Artificial Intelligence for

Communications and Networks: First

EAI International Conference, AICON
2019, Harbin, China, May 25–26, 2019,
Proceedings, Part II Shuai Han
Visit to download the full and correct content document:
https://textbookfull.com/product/artificial-intelligence-for-communications-and-network
s-first-eai-international-conference-aicon-2019-harbin-china-may-25-26-2019-proceed
ings-part-ii-shuai-han/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Artificial Intelligence for Communications and

Networks: First EAI International Conference, AICON
2019, Harbin, China, May 25–26, 2019, Proceedings, Part
II Shuai Han
https://textbookfull.com/product/artificial-intelligence-for-
communications-and-networks-first-eai-international-conference-
aicon-2019-harbin-china-may-25-26-2019-proceedings-part-ii-shuai-
han/

Artificial Intelligence for Communications and

Networks: First EAI International Conference, AICON
2019, Harbin, China, May 25–26, 2019, Proceedings, Part
I Shuai Han
https://textbookfull.com/product/artificial-intelligence-for-
communications-and-networks-first-eai-international-conference-
aicon-2019-harbin-china-may-25-26-2019-proceedings-part-i-shuai-
han/

Security and Privacy in Communication Networks 15th EAI

International Conference SecureComm 2019 Orlando FL USA
October 23 25 2019 Proceedings Part II Songqing Chen

https://textbookfull.com/product/security-and-privacy-in-
communication-networks-15th-eai-international-conference-
securecomm-2019-orlando-fl-usa-october-23-25-2019-proceedings-
part-ii-songqing-chen/

Broadband Communications, Networks, and Systems: 10th

EAI International Conference, Broadnets 2019, Xi’an,
China, October 27-28, 2019, Proceedings Qingshan Li

https://textbookfull.com/product/broadband-communications-
networks-and-systems-10th-eai-international-conference-
broadnets-2019-xian-china-october-27-28-2019-proceedings-
Advanced Hybrid Information Processing Third EAI
International Conference ADHIP 2019 Nanjing China
September 21 22 2019 Proceedings Part II Guan Gui

https://textbookfull.com/product/advanced-hybrid-information-
processing-third-eai-international-conference-adhip-2019-nanjing-
china-september-21-22-2019-proceedings-part-ii-guan-gui/

Artificial Neural Networks and Machine Learning ICANN

2019 Deep Learning 28th International Conference on
Artificial Neural Networks Munich Germany September 17
19 2019 Proceedings Part II Igor V. Tetko
https://textbookfull.com/product/artificial-neural-networks-and-
machine-learning-icann-2019-deep-learning-28th-international-
conference-on-artificial-neural-networks-munich-germany-
september-17-19-2019-proceedings-part-ii-igor-v-tet/

PRICAI 2019 Trends in Artificial Intelligence 16th

Pacific Rim International Conference on Artificial
Intelligence Cuvu Yanuca Island Fiji August 26 30 2019
Proceedings Part II Abhaya C. Nayak
https://textbookfull.com/product/pricai-2019-trends-in-
artificial-intelligence-16th-pacific-rim-international-
conference-on-artificial-intelligence-cuvu-yanuca-island-fiji-
august-26-30-2019-proceedings-part-ii-abhaya-c-nayak/

Image and Graphics 10th International Conference ICIG

2019 Beijing China August 23 25 2019 Proceedings Part
III Yao Zhao

https://textbookfull.com/product/image-and-graphics-10th-
international-conference-icig-2019-beijing-china-
august-23-25-2019-proceedings-part-iii-yao-zhao/

Advanced Hybrid Information Processing Third EAI

International Conference ADHIP 2019 Nanjing China
September 21 22 2019 Proceedings Part I Guan Gui

https://textbookfull.com/product/advanced-hybrid-information-
processing-third-eai-international-conference-adhip-2019-nanjing-
china-september-21-22-2019-proceedings-part-i-guan-gui/
Shuai Han
Liang Ye
Weixiao Meng (Eds.)

287

Artificial Intelligence
for Communications
and Networks
First EAI International Conference, AICON 2019
Harbin, China, May 25–26, 2019
Proceedings, Part II

Part 2

123
Lecture Notes of the Institute
for Computer Sciences, Social Informatics
and Telecommunications Engineering 287

Editorial Board Members

Ozgur Akan
Middle East Technical University, Ankara, Turkey
Paolo Bellavista
University of Bologna, Bologna, Italy
Jiannong Cao
Hong Kong Polytechnic University, Hong Kong, China
Geoffrey Coulson
Lancaster University, Lancaster, UK
Falko Dressler
University of Erlangen, Erlangen, Germany
Domenico Ferrari
Università Cattolica Piacenza, Piacenza, Italy
Mario Gerla
UCLA, Los Angeles, USA
Hisashi Kobayashi
Princeton University, Princeton, USA
Sergio Palazzo
University of Catania, Catania, Italy
Sartaj Sahni
University of Florida, Gainesville, USA
Xuemin (Sherman) Shen
University of Waterloo, Waterloo, Canada
Mircea Stan
University of Virginia, Charlottesville, USA
Jia Xiaohua
City University of Hong Kong, Kowloon, Hong Kong
Albert Y. Zomaya
University of Sydney, Sydney, Australia
More information about this series at http://www.springer.com/series/8197
Shuai Han Liang Ye Weixiao Meng (Eds.)
• •

Artificial Intelligence
for Communications
and Networks
First EAI International Conference, AICON 2019
Harbin, China, May 25–26, 2019
Proceedings, Part II

123
Editors
Shuai Han Liang Ye
Harbin Institute of Technology Harbin Institute of Technology
Harbin, China Harbin, China
Weixiao Meng
Harbin Institute of Technology
Harbin, China

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

Lecture Notes of the Institute for Computer Sciences, Social Informatics
and Telecommunications Engineering
ISBN 978-3-030-22970-2 ISBN 978-3-030-22971-9 (eBook)
https://doi.org/10.1007/978-3-030-22971-9

© ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2019
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the
material is concerned, specifically 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 specific 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, expressed 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 affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
 Preface

We are delighted to introduce the proceedings of the first edition of the 2019 European
Alliance for Innovation (EAI) International Conference on Artificial Intelligence for
Communications and Networks (AICON). This conference brought together
researchers, developers, and practitioners from around the world who are leveraging
and developing artificial intelligence technology for communications and networks.
The theme of AICON 2019 was “Artificial Intelligence for Communications and
Networks: Applying Artificial Intelligence to Communications and Networks.”
The technical program of AICON 2019 consisted of 93 full papers, including six
invited papers in oral presentation sessions during the main conference tracks. The
conference tracks were: Track 1—AI-Based Medium Access Control;
Track 2—AI-based Network Intelligence for IoT; Track 3—AI-enabled Network Layer
Algorithms and Protocols; Track 4—Cloud and Big Data of AI-enabled Networks;
Track 5—Deep Learning/Machine Learning in Physical Layer and Signal Processing;
and Track 6—Security with Deep Learning for Communications and Networks. Aside
from the high-quality technical paper presentations, the technical program also featured
four keynote speeches and four invited talks. The four keynote speeches were by Prof.
Moe Win from the Laboratory for Information and Decision Systems, Massachusetts
Institute of Technology, USA, Prof. Mohsen Guizani from the Department of Computer
Science and Engineering, Qatar University, Qatar, Prof. Guoqiang Mao from the Center
for Real-Time Information Networks, University of Technology Sydney, Australia, and
Prof. Byonghyo Shim from the Department of Electrical and Computer Engineering,
Seoul National University, South Korea. The invited talks were presented by Prof.
Jinhong Yuan from the University of New South Wales, Australia, Prof. Shui Yu from
the University of Technology Sydney, Australia, Prof. Bo Rong from the Communi-
cations Research Centre, Canada, and Prof. Haixia Zhang from Shandong University,
China.
Coordination with the steering chair, Imrich Chlamtac, the general chairs, Xuemai
Gu and Cheng Li, and the executive chairs, Qing Guo and Hsiao-Hwa Chen, was
essential for the success of the conference. We sincerely appreciate their constant
support and guidance. It was also a great pleasure to work with such an excellent
Organizing Committee team and we thank them for their hard work in organizing and
supporting the conference. In particular, we thank the Technical Program Committee,
led by our TPC co-chairs, Prof. Shuai Han and Prof. Weixiao Meng, who completed
the peer-review process of technical papers and compiled a high-quality technical
program. We are also grateful to the conference manager, Andrea Piekova, for her
support and all the authors who submitted their papers to the AICON 2019 conference.
vi Preface

We strongly believe that the AICON conference provides a good forum for all
researchers, developers, and practitioners to discuss all scientific and technological
aspects that are relevant for artificial intelligence and communications. We also expect
that future AICON conferences will be as successful and stimulating as indicated by the
contributions presented in this volume.

