeling.

Bo Jiang, Yuqing Lan

Network Load Minimization-Based Virtual Network Embedding Algorithm for Software-Defined Networking

In a network virtualization-enabled software-defined networking (SDN), the problem of virtual network embedding (VNE) is a major concern. Although a number of VNE algorithms have been proposed, they fail to consider the efficient utilization of substrate resources or the network load extensively, thus resulting in less efficient utilization of substrate resources or higher blocking ratio of the virtual networks. In this paper, we study the problem of mapping a number of virtual networks in SDN and formulate the VNE problem as a network load minimization problem. Since the formulated optimization problem is NP-hard and it cannot be solved conveniently, we propose a two-stage VNE algorithm consisting of node mapping stage and link mapping stage. Numerical results demonstrate that the effectiveness of our proposed algorithm.

Desheng Xie, Rong Chai, Mengqi Mao, Qianbin Chen, Chun Jin

Joint User Association and Content Placement for D2D-Enabled Heterogeneous Cellular Networks

The explosive increase of the multimedia traffic poses challenges on mobile communication systems. To stress this problem, caching technology can be exploited to reduce backhaul transmissions latency and improve content fetching efficiency. In this paper, we study the user association and content placement problem of device-to-device-enabled (D2D-enabled) heterogeneous cellular networks (HCNs). To stress the importance of the service delay of all the users, we formulate the joint user association and content placement problem as an integer-nonlinear programming problem. As the formulated NP-hardness of the problem, we apply the McCormick envelopes and the Lagrangian partial relaxation method to decompose the optimization problem into three subproblems and solved it by using Hungarian method and unidimensional knapsack algorithm. Simulation results validate the effectiveness of the proposed algorithm.

Yingying Li, Rong Chai, Qianbin Chen, Chun Jin

Hybrid Caching Transmission Scheme for Delay-sensitive Service in Vehicular Networks

With the inspiring development of vehicular networks, caching popular contents in the network edge nodes could greatly enhance the quality of user experience. However, the highly dynamic movements of vehicles make it difficult to maintain stable wireless transmission links between vehicle-to-vehicle pairs or vehicle-to-road side units (RSUs) pairs, and then resulting in unbearable transmission delays or even transmission interruptions. In this paper, we proposed a predictive and hybrid caching transmission scheme for delay-sensitive services in vehicular networks. In order to select the most proper node for transmitting desired content from the nearby RSUs or vehicles, we evaluate the candidate nodes from the prediction on effective communication range and connection time based on the relative velocities and SINR threshold. Then the end-to-end delay for respective nodes is compared which includes two parts: waiting period and transmitting period. Waiting period is predicted based on the relative distance and relative velocities between two nodes at the starting position. Transmitting period is calculated from the transmission rate and effective communication range. The candidate node with the lowest delay is selected to transmit the desired content to the destination vehicle. Simulation results show that the proposed scheme could significantly reduce time delays in data transmission, especially when the requesting vehicle is far from the nearest RSU.

Rui Shi, Xi Li, Hong Ji, Heli Zhang

Predictive Time Division Transmission Algorithm for Segmented Caching in Vehicular Networks

With the increasing number of different types of applications for road safety and entertainment, it demands more flexible solutions for caching and transmitting large files in vehicular networks. In order to decrease the transmission delay and raise the hit ratio of cached files, there is already a lot of research on caching technology, including segmented caching technology. But the problem of long transmission delay and low successful transmission ratio caused by the high dynamic of vehicles still needs to be solved. In this paper, we proposed an algorithm named Predictive Time Division Transmission (PTDT) to reduce transmission delay and raise the ratio of successful transmission for segmented cached file in vehicular networks. Our algorithm predicts the link duration between requesting vehicle and neighboring vehicles according to the relative inter-vehicle distances and velocities. By predicting the transmit rate of each vehicle on different time point, we divide the link duration into slices for subsequent transmitter selections. And finally we compare those time points and select the vehicles that make the transmitting delay the lowest. In the mean time, we arrange the transmitting order of those vehicles to guarantee the success of full file transmission process. The simulation results show that after applying our algorithm, transmission delay has reduced and successful transmission rate has increased substantially.

Rui Shi, Xi Li, Hong Ji, Heli Zhang

Secrecy Sum Rate Optimization in MIMO NOMA OSTBC Systems with Imperfect Eavesdropper CSI

In this research, we investigate the secrecy sum rate optimization problem for a multiple-input multiple-output (MIMO) non-orthogonal multiple access (NOMA) system with orthogonal space-time block codes (OSTBC). We construct a model where the transmitter and the relay send information by employing OSTBC, while both the source and the relay have imperfect channel state information (CSI) of the eavesdropper. The precoders and the power allocation scheme are jointly designed to maximize the achievable secrecy sum rate subject to the power constraints and the minimum transmission rate requirements of the weak user. To solve this non-convex problem, we propose the constrained concave convex procedure (CCCP)-based iterative algorithm and the alternative optimization (AO) method, where the closed-form expression for power allocation is derived. The simulation results demonstrate the superiority of our proposed scheme.

Jianfei Yan, Zhishan Deng, Qinbo Chen

Network-coding-based Cooperative V2V Communication in Vehicular Cloud Networks

We investigate the potential of applying cooperative relaying and network coding techniques to support vehicle-to-vehicle (V2V) communication in vehicular cloud networks (VCN). A reuse-mode MIMO content distribution system with multiple sources, multiple relays, and multiple destinations under Nakagami-m fading is considered. We apply a class of finite field network codes in the relays to achieve high spatial diversity in an efficient manner and derive the system communication error probability that the destinations fail to recover the desired source messages. The results show that our method can improve the performance over conventional data transmission solutions.

