Machine learning and intelligent communications part II 2017

The conference tracks were: MainTrack, Machine Learning; Track 1, Intelligent Positioning and Navigation; Track 2,Intelligent Multimedia Processing and Security; Track 3, Intelligent Wir

Second International Conference, MLICOM 2017

Weihai, China, August 5–6, 2017

Proceedings, Part II

227

for Computer Sciences, Social Informatics

University of Florida, Florida, USA

Xuemin Sherman Shen

University of Waterloo, Waterloo, Canada

Bo Li (Eds.)

Machine Learning

and Intelligent

Communications

Second International Conference, MLICOM 2017

Proceedings, Part II

123

ISSN 1867-8211 ISSN 1867-822X (electronic)

Lecture Notes of the Institute for Computer Sciences, Social Informatics

and Telecommunications Engineering

ISBN 978-3-319-73446-0 ISBN 978-3-319-73447-7 (eBook)

https://doi.org/10.1007/978-3-319-73447-7

Library of Congress Control Number: 2017963764

© ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2018 This work is subject to copyright All rights are reserved by the Publisher, whether the whole or part of the material is concerned, speci fically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed.

The use of general descriptive names, registered names, trademarks, service marks, etc in this publication does not imply, even in the absence of a speci fic statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made The publisher remains neutral with regard to jurisdictional claims in published maps and institutional af filiations.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG

The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

We are delighted to introduce the proceedings of the second edition of the 2017European Alliance for Innovation (EAI) International Conference on Machine Learningand Intelligent Communications (MLICOM) This conference brought togetherresearchers, developers, and practitioners from around the world who are leveragingand developing machine learning and intelligent communications.

The technical program of MLICOM 2017 consisted of 141 full papers in oralpresentation sessions at the main conference tracks The conference tracks were: MainTrack, Machine Learning; Track 1, Intelligent Positioning and Navigation; Track 2,Intelligent Multimedia Processing and Security; Track 3, Intelligent Wireless MobileNetwork and Security; Track 4, Cognitive Radio and Intelligent Networking; Track 5,Intelligent Internet of Things; Track 6, Intelligent Satellite Communications and Net-working; Track 7, Intelligent Remote Sensing, Visual Computing andThree-Dimensional Modeling; Track 8, Green Communication and Intelligent Net-working; Track 9, Intelligent Ad-Hoc and Sensor Networks; Track 10, IntelligentResource Allocation in Wireless and Cloud Networks; Track 11, Intelligent SignalProcessing in Wireless and Optical Communications; Track 12, Intelligent RadarSignal Processing; Track 13, Intelligent Cooperative Communications and Networking.Aside from the high-quality technical paper presentations, the technical program alsofeatured three keynote speeches The three keynote speeches were by Prof HaijunZhang from the University of Science and Technology Beijing, China, Prof YongWang from Harbin Institute of Technology, China, and Mr Lifan Liu from NationalInstruments China

Coordination with the steering chairs, Imrich Chlamtac, Xuemai Gu, and GongliangLiu, was essential for the success of the conference We sincerely appreciate theirconstant support and guidance It was also a great pleasure to work with such anexcellent Organizing Committee who worked hard to organize and support the con-ference, and in particular, the Technical Program Committee, led by our TPC co-chairs,Prof Xin Liu and Prof Mingjian Sun, who completed the peer-review process oftechnical papers and created a high-quality technical program We are also grateful tothe conference manager, Katarina Antalova, for her support and to all the authors whosubmitted their papers to MLICOM 2017

We strongly believe that the MLICOM conference provides a good forum forresearchers, developers, and practitioners to discuss all the science and technologyaspects that are relevant to machine learning and intelligent communications We alsohope that future MLICOM conferences will be as successful and stimulating, asindicated by the contributions presented in this volume

Gongliang Liu

Bo Li

Steering Committee

Steering Committee Chair

Imrich Chlamtac University of Trento, Create-Net, Italy

Steering Committee

Organizing Committee

General Chairs

Z Jane Wang The University of British Columbia, CanadaGongliang Liu Harbin Institute of Technology (Weihai), ChinaGeneral Co-chairs

Jianjiang Zhou Nanjing University of Aeronautics and Astronautics,

Aijun Liu Harbin Institute of Technology (Weihai), ChinaSponsorship and Exhibits Chair

Chenxu Wang Harbin Institute of Technology (Weihai), ChinaPublications Chairs

Posters and PhD Track Chair

Xiuhong Wang Harbin Institute of Technology (Weihai), China

Local Chair

Conference Manager

Katarina Antalova EAI - European Alliance for Innovation

Technical Program Committee

Technical Program Committee Chairs

Z Jane Wang University of British Columbia, Canada

Mingjian Sun Harbin Institute of Technology (Weihai), ChinaTPC Track Chairs

Machine Learning

Intelligent Positioning and Navigation

and Telecommunications, China

Intelligent Multimedia Processing and Security

Fangjun Huang Sun Yat-Sen University, China

Wireless Mobile Network and Security

Cognitive Radio and Intelligent Networking

Yulong Gao Harbin Institute of Technology, China

Weidang Lu Zhejiang University of Technology, China

Huiming Wang Xi’an Jiaotong University, China

Intelligent Internet of Things

Xiangping Zhai Nanjing University of Aeronautics and Astronautics,

ChinaChunsheng Zhu The University of British Columbia, CanadaYongliang Sun Nanjing Tech University, China

Intelligent Satellite Communications and Networking

Zhiqiang Li PLA University of Science and Technology, China

Intelligent Remote Sensing, Visual Computing, and Three-Dimensional ModelingJiancheng Luo Institute of Remote Sensing and Digital Earth,

Chinese Academy of Sciences, China

Bo Wang Nanjing University of Aeronautics and Astronautics,

ChinaGreen Communication and Intelligent Networking

Jingjing Wang Qingdao University of Science and Technology, China

Intelligent Ad-Hoc and Sensor Networks

Bao Peng Shenzhen Institute of Information Technology, China

Intelligent Resource Allocation in Wireless and Cloud Networks

Intelligent Signal Processing in Wireless and Optical Communications

Enxiao Liu Institute of Oceanographic Instrumentation,

Shandong Academy of Sciences, ChinaGuanghua Zhang Northeast Petroleum University, China