May 2019 Shuai Han

Liang Ye
Weixiao Meng
 Organization

Steering Committee
Imrich Chlamtac University of Trento, Italy

Organizing Committee
General Chairs
Xuemai Gu Harbin Institute of Technology, China
Cheng Li Memorial University of Newfoundland, Canada

Executive Chairs
Qing Guo Harbin Institute of Technology, China
Hsiao-Hwa Chen National Cheng Kung University, Taiwan

Program Chairs
Shuai Han Harbin Institute of Technology, China
Weixiao Meng Harbin Institute of Technology, China

Sponsorship and Exhibits Chair

Rose Hu Utah State University, USA

Local Chair
Shuo Shi Harbin Institute of Technology, China

Workshop Chairs
Yahong Zheng Missouri University of Science and Technology, USA
Shaochuan Wu Harbin Institute of Technology, China

Publicity and Social Media Chairs

Zhensheng Zhang Retired
Lin Ma Harbin Institute of Technology, China

Publications Chair
Liang Ye Harbin Institute of Technology, China

Web Chairs
Xuejun Sha Harbin Institute of Technology, China
Chenguang He Harbin Institute of Technology, China
viii Organization

Posters and PhD Track Chair

Chau Yuen Singapore University of Technology and Design,
Singapore

Panels Chair
Xianbin Wang University of Western Ontario, Canada

Demos Chair
Yi Qian University of Nebraska Lincoln, USA

Tutorials Chair
Mugen Peng Beijing University of Posts and Telecommunications,
China

Conference Manager
Andrea Piekova EAI

Technical Program Committee

Deyue Zou Dalian University of Technology, China
Fan Jiang Memorial University of Newfoundland, Canada
Lingyang Song Peking University, China
Bo Rong Communications Research Center, Canada
Wei Li Northern Illinois University, USA
Xi Chen Flatiron Institute, Simons Foundation, USA
Baoxian Zhang University of China Academy of Sciences, China
Jalel Ben-Othman University of Paris 13, France
Jun Shi Harbin Institute of Technology, China
Ruiqin Zhao Northwestern Polytechnical University, China
Feng Ye University of Dayton, USA
Tao Jiang Huazhong University of Science and Technology,
China
Hossam Hassanein Queen’s University, Canada
Ruofei Ma Harbin Institute of Technology, China
Jun Zheng Southeast University, China
Wei Shi Carleton University, Canada
Kun Wang Nanjing University of Posts and Telecommunications,
China
Xianye Ben Shandong University, China
Shiwen Mao Auburn University, USA
Mianxiong Dong Muroran Institute of Technology, Japan
Wei Xiang James Cook University, Australia
Yulong Gao Harbin Institute of Technology, China
 Contents – Part II

Security with Deep Learning for Communications and Networks

VITEC: A Violence Detection Framework . . . . . . . . . . . . . . . . . . . . . . . . . 3

Hany Ferdinando, Tuija Huuki, Liang Ye, Tian Han, Zhu Zhang,
Guobing Sun, Tapio Seppänen, and Esko Alasaarela

Software Defect Prediction Model Based on Stacked Denoising

Auto-Encoder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Yu Zhu, Dongjin Yin, Yingtao Gan, Lanlan Rui, and Guoxin Xia

Deep&Cross Network for Software-Intensive System Fault Prediction . . . . . . 28

Guoxin Xia, Dongjin Yin, Yingtao Gan, Lanlan Rui, and Yu Zhu

Research on Evaluation Method of Cooperative Jamming Effect

in Cognitive Confrontation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Jing Ma, Bin Shi, Fei Che, and Sitong Zhang

Navigation Performance Comparison of ACE-BOC Signal

and TD-AltBOC Signal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Chunxia Li, Jianjun Fan, Min Li, and Yang Gao

A Video-Selection-Encryption Privacy Protection Scheme Based

on Machine Learning in Smart Home Environment . . . . . . . . . . . . . . . . . . . 65
Qingshui Xue, Haozhi Zhu, Xingzhong Ju, Haojin Zhu, Fengying Li,
Xiangwei Zheng, and Baochuan Zuo

A Multi-agent Reinforcement Learning Based Power Control Algorithm

for D2D Communication Underlaying Cellular Networks . . . . . . . . . . . . . . . 77
Wentai Chen and Jun Zheng

An Adaptive Window Time-Frequency Analysis Method Based

on Short-Time Fourier Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Zhiqiang Li, Xiao Wang, Ming Li, and Shuai Han

Research on the Security of Personal Information in the Era of Big Data . . . . 107
Cheng Chi, Tengyu Liu, Xiaochen Yu, Shuo Zhang, and Shuo Shi

Secure Access and Routing Scheme for Maritime Communication Network . . . 115
Yuzhou Fu, Chuanao Jiang, Yurong Qin, and Liuguo Yin

Multipath Based Privacy Protection Method for Data Transmission in SDN . . . 127
Na Dong, Zhigeng Han, and Liangmin Wang
x Contents – Part II

A Biometric-Based IoT Device Identity Authentication Scheme . . . . . . . . . . 139

Qingshui Xue, Xingzhong Ju, Haozhi Zhu, Haojin Zhu, Fengying Li,
and Xiangwei Zheng

A Physical and Verbal Bullying Detecting Algorithm Based on K-NN

for School Bullying Prevention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
Shangbin Gao and Liang Ye

Speech Bullying Vocabulary Recognition Algorithm in Artificial Intelligent

Child Protecting System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Tong Liu, Liang Ye, Tian Han, Yue Li, and Esko Alasaarela

Cross Station-Voltage-Level Route Planning Algorithm for System

Protection Services in UHV Grid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
Danyang Xu, Peng Yu, Wenjing Li, Xuesong Qiu, and Luoming Meng

An Efficient Implementation of Multi-interface SSD Controller on SoC . . . . . 181

Yifei Niu, Hai Cheng, Songyan Liu, Huan Liu, Xiaowen Wang,
and Yanlin Chen

Joint BP and RNN for Resilient GPS Timing Against Spoofing Attacks . . . . 190
Sriramya Bhamidipati, Kyeong Jin Kim, Hongbo Sun, Philip Orlik,
and Jinyun Zhang

Cloud and Big Data of AI-Enabled Networks

Regional Landslide Sensitivity Analysis Based on CPSO-LSSVM. . . . . . . . . 213

Yanze Li, Zhenjian Yang, Yunjie Zhang, and Zhou Jin

Multitasks Scheduling in Delay-Bounded Mobile Edge Computing . . . . . . . . 224

Longyu Zhou, Supeng Leng, Ning Yang, and Ke Zhang

Edge Computing Based Traffic Analysis System Using Broad Learning. . . . . 238
Xiting Peng, Kaoru Ota, and Mianxiong Dong

The Performance of DF Relaying System Based on Energy Harvesting

and Dual-Media Channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
Zhixiong Chen, Lijiao Wang, Cong Ye, and Dongsheng Han

School Violence Detection Based on Multi-sensor Fusion and Improved

Relief-F Algorithms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
Liang Ye, Jifu Shi, Hany Ferdinando, Tapio Seppänen,
and Esko Alasaarela

A UAV Based Multi-object Detection Scheme to Enhance Road

Condition Monitoring and Control for Future Smart Transportation . . . . . . . . 270
Jian Yang, Jielun Zhang, Feng Ye, and Xiaohui Cheng
 Contents – Part II xi

Sentiment Analysis for Tang Poetry Based on Imagery Aided

and Classifier Fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
Yabo Shen, Yong Ma, Chunguo Li, Shidang Li, Mingliang Gu,
Chaojin Zhang, Yun Jin, and Yingli Shen

Detection of Insulator Defects Based on YOLO V3. . . . . . . . . . . . . . . . . . . 291