Rui Chen, Weijun Xing, Chao Wang, Ping Wang, Fuqiang Liu, Yusheng Ji

Cluster-Based Dynamic FBSs On/Off Scheme in Heterogeneous Cellular Networks

Recent years, with the explosive growth of mobile data traffic, cellular communication system is faced with enormous challenges. The ultra-dense deployment of small cells will increase the network capacity while increasing the energy consumption. In this paper, we study a cluster-based dynamic FBSs on/off scheme in heterogeneous cellular networks, where the overall objective is to maximize the network energy efficiency by optimizing jointly the cell association, the base station on/off strategies and the cluster division, taking into account the load balancing and the QoS requirement of heterogenous cellular networks. The optimization problem is divided into three processes: the base station and the user equipment (UE) association scheme, the femtocell base station (FBS) clustering, and the FBS on/off scheme according to the current traffic load. A cluster-based dynamic FBSs on/off scheme is proposed to improve EE in HCNs while ensuring the load balancing, the probability of outage, and the communication requirement of UEs in the core area. Simulation result shows that the proposed algorithm could achieve significant improvement of the network energy efficiency in all aspects than comparison algorithms in literature.

Xiaoge Huang, She Tang, Dongyu Zhang, Qianbin Chen

Application Identification for Virtual Reality Video with Feature Analysis and Machine Learning Technique

Immersive media services such as Virtual Reality (VR) video have attracted more and more attention in recent years. They are applications that typically require large bandwidth, low latency, and low packet loss ratio. With limited network resources in wireless network, video application identification is crucial for optimized network resource allocation, Quality of Service (QoS) assurance, and security management. In this paper, we propose a set of statistical features that can be used to distinguish VR video from ordinary video. Six supervised machine learning (ML) algorithms are explored to verify the identification performance for VR video application using these features. Experimental results indicate that the proposed features combined with C4.5 Decision Tree algorithm can achieve an accuracy of 98.6% for VR video application identification. In addition, considering the requirement of real-time traffic identification, we further make two improvements to the statistical features and training set. One is the feature selection algorithm to improve the computational performance, and the other is the study of the overall accuracy in respect to training set size to obtain the minimum training set size.

Xiaoyu Liu, Xinyu Chen, Yumei Wang, Yu Liu

28-GHz RoF Link Employing Optical Remote Heterodyne Techniques with Kramers–Kronig Receiver

We propose and demonstrate a 28-GHz optical remote heterodyne RoF link using KK receiver for the first time. An optical SSB modulated signal is obtained utilizing an IQ modulator and two free-running lasers. Due to its minimum phase property, KK algorithm can be adopted to reconstruct the complex 16-QAM signal from the received intensity signal. This scheme is effective in eliminating the SSBI penalty introduced by square-law detection. Through the use of the KK receiver, the power penalty caused by the 80 km SSMF transmission is found to be less than 1 dB with digital CDC post processing. The KK-based receiver can also provide about 2 dB advantage over the traditional receiver at the 7% HD-FEC threshold in the case of 28 GBaud rate transmission over 80 km fiber. Furthermore, as the baud rate increases, the benefit of KK receiving scheme is more obvious and superior than that of the traditional receiving scheme.

Yuancheng Cai, Xiang Gao, Yun Ling, Bo Xu, Kun Qiu

Fairness-Based Distributed Resource Allocation in Cognitive Small Cell Networks

In this paper, we aim to maximize the total throughput of the cognitive small cell networks by jointly considering interference management, fairness-based resource allocation, average outage probability and channel reuse radius. In order to make the optimization problem tractable, we decompose the original problem into three sub-problems. Firstly, we derive the average outage probability function of the system with respect to the channel reuse radius. With a given outage probability threshold, the associated range of the channel reuse radius is obtained. In addition, a fairness-based distributed resource allocation (FDRA) algorithm is proposed to guarantee the fairness among cognitive small cell base stations (CSBSs). Finally, based on the channel reuse range we could find the maximum throughput of the small cell network tire. Simulation results demonstrate that the proposed FDRA algorithm could achieve a considerable performance improvement relative to the schemes in literature, while providing a better fairness among CSBSs.

Xiaoge Huang, Dongyu Zhang, She Tang, Qianbin Chen

A Distributed Self-healing Mechanism Based on Cognitive Radio and AP Cooperation in UDN

Self-healing is considered as an indispensable function to achieve intelligent network management in future wireless communication systems. However, in ultra-dense networks (UDNs), it’s a great challenge to realize efficient self-healing due to the massive and diverse network nodes, as well as complex transmission environment. The failed network access point (AP) may result in sudden traffic outage and severe user service degrading. In this paper, we propose an effective self-healing mechanism for UDNs with complete procedure of intelligent failure detection, diagnosis and recovery. Cognitive technology has been introduced to realize the effective detection of the AP working status. Then the processed information are analyzed based on multi-armed bandit model for possible AP failure judgement. After it is confirmed that an AP is failed, the impacted users, which are served originally by the failed AP, would be accessed to the proper neighbor APs. Furthermore, the corresponding resource allocation based on Non-Orthogonal Multiple Access (NOMA) is proposed. Simulation results show that the proposed mechanism could detect the AP failure effectively and realize quick self-healing for the network.