Intelligent Radar Signal Processing

Weijie Xia Nanjing University of Aeronautics and Astronautics,

ChinaXiaolong Chen Naval Aeronautical and Astronautical University,

ChinaIntelligent Cooperative Communications and Networking

Jiancun Fan Xi’an Jiaotong University, China

Contents – Part II

Intelligent Resource Allocation in Wireless and Cloud Networks

The Application of Equivalent Mean Square Error Method in Scalable

Video Perceptual Quality 3Daxing Qian, Ximing Pei, and Xiangkun Li

Spectrum Allocation in Cognitive Radio Networks by Hybrid Analytic

Hierarchy Process and Graph Coloring Theory 8Jianfei Shi, Feng Li, Xin Liu, Mu Zhou, Jiangxin Zhang,

and Lele Cheng

Spectrum Pricing in Condition of Normally Distributed User Preference 15

Li Wang, Lele Cheng, Feng Li, Xin Liu, and Di Shen

Allocation Optimization Based on Multi-population Genetic

Algorithm for D2D Communications in Multi-services Scenario 23Xujie Li, Xing Chen, Ying Sun, Ziya Wang, Chenming Li,

and Siyang Hua

Agricultural IoT System Based on Image Processing

and Cloud Platform Technology 33Yaxin Zheng and Chungang Liu

Extension of 2FSK Signal Detection Utilizing Duffing Oscillator 43Dawei Chen, Enwei Xu, Shuo Shi, and Xuemai Gu

An Efficient DOA Estimation and Network Sorting Algorithm

for Multi-FH Signals 53Xin-yong Yu, Ying Guo, Kun-feng Zhang, Lei Li, Hong-guang Li,

and Ping Sui

Study on Correlation Properties of Complementary Codes and the

Design Constraints of Complementary Coded CDMA Systems 61Siyue Sun, Guang Liang, and Kun Wang

A Novel Structure Digital Receiver 71Zijian Zhang, Dongxuan He, and Yulei Nie

Analysis of Passive Intermodulation Effect

on OFDM Frame Synchronization 79

Yi Wang, Xiangyuan Bu, Xiaozheng Gao, and Lu Tian

Variable Tap-Length Multiuser Detector for Underwater

Acoustic Communication 87Zhiyong Liu, Yinghua Wang, and Yinyin Wang

Two-Phase Prototype Filter Design for FBMC Systems 96Jiangang Wen, Jingyu Hua, Zhijiang Xu, Weidang Lu,

and Jiamin Li

A Fine Carrier Phase Recovery Method for 32APSK 106Yulei Nie, Zijian Zhang, and Peipei Liu

Intelligent Radar Signal Processing

Interferometric-Processing Based Small Space Debris Imaging 117Yuxue Sun, Ying Luo, and Song Zhang

Sparse Representation Based SAR Imaging Using Combined Dictionary 124Han-yang Xu and Feng Zhou

Parametric Sparse Recovery and SFMFT Based M-D Parameter Estimation

with the Translational Component 132Qi-fang He, Han-yang Xu, Qun Zhang, and Yi-jun Chen

A New Radar Detection Effectiveness Estimation Method Based

on Deep Learning 142Feng Zhu, Xiaofeng Hu, Xiaoyuan He, Kaiming Li, and Lu Yang

A Novel Parameter Determination Method for Lq Regularization Based

Sparse SAR Imaging 150Jia-cheng Ni, Qun Zhang, Li Sun, and Xian-jiao Liang

Downward-Looking Sparse Linear Array Synthetic Aperture Radar 3-D

Imaging Method Based on CS-MUSIC 160Fu-fei Gu, Le Kang, Jiang Zhao, Yin Zhang, and Qun Zhang

Adaptive Scheduling Algorithm for ISAR Imaging Radar

Based on Pulse Interleaving 169

Di Meng, Han-yang Xu, Qun Zhang, and Yi-jun Chen

Direction of Arrive Estimation in Spherical Harmonic Domain Using Super

Resolution Approach 179Jie Pan, Yalin Zhu, and Changling Zhou

Adaptive Mainlobe Interference Suppression in Sparse Distributed Array

Radar Based on Synthetic Wideband Signal 188Jian Luo, Honggang Zhang, and Yuanyuan Song

Wideband MIMO Radar Waveform Optimization Based on Dynamic

Adjustment of Signal Bandwidth 198Yi-shuai Gong, Qun Zhang, Kai-ming Li, and Yi-jun Chen

Learning Algorithm for Tracking Hypersonic Targets in Near Space 206Luyao Cui, Aijun Liu, Changjun Yv, and Taifan Quan

Coherent Integration Algorithm for Weak Maneuvering Target Detection

in Passive Radar Using Digital TV Signals 215Ying Zhou, Weijie Xia, Jianjiang Zhou, Linlin Huang,

and Minling Huang

High-Resolution Sparse Representation of Micro-Doppler Signal in Sparse

Fractional Domain 225Xiaolong Chen, Xiaohan Yu, Jian Guan, and You He

Estimating of RCS of Ionosphere for High Frequency Surface

Wave Radar 233Yang Xuguang, Yu Changjun, Liu Aijun, and Wang Linwei

Intelligent Cooperative Communications and Networking

Joint Mode Selection and Beamformer Optimization for Full-Duplex

Cellular Systems 243Fangni Chen, Jingyu Hua, Weidang Lu, and Zhongpeng Wang

Construction of Emergency Communication Network with Multi

Constraints Based on Geographic Information 254Yuan Feng, Fu-sheng Dai, and Ji Zhou

Design of Turntable Servo Control System Based on Sliding Mode

Control Algorithm 263Zongjie Bi, Zhaoshuo Tian, Pushuai Shi, and Shiyou Fu

Joint Power Allocation and Relay Grouping for Large MIMO Relay

Network with Successive Relaying Protocol 273Hong Peng, Changran Su, Yu Zhang, Linjie Xie, and Weidang Lu

The Second Round

Generation of Low Power SSIC Sequences 285Bei Cao and Yongsheng Wang

Intrusion Detection with Tree-Based Data Mining Classification

Techniques by Using KDD 294Mirza Khudadad and Zhiqiu Huang

Night Time Image Enhancement by Improved Nonlinear Model 304Yao Zhang, Chenxu Wang, Xinsheng Wang, Jing Wang,

and Le Man

Research on Non-contact Heart Rate Detection Algorithm 316Chenguang He, Yuwei Cui, and Shouming Wei