Fei Guo, Kun Hao, Mengqi Xia, Lu Zhao, Li Wang, and Qi Liu

MCU-Based Isolated Appealing Words Detecting Method

with AI Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
Liang Ye, Yue Li, Wenjing Dong, Tapio Seppänen, and Esko Alasaarela

Performance Analysis Based on MGF Fading Approximation

in Hybrid Cooperative Communication . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
Cong Ye, Zhixiong Chen, Jinsha Yuan, and Lijiao Wang

Age and Gender Classification for Permission Control of Mobile Devices

in Tracking Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318
Merahi Choukri and Shaochuan Wu

Research on Fusion of Multiple Positioning Algorithms Based

on Visual Indoor Positioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
Guangchao Xu, Danyang Qin, Min Zhao, and Ruolin Guo

Social Aware Edge Caching in D2D Enabled Communication . . . . . . . . . . . 335

Jingyi Chen

AI-Based Network Intelligence for IoT

An Emergency Resource Scheduling Model Based on Edge Computing. . . . . 353

Songyan Zhong, Tao He, Mengyu Li, Lanlan Rui, Guoxin Xia,
and Yu Zhu

Mobility-Aware Task Parallel Offloading for Vehicle Fog Computing . . . . . . 367

Jindou Xie, Yunjian Jia, Zhengchuan Chen, and Liang Liang

A Floor Distinction Method Based on Recurrent Neural Network

in Cellular Network. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
Yongliang Zhang, Lin Ma, Danyang Qin, and Miao Yu

A Multi-classifier Approach for Fuzzy KNN Based

WIFI Indoor Localization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393
Yuanfeng Du and Dongkai Yang

Research on Application and Development of Key Technologies

of Dynamic Wireless Charging System in New Intelligent
Transportation System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402
Lina Ma, Shi An, and Wanlong Zhao
xii Contents – Part II

Binocular Vision-Based Human Ranging Algorithm Based on Human

Faces Recognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 414
Xiaolin He, Lin Ma, and Weixiao Meng

Deep Reinforcement Learning Based Task Offloading in SDN-Enabled

Industrial Internet of Things . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425
Jiadai Wang, Yurui Cao, Jiajia Liu, and Yanning Zhang

A Reinforcement Learning Based Task Offloading Scheme for Vehicular

Edge Computing Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438
Jie Zhang, Hongzhi Guo, and Jiajia Liu

Sensor-Based Human Activity Recognition for Smart Healthcare:

A Semi-supervised Machine Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
Abrar Zahin, Le Thanh Tan, and Rose Qingyang Hu

CoLoRSim: A Highly Scalable ICN Simulation Platform . . . . . . . . . . . . . . . 473

Hongyi Li and Hongbin Luo

Manifold Learning Based Super Resolution for Mixed-Resolution

Multi-view Video in Visual Internet of Things . . . . . . . . . . . . . . . . . . . . . . 486
Yuan Zhou, Ying Wang, Yeda Zhang, Xiaoting Du, Hui Liu, and Chuo Li

Posture Recognition and Heading Estimation Based on Machine Learning

Using MEMS Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 496
Boyuan Wang, Xuelin Liu, Baoguo Yu, Ruicai Jia, and Lu Huang

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 509

 Contents – Part I

Deep Learning/Machine Learning in Physical Layer

and Signal Processing

Dual-Mode OFDM-IM by Encoding All Possible Subcarrier

Activation Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Xiaoping Jin, Zheng Guo, Mengmeng Zhao, Ning Jin,
and Dongxiao Chen

Energy Efficiency Optimization Based SWIPT in OFDM

Relaying Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Dan Huang, Weidang Lu, and Mengshu Hou

An Efficient Resource Allocation Algorithm for LTE Uplink

VMIMO Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Yang Cai, Shaojun Qiu, Jia Cai, Wenchi Cheng, and Xiaofeng Lu

Research on UAV Swarm Interference Based on Improved Invasive

Weed Optimization Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Lijiao Wang, Yanping Liao, and Xiaoming Luan

Sparse Decomposition Algorithm Based on Joint Sparse Model . . . . . . . . . . 47

Qiyun Xuan, Si Wang, Yulong Gao, and Junhui Cheng

Millimeter Wave Massive MIMO Channel Estimation and Tracking . . . . . . . 57

Yanwu Song, Shaochuan Wu, Wenbin Zhang, and Huafeng Zhang

Deep Learning-Based Space Shift Keying Systems . . . . . . . . . . . . . . . . . . . 70

Yue Zhang, Xuesi Wang, Jintao Wang, Yonglin Xue, and Jian Song

Non-orthogonal Multiple Access Enabled Power Allocation

for Cooperative Jamming in Wireless Networks . . . . . . . . . . . . . . . . . . . . . 79
Yuan Wu, Weicong Wu, Daohang Wang, Kejie Ni, Li Ping Qian,
Weidang Lu, and Limin Meng

Joint Time-Frequency Diversity in the Context

of Spread-Spectrum Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Qiuhan Teng, Xuejun Sha, and Cong Ma

Design of Radar-Communication Integrated Signal Based on OFDM. . . . . . . 108

Tianqi Liang, Zhuoming Li, Mengqi Wang, and Xiaojie Fang
xiv Contents – Part I

A Multicast Beamforming Algorithm to Improve the Performance

of Group Service for Multicell B-TrunC System . . . . . . . . . . . . . . . . . . . . . 120
Zheming Zhang, Chengwen Zhang, Yutao Liu, and Bin Wang

Neural Networks in Hybrid Precoding for Millimeter Wave Massive

MIMO Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
Jing Yang, Kai Chen, Xiaohu Ge, Yonghui Li, and Lin Tian

A Reinforcement Learning Based Joint Spectrum Allocation

and Power Control Algorithm for D2D Communication Underlaying
Cellular Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
Wentai Chen and Jun Zheng

Improved Neural Machine Translation with POS-Tagging Through

Joint Decoding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
Xiaocheng Feng, Zhangyin Feng, Wanlong Zhao, Nan Zou, Bing Qin,
and Ting Liu

Distance Measurement Based on Linear Phase Correlation in WiFi CSI . . . . . 167

Qingfei Kang, Liangbo Xie, Mu Zhou, and Zengshan Tian

A Transfer Learning Method for Aircrafts Recognition . . . . . . . . . . . . . . . . 175

Hongbo Li, Bin Guo, Tong Gao, and Hao Chen

A New Two-Microphone Reduce Size SMFTF Algorithm for Speech

Enhancement in New Telecommunication Systems . . . . . . . . . . . . . . . . . . . 186
Zineddine Guernaz and Xuanli Wu

Adaptive Beamforming of Vertical Frequency Diverse Array

for Airborne Radar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
Xuzi Wu and Yongliang Sun

2D DOA Estimation of PR-WSF Algorithm Based on Modified

Fireworks Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
Yanping Liao, Chang Fu, and Emmanuel Milambo Mung’onya

Research on Indoor and Outdoor Seamless Positioning Based

on Combination of Clustering and GPS . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Jingqiu Ren, Ke Bao, Siyue Sun, and Weidang Lu

Angle-of-Arrival Positioning System Based on CSI Virtual

Antenna Array . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
Lu Yin, Ziyang Wang, Zhongliang Deng, Tianrun Jiang, and Yuan Sun

Cross-Sensor Image Change Detection Based on Deep Canonically

Correlated Autoencoders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
Yuan Zhou, Hui Liu, Dan Li, Hai Cao, Jing Yang, and Zizi Li
 Contents – Part I xv

An Innovative Weighted KNN Indoor Location Technology. . . . . . . . . . . . . 258

Lu Huang, Xingli Gan, Dan Du, Boyuan Wang, and Shuang Li

A Resistance Frequency Offset Synchronization Scheme Based

on the Zadoff-Chu Conjugate Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
Cong Ma, Xuejun Sha, Yong Li, and Xu Lin

AI-Based Medium Access Control

Compressed Sensing ISAR 3D Imaging Methods Based

on OMP Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
Jingcheng Zhao, Zongkai Yang, and Shaozhu Gu

Ambiguity Function Analysis of Radar-Communication Integrated

Waveform Based on FDM and TDM Technologies . . . . . . . . . . . . . . . . . . . 293
Hongzhi Men, Zhiqun Song, and Guisheng Liao

Transmission Quality Improvement Algorithms for Multicast

Terrestrial-Satellite Cooperation System . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
Yuandong Zhang and Liuguo Yin