Zhongming Gao, Xi Li, Hong Ji, Heli Zhang

A High-Speed Large-Capacity Packet Buffer Scheme for High-Bandwidth Switches and Routers

Today’s switches and routers require high-speed and large-capacity packet buffers to guarantee a line rate up to 100 Gbps as well as more fine-grained quality of service. For this, this paper proposes an efficient parallel hybrid SRAM/DRAM architecture for high-bandwidth switches and routers. Tail SRAM and head SRAM are used for guaranteeing the middle DRAMs are accessed in a larger granularity to improve the bandwidth utilization. Then, a simple yet efficient memory management algorithm is designed. The memory space is dynamically allocated when a flow arrives, and a hard timeout is assigned for each queue. Hence, the SRAM space is utilized more efficiently. A queueing system is used to model the proposed method, and theoretical analysis is performed to optimize the timeout value. Simulation shows that the proposed architecture can reduce packet loss rate significantly compared with previous solutions with the same SRAM capacity.

Ling Zheng, Zhiliang Qiu, Weitao Pan, Ya Gao

Non-stationary Characteristics for Indoor Massive MIMO Channels

Massive Multiple Input Multiple Output (MIMO) has been widely considered as one of the most promising technologies for the fifth-generation (5G) wireless communication. In massive MIMO system, the research on channel characteristics is important. In this paper, the characteristics for massive MIMO channels at both 2 GHz and 6 GHz are investigated. Based on the real-world measurements, the channel parameters in the delay and frequency domains are extracted to show the non-stationary phenomenon over the large-scale antenna array. Furthermore, the characteristics of the angular parameters extracted by space-alternating generalized expectation-maximization (SAGE) algorithm are investigated and the fluctuations are modeled. The results for different frequencies are useful for deep understanding of massive MIMO channels in the future.

Qi Wang, Jiadong Du, Yuanyuan Cui

DSP Implementation and Optimization of Pseudo Analog Video Transmission Algorithm

With the development of wireless video technology and embedded technology, a dedicated digital signal processor (DSP) can achieve the video transmission stably and flexibly. Some existing wireless video transmission algorithms do not perform well in response to complex channel environments. A pseudo-analog video algorithm that can be run in a dedicated instruction set was proposed. At the transmitter, the image data which are removed spatially redundant are divided into L-shaped blocks for power allocation, and the digital signal are sent to CRC and Turbo coding. Finally, the modulated digital signal and the pseudo-analog data after power allocation are sent to framing. The receiver includes channel estimation and de-framing, recovers digital signal and pseudo-analog signal through error detection and decoding. We have optimized the algorithm at the assembly level, so that the entire system is more flexible. The entire transfer system will run on the FPGA and hardware DSP boards for debugging.

Chengcheng Wang, Pengfei Xia, Haoqi Ren, Jun Wu, Zhifeng Zhang

Spectrum Modulation of Smart-Surfaces for Ultra High Frequency Radars

Smart surfaces are reconfigurable meta-materials whose electromagnetic characteristics can be altered for applications such as remote identification, stealth, etc. This paper introduces the spectrum modulation of smart-surfaces for long range radars. Applying controlling signals onto the tunable lumped elements loaded on smart-surfaces, modulations can be achieved on reflecting signals illuminating on smart surfaces. Changing the spectral characteristics of the modulated signals, radar receivers can only detect the limited information of the target. This paper introduces the operation mechanism of smart surfaces and analyzes two specific modulating signals, the square wave signal and the pseudo-random Gaussian white noise signal. The spectrum of reflecting signals will change accordingly, making it difficult for the radar receiver to detect. Simulations and results show that the proposed method can change the reflecting echo of the radar and reduce the probability of the target being detected.

Kai Liu, Yang Wang, Qilong Song, Xi Liao

Two Stage Detection for Uplink Massive MIMO MU-SCMA Systems

In this paper, we propose a two stage multiuser detection scheme: a linear pre-filtering and iteration removal based message passing algorithm (RM-MPA). As the first stage of the proposed detection, a linear pre-filtering based on Richardson method is proposed to avoid the complicated matrix inversion in an iterative way. Meanwhile, we also present a sub-optimum relaxation parameter to Richardson for lower-complexity. Then the RM-MPA is used for multiuser decoding, which compared the decoding advantages of users and sorted users according to decoding advantages. After the each iteration, the users with higher decoding advantages directly are decoded and removed. The removed users do not participate in the subsequent iterations, therefore, the complexity of subsequent iterations decrease gradually. Simulation results show that the proposed two stages multiuser detection can significantly reduce the computational complexity with better symbol error rate performance.

Cuitao Zhu, Ning Wei, Zhongjie Li, Hanxin Wang

Robust Spectrum Sensing for Cognitive Radio with Impulsive Noise

Spectrum sensing plays an important role in cognitive radio. In this paper, a robust spectrum sensing method via empirical characteristic function based on goodness-of-fit testing is proposed, named as ECF detector. The test statistic is derived from the empirical characteristic function of the observed samples, thus the secondary users do not require any prior knowledge of the primary signal and the noise distribution. Extensive simulations are performed and compared with the existing spectrum sensing methods, such as energy detector, eigenvalue-based detector, AD detector and KS detector. The results show that, the proposed ECF detector can offer superior detection performance under both the Gaussian noise and the impulsive noise environments.

Liping Luo

Resource Allocation for Mobile Data Offloading Through Third-Party Cognitive Small Cells

Mobile data offloading is considered as an effective way to solve the network overloading issue. In this paper, we study the mobile data offloading problem through a third-party cognitive small cell providing data offloading service to a macrocell. Particularly, four scenarios, namely, successive interference cancellation (SIC) available at neither the macrocell base station (MBS) nor the small cell BS (SBS), SIC available at both the MBS and the SBS, SIC available at only the MBS, and SIC available at only the SBS are considered. For all the four scenarios, iterative optimization based data offloading schemes are proposed. We show that the proposed data offloading schemes outperform the corresponding schemes without data offloading. We also show that equipping SIC at the SBS is more beneficial compared to equipping SIC at the MBS.