Lorentzian Norm Based Super-Resolution Reconstruction

of Brain MRI Image 326Dongxing Bao, Xiaoming Li, and Jin Li

A Virtual Channel Allocation Algorithm for NoC 333Dongxing Bao, Xiaoming Li, Yizong Xin, Jiuru Yang, Xiangshi Ren,

Fangfa Fu, and Cheng Liu

A Two-Layered Game Approach Based Relay’s Source Selection

and Power Control for Wireless Cooperative Networks 343Yanguo Zhou, Hailin Zhang, Ruirui Chen, and Tao Zhou

A Novel Method of Flight Target Altitude Attributes

Identification for HFSWR 351Shuai Shao, Changjun Yu, and Kongrui Zhao

A Minimum Spanning Tree Clustering Algorithm Inspired by P System 361Xiaojuan Guo and Xiyu Liu

Transfer Learning Method for Convolutional Neural Network in Automatic

Modulation Classification 371

Yu Xu, Dezhi Li, Zhenyong Wang, Gongliang Liu, and Haibo Lv

Pulse Compression Analysis for OFDM-Based Radar-Radio Systems 381Xuanxuan Tian, Tingting Zhang, Qinyu Zhang, Hongguang Xu,

and Zhaohui Song

Implementation of Video Abstract Algorithm Based on CUDA 391Hui Li, Zhigang Gai, Enxiao Liu, Shousheng Liu, Yingying Gai,

Lin Cao, and Heng Li

Realization of Traffic Video Surveillance on DM3730 Chip 402Xin Zhang and Hang Dong

Fertilization Forecasting Algorithm Based on Improved BP

Neural Network 410Tong Xue and Yong Liu

Green Resource Allocation in Intelligent Software Defined

NOMA Networks 418Baobao Wang, Haijun Zhang, Keping Long, Gongliang Liu,

and Xuebin Li

An Algorithm for Chaotic Masking and Its Blind Extraction of Image

Information in Positive Definite System 428Xinwu Chen, Yaqin Xie, Erfu Wang, and Danyang Qin

Instruction Detection in SCADA/Modbus Network Based

on Machine Learning 437Haicheng Qu, Jitao Qin, Wanjun Liu, and Hao Chen

A Joint Source-Channel Error Protection Transmission Scheme

Based on Compressed Sensing for Space Image Transmission 455Dongqing Li, Junxin Luo, Tiantian Zhang, Shaohua Wu,

and Qinyu Zhang

Local Density Estimation Based on Velocity and Acceleration Aware

in Vehicular Ad-Hoc Networks 463Xiao Luo, Xinhong Wang, Ping Wang, Fuqiang Liu,

and Nguyen Ngoc Van

Research on Millimeter Wave Communication Interference Suppression

of UAV Based on Beam Optimization 472Weizhi Zhong, Lei Xu, Xiaoyi Lu, and Lei Wang

Global Dynamic One-Step-Prediction Resource Allocation Strategy

for Space Stereo Multi-layer Data Asymmetric Scale-Free Network 482Weihao Xie, Zhigang Gai, Enxiao Liu, and Dingfeng Yu

Machine Learning Based Key Performance Index Prediction Methods

in Internet of Industry 490Haowei Li, Liming Zheng, Yue Wu, and Gang Wang

An Auction-Gaming Based Routing Model for LEO Satellite Networks 498Ligang Cong, Huamin Yang, Yanghui Wang, and Xiaoqiang Di

Parameters Estimation of Precession Cone Target Based

on Micro-Doppler Spectrum 509MingFeng Wang, AiJun Liu, LinWei Wang, and ChangJun Yu

Automated Flowering Time Prediction Using Data Mining

and Machine Learning 518Runxuan Li, Yu Sun, and Qingquan Sun

Spatial Crowdsourcing-Based Sensor Node Localization in Internet

of Things Environment 528Yongliang Sun, Yejun Sun, and Kanglian Zhao

Influence of Inter-channel Error Distribution on Mismatch

in Time-Interleaved Pipelined A/D Converter 537Yongsheng Wang, Chen Yin, and Xunzhi Zhou

Distributed Joint Channel-Slot Selection for Multi-UAV Networks:

A Game-Theoretic Learning Approach 546Jiaxin Chen, Yuhua Xu, Yuli Zhang, and Qihui Wu

Ship Detection in SAR Using Extreme Learning Machine 558Liyong Ma, Lidan Tang, Wei Xie, and Shuhao Cai

Obtaining Ellipse Common Tangent Line Equations by the Rolling Tangent

Line Method 569Naizhang Feng, Teng Jiang, Shiqi Duan, and Mingjian Sun

On Sampling of Bandlimited Graph Signals 577

Mo Han, Jun Shi, Yiqiu Deng, and Weibin Song

Data Association Based Passive Localization in Complex

Multipath Scenario 585Bing Zhao and Ganlin Hao

Design and Implementation of Multi-channel Burst Frame Detector 595Bing Zhao

Research on Cache Placement in ICN 603

Yu Zhang, Yangyang Li, Ruide Li, and Wenjing Sun

The Digital Chaos Cover Transport and Blind Extraction

of Speech Signal 612Xinwu Chen, Yaqin Xie, and Erfu Wang

A Multi-frame Image Speckle Denoising Method Based on Compressed

Sensing Using Tensor Model 622Ruofei Zhou, Gang Wang, Wenchao Yang, Zhen Li, and Yao Xu

Frequency-Hopped Space-Time Coded OFDM over Time-Varying

Multipath Channel 634Fangfang Cheng, Jiyu Jin, Guiyue Jin, Peng Li, and Jun Mou

Dynamic Characteristic Analysis for Complexity of Continuous Chaotic

Systems Based on the Algorithms of SE Complexity and C0Complexity 647Xiaolin Ye, Jun Mou, Zhisen Wang, Peng Li, and Chunfeng Luo

Design and Implemention of an Emulation Node for Space Network

Protocol Testing 658Sichen Zhao, Yuan Fang, Wenfeng Li, and Kanglian Zhao

Optimization Spiking Neural P System for Solving TSP 668Feng Qi and Mengmeng Liu

Author Index 677

Contents – Part I

Machine Learning

An Effective QoS-Based Reliable Route Selecting Scheme

for Mobile Ad-Hoc Networks 3Jiamei Chen, Yao Wang, Xuan Li, and Chao Gao

Space Encoding Based Compressive Tracking with Wireless

Fiber-Optic Sensors 12Qingquan Sun, Jiang Lu, Yu Sun, Haiyan Qiao,

and Yunfei Hou

Moving Object Detection Algorithm Using Gaussian Mixture Model

and SIFT Keypoint Match 22Hang Dong and Xin Zhang

Low-Complexity Signal Detection Scheme Based on LLR for Uplink

Massive MIMO Channel 30Xifeng Chen, Liming Zheng, and Gang Wang

Accurate Scale-Variable Tracking 40Xinyou Li, Wenjing Kang, and Gongliang Liu

Sparse Photoacoustic Microscopy Reconstruction Based on Matrix

Nuclear Norm Minimization 49Ying Fu, Naizhang Feng, Yahui Shi, Ting Liu, and Mingjian Sun