Application of Wavelet Analysis Method in Radar Echo

Signal Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
Qiuyue Li and Xiangyu Tong

Chinese News Keyword Extraction Algorithm Based on TextRank

and Topic Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334
Ao Xiong and Qing Guo

An Adaptive Threshold Decision Algorithm in Non-cooperative

Signal Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
Ziheng Li, Shuo Shi, and Xuemai Gu

Trajectory Optimization Under Constrained UAV-Aided Wireless

Communications with Ground Terminals . . . . . . . . . . . . . . . . . . . . . . . . . . 350
Kun Chen, Hong Lu, Xiangping Bryce Zhai, Congduan Li,
Yunlong Zhao, and Bing Chen

IOT-Based Thermal Comfort Control for Livable Environment . . . . . . . . . . . 363

Miao Zang, Zhiqiang Xing, and Yingqi Tan

Context Adaptive Visual Tracker in Surveillance Networks . . . . . . . . . . . . . 374

Wei Feng, Minye Li, Yuan Zhou, Zizi Li, and Chenghao Li

Research on Indoor Localization Based on Joint Coefficient APIT . . . . . . . . 383

Min Zhao, Danyang Qin, Ruolin Guo, and Lin Ma
xvi Contents – Part I

A Cross-Layer Approach to Maximize the Lifetime of Underwater

Wireless Sensor Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391
Yuan Zhou, Hui Liu, Hai Cao, Dan Li, Hongyu Yang, and Tao Cao

An Efficient Indoor Localization Method Based on Visual Vocabulary . . . . . 398

Ruolin Guo, Danyang Qin, Min Zhao, and Guangchao Xu

Realization and Performance Simulation of Spectrum Detection Based

on Cyclostationarity Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407
Zhiqun Song, Yujing Lv, and Zhongzhao Zhang

AI-Enabled Network Layer Algorithms and Protocols

An Improved TDoA Localization Algorithm Based on AUV

for Underwater Acoustic Sensor Networks . . . . . . . . . . . . . . . . . . . . . . . . . 419
Kaicheng Yu, Kun Hao, Cheng Li, Xiujuan Du, Beibei Wang,
and Yonglei Liu

Fuzzy Probabilistic Topology Control Algorithm for Underwater

Wireless Sensor Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
Wenhao Ren, Kun Hao, Cheng Li, Xiujuan Du, Yonglei Liu,
and Li Wang

Naive Bayes Classifier Based Driving Habit Prediction Scheme

for VANET Stable Clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445
Tong Liu, Shuo Shi, and Xuemai Gu

Path Optimization with Machine-Learning Based Prediction

for Wireless Sensor Networks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453
Jianxin Ma, Shuo Shi, and Xuemai Gu

Research on the Maximization of Total Information Rate Based

on Energy Allocation in Multi-user SWIPT Relaying System . . . . . . . . . . . . 460
Jianxiong Li, Xuelong Ding, Xianguo Li, Kunlai Li, Ke Zhao,
and Weiguang Shi

Optimization of AODV Routing Protocol in UAV Ad Hoc Network . . . . . . . 472

Jianze Wu, Shuo Shi, Zhongyue Liu, and Xuemai Gu

A KFL-TDOA UWB Positioning Method Based on Hybrid

Location Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479
Shuo Shi, Meng Wang, and Kunqi Hong

Simultaneous Wireless Information and Power Transfer Protocol

Under the Presence of Node Hardware Impairments . . . . . . . . . . . . . . . . . . 488
Yanlin Liu, Juan Li, Fengye Hu, and Qiao Qiao
 Contents – Part I xvii

An Improved A* Algorithm Based on Divide-and-Conquer Method

for Golf Unmanned Cart Path Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . 497
Yi Chen, Liangbo Xie, Wei He, Qing Jiang, and Junxing Xu

Power Allocation Method for Satellite Communication Based

on Network Coding. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 506
Jin Liu, Zhuoming Li, and Gongliang Liu

Seamless Positioning and Navigation System Based on GNSS,

WIFI and PDR for Mobile Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 522
Yuanfeng Du and Dongkai Yang

Adaptive Routing Protocol for Underwater Wireless Sensor

Network Based on AUV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 541
Yuying Ding, Cheng Li, Kun Hao, Xiujuan Du, Lu Zhao, and Qi Liu

Performance Analysis of Energy-Efficient Cell Switch

off Scheme for CoMP Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 554
Fei Ding, Yan Lu, Jialu Li, Ruoyu Su, Dengyin Zhang, and Hongbo Zhu

An Energy-Efficient Routing Protocol for Internet

of Underwater Things . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 563
Ruoyu Su, Fei Ding, Dengyin Zhang, Hongbo Zhu, and Xiaohong Wang

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 571

Security with Deep Learning for
Communications and Networks
 VITEC: A Violence Detection Framework

Hany Ferdinando1,2(&) , Tuija Huuki1 , Liang Ye1,3 ,

Tian Han1,4, Zhu Zhang1,4, Guobing Sun1,5, Tapio Seppänen1 ,
and Esko Alasaarela1
1
University of Oulu, Oulu, Finland
hany.ferdinando@oulu.fi
2
Petra Christian University, Surabaya, Indonesia
3
Harbin Institute of Technology, Harbin, China
4
Harbin University of Science and Technology, Harbin, China
5
Heilongjiang University, Harbin, China

Abstract. Hundreds of millions of youths suffer from various violence each

year. The negative impacts motivate much research and numerous studies on
violence. However, those attempts went their own way, making the achieved
results, especially from engineering, not so useful. Based on the Sensor and
Social Web (SEWEB) concept, Violence Detection (VITEC) was proposed as a
possible framework to facilitate multi-disciplinary researchers in their fight
against violence. At its core, it consists of a primary agent, which is violence
detection using physiological signals and activity recognition, and a secondary
agent, which is violence detection using surveillance video. The second layer of
the proposed framework contains a cloud computing service with a Personal
Safety Network (PSN) database. The cloud computing service manages all data,
notifications, and some more thorough processing. The upper layer is for both
observed young persons and members of the PSN. The proposed framework
offers business opportunities. The existing school violence/bullying intervention
programs can take advantage of VITEC by providing almost instant notifications
of violent events, enabling the victims to get immediate help and intensifying
coordination among different sectors to fight against violence. In the long run,
VITEC may provide an answer related to the vision of having a world free from
violence in 2030, as addressed by the UN Special Representative of Secretary-
General on violence against children.

Keywords: Violent detection
Bullying detection
VITEC

Surveillance video
Physiological signals
Activity recognition

1 Introduction

Our societies change dramatically and drastically due to advancement in technologies.

Information and news spread easily and quickly to enable fast learning and engagement.
There are many online tutorials we can find not only in text form but also in
audio/video/multimedia format. We can know what happened on the other side of the
world almost instantly. Intercontinental video calls are now very affordable, with better
quality. However, the negative side is unavoidable. Abuses of power emerge in new and

© ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2019
Published by Springer Nature Switzerland AG 2019. All Rights Reserved
S. Han et al. (Eds.): AICON 2019, LNICST 287, pp. 3–17, 2019.
https://doi.org/10.1007/978-3-030-22971-9_1
4 H. Ferdinando et al.