Qun Li, Zheng Yin, Ding Xu

Performance Analysis of Non-coherent Massive SIMO Systems with Antenna Correlation

Recently, energy detection (ED) has been investigated in massive single-input multiple-output (SIMO) systems, where transmit symbols can be decoded by averaging the received power across all receive antennas. In this paper, we concentrate on the performance of non-coherent massive SIMO in the presence of antenna correlation. Specifically, closed-form expressions of symbol error rate (SER) and achievable rate are derived. Furthermore, asymptotic behaviors of SER and achievable rate in regimes of a large number of receive antennas, high antenna correlation and large signal-to-noise ratio (SNR) are investigated. Interestingly, the results show that antenna correlation poses a great impact to SER, but has little effect on the achievable rate. Numerical results are presented to verify our analytical results.

Weiyang Xu, Huiqiang Xie, Shengbo Xu

Coalition Formation Game Based Energy Efficiency Oriented Cooperative Caching Scheme in UUDN

It is generally considered that Ultra-Dense Network (UDN) is a promising solution for 5G and the network is going to turn into user centric. Caching popular contents at the edge of network is an efficient way to reduce the energy consumption and data traffic of backhaul link. But most of current researches on caching in UDN fail to take into account of user centric and energy efficiency performance during caching files delivery process. In this paper, we consider an User-centric Ultra-Dense Network (UUDN) with cache-enabled Small Base Stations (SBSs) and investigate the energy efficiency of cooperative caching in UUDN. In order to achieve energy efficiency during delivery, we design a novel SBS grouping rule and a cooperative caching scheme based fragmentation with the consideration of user mobility. We formulate an energy optimization problem on caching and introduce coalition formation game to simplify and solve our optimization objective. Then we analyze the impacts of system parameters on the overall performance and compare our scheme to some other schemes. Numerical results demonstrate our scheme is energy efficient and outperforms the others.

Yu Li, Heli Zhang, Hong Ji, Xi Li

A Joint Frequency Offset Estimation Method Based on CP and CRS

In order to solve the problem that Fraction Frequency Offset (FFO) estimation algorithm has the problem of low estimation precision, small range and high occupancy rate of spectrum resource, this paper proposed a FFO estimation method based on the combination of Cyclic Prefix (CP) and Cell-specific reference signals (CRS). First, judging the range of the true frequency offset value according to the results of the frequency offset estimation algorithm based on CP. Then the possible true frequency offset value obtained by adding value calculated by frequency offset estimation algorithm based on CRS and the possible rotation value of 2000 nHz. Finally, comparison the results of the frequency offset estimation algorithm based on CP and the possible true frequency offset, the minimum deviation is its true. The accuracy is the same as that frequency offset estimation algorithm that based on CRS. The range is the same as frequency offset estimation algorithm based on CP, which is $$\left[ -7500\,\mathrm{Hz}, 7500\,\mathrm{Hz}\right] $$ . The principle of the algorithm is simple and does not occupy additional bandwidth resources.

Xiaoling Hu, Zhizhong Zhang, Yajing Zhang

Mobility-Aware Caching Specific to Video Services in Hyper-Dense Heterogeneous Networks

Caching at the network edge has emerged as a promising technique to cope with the dramatic increase of mobile data traffic. It is noted that different types of video applications on mobile devices have different requirements for cached contents, thus corresponding caching policies should be developed accordingly. In hyper-dense heterogeneous networks, due to the user mobility and limited connection duration, the user often could not download the complete cached contents from an associated SBS before it moves away, which makes the design of caching strategy more challenging. In this paper, we propose two different caching strategies to adapt to multimedia applications of different video contents. For ordinary network video files, coded caching is used to increase the efficiency of content access. The caching problem is formulated as an optimization problem to minimize the average transmission cost of cached contents. We first present an optimal caching strategy based on the critical value of validity period of user requests. Then, for the validity period greater than its critical value, an iterative optimization on the basis of the above optimal solution is performed. For typical streaming video, uncoded video fragments is considered to be stored in the caches to meet the needs of online viewing. The principle of the proposed caching scheme is to cache data chunks in advance according to the sequences of SBSs passed by the user based on the mobility prediction results. Simulation results indicate that the proposed mobility-based caching performs better than the existing popularity-based caching scheme.

Zhenya Liu, Xi Li, Hong Ji, Heli Zhang

Long-Reach PON Based on SSB Modulated Frequency-Shifted QAM and Low-Cost Direct-Detection Receiver with Kramers–Kronig Scheme

As PON systems move towards terabit/s aggregated data rates with longer transmission distance, optical coherent receivers become preferred due to their high tolerance to power fading from fiber transmission. To solve the high complexity and high cost problems of optical coherent receivers, a scheme for complex QAM signal transmission with simple direct detection is recommended in this paper. The scheme based on optical SSB modulation with frequency-shifted QAM signals and low-cost single-ended PD provides an efficient low-cost solution for long reach coherent PON. Due to its minimum phase property of the optical SSB modulated signal, Kramers-Kronig scheme can be used to reconstruct the complex QAM signal from the received intensity signal. The efficiency of the proposed scheme is validated by both numerical simulations and experiments for both QPSK and 16-QAM modulated signals. By using standard commercially available components, the experiments demonstrated that the combination of SSB modulation of frequency-shifted QAM signal and its single-ended PD receiver with KK scheme can support SSMF transmission over 75 km for both QPSK and 16-QAM signals with receiver optical power penalty less than 1.5 dB.