Clustering Analysis Based on Segmented Images 57Hongxu Zheng, Jianlun Wang, and Can He

Channel Estimation Based on Approximated Power Iteration Subspace

Tracking for Massive MIMO Systems 76Liming Zheng, Donglai Zhao, Gang Wang, Yao Xu, and Yue Wu

BER Performance Evaluation of Downlink MUSA over Rayleigh

Fading Channel 85Yao Xu, Gang Wang, Liming Zheng, Rongkuan Liu,

and Donglai Zhao

Intelligent Positioning and Navigation

Privacy Protection for Location Sharing Services in Social Networks 97Hui Wang, Juan Chen, Xianzhi Wang, Xin Liu, and Zhenyu Na

A Non-line-of-Sight Localization Method Based on the Algorithm

Residual Error Minimization 103Sunan Li, Jingyu Hua, Feng Li, Fangni Chen, and Jiamin Li

WLAN Indoor Localization Using Angle of Arrival 112Zengshan Tian, Yong Li, Mu Zhou, and Yinghui Lian

Defect Detection of Photovoltaic Modules Based on Convolutional

Neural Network 122Mingjian Sun, Shengmiao Lv, Xue Zhao, Ruya Li,

Wenhan Zhang, and Xiao Zhang

An Effective BLE Fingerprint Database Construction Method

Based on MEMS 133

Mu Zhou, Xiaoxiao Jin, Zengshan Tian, Haifeng Cong,

and Haoliang Ren

Intelligent Multimedia Processing and Security

A New Universal Steganalyzer for JPEG Images 145

Ge Liu, Fangjun Huang, Qi Chen, and Zhonghua Li

Double JPEG Compression Detection Based on Fusion Features 158Fulong Yang, Yabin Li, Kun Chong, and Bo Wang

Complexity Based Sample Selection for Camera Source Identification 168Yabin Li, Bo Wang, Kun Chong, and Yanqing Guo

Wireless Mobile Network and Security

Lattice Reduction Aided Linear Detection for Generalized

Spatial Modulation 181Chungang Liu, Chen Wang, and Wenbin Zhang

Radio Frequency Fingerprint Identification Method

in Wireless Communication 195Zhe Li, Yanxin Yin, and Lili Wu

Cognitive Radio and Intelligent Networking

Short Term Prediction Models of Mobile Network Traffic Based

on Time Series Analysis 205Yunxue Gao, Liming Zheng, Donglai Zhao, Yue Wu,

and Gang Wang

Calculation Method of Field Strength in the Case of Side Obstacles 212

Lu Chen, Fusheng Dai, Yonggang Chi, and Ji Zhou

Variable Dimension Measurement Matrix Construction for Compressive

Sampling via m Sequence 221Jingting Xiao, Ruoyu Zhang, and Honglin Zhao

Signal Quality Assessment of Wireless Signal Based

on Compressed Sensing 231Fei An and Fusheng Dai

Distributed Compressive Sensing Based Spectrum Sensing Method 239Yanping Chen, Yulong Gao, and Yongkui Ma

Recent Advances in Radio Environment Map: A Survey 247Jingming Li, Guoru Ding, Xiaofei Zhang, and Qihui Wu

Elimination of Inter-distract Downlink Interference Based on

Autocorrelation Technique 258Hui Kang, Hongyang Xia, and Fugang Liu

Intelligent Internet of Things

Application of Cooperative Communications with Dual-Stations

in Wireless Mobile Environments 271Ershi Xu, Xiangping Zhai, Weiyi Lin, and Bing Chen

Design for Attendance System with the Direction Identification

Based on RFID 282Hongyuan Wang

A Geo-Based Fine Granularity Air Quality Prediction Using

Machine Learning and Internet-of-Things 291Hang Wang, Yu Sun, and Qingquan Sun

Research on Key Technology in Traditional Chinese Medicine (TCM)

Smart Service System 300Yongan Guo, Tong Liu, Xiaomin Guo, and Ye Yang

Application of Wireless Sensor Network in Smart Buildings 315Mingze Xia and Dongyu Song

Distributed System Model Using SysML and Event-B 326

Qi Zhang, Zhiqiu Huang, and Jian Xie

Intelligent Satellite Communications and Networking

A Full-Protocol-Stack Testbed for Space Network Protocol Emulation 339Xiaoqin Ni, Kanglian Zhao, and Wenfeng Li

Application Layer Channel Coding for Space DTN 347Dongxu Hou, Kanglian Zhao, and Wenfeng Li

Routing Optimization of Small Satellite Networks Based

on Multi-commodity Flow 355Xiaolin Xu, Yu Zhang, and Jihua Lu

Modeling of Satellite-Earth Link Channel and Simulating in Space-Ground

Integrated Network 364Beishan Wang and Qi Guo

A Deep Learning Method Based on Convolutional Neural Network

for Automatic Modulation Classification of Wireless Signals 373

Yu Xu, Dezhi Li, Zhenyong Wang, Gongliang Liu, and Haibo Lv

Modeling and Performance Analysis of Multi-layer Satellite Networks

Based on STK 382

Bo Li, Xiyuan Peng, Hongjuan Yang, and Gongliang Liu

Artificial-Neural-Network-Based Automatic Modulation Recognition

in Satellite Communication 394Yumeng Zhang, Mingchuan Yang, and Xiaofeng Liu

Licklider Transmission Protocol for GEO-Relayed Space Networks 405Wenrui Zhang, Chenyang Fan, Kanglian Zhao, and Wenfeng Li

Intelligent Remote Sensing, Visual Computing

and Three-Dimensional Modeling

Design of LED Collimating Optical System 417Yihao Wang, Yuncui Zhang, Xufen Xie, and Yuxuan Zhang