unexpected ways, such as harassment in social media and cyberbullying. Consequently,

feelings of insecurity may arise not only among adults but also among young people.
Indeed, hundreds of millions of youths suffer from different kinds of violence in
educational environments [1–3] and on the streets [4] every year all over the globe.
Addressing problems of violence in the lives of children has never been more relevant
than in 2018. Highly valued research reviews inform us that the consequences of
violence are diverse, immediate, and long-lasting. Young persons who experience
violence are likely to have stomach pains, headaches, and difficulties eating and
sleeping; they may become afraid of attending school, which may in turn interfere with
their ability to concentrate in class or participate in school activities. Furthermore, a
violent experience can cause depression, loneliness, low self-esteem, suicidal thoughts,
interpersonal difficulties, antisocial and criminal behavior, and attempted suicide [5, 6].
Research has also shown how violence slows social progress by generating huge
economic costs, hindering sustainable development, and eroding human capital.
A study found that the global costs of violence against children could be as high as US
$7 trillion per year [7]. All forms of violence infringe on the fundamental right to a safe
life, create unsafe learning environments, and reduce the quality of lives for all children
and youths.
Concerning these impacts, United nation children’s fund (UNICEF) launched
#ENDviolence in 2013 to bring them to an end. The UN Special Representative of the
Secretary-General on violence against children (SRSG-VAC), Marta Santos Pais,
emphasized the vision of building a world free from fear and violence as the result of
adopting the 2030 Agenda for Sustainable Development in September 2015 [8].
Quoted from Article 19 [9], “the child must be protected from all forms of physical or
mental violence, injury or abuse, neglect or negligent treatment, maltreatment, or
exploitation, including sexual abuse, while in the care of parent(s), legal guardian(s)
or any other person who has the care of the child”.
However, the progress on protecting children from violence around the world has
been slow, uneven, and fragmented [10]. Furthermore, most of the promising inter-
vention models rely on education systems, schools, and teachers, which often have
limited capacities and resources to work efficiently enough to address violence.
Coordination between the education sector and other sectors such as health, social
service, and child protection, might not work efficiently in practice, making the action
too slow. The reporting system must be anonymous and confidential to protect both
victims and witnesses.
Recently, it has been discussed whether or not technological solutions might assist
in tackling violence in young persons’ lives. Technology has evolved to a level that
would have been impossible to achieve a couple of decades ago. Recently, semicon-
ductor technology, including sensor development has enabled systems to become
smaller and more powerful than before. A smartphone, its apps and the internet have
become part of contemporary life, which promote mobility for the users.
A smartphone can be considered a small computer. It is equipped with a processor,
memory, and operating system. Moreover, a smartphone is also armed with several
sensors, the most common one being the 3D acceleration sensor. It delivers information
about the position of the smartphone, for example detecting if a smartphone is in a
 VITEC: A Violence Detection Framework 5

landscape or portrait position, so that the display can be adjusted. It also enables the
possibility for developers to write apps, providing users with various applications that
can be downloaded.
In terms of protecting young persons against violence, apps such as STOP!t, Speak
UP!, BullyButton, and BullyTag have been developed to help victims and witnesses
report violent events anonymously. They provide a safe way to send audio, video, and
text about an acute violent situation to a corresponding database server. Investigation of
the database report may lead to new insights about violence. Violence, however, is an
iceberg phenomenon, where only limited cases are reported, due to, for example,
feeling of ashamed or fear of revenge from the perpetrators.
However, those apps, for example Stop!t got about three out of five stars based on
the users’ rating [11]. Collecting from various apps ratings, a few people reported that
they were helpful to report the violent event with no fear and got assistance from the
school several days later [11] and one person put his hope in anti-bullying apps [12].
There was also a comment that someone might misuse the apps to make a joke or false
claim [13]. In general, all these apps require a manual activation to generate reports. In
a threatening situation, most likely the victim has no time, courage, or possibility to tap
a button. Also, the bystanders may face similar barriers to reporting, which makes these
apps less useful. The apps for reporting violence must not only be anonymous, but also
be able to autonomously detect the violent event and automatically perform the
required action.
In addition to the apps-based reporting systems, there have been several attempts to
detect violent situations using various modalities such as video, audio, and physio-
logical signals. Although they have offered some promising results, no significant
impact to reduce violence rate has been delivered, because each modality has worked
on its own with no specific framework on which the information and achievements
from various resources would be gathered and shared easily. In an attempt to connect
different modalities and a workable reporting system, in this paper we propose VITEC,
a violence detection framework based on the Sensor and Social Web (SEWEB) concept
[14], as a contribution from various disciplines to address violence among young
people. To the best of our knowledge, there is not yet available such an inter-
disciplinary designed, research-based framework to address physical violence targeted
at young people.
SEWEB has been offered to protect children and youth from various kinds of
dangerous situations, including health issues and violence [14]. It can be extended to
not only school-related violence [3] but also street violence [4] and rape [15]. It is about
connecting wearable and mobile sensors to social media in human well-being appli-
cations. It involves technologies such as smartphone apps, cloud computing services,
social media, wearable safety devices to process the activity, location, voice, health-
related and well-being information of the subjects. The system identifies and catego-
rizes phenomena such as stress level, emotional and mood states, physical activity level
in various modalities, and location. The results are reported to some pre-defined
contacts, for example parents, school staffs, and authorities.
6 H. Ferdinando et al.

2 Research in Violence Detection

In general, two approaches or sets of system exist in violence detection research.

Firstly, the systems work as a witness to violent events. Included in this approach are
violence detection using surveillance video signals [16–18] and physiological signals
measured from the subjects who were watching the violent scenes, e.g. EDA [19], EEG
[20]. Secondly, the systems use signals measured from the victims directly, for
example, their human body movement [21] and ECG signal [22].
Nam, Alghoniemy, and Tewfik pioneered a study on violent scene characterization
based on flame and blood detection [16]. It opened a research pathway on violence
detection using video. In general, studies using video to detect violence can be sepa-
rated into two groups based on the source of video signals, either from movies or
surveillance cameras. The main differences reside in the availability of sound and
colour, because videos from surveillance cameras have no sound and are in a grey scale
format. Video from surveillance camera usually has low resolution. The VITEC
framework takes more advantages from the studies using surveillance videos. Table 1
provides the most promising attempts to detect violence using surveillance video.
Various methods have been proposed with encouraging results and contributions.

Table 1. Attempts in violence detection using surveillance-like video

Authors Results/Contributions Note
Souza et al. Accuracies were up to 99%, depending It used the collected grey scale videos
[25] on the methods from various sources without audio
Nievas et al.Accuracies were up to 92%, depending It ignored color and audio to detect
[26] on the methods either fight or non-fight scene
Bilinski Accuracies were up to 99%, depending Videos were from various data sets:
et al. [27] on the methods and data sets. The Violent Flow, Hockey Fight, and
proposed method offered a fast- Movie
computational algorithm
Zhang et al. Accuracies up to 87%, depending on the Surveillance videos were from
[28] data sets. The proposed method offered various data sets: BEHAVE, the
a moderately fast computational CAVIAR, and Crowded Violence
algorithm
Ribeiro Accuracies were around 90% using Videos were from both simulation
et al. [17] various databases. The system can be and real surveillance
used in various contexts

The relationship between electrodermal activities (EDA), which is represented as

either skin conductance or resistance, and motion picture violence has been studied as a
pilot several decades ago [23]. A correlation was found between EDA and reaction to
violent scenes. Lorber conducted a meta-analysis of 95 studies to investigate the
relationship of EDA and heart rate (HR) – the number of heart beats in a minute – with
aggression, psychopathy, and conduct problems [24]. It was suggested that both HR
and EDA were reliably though modestly associated to each other in many cases.
 VITEC: A Violence Detection Framework 7

Furthermore, violent scene annotation using EDA signal was proposed based on the
hypothesis that high-level information in EDA may reveal the affective state of the
audiences, especially their reactions to violence [19]. This study found that EDA offers
objective, reliable, and robust measurements of user reaction to help in violent scenes
annotation. However, since the signal is sensitive to any affective stimulation, e.g.
pleasure, it was recommended to be used with other modalities.
Violent events could also be detected using brain signals (EEG) measured from
subjects watching violent scenes [20]. It was a breakthrough to address the problem of
violence, as compared to the usual questionnaires, interviews, and psychological tests.
It achieved an accuracy up to 98.7% and strengthened the idea of using physiological
signals to detect violent events.
Another study found that the highest average HR and EDA changes occurred in
subjects who played a violent video game, as compared to watching others playing a
violent video game or watching a violent movie [29]. Although EEG was not included
in this study, perhaps this type of signal also provides the same response. This work
empirically revealed that two approaches to attack violence were available, one from
the witness’ and the other from the victim’s point of view. Research on violence
detection using surveillance video belongs to the first approach, as well as those using
physiological signals measured from the subjects watching violent scenes.
Attempting to detect violence from the victim’s point of view, Ye et al. used 3D
acceleration and 3D gyroscope signals measured from the subject involved in violence
simulations to detect violence based on their body movements [21, 30]. The study
relied on the main idea that during a violent situation the victim’s body movement are
different than during anon-violent situation. The proposed system successfully distin-
guished some normal daily life body movements from the ones related to violence,
achieving an average accuracy of 92%.
Using single channel ECG (electrical activity of the heart) signals measured in
pupils involved in violence simulation, violent events could also be identified [22].
This work also emphasized on the violence detection from the victim’s point of view. It
required a wearable sensor able to measure ECG signals. Accuracies to classify violent
and non-violent events using 6-s ECG signal were up to 87%.