Xiang Gao, Bo Xu, Yuancheng Cai, Mingyue Zhu, Jing Zhang, Kun Qiu

Two-Layer FoV Prediction Model for Viewport Dependent Streaming of 360-Degree Videos

As the representative and most widely used content form of Virtual Reality (VR) application, omnidirectional videos provide immersive experience for users with 360-degree scenes rendered. Since only part of the omnidirectional video can be viewed at a time due to human’s eye characteristics, field of view (FoV) based transmission has been proposed by ensuring high quality in the FoV while reducing the quality out of that to lower the amount of transmission data. In this case, transient content quality reduction will occur when the user’s FoV changes, which can be improved by predicting the FoV beforehand. In this paper, we propose a two-layer model for FoV prediction. The first layer detects the heat maps of content in offline process, while the second layer predicts the FoV of a specific user online during his/her viewing period. We utilize a LSTM model to calculate the viewing probability of each region given the results from the first layer, the user’s previous orientations and the navigation speed. In addition, we set up a correction model to check and correct the unreasonable results. The performance evaluation shows that our model obtains higher accuracy and less undulation compared with widely used approaches.

Yunqiao Li, Yiling Xu, Shaowei Xie, Liangji Ma, Jun Sun

Energy Efficient Caching and Sharing Policy in Multihop Device-to-Device Networks

Caching content at the user device and sharing files via multihop Device-to-Device link can offload the traffic from the Base Station, which is inevitable to consume the user’s energy. But most works usually assume that the battery capacity is implicitly infinite and rarely consider the impact of the user’s remaining battery energy on the file transmission. In fact, the user device has limited battery capacity and the transmission may be not completed due to the insufficient battery energy. So it is important to utilize the limited battery energy to ensure more successful transmission and traffic offloading. In this paper, we firstly optimize the caching policy and obtain the minimum energy cost of cache-enabled multihop D2D communications. For this purpose, we classify users into different clusters and use a weighted undirected graph to represent the topological relationship of users in one cluster. Then, we propose a novel algorithm to find the optimal path to transmit files via multihop D2D link. Finally, we obtain the minimum energy cost and optimal caching policy. Simulation results show that the proposed caching policy performs better than other general caching strategies in terms of energy conservation.

Yuling Zuo, Heli Zhang, Hong Ji, Xi Li

A Task Scheduling Algorithm Based on Q-Learning for WSNs

In industrial Wireless Sensor Networks (WSNs), the transmission of packets usually have strict deadline limitation and the problem of task scheduling has always been an important issue. The problem of task scheduling in WSNs has been proved to be an NP-hard problem, which is usually scheduled using a heuristic algorithm. In this paper, we propose a task scheduling algorithm based on Q-Learning for WSNs called Q-Learning Scheduling on Time Division Multiple Access (QS-TDMA). The algorithm considers the packet priority in combination with the total number of hops and the initial deadline. Moreover, according to the change of the transmission state of packets, QS-TDMA designs the packet transmission constraint and considers the real-time change of packets in WSNs to improve the performance of the scheduling algorithm. Simulation results demonstrate that QS-TDMA is an approximate optimal task scheduling algorithm and can improve the reliability and real-time performance of WSNs.

Benhong Zhang, Wensheng Wu, Xiang Bi, Yiming Wang

Performance Analysis of Task Offloading in Double-Edge Satellite-Terrestrial Networks

With the rapid development of wireless networks, the growing number of mobile applications results in massive computation task to be processed. Multi-access edge computing (MEC) can efficiently minimize computational latency, reduce response time, and improve quality of service (QoS) by offloading tasks in the access network. Although lots of MEC task offloading schemes have been proposed in terrestrial networks, the integrated satellite-terrestrial communication, as an emerging trend for the next generation communication, has not taken MEC offloading into consideration. In this paper, we proposed a cooperative offloading scheme in a double-edge satellite-terrestrial (DESTN) network. Performance of offloading efficiency and energy consumption are derived analytically. Simulations show that the proposed offloading scheme in the double-edge satellite-terrestrial outperforms the traditional terrestrial-only offloading scheme by approximately 18.7%. Our research provides an insight for following studies in task offloading of double-edge satellite-terrestrial networks.

Peng Wang, Xing Zhang, Jiaxin Zhang, Zhi Wang

Performance Analysis of Relay-Aided D2D Communications with Traffic Model

In this paper, we consider a communication scenario where relay users assist nearby a pair of D2D users underlaying cellular network. In our communication scenario, we analyze not only fading channel model but also different traffic models. In order to jointly consider the impact of interference level and network traffic condition, the packet loss probability (PLP) of D2D link is carefully orchestrated from two perspectives, i.e., link outage probability and packet delivery failure probability. The closed-form expressions of them are respectively obtained based on a Rician–Rayleigh fading model and different traffic models, and then the performance of our relay-aided D2D communication scenario is evaluated by the PLP of D2D link. Finally, the PLP of D2D link with three representative traffic models including Pareto, FBM, and Poisson traffic models are compared, respectively. We believe that the proposed analytical approach can provide a useful insight into the application of traffic model in relay-aided D2D communications.

Jun Huang, Yong Liao, Yide Zhou

Phase Noise Estimation and Compensation Algorithms for 5G Systems

In the 5G system, orthogonal frequency division multiplexing (OFDM) waveform survived for its superior performance. As is well known, OFDM systems are sensitive to the phase noise introduced by local oscillators and it may get worse for likely higher carrier frequency in 5G systems. There are two aspects of the impact of phase noise, namely the common phase error (CPE) and the inter-carrier-interference (ICI). In this paper, first, we propose a more accurate way to estimate CPE. Then, we focus on ICI cancellation. To simplify the ICI model, we only consider the interference from adjacent sub-carriers. Based on the simplified model, we propose two schemes to estimate ICI. The performance of phase noise compensation algorithm we proposed is presented. Simulation results show that the algorithm we proposed can significantly reduce the impact of phase noise and improve the throughput of 5G systems.