Global Depth Refinement Based on Patches 423

Xu Huang, Yanfeng Zhang, Gang Zhou, Lu Liu,

and Gangshan Cai

3D Surface Features Scanning System with UAV-Carried Line Laser 434Yilang Sun, Shuqiao Sun, Zihao Cui, Yanchao Zhang,

and Zhaoshuo Tian

Contourlet Based Image Denoising Method Combined Recursive

Cycle-Spinning Algorithm 444Hongda Fan, Xufen Xie, Yuncui Zhang, and Nianyu Zou

Green Communication and Intelligent Networking

A Resource Allocation Algorithm Based on Game Theory in UDN 453Changjun Chen, Jianxin Dai, Chonghu Cheng, and Zhiliang Huang

Optimal Relay Selection Algorithm for Combining Distance and Social

Information in D2D Cooperative Communication Networks 463Kaijian Li, Jianxin Dai, Chonghu Cheng, and Zhiliang Huang

Linear Massive MIMO Precoding Based on Nonlinear

High-Power Amplifier 475Xudong Yin, Jianxin Dai, Chonghu Cheng, and Zhiliang Huang

Linear Precoding for Massive MIMO Systems with IQ Imbalance 484Juan Liu, Jianxin Dai, Chonghu Cheng, and Zhiliang Huang

Research on Insurance Data Analysis Platform Based on the

Hadoop Framework 494Mingze Xia

SNR Analysis of the Millimeter Wave MIMO with Lens Antenna Array 505Min Zhang, Jianxin Dai, Chonghu Cheng, and Zhiliang Huang

Cross-Entropy Optimization Oriented Antenna Selection for Clustering

Management in Multiuser MIMO Networks 516Xinyu Zhang, Jing Guo, Qiuyi Cao, and Nan Zhao

Subcarrier Allocation-Based Simultaneous Wireless Information

and Power Transfer for Multiuser OFDM Systems 524Xin Liu, Xiaotong Li, Zhenyu Na, and Qiuyi Cao

Intelligent Ad-Hoc and Sensor Networks

A 100 MHz SRAM Design in 180 nm Process 535Zhuangguang Chen and Bei Cao

A Modified AODV Protocol Based on Nodes Velocity 545Tong Liu, Zhimou Xia, Shuo Shi, and Xuemai Gu

RSA Encryption Algorithm Design and Verification Based

on Verilog HDL 555Bei Cao, Tianliang Xu, and Pengfei Wu

A Novel High Efficiency Distributed UEP Rateless Coding Scheme

for Satellite Network Data Transmission 564Shuang Wu, Zhenyong Wang, Dezhi Li, Qing Guo,

and Gongliang Liu

A New Class of Unequal Error Protection Rateless Codes with Equal

Recovery Time Property 574Shuang Wu, Zhenyong Wang, Dezhi Li, Gongliang Liu,

and Qing Guo

Stochastic Geometry Analysis of Ultra Dense Network

and TRSC Green Communication Strategy 584Guoqiang Wang and Bai Sun

Reputation-Based Framework for Internet of Things 592Juan Chen, Zhengkui Lin, Xin Liu, Zhian Deng,

and Xianzhi Wang

Gain-Phase Error Calculation in DOA Estimation for Mixed

Wideband Signals 598Jiaqi Zhen, Yong Liu, and Yanchao Li

Mutual Coupling Estimation in DOA Estimation for Mixed

Wideband Signals 606Jiaqi Zhen, Yong Liu, and Yanchao Li

Efficient Data Gathering with Compressed Sensing Multiuser Detection

in Underwater Wireless Sensor Networks 614Rui Du, Wenjing Kang, Bo Li, and Gongliang Liu

An Efficient Data Collection and Load Balance Algorithm in Wireless

Sensor Networks 626Danyang Qin, Ping Ji, Songxiang Yang, and Qun Ding

RFID Based Electronic Toll Collection System Design

and Implementation 635Yang Li and Peidong Zhuang

Design and Implementation of Survey Vehicle Based on VR 641Weiguang Zhao and Peidong Zhuang

Development of the Embedded Multi Media Card Platform

Based on FPGA 648Songyan Liu, Ting Chen, Shangru Wu,

and Cheng Zhang

An Implementation of Special Purpose SSD Device 657Songyan Liu, Shangru Wu, Ting Chen, and Cheng Zhang

Performance Evaluation of DTN Routing Protocols in Vehicular

Network Environment 666Yongliang Sun, Yinhua Liao, Kanglian Zhao, and Chenguang He

Benefits of Compressed Sensing Multi-user Detection

for Spread Spectrum Code Design 675Yan Wu, Wenjing Kang, Bo Li, and Gongliang Liu

Application of Time-Varying Filter in Time-Frequency

Resource Allocation 682Zhongchao Ma, Liang Ye, and Xuejun Sha

Secure Communication Mechanism Based on Key Management

and Suspect Node Detection in Wireless Sensor Networks 692Danyang Qin, Songxiang Yang, Ping Ji, and Qun Ding

Research on the Pre-coding Technology of Broadcast Stage

in Multi-user MIMO System 701Guoqiang Wang and Shangfu Li

Author Index 711

Intelligent Resource Allocation in Wireless and Cloud Networks

Error Method in Scalable Video Perceptual

QualityDaxing Qian(&), Ximing Pei, and Xiangkun LiDalian Neusoft University of Information,Software Park Road 8, Dalian 116023, Liaoning, China{qiandaxing,peiximing,lixiangkun}@neusoft.edu.cn

Abstract Scalable video is a stream video over heterogeneous networks todifferent clients To provide the better quality of service (QoS) or quality ofexperience (QoE) to customer, we propose an Equivalent Mean Square Error(Eq-MSE) method which is developed based on spatial and temporal frequencyanalysis of input video content Eq-MSE is used to calculate minimal frame rate(MinFR) for different videos to guarantee motion without jitter The proposedscheme in this paper can provide better perceptual video quality than withoutconsidering the video content impact