3 VITEC Framework

Based on the SEWEB concept, an online multi-modal VIolence deTECtion (VITEC)

framework is proposed using several components working together to provide reliable
performance: (1) violence detection system based on the witnesses’ and victims’ point
of view, (2) geographical location acquired from the GPS of a smartphone, (3) indoor
location acquired from an RFID-based system, (4) reliable communication channel,
and (5) processing unit: both local and cloud computing services.
Table 2 compares the original SEWEB concept to VITEC, providing general
overview of the proposed framework in relation to the SEWEB concept. Embedded
sensors used in the proposed framework are a 3D accelerometer and 3D gyroscope [21]
to measure body movement or activity. This choice is a bit problematic, because not all
smartphones have a gyroscope sensor. So, research on body movement recognition
8 H. Ferdinando et al.

using an accelerometer only is recommended. Ambient information is not relevant, but

the physiological signals of the subject measured using a wireless wearable sensor are
needed. The SEWEB used no CCTV, but VITEC takes advantage of research on
violence detection using surveillance video signals. Both SEWEB and VITEC need
location data from either the GPS or RFID-based system.

Table 2. Comparing the SEWEB’s and VITEC’s component side-by-side

SEWEB VITEC
Technologies used to identify and measure relevant signals
Embedded sensors of smartphones ✓ ✓
Microphone of smartphones ✓
Cameras of smartphones ✓
Processing capacity of smartphones ✓ ✓
Wearable sensors ✓ ✓
Ambient sensors ✓
CCTV ✓
Data measured and collected by sensors
Activity data ✓ ✓
Location data ✓ ✓
Voice data ✓
Health-related data ✓ ✓*
Well-being data ✓
Video data ✓
Technologies used to collect, process, and share/display data
Smartphone apps ✓ ✓
Cloud computing service ✓ ✓
Social media solutions ✓ ✓
Wearable safety devices ✓ ✓
Phenomena to identify, measure, and categorize
Change in stress level ✓
Emotional state changes ✓
Fast and slow changes in mood ✓
Changes in physical activity level and modality ✓ ✓
Detailed location information ✓ ✓
* Physiological signals such as ECG, EDA, EEG, temperature, etc.

Although we cannot avoid verbal bullying, voice data is not relevant to VITEC
because recording voice outside an isolated area is problematic. When someone is
shouting the voices are recorded through all surrounding microphones, and the trace-
ability of the voice data to a specific subject is questionable. VITEC does not use the
smartphone camera, because users have to direct the camera manually. It is unrealistic
to assume that the victim would have the possibility to direct the camera towards the
perpetrator(s) or himself/herself.
 VITEC: A Violence Detection Framework 9

VITEC does not identify stress level and emotional/mood changes. Psychologists
argue about what kind of emotions/moods emerge during violent events, but to the best
of our knowledge no empirical study has yet been found, which map those changes to
violent events. Perhaps in the future these modalities can be added.
The four main components that make up the VITEC framework are a primary and
secondary detection system, a Personal Safety Network, and a cloud computing ser-
vice. The primary detection system is a collection of numerous primary agents, which
work on the young people and detect violence based on the victim’s point of view. The
secondary detection system contains several of CCTV cameras connected to local
video processing that detect a violent event based on the witness’ point of view. They
serve as secondary agents. Figure 1 shows the connectivity diagram of the VITEC
framework.
As mentioned above, pupil welfare systems and other child protection services have
limited capacities to work efficiently to address violence. The primary and secondary
systems together with the cloud service are able to track and report violence auto-
matically, delivering an additional resource for child safeguarding and combating
violence. The VITEC framework also provides an automatic reporting feature. Both
primary and secondary agents initiate the report by sending a message about suspicious
events to the cloud service. When the cloud service confirms that a violent act, that
event is recorded in the database together with the ID of all agents, the geographical
location, date, and time. Thus, the number of unreported violent events can be reduced.
Furthermore, the database in the cloud service can be used to study phenomena related
to violence, which can benefit schools, other child welfare services, policy makers, and
parents. It also enables better coordination between education and other child welfare
sectors, such as health and social services and child protection. The primary and
secondary systems depend upon various research activities, and so does the cloud
service. As a result, the VITEC framework arranges such spaces for researchers from
various fields to work together to fight against violence. It also encourages more
collaborative work to deliver more valuable results.

3.1 Primary and Secondary Agents

Each primary agent consists of a wireless wearable multi-modal sensor that measures
signals from the body of a young person and regularly sends appropriate signals to their
smartphone for further processing. Each young person in VITEC is equipped with a
primary agent. The smartphone employs a pre-detection algorithm, which must be
simple but powerful enough to run within the limited smartphone environment and
resources, to identify whether or not a suspicious event has occurred. Once a suspicious
event is identified, the primary agent sends signals carrying that suspicious event along
with the primary agent’s ID and geographical location acquired from GPS to the cloud
for further processing. The contribution of detecting violence using human body
movements [21, 30] and physiological signal [22] fit for this role.
The secondary system contains several CCTV cameras connected to local video
processing. CCTV cameras are security cameras that monitor the area of interest, for
example a school yard, parking area, or street corner. Each CCTV sends recorded
surveillance video with its ID to a local computer server for further processing.
10 H. Ferdinando et al.

This local server sends the detection result to the cloud. Results presented in Table 1
are promising for this role, because installing CCTV cameras at every geographical
spot is impossible. According to the SEWEB concept, the secondary system is an
additional system.

Fig. 1. VITEC, an online multimodal framework, consists of several primary agents (a smart-
phone and wearable sensors) and a number of secondary agents (CCTV connected to local video
processing), and specific members of the Personal Safety Network connected to a cloud
computing service.

Consequently, in the cloud service there are three possible cases based on the
availability of the primary and secondary agents at a certain location. Upon receiving
data from the primary agents, the cloud service must clarify if a violent event has
occurred. It starts identifying the violent event using the signals from the primary
agents. If a violent event is confirmed, the cloud service seeks for some notifications
from the secondary agents in the area based on the location sent by the primary agents.
The cloud services then decide which notifications should be fused to the result from
the primary agents. This is the case when both agents are available. When there is no
secondary agent available in the area of interest, the end result depends on the primary
agents only. Another case is when there is no notification from the primary agents, but
the secondary agents report a suspicious event at a certain location. The reason for this
might be that the victim is one of the primary agents, but the battery power is low or
 VITEC: A Violence Detection Framework 11

that victim is not in the system yet. In this case the cloud service sends notification
of the event to local authorities, for example the nearest police station, so the victim
gets help immediately.
The primary agent solves problem about manual activation in the existing anti
bullying apps. With automatic activation, the victims do not have to remember to press
certain button when they experience violence. Furthermore, the system also sends
notification to the authority on the victims’ behalf. Some victims decide to report the
violence events later, but most of them cancel or forget it.

3.2 Personal Safety Network

The Personal Safety Network is defined as the network connections a user wants to be
in touch with in case of emergency or a socially challenging situation [14]. In other
words, to whom should the notification be sent, depending on the context, e.g. location,
type of challenging situation, time, type of help, etc. Those determine the members of
the Personal Safety Network, so it is dynamically changing. For example, when a
young person experiences violence on the street or on the way home from school,
parents and the nearest authority can be notified. So, VITEC includes parents in all
cases by default and dynamically changes other members of the Personal Safety
Network based on the context.

3.3 Notifications
Notifications are sensor-initiated and user-initiated [14]. Within the VITEC framework,
the sensor-initiated notification is a notification generated by the cloud services,
because it is the end result after processing signals from sensors. The cloud sends a
notification about a suspicious event to the members of the Personal Safety Network
retrieved from the database as an alarm. The corresponding young person also receives
a confirmation notification, indicating that the cloud has acknowledge the event. When
a young person thinks that a confirmation notification should be received but he/she
receives none, a user-initiated notification can be sent to the cloud manually, similar to
the existing smartphone apps related to violence. Perhaps some situations cannot be
measured by sensors, so that no notification is generated by the cloud services. This
kind of situation is defined as a challenging situation. VITEC also sends notification
about important messages, e.g. low battery power. Notification can be in a text message
and/or as a status displayed in social media.