Shuangshuang Gu, Hang Long, Qian Li

A Machine Learning Based Temporary Base Station (BS) Placement Scheme in Booming Customers Circumstance

Explosive increase of terminal users and the amount of data traffic give a great challenge for Internet service providers (ISPs). At the same time, this big data also brings an opportunity for ISPs. How to solve network planning problem in emergency or clogging situation, based on big data? In this paper, we try to realize effective and flexible temporary base station (BS) placement through machine learning in a booming customers situation, with ISPs’ massive data. A machine learning based temporary BS placement scheme is presented. A K-means based model training algorithm is put forward, as a vital part of machine learning based temporary BS placement scheme. K-means algorithm is selected as a representative example of machine learning algorithm. The performances of BS position with random starting point, BS position iteration, average path length with different parameters, are conducted to prove the availability of our work.

Qinglong Dai, Li Zhu, Peng Wang, Guodong Li, Jianjun Chen

An Improved Preamble Detection Method for LTE-A PRACH Based on Doppler Frequency Offset Correction

In the random access process of the Long Term Evolution Advanced (LTE-A) system, the Doppler shift influences the detection of the Physical Random Access Channel (PRACH) signal, resulting in the appearance of the pseudo correlation peaks at the receiving end. In the 3GPP protocol, the frequency offset in the mid-speed and low-speed modes is not processed, and the frequency offset processing algorithm in the high-speed mode only applies to the case where the Doppler frequency offset does not exceed the unit sub-carrier. For solve the problem, a three-step improvement method is proposed. The first step is to perform the maximum likelihood (ML) offset estimation to do the frequency offset correction; the second step is to perform the sliding average filter processing to eliminate the influence of multipath; the third step is to use multiple sliding window peak detection algorithm. Compared with the traditional algorithm, the performance of the proposed method is better. And the false alarm performance under the AWGN channel is at least 3.8 dB better, and the false alarm performance under ETU channel is at least 1 dB better.

Yajing Zhang, Zhizhong Zhang, Xiaoling Hu

Cryptographic Algorithm Invocation in IPsec: Guaranteeing the Communication Security in the Southbound Interface of SDN Networks

Due to the static configuration of IPsec cryptographic algorithms, the invocation of these algorithms cannot be dynamically self-adaptable to the traffic fluctuation of software-defined networking (SDN) southbound communication. In this paper, an invocation mechanism, based on the Free-to-Add (FTA) scheme, is proposed to optimize the invocation mode of cryptographic algorithms in traditional IPsec. To balance the link security and communication performance, a feedback-based scheduling approach is designed for the controller of IPsec-applied SDN to replace flexibly and switch synchronously the IPsec cryptographic algorithms in use according to the real-time network status. The feedback information is applied to decide which appropriate algorithm(s) should be employed for the cryptographic process in a special application scenario. The validity and effectiveness of the proposed invocation mechanism are verified and evaluated on a small-scale SDN/OpenFlow platform with the deployed IPsec security gateway. The results show that the FTA-based mechanism invokes IPsec encryption algorithms consistently with the requirement for communication security in the SDN southbound interface, and the impact of the IPsec cryptographic process on the network performance will be reduced even if the network traffic fluctuates markedly.

Deqiang Wang, Wan Tang, Ximin Yang, Wei Feng

HetWN Selection Scheme Based on Bipartite Graph Multiple Matching

Next generation communication networks will be a heterogeneous wireless networks (HetWN) based on 5G. Studying the reasonable allocation of new traffics under the new scenario of 5G is helpful to make full use of the network resources. In this paper, we propose a HetWN selection algorithm based on bipartite graph multiple matching. Firstly, we use the AHP-GRA method to calculate the user’s preference for network and the network’s preference for user. After these two preferences are traded off as the weights of edges in bipartite graph, we can extend the bipartite graph to a bipartite graph network. The minimum cost maximum flow algorithm is used to obtain the optimal matching result. Simulations show that our scheme can balance the traffic dynamically. And it is a tradeoff between user side decision and network side decision.

Xiaoqian Wang, Xin Su, Bei Liu

Hybrid Deep Neural Network - Hidden Markov Model Based Network Traffic Classification

Traffic classification has been well studied in the past two decades, due to its importance for network management and security defense. However, most of existing work in this area only focuses on protocol identification of network traffic instead of content classification. In this paper, we present a new scheme to distinguish the content type for network traffic. The proposed scheme is based on two simple network-layer features that include relative packet arrival time and packet size. We utilize a new model that combines deep neural network and hidden Markov model to describe the network traffic behavior generated by a given content type. For a given model, deep neural network calculates the posterior probabilities of each hidden state based on given traffic feature sequence; while the hidden Markov model profiles the time-varying dynamic process of the traffic features. We derive the parameter learning algorithm for the proposed model and conduct experiments by using real-world network traffic. Our results show that the proposed approach is able to improve the accuracy of conventional GMM-HMM from 77.66% to 96.11%.

Xincheng Tan, Yi Xie

A New Clustering Protocol for Vehicular Networks in ITS

The dissemination of information is the main application of vehicular networks. The cost of network data transmission is significantly increased and the overall performance of the network routing protocol is deteriorated obviously in the vehicle networks without cluster. In order to ensure the reliability and timeliness of information transmission in VANET, it is necessary to cluster the network in most cases. The existed clustering protocols cannot solve the problem of the stability of the cluster due to the high dynamic mobility in VANET. In this work, we improve a clustering protocol based on the optimization of number of cluster heads, the distance from cluster head to cluster members, and the relative velocity of vehicles within the cluster. The optimized cost function contains three weighting factors for adjusting the specific gravity. The weighting factors can be decided with different demand. Simulation results show that our improved clustering protocol has very good performance for the stability of the cluster. It can provide a good support for the future network routing protocol.