Keywords: SVC
Eq-MSE
MinFR
Perceptual quality

1 Introduction

With the advances of semi-conductor and access network technologies, real-time videostreaming becomes more and more popular in our daily life We can enjoy the videosservice at famous website through different networks using heterogeneous devices.How to provide the high quality of service (QoS) or quality of experience (QoE) todifferent users over heterogeneous networks is a crucial problem for the success ofvideo streaming application Scalable video coding (SVC) [1,2] is a full resolutionscalable video stream which can be truncated to adapt different requirements imposed

by the subscribed users and underlying access networks

SVC includes temporal, spatial, SNR and combined scalabilities Temporal bility is realized by the hierarchical-B prediction [3] Spatial scalability is achieved byencoding each supported spatial resolution into one layer SNR scalability includescoarse grain scalability (CGS) and medium grain scalability (MGS) [4] To achieve theSNR refinement, we usually use different quantization steps at different SNR layers Inthis paper, we study the temporal and SNR joint scalability, and the spatial scalability isnot mentioned

scala-Video content have a significant impact on the perceptual quality For example, amotion intensive video need a larger frame rate to maintain the continuity of the objectmovement and avoid jitter and guarantee the motion smoothness, while for stationaryvideo, a relatively lower frame rate is enough to provide the decent video quality Formotion-intensive content, bit stream extracted at higher frame rate is favored On the

© ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2018

X Gu et al (Eds.): MLICOM 2017, Part II, LNICST 227, pp 3–7, 2018.

https://doi.org/10.1007/978-3-319-73447-7_1

other hand, if there are larger high-frequency components (i.e., rich texture) in a singleframe of the video, a finer quantization to reach better spatial quality is typicallypreferred To solve the problem, we study the spatial and temporal frequency of theinput video content, and propose an Equivalent Mean Square Error (Eq-MSE) scheme

to derive the minimal frame rate (MinFR) for different video sources to guarantee themotion smoothness and excellent QoE of the decoded video

This paper is organized as follows Section2introduces the temporal frequency in

a video sequence (i.e., motion) In Sect.3we introduce the Eq-MSE method to derivethe minimal frame rate without jitter for different input video sources Subjective testevaluation and experimental results are shown in Sect.4 Section5concludes the paperand discusses the future directions

2 Temporal Frequency

The concept of spatial frequency is introduced in [5]

We can use the function [6,7]:

From (2) we can draw a conclusion that the temporal frequency depends on notonly the motion, but also the spatial frequency [6] of the object

3 Equivalent Mean Square Error (Eq-MSE)

We propose an Eq-MSE method to calculate the SF of general objects in a picture andfind the appropriate frame rate [6,7]

Figure1illustrates that the size of black column is w  h and picture size is W  H.

We use ftB to represent the induced frame rate by the object, which is defined as:

ftB¼ MSEf
Hh; 0MSEf 1; 0ð Þ
vxMSEf 0;Ww

MSEf 0;
WwMSEf 0; 1ð Þ

H;wW

is all the MBs in the picture vxand vyare velocities in horizontal and verticaldirections of corresponding MB We get the mode and number of MB in a picture, andthen choose the other picture within the same GOP to get MVs according to every MB.The ratio between MVs number in MB and the time interval between two frames are vxand vy For example, the Ph

Hvx and Pw

Wvy of sequence Mobile are 12.2, 9.4,respectively Its MinFR is 12.2 + 9.4 = 21.6 Note that with a real signal, the CSFT issymmetric, so that for every frequency component at fx; fy

, there is also a component

at fx; fy

with the same magnitude The corresponding temporal frequency caused

by this other component is fxvxþ fyvy[5]

Equation (5) is the function of minimal frame rate (MinFR) that makes the videomotion smoothness without jitter

Fig 1 Illustrativefigure for object in general picture

4 Experimental Results

We invite 15 experimenters to give the decoded video subjective ratings for evaluatethe subjective quality Sub0 is the default scalable video adaptation without consideringthe video content impact, while sub1 is scalable adaptation with dependent videocontent We use 11 ranks (i.e., 0–10) for the subjective tests ranging The worst is 0 andthe best is 10 The subjective assessment follows [8] The results show in Table1

Table1 depicts the subjective test results of four sequences It is obviously that

“City”, “Mobile” and “Football” have better perceptual rating for sub1 session, while

“Akiyo” is quite similar between sub1 and sub0 We can draw a conclusion that theEq-MSE method is providing better-decoded video quality at a given bit rate

5 Conclusions

In this paper, we propose the Eq-MSE scheme, which is developed based on the spatialand temporal frequency analysis of the video content This scheme is used to derive theMinFR for different videos and in consequence, so as to guarantee the motionsmoothness for decent decoded video quality Compared with the default scalablevideo adaptation without considering the video content impact, our proposed schemecan provide better perceptual video quality at the same bit rate according to the sub-jective quality assessments

5 Wang, Y., Ostermann, J., Zhang, Y.-Q.: Video Processing and Communications (2001)

Table 1 Sequences subjective test comparative results

Sequences Sub0 Sub1Akiyo 6.6 6.8City 4.9 7.7Mobile 5.7 7.1Football 6.1 7.9

6 Qian, D., Wang, H., Niu, F.: Scalable video coding bit stream extraction based on equivalentMSE method In: Advanced Materials Research, vol 204–210, pp 1728–1732 (2011)

7 Qian, D., Wang, H., Sun, W., Zhu, K.: Bit stream extraction based on video content method inthe scalable extension of H.264/AVC J Softw 6, 2090–2096 (2011)

8 ITU-R Rec BT.500-11: Methodology for the subjective assessment of the quality oftelevision pictures (2002)

Networks by Hybrid Analytic Hierarchy Process

and Graph Coloring Theory

Jianfei Shi1, Feng Li1(B), Xin Liu2, Mu Zhou3,Jiangxin Zhang1, and Lele Cheng1

Abstract In this paper, a graph coloring-based spectrum allocation

algorithm in cognitive radio networks combined with analytic hierarchyprocess is proposed By analyzing several key factors that affect the qual-ity of the leased spectrum, the algorithm combines the graph algorithmand analytic hierarchy process to assign the optimal spectrum to cogni-tive users orderly Simulation results show that the proposed algorithmcan effectively improve the network efficiency compared with originalalgorithms and arose inconspicuous loss to the whole network’s fairness.The proposal not only improves the efficiency of spectrum allocation, butalso balances the requirements of the overall fairness of cognitive radionetworks

Keywords: Cognitive radio·Graph coloring·Spectrum allocation

Analytic hierarchy process

Spectrum sharing is the key technology in cognitive radio which attracts theincreasing interest [1 3] Graph theory, as a classical optimization theory, hasbeen introduced to solve the difficulty of spectrum allocation in cognitive radio.[4] proposed a list-coloring algorithm based on the graph coloring theory, whichincludes distributed greedy algorithm and distributed fairness algorithm In [5],

as the list coloring algorithm can allocate only one spectrum once, the authorsproposed a spectrum allocation algorithm to assign channels to multiple users

at the same time without incurring interference Based on [5], authors in [6]proposed an improved graph coloring algorithm for spectrum allocation withregards to the maximum weighted independent set to improve the spectrumutilization by combining the power control technology

c

 ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2018

X Gu et al (Eds.): MLICOM 2017, Part II, LNICST 227, pp 8–14, 2018.