3.4 Context and Sensor Data

According to SEWEB, sensor data must be interpreted based on the context [14]. For
example, in violence detection a primary agent interprets a violent event on the basis of
pushing or tackling, also in cases of equally rough and tumble play. To decrease
these kinds of false alarms, a multi-modal approach – including for example EDA and
ECG – is needed to aid in violence detection. Both physiological signals are sensitive
enough to affect stimulation, such that it may indicate that the subject is not in danger
but having fun. Ferdinando et al. (2017) showed that it is possible, although not always,
12 H. Ferdinando et al.

to separate some fun activities denoted as non-violence, e.g. playing, from simulated
violence, e.g. pushing and hitting. It will require deeper investigation to get a more
accurate algorithm to extract relevant information from the raw signals.
Figure 2 displays the position of each component in the VITEC framework in a
layer diagram. This is a general layer diagram, as detailed layers on communication
channels between layers are not provided. However, these follow a general data
communication layer, consisting of certain standard internet protocols between clients
and servers. The core of VITEC, as shown in Fig. 2, contains the primary and
secondary agents, which detect violent events from the victims’ and witnesses’ points
of view, respectively. Each agent only sends a message to the cloud when a suspicious
event occurs.

Fig. 2. Layer diagram of the VITEC framework presenting the position of each component and
how each component interacts with the others.

The cloud computing service analyses the signals sent by primary agents more
thoroughly and then combines with information from secondary agents to decide
whether a violent event is occurred or not. If the violent event happens, the cloud
service send notification to contact retrieved from database based on the context of the
young person. A double arrow between the young person and the cloud represents a
confirmation notification sent by the cloud to the young person and a user-initiated
notification from the young person to the cloud, described above as a challenging
situation.
Another random document with
no related content on Scribd:
My dear Father:
I have to thank you for three letters, for two poems, for the proof
sheets of your book and for your beautiful Christmas gift, which I
treasure and have with me always. The poem “To Rose in Heaven” is
so exquisite that I cannot write or speak all my deep appreciation of it.
But I know that it is not my personal feeling alone that makes me
consider it one of the noblest elegiac poems in our language. I wish
that I were still in the book-reviewing game—I’d like to express my
opinion of your new book. My column in the Literary Digest is now
conducted by Mrs. Edwin Markham. However, I occasionally send in
some clippings and comment, and I’ll try to get some of your book
quoted. I congratulate you on your book; it is an achievement for which
you must be deeply grateful; of course none of us rhymers can really
be proud if we’re true to the traditions of our craft. You see, I’m not in
“The Proud Poet” mood now. I’m neither proud nor a poet—I’ve written
only one poem since I sailed, a little thing called “Militis Meditatio.”
We all are well and happy—I do not think that any one of us can
honestly say that he has not enjoyed this Winter. The trials and
hardships of the life are so novel as to be interesting. But I imagine
that one enjoys very little his tenth year of soldiering. It is especially
pleasant to be in France and the part of France that we are in. I am
surprised, I acknowledge, by the passionate Catholicity of the people.
Even “Holy Ireland” can scarcely be more Catholic than rural France.
Tell my friends my address and urge them to write to me. Pray for
me, and believe me always
Yours sincerely,
Joyce Kilmer.

* * * * *

Headquarters Company, 165th Infantry,

A. E. F., France.
May 6, 1918.
My dear Father:
 It is very comfortable to dwell in so genuinely Catholic a land as
this; to be reminded in every room of every house, and at every cross-
road of the Faith. I do not know when you last visited France, and
perhaps you are already familiar with conditions here, but I do not
cease to be surprised and delighted to see the number of people
receiving Holy Communion not only on Sunday, but every day. And you
will find soldiers of this Regiment in church whenever and wherever
there is a priest to say Mass. I think that most of us are better
Catholics now than when we were at home—certainly we should be.
My own work is growing steadily more interesting. For a while I
worked, as you know, in the Adjutant’s Office, having special charge of
recording and reporting statistics. But this I gave up recently to enter
(as a Sergeant) the Regimental Intelligence Section. This is much
better for me, since it is open-air work of an intensely interesting kind,
and I think that I am now more useful to the Regiment. My newspaper
training, you see, has made me a competent observer, and that is what
I am nowadays—an observer of the enemy’s activities. I have already
some strange stories to tell, but for telling them I must await a time
when censorship rules are abrogated. I have written very little—two
prose sketches and two poems since I left the States—but I have a
rich store of memories. Not that what I write matters—I have
discovered, since some unforgettable experiences, that writing is not
the tremendously important thing I once considered it. You will find me
less a bookman when you next see me, and more, I hope, a man.
Pray for me, my dear Father, that I may love God more and that I
may be unceasingly conscious of Him—that is the greatest desire I
have.
Your affectionate friend,
Joyce Kilmer.

* * * * *

To Reverend James J. Daly, S.J.

 Cragmere, Mahwah, New Jersey.
Oct. 7, 1912.
My dear Father Daly:
I hope that I interpret your letter of September 20th correctly in
thinking that it means that you will permit me to write to you
occasionally, and that you will write me in return. Of course I know that
your work occupies much of your time, and that you have pleasanter
uses for your leisure than writing letters to some one you have not
seen. So do not feel obliged by courtesy to answer me at length.
I am glad that you like some of my work. My present occupation is
that of assistant editor of The Churchman, an Anglican weekly paper. It
is a Church Newspaper, with some literary features. Did Grey tell you
that I was not of your Communion? I hope that this knowledge will not
shut me out of your regard.
Your remarks, in your letter, on the fact that many of our most
famous writers to-day are anti-Christian, are certainly justified. Still, do
you not think that a reaction is coming? Already we have Chesterton,
and Belloc, and Bazin, and Miss Guiney, and Fr. Vincent McNabb, and
a number of other brilliant writers who not as theologians but purely as
literary artists express a fine and wholesome faith. People are
beginning to tire of cheap eroticism and “realism” and similar
absurdities.
Yours sincerely,
Joyce Kilmer.
 JOYCE KILMER, AGE 5
FROM A PHOTOGRAPH
IN THE POSSESSION OF
MRS. KILBURN KILMER
Dec. 14, 1912.
My dear Father Daly:
The cards are rather laconic, but they are supposed to convey the
information, as you thought, that my family is the larger by one. I have
already a little boy, who will be four next March. His name is Kenton
Sinclair. I married when I was twenty-one, immediately after my
graduation from Columbia. Then—since you have twice asked for
biographical information—I taught school in Morristown for a year,
came to New York, went to work on the Standard Dictionary, and after
three years took my present position on The Churchman.
Do you know the Rev. Michael Earls? I recently received a copy of
his book of verse. He is a member of your Society—I think he is at
Georgetown. The poems about children are very delicate and
sympathetic—one called, I believe, “An Autumn Rose-bush” is
admirable.
I see Grey frequently—I hope to take breakfast with him to-morrow.
A number of us have the custom of breakfasting on Sunday mornings
on the excellent Irish bacon to be procured at Healy’s, on 91st St.
When you return to New York, you must join us. Do you know Thomas
Walsh? He is with us occasionally—you should know him, and his
translations from the Spanish, and his verse. I hope you are coming
East soon.
Yours sincerely,
Joyce Kilmer.

* * * * *

Cragmere, Mahwah, N. J.
July 24, 1913.
Dear Father Daly:
I do not like to burden my friends with my troubles, but you have
certain opportunities that I lack, so I am asking you the greatest favour.
Please pray for the healing of my little daughter Rose. She is
dangerously ill with infantile paralysis. This is a disease that has
appeared among mankind only recently, and physicians are uncertain
how to treat it. She is staying in New York with her mother to be near
the doctor and I am staying here nights to take care of my other child.
Of course the maid is here during the day so the house is kept up. But
Rose cannot move her legs or arms—She was so active and happy
only last week—she cannot even cry—her voice is just a little whimper
—the danger is of its reaching her lungs and killing her. I cannot write
any more. You know how I feel. Pray for her.
Joyce Kilmer.

* * * * *

Aug. 30, 1913.