Mengmeng Liu, Xuming Zeng, Mengyao Tao, Zhuo Cheng

Accelerated Matrix Inversion Approximation-Based Graph Signal Reconstruction

Graph signal processing (GSP) is an emerging field which studies signals lived on graphs, like collected signals in a sensor network. One important research point in this area is graph signal reconstruction, i.e., recovering the original graph signal from its partial collections. Matrix inverse approximation (MIA)-based reconstruction has been proven more robust to large noise than the conventional least square recovery. However, this strategy requires the K-th eigenvalue of Laplacian operator $$\varvec{\mathcal {L}}$$ . In this paper, we propose an efficient strategy for approximating the K-th eigenvalue in this GSP filed. After that, the MIA reconstruction method is modified by this proposed substitution, and thereby accelerated. Consequently, we apply this modified strategy into artificial graph signal recovery and real-world semi-supervised learning field. Experimental results demonstrate that the proposed strategy outperforms some existed graph reconstruction methods and is comparable to the MIA reconstruction with lower numerical complexity.

Qian Dang, Yongchao Wang, Fen Wang

A Method of Interference Co-processing in Software-Defined Mobile Radio Networks

The intensive network technology that is one of the key technologies of 5G is the main means to solve the explosive growth of data traffic demands in the future. However, with the implementation of network-inte-nsive deployments, serious interference problems are generated. The software-defined network (SDN) allows the separation of the control plane from forward plane, which provides the flexibility of dynamic network programming. Combining SDN with 5G is an effective method to copy with the interference management. This paper proposes an SDN-based mobile wireless network architecture to solve the problem of interference coordination in mobile wireless networks. Through the advantages of the SDN controller, the underlying wireless network topology information is centralized to the control layer that is running the optimization algorithm of resource allocation, which solves the problem of interference coordination. We introduces an Integer Programming to sovle the problem of formulation. A Tabu heuristic algorithm is used to solve the problem of interference coordination. The experimental results show that compared with the algorithm of non interference coordination, the proposed algorithm evidently reduces the interference value of the whole network.

RenGui Gao, Dong Zhang

Detecting Network Events by Analyzing Dynamic Behavior of Distributed Network

Detecting network events has become a prevalent task in various network scenarios, which is essential for network management. Although a number of studies have been conducted to solve this problem, few of them concern about the universality issue. This paper proposes a General Network Behavior Analysis Approach (GNB2A) to address this issue. First, a modeling approach is proposed based on hidden Markov random field. Markovianity is introduced to model the spatio-temporal context of distributed network and stochastic interaction among interconnected and time-continuous events. Second, an expectation maximum algorithm is derived to estimate parameters of the model, and a maximum a posteriori criterion is utilized to detect network events. Finally, GNB2A is applied to three network scenarios. Experiments demonstrate the generality and practicability of GNB2A.

Haishou Ma, Yi Xie, Zhen Wang

Misbehavior Constraint MAC Protocol (MC-MAC) for Wireless Networks

The IEEE 802.11 protocol assumes that all wireless network nodes will abide by the protocol and cooperate well with it. However, in order to obtaining more channel resources or destroying network performance, some selfish nodes will be in misbehaviors when they are in the certain condition wireless contention-sharing channels, such as, Backoff Value Manipulation is a kind of misbehavior. And for this kind of misbehavior, this paper proposes a Misbehavior Constraint MAC protocol (MC-MAC), which can detect and penalize the backoff value manipulation, and it includes a new backoff value generating function with penalty function and a reputation model. Simulation experiments shows that the MC-MAC protocol has a significant inhibitory effect on misbehavior and can improve system throughput.

Yupeng Ma, Yonggang Li, Zhizhong Zhang, Haixing Li

Blind Channel Estimation of Doubly Selective Fading Channels

Blind channel identification methods based on second-order statistics (SOS), have attracted much attention in the literature. However, these estimators suffer from the phase ambiguity problem, until additional diversity can be exploited. In this paper, with the aid of the cyclic prefix (CP) induced periodicity, a channel identification algorithm based on the time varying autocorrelation function (TVAF) is proposed for doubly selective fading channels in Orthogonal Frequency Division Multiplexing (OFDM) systems. The closed-form expression for time-varying channel identification is derived within the restricted support set of time index. Particularly, the CP-induced TVAF components and their corresponding channel-spread correlation elements implicitly carry rich channel information and are not perturbed by additive noise. These advantageous peaks can be employed to address the phase uncertainty problem, offering an alternative way of increasing the rank of signal matrix to achieve complementary diversity. Simulation results demonstrate the proposed method can provide distinctly higher accurate of channel estimation over the classical scheme.

Jinfeng Tian, Ting Zhou, Tianheng Xu, Honglin Hu, Mingqi Li

User Scheduling for Large-Scale MIMO Downlink System Over Correlated Rician Fading Channels

In this paper, we investigate the downlink transmission, especially the user scheduling algorithm for single-cell multiple-input multiple-output (MIMO) system under correlated Rician fading channels. Under the assumption of only statistical channel state information (CSI) at the base station (BS), the statistical beamforming transmission is derived by maximizing the lower bound of the average signal-to-leakage-plus-noise ratio ( $$\text {SLNR}$$ ). Based on this beamforming transmission algorithm, three user scheduling algorithms are proposed exploiting only statistical CSI: (1) maximum SLNR: schedule the user with the maximum SLNR; (2) most dissimilar: schedule the user that is most dissimilar to the already selected users; (3) modified-treating interference as noise (TIN): treat the inter-user interference as uncorrelated noise to each user’s useful signal and schedule the user with the largest signal-to-noise factor. Simulation results show that the proposed user scheduling algorithms perform well in achieving considerable sum rate.