In this paper, we apply analytic hierarchy process (AHP), one of the objective decision method to proceed spectrum decision and provide a reasonablespectrum access strategy according to the heterogeneous idle spectrum Theimproved proposal takes into account diverse spectral characteristics to meet theactual needs of the spectrum allocation in cognitive radio networks In addition,

multi-by combining the advantages of the methods of AHP and coloring theory, theproposal not only improves the efficiency of spectrum allocation, but also satisfiesthe requirements of the overall benefits of cognitive radio networks

2.1 Spectrum Selection Based on AHP

In spectrum selection, secondary users always want to switch to the spectrumwith high bandwidth, low delay, low jitter and packet loss rate, etc Therefore,this paper selects four indexes of bandwidth, delay, jitter and packet loss rate

as the judgment criterion of spectrum selection During the course, secondaryusers should also take their own preferences into account When the number ofconsidering factors become increasing, secondary users will struggle to make arational choice by qualitative analysis We thus introduce the method of AHP

to analyze this problem The selection problem can decomposed into three levels

as shown in Fig.1 In more complex environment, more evaluation criteria can

be introduced to make it closer to realize

Fig 1 Hierarchical graph of optimal spectrum decision based on AHP algorithm

To compare the impacts of factors C1, C2, · · · , C n of one layer on a factor

S iof another layer, such as the importance of different choice criterions on final

channel selection, it is essential to make a comparison between only two factorsrather than multiple factors at the same time Selecting two factors C i andC j

each time, we use a ij to denote the impact ratio of C i and C j on S i, all thecomparison results can be expressed in matrix as following

2.2 Single Hierarchical Arrangement and Consistency Check

Single hierarchical arrangement refers to the same level of the correspondingfactors for the relative importance of the upper level of a factor ranking weight,

it can be obtained by normalizing the eigenvector (weighted vector) W of the

largest eigenvalueλ maxof judgment matrixA So the essence is to calculate the

weight vector The calculation of the weight vector has the characteristic rootmethod, the sum method, the root method, the power method and so on In thispaper, we use the sum method The steps of the sun method are as follows:

1 Normalize each column vector ofA : ˜ W ij =a ij /n

The procedure for the consistency check is as follows:

Step 1: Calculate the consistency index (CI).

Step 2: Seek table to determine the corresponding random index (RI).

According to the different order of the judgment matrix, we get the averagerandom index RI

Table 1 Random index RI

Matrix order 1 2 3 4 5 6 7 8 9 10

RI 0 0 0.58 0.9 1.12 1.24 1.32 1.41 1.45 1.40

Step 3: Calculate the consistency ratio (CR) and make judgments.

Fig 2 Diagram of greedy spectrum allocation algorithm combined with AHP

CR > 0.1, it is considered that the judgment matrix does not meet the

consistency requirement, the judgment matrix should be revised again To testthe consistency We take matrix A as an example First, we use (3) to calcu-late the consistency index CI = 0.0083, and then obtain RI = 0.9 via Table1.Finally, due to the fact that CR = 0.0092 < 0.1, we can conclude that A is

verified to pass consistency check

2.3 Improved Spectrum Allocation Algorithm Model

According to the analysis above-mentioned, by selecting the optimal spectrumand then using the graph algorithm to allocate idle spectrum, we can make fulluse of spectrum resources, enhance the spectrum utilization and improve theoverall efficiency

Combining the distributed greedy algorithm and the method of AHP, thespectrum decision diagram is shown in Fig.2

The improved spectrum allocation algorithm can be described as follows:After initializing the system and updating the node information, according tothe measured values of bandwidth, delay, jitter and packet loss rate of eachspectrum, the hierarchical structure is established by using AHP; Set up thecorresponding comparison matrix, and obtain the spectrum efficiency; Finally,the selected optimal spectrum is allocated using the coloring algorithm

WDGA AHP-DGA DGA

Fig 3 Network utility curves of the three algorithms in greed mode

40 60 80 100 120

Number of secondary users 1

1.5 2 2.5 3 3.5 4

WDGA AHP-DGA DGA

Fig 4 Changing curves of the three algorithms in greedy mode

U(R) to measure the network efficiency, with the variance to measure the

fair-ness between users In the simulation, randomly generate the network topologydiagram Furthermore, we randomly set the values in available spectrum matrix

L, interference matrix C and utility matrix B within [0, 1].

The parameters including bandwidth, delay, jitter and packet loss rate meetthe data transmission standard of wireless networks proposed by ITU-T [7] FromFigs.3and4, our proposed method using AHP and distributed greed algorithm(AHP-DGA) is compared with the results obtained by original distributed greedyalgorithm (DGA) and weighted distributed greedy algorithm (WDGA) It can

be concluded that AHP-DGA can receive high network efficiency and decentnetwork fairness

References

1 Fadeel, K.Q.A., Elsayed, D., Khattab, A., Digham, F.: Dynamic spectrum accessfor primary operators exploiting LTE-A carrier aggregation In: IEEE ICNC, pp.143–147 (2015)

2 Li, F., Tan, X., Wang, L.: A new game algorithm for power control in cognitive

radio networks IEEE Trans Veh Technol 60(9), 4384–4392 (2011)

3 Liu, X., Jia, M., Tan, X.: Threshold optimization of cooperative spectrum sensing

in cognitive radio network Radio Sci 48(1), 23–32 (2013)

4 Wang, W., Liu, X.: List-coloring based channel allocation for open-spectrum wirelessnetworks In: IEEE VTC-Fall, pp 690–694 (2005)

5 Liu, Y., Jiang, M., Tan, X., et al.: Maximal independent set based channel cation algorithm in cognitive radios In: IEEE Youth Conference on Information,Computing and Telecommunication, pp 78–81 (2009)

allo-6 Bao, Y., Wang, S., Yan, B., et al.: Research on maximal weighted independent based graph coloring spectrum allocation algorithm in cognitive radio networks In:Proceedings of the International Conference on Communications, Signal Processingand Systems, pp 263–271 (2016)

set-7 ITU-T (2016).https://en.wikipedia.org/wiki/ITU-T

8 Li, M.: Research of cognitive radio spectrum allocation algorithm based on graphtheory Southwest Jiaotong University, pp 45–47 (2012)