Tingting Sun, Xiao Li, Xiqi Gao

A Blind Detection Algorithm for Modulation Order in NOMA Systems

The blind detection algorithm for modulation order (MOD) of interference user in power-domain non-orthogonal multiple access (NO-MA) is studied by academics. Maximum likelihood method is the optimal approach, but with huge computational complexity. A sub-optimal approach based on max-log approximation is deduced which can reduce computational complexity, but with performance degradation. This paper investigates an improved blind detection algorithm for modulation order based on max-log likelihood approach in NOMA systems. Unlike the other two algorithms, the proposed algorithm takes the statistical characteristics of the received signal into consideration. The complexity analysis and link-level simulation results are provided to verify that the proposed method outperforms the max-log likelihood method with a little additional computational complexity, and it is a good trade-off between complexity and performance.

Kai Cheng, Ningbo Zhang, Guixia Kang

A Novel Non-WSSUS Statistical Model of Vehicle-Vehicle Radio Channel for the 5-GHz Band

In recent years, with the dramatic development in intelligent transportation systems (ITS), vehicle-vehicle (V2V) radio channel models have drawn much attention. With the analysis of the preceding statistical models of V2V channel, it is obvious that the critical works in developing statistical channel models focus on two aspects, the modeling of the time-variant properties and the modeling of the severe multipath fading. In this paper, we discuss an innovative method to model the fading dispersive channels that do not satisfy the assumption of wide-sense stationary uncorrelated scattering (WSSUS). And the Weibull distribution is integrated to mimic the severe multipath fading of V2V radio channel. Moreover, based on the tapped-delay like (TDL) model, the non-WSSUS channel impulse response (CIR) function has been formulated. There are several statistical properties characterized to evaluate the performance of the proposed model, such as, Power delay profile (PDP), Temporal autocorrelation function (ACF), Local scattering function (LSF) and Power spectrum density (PSD). The simulation results demonstrate that the proposed model has a good performance in the characterization of the non-WSSUS V2V radio channel. Hence, the channel model presented will be beneficial in future V2V communications systems.

Tao He, Ye Jin, Weiting Fu, Mingshuang Lian

GPP-SDR Based GSM-R Air Interface Monitoring System and Its Big Data Interference Analysis

In the railway transportation industry, the monitoring of Global System for Mobile Communications for Railway (GSM-R) network is essential, which plays an important role in safety of the train. The traditional monitoring systems are mainly based on the A/Abis or PRI interfaces. Therefore, the traditional ways are difficult to monitor random interferences and faults occurred over wireless channels, which may causes the potential security menace. In this paper, we propose a GSM-R monitoring system based on the Um interface. Adopting the General Purpose Processor (GPP)-Software Defined Radio (SDR) framework, the GSM-R network can be monitored by full Um interface information, including spectrum, signaling and traffic information. We use the GPP-SDR based front-end processors to obtain the data from Um interface, which are transmitted to the center servers in a railway bureau data center. After receiving the original data, the C/S structure based center servers will process the data for users to monitor. The whole system design has been implemented and deployed in Guangzhou Railway Bureau, including Guang-Shen Line and Guang-Shen-Gang Line. Furthermore, a big data interference analysis framework is proposed based on the Um interface monitoring database, which has also been verified to successfully capture and classify traditional types of interferences in field tests.

Xiang Chen, Zhongfa Li

Shared Buffer-Based Reverse Scheduling for Onboard Clos-Network Switch

Onboard switching (OBS) is facing resource constraints and special requirements of hardware complexity and scheduling efficiency. By studying the existing OBS fabrics and scheduling algorithms, the Shared Buffer-based Reverse Scheduling (SB-REV) Algorithm is proposed, adopting the shared buffer in the input module (IM) and guiding the IM scheduling with the matching result of the central modules. Theoretical and experimental analysis shows that the SB-REV algorithm greatly improves the resource utilization and scheduling efficiency, while guaranteeing the cell delay and the throughput performance. The SB-REV Algorithm is highly suitable for resource-constrained OBS environment.

Wanli Chen, Kai Liu, Xiang Chen, Xiangming Kong

Ergodic Capacity and Throughput Analysis of Two-Way Wireless Energy Harvesting Network with Decode-and-Forward Relay

In this paper, we consider a wireless energy harvesting network, where two source nodes exchange information via a decode-and-forward (DF) relay node. The network adopts the time switching relaying (TSR) or power splitting relaying (PSR) protocols. In the TSR protocol, transmitting process is split into three time slots. In the first time slot, two source nodes send the signals to the relay node simultaneously and the relay node harvests energy from the radio frequency (RF) signals. In the second time slot, two source nodes send the information signals to the relay node simultaneously. In the third time slot, the relay node decodes the signals and then forwards the regenerated signal to two source nodes using all harvested energy. In the PSR protocol, every transmission frame is divided into two equal time duration slots. The energy constrained relay node splits the received power into two parts for energy harvesting (EH) and information processing in the first time slot, respectively, and forwards the reproduced information signal to the source nodes in the second time slot. We derive the analytical expressions of the ergodic capacity and ergodic throughput of the network both for the TSR and PSR protocols. Numerical results verify the theoretical analysis and exhibit the performance comparisons of two proposed schemes.

Yingting Liu, Jianmei Shen, Hongwu Yang, Chunman Yan, Li Cong


Weitere Informationen

Premium Partner