Distributed User Preference

Li Wang1, Lele Cheng1, Feng Li1(B), Xin Liu2, and Di Shen1

1 Zhejiang University of Technology, Hangzhou 310023, Zhejiang, China

{liwang2002,fenglzj}@zjut.edu.cn, 478708892@qq.com, 814120631@qq.com

2

Dalian University of Technology, Dalian 116024, Liaoning, China

liuxinstar1984@dlut.edu.cn

Abstract During secondary user’s dynamic access to authorized

spec-trum, a key issue is how to ascertain an appropriate spectrum price so

as to maximize primary system’s benefit and satisfy secondary user’sdiverse spectrum demands In this paper, a scheme of pricing-baseddynamic spectrum access is proposed According to the diverse qualities

of idle spectrum, the proposal applies Hotelling game model to describethe spectrum pricing problem Firstly, establish a model of spectrumleasing, among which the idle spectrum with different qualities forms aspectrum pool Then, divide the idle spectrum into equivalent width ofleased channels, which will be uniformly sold in order Secondary userscan choose proper channels to purchase in the spectrum pool according

to their spectrum usage preferences which are subject to normal bution and affected by the spectrum quality and market estimation Thispaper analyzes the effect of spectrum pricing according to the primarysystem’s different tendencies to spectrum usage and economic income

distri-Keywords: Spectrum pricing·Cognitive radio·User preference

Spectrum quality

With the rapid development of wireless communication technology and the lishment of next-generation 5g communication standard, high-quality idle spec-trum is more scarce which has become one of the bottlenecks restricting thedevelopment of wireless communication technology [1] Cognitive radio which isbased on dynamic spectrum access has attracted more and more attention ofacademe and engineering recent years [2] Various kinds of emerging networktechnology have begun to adopt dynamic spectrum detection and dynamic spec-trum access to improve the efficiency of spectrum utilization In the process

estab-of dynamic spectrum access, primary users owning licensed spectrum can leasethe idle channels to secondary user to gain incomes For primary users, how

to identify an optimal channel pricing to maximize its own profit has become

a significant issue In this paper, we directly price the idle spectrum of rized users according to the secondary user’s diverse preferences The spectrum

autho-c

 ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2018

X Gu et al (Eds.): MLICOM 2017, Part II, LNICST 227, pp 15–22, 2018.

pricing scheme has a prior estimate to the spectrum market Compared with thespectrum auction, it doesn’t need many overheads and improves the convenience

of the spectrum access

Spectrum trading provides an efficient way for secondary users to dynamicallyaccess licensed bands while the financial gains can encourage primary users tolease unused spectrum temporarily Generally, the participants can perform thedeal by auction-based method or pricing-based method The spectrum auctionmechanism can be divided into many kinds according to different applicationcircumstances, such as trust-based auction which relaxes the credit limit appro-priately in return for a higher economic efficiency to balance the honesty andthe efficiency [3,4] On the other hand, to lower the overhead and time costfor spectrum pricing, pricing-based spectrum trading has also been widespreadconcerned either [5,6]

In this paper, we investigate how to price the spectrum when heterogeneousspectrum and stochastic secondary user’s preference are under consideration

A concept of spectrum pool is introduced to facilitate the following spectrumdeal A secondary spectrum customer will pick a high-quality channel for usagewhen its capital is ample or wide band is required to support essential service

We adopt Hotelling model which is proper to describe the product pricing issue

in heterogeneous market By analyzing the secondary user’s preference eter, an iterative algorithm for spectrum pricing is obtained by fixing the Nashequilibrium Numerical results are further provided to evaluate how the pricingparameters affect the primary system’s profits

Suppose the idle spectrum leased by the primary system consists a spectrumsharing pool, where the spectrum can be divided into many uniform channels forselling Besides, the qualities of these channels are not homogeneous For high-quality channels, the secondary users suffers lower channel fading or adjacentchannel interference Thus, secondary users choose these channels according totheir diverse preferences The preference parameter is determined by the channelquality and channel price

2.1 Utility Functions

In this paper, we consider the spectrum trading is performed without auctionactivities During the course, primary systems have no prior knowledge of thesecondary customer’s spectrum preference In spectrum trading, the utility func-tion of a secondary user can be expressed as

where θ denotes the secondary user’s preference, s denotes the channel quality and p denotes the channel price In the spectrum sharing pool, it is assumed that

two kinds of channels with diverse qualities can be chosen by secondary users asshown in Fig.1 We use s1to denote the channel quality of high-quality channel

and s2the low-quality channel Then, we have s1> s2> 0 Here, different

chan-nel qualities means various transmission capacities Furthermore, we suppose

the secondary user’s preference parameter θ is subject to normal distribution expressed as g(θ) θ locates in the region of [θ L , θ H ], and ρ is the corresponding probability distribution function denoted as ρ = G(θ) We adopt θ0 to expressthe non-preference parameter of a cognitive user which means no demand differ-ence existing between the high-quality channel and low-quality channel Then,

it can be calculated as θ0 = p1−p2

s1−s2, where p1 and p2 represent the two kinds

of channels’ prices When a secondary user’s spectrum preference θ i is higher

than θ0, the user prefers to choose the high-quality channel Otherwise, it wouldrather to choose the low-quality channel to lease

Fig 1 Spectrum pool

2.2 Spectrum Pricing

Secondary user’s preference parameter is considered to be non-uniform and obeynormal distribution in practical application Figure2shows the density curve ofthe standard normal distribution The probability density can be given as

2πe − x22

(2)Then, the distribution function is

-3 -2 -1 0 1 2 3 0

0.1 0.2 0.3 0.4

2μ-a

(μ)

f(x)

Fig 3 General normal distribution

Thus, the probability can be approximately calculated in given region [−a, a].

The conclusion can also be applied to the case of general normal distribution

as shown in Fig.3 When the distribution mean is μ, the probability calculated approximately in [a, 2μ − a] is obtained as

f(a) =



1− e − (u−a)2

Furthermore, as shown in Fig.4, the secondary customer whose preference

parameters θ locates in [θ L , θ0], will purchase low-quality channels The user

with preference parameters θ ∈ [θ0, θ H] chooses a high-quality channel

Then, in order to obtain the specific solution, divide the red shadow part inFig.4 into two parts, where we have θ0 = θ0+ μ and θ H − θ