Dear Father Daly:
There is not, it seems, any danger of the death of Rose now. And
she has regained her strength. Her arms and legs have been left
paralysed but her hands and feet move. It will take a year’s time for us
to know whether or not this paralysis is permanent.
I am deeply grateful to you for your compliance with my request. I
know that your prayers were of value in keeping her alive. Apologies
for the brusqueness of my letter asking your aid, are, I think, not
necessary—you were the only man to whom I could appeal for the
special help needed, and I was in bitter grief and anxiety.
My wife asks me to express her sincere gratitude to you.
Now that the crisis is past, we have a strange tranquillity.
And we have acquired a humility that is, I think, good for us. Well,
these are things not readily expressed—accept our heart-felt thanks.
 Yours sincerely,
Joyce Kilmer.

* * * * *

New York.
Oct. 6, 1913.
Dear Father Daly:
This is to express our gratitude for your prayers. They have helped
us in our trial—they are helping—and I know that in part to your efforts
Rose owes her life and we our peace of mind. For fifteen or sixteen
years she will, the doctors say, remain paralysed, unable to move her
arms or legs. But her mind is active and we will keep her happy. I think
that there are compensations, spiritual and mental, for the loss of
physical power.
My wife and I are studying Catholic doctrine and we hope to be
received this Autumn.
Yours sincerely,
Joyce Kilmer.

* * * * *

New York.
Dec. 5, 1913.
Dear Father Daly:
You may be amused by the circular which I enclose. You know, we
have been having a plague of white-slave plays in New York this
season—“The Lure,” “The Fight” and the rest; and now they are giving
white-slave moving pictures—“The Traffic in Souls” they are called. I
think that “The Drama as a Factor in Sex Education” is ridiculous—and
I’m going to say so. Most of the other people who are to speak at the
Academy of Medicine are very serious-minded settlement workers,
ready to save the world by means of eugenics and inexpensive
divorce. So I don’t think I’ll make myself very popular. But I hope to
have a lot of fun.
My wife and I are very comfortable now that we are Catholics. I
think we rather disappointed Father Cronyn (the Paulist who received
us) by not showing any emotion during the ceremony. But our chief
sensation is simply comfort—we feel that we’re where we belong, and
it’s a very pleasant feeling.
I wonder if the enclosed verses will shock you. I am very much
bored by the praise of suicide so common in “æsthetic” circles. Many
of my friends like to think themselves “decadents” and they are most
enthusiastic over the work of such third-rate versifiers as the late
Richard Middleton, who killed himself at the age of thirty. It seems to
me that suicide is the most absurd of all sins. There is, undoubtedly,
pleasure in many sorts of evil; many crimes are reprehensible but not
ridiculous. But suicide is a stupid sort of sin—the criminal gets no fun
at all out of it. I have tried to put this idea into verse—and you should
have heard the criticisms when the poem was read aloud at a meeting
of the Poetry Society!
Yours sincerely,
Joyce Kilmer.

* * * * *

New York, Dec. 22, 1913.

Dear Father Daly:
I am grateful to you for your sound criticism of “To a Young Poet
Who Killed Himself.” It had previously been criticised by “pagans” who
thought that suicide was romantic and charming. You gave me just
what I needed—an analysis of its “philosophy.” Of course I’ll leave it
out of my new book of verse, or else, as you suggest, make it the
utterance of someone not myself.
You will be glad to hear of our Christmas present from Soeur
Therese. Rose can lift her left forearm. I know that you have been
getting her aid for us. Please tell us how to show our gratitude to her
and to you.
A picture of Rose and her mother is going to you for Christmas. I
hope that it may be a mirthful feast for you. We can honestly say that
we have never before been so happy.
Yours sincerely,
Joyce Kilmer.

* * * * *

Mahwah, N. J.
Jan. 9, 1914.
Dear Father Daly:
At last I have leisure to thank you for your Christmas gifts. I did not
know that such beautiful cards were made. How is it that in Prairie du
Chien—a place of which the name suggests Indians and tomahawks
and Deadwood stage-coaches—you can procure better cards than I
can get in New York? The medals are highly valued—the workmanship
on them is admirable—they will be worn properly as soon as I can
afford to buy some silver chains—and that will be next Tuesday.
Of course you understand my conversion. I am beginning to
understand it. I believed in the Catholic position, the Catholic view of
ethics and æsthetics, for a long time. But I wanted something not
intellectual, some conviction not mental—in fact I wanted Faith.
Just off Broadway, on the way from the Hudson Tube Station to the
Times Building, there is a Church, called the Church of the Holy
Innocents. Since it is in the heart of the Tenderloin, this name is
strangely appropriate—for there surely is need of youth and
innocence. Well, every morning for months I stopped on my way to the
office and prayed in this Church for faith. When faith did come, it came,
I think, by way of my little paralysed daughter. Her lifeless hands led
me; I think her tiny still feet know beautiful paths. You understand this
and it gives me a selfish pleasure to write it down.
 I was very glad of your criticism of “The Young Poet Who Killed
Himself.” I need some stricter discipline, I think, and it’s hard to get it. I
enjoy Father Cullem’s direction very much, he is a fine old Irishman
with no nonsense about him. But I need to be called a fool, I need to
have some of the conceit and sophistication knocked out of me. I
suppose you think this is “enthusiasm”—that much heralded danger of
converts. Perhaps it is, but I don’t think so. I know I’m glad I live two
miles from the Church, because it’s excellent for a lazy person like
myself to be made to exert himself for religion. And I wish I had a stern
mediæval confessor—the sort of person one reads about in anti-
Catholic books—who would inflict real penances. The saying of Hail
Marys and Our Fathers is no penance, it’s a delight.
Forgive this egotistical letter! I am praying that the New Year may
bring you much happiness.
Yours sincerely,
Joyce Kilmer.

* * * * *

Mahwah, N. J.
July 14, 1914.
Dear Father Daly:
If you read my “Some Mischief Still” I hope the strong language of
my janitor didn’t shock you. I’ve written to my long-suffering confessor
to ask whether or not I can let characters in my plays and stories
“cuss” when they feel so disposed. I fear that I introduced a new sin to
him—vicarious profanity.
You asked me several questions recently. Let me see if I can
answer them. In the first place, you asked me why I addressed the
Holy Name Society of Suffern, New York. The answer to that is, why
not? I go to Sacred Heart Church in Suffern. My parish church is in
Ramsey, New Jersey, but we Mahwah Catholics (those of us who go to
Church) are allowed to go to Suffern, which is much nearer. Ramsey
has only a small Mission Church, to which comes the priest from
Hohokus.
 Furthermore, I tolerate no levity on the name Suffern! especially
from a dweller in Prairie du Chien (named after a ridiculous vermin)
who has been visiting a place called Pilsen (named after a beer) and
conducting a retreat in a place called Peoria (obviously a corruption of
Peruna, a patent medicine of ill repute).
You ask me if I know what a retreat is. I do, because long before I
was a Catholic, before I was a wild-eyed Socialistic revolutionary, I was
a ritualistic Anglican, and I went twice up to Holy Cross Monastery at
West Park. This is an Anglican institution, which observes a
modification of the Benedictine rule. And there I learned about imitation
retreats, anyway. I must go to a regular retreat sometime.
I don’t expect to take any vacation this Summer, but I’m going to
spend several Saturdays fishing. I hope you do not scorn that
wholesome and unliterary diversion. Most writers of my acquaintance
consider it lamentably base.
I hope you were satisfied with your efforts at Peoria. I did what you
asked. By the way, when you pray for the Kilmers, please include a
new Kilmer, who is coming, we hope, in October. He—or she—will
have Barry among his or her names, after the family from whom my
wife is descended.
Yours affectionately,
Joyce Kilmer.

* * * * *

Mahwah, New Jersey.

Nov., 1914.
Dear Father Daly:
We are now back in Mahwah, thank God! It’s the chief thing I thank
Him for, of all the splendid things He’s given me—this home of mine.
My children are all well, and my wife. Prayer has given Rose the
almost normal use of one arm and the power to sit up. And prayer will
do more.
 Yours affectionately,
Joyce.
This is a postscript, to say that the smug young person portrayed
in the Citizen resembles me only “as the dew resembles the rain.” I am
fat and gross, and I have “elastic and rebellious hair,” like M.
Baudelaire’s cat. Someday I’ll have a decent picture and courageously
send it to you. In my young youth, I was slightly decorative. But now I
drink beer instead of writing about absinthe. And therefore—

* * * * *

June 15, 1915.

Dear Father James:
Your suggestion as to a lecture tour is characteristically kind. I am
already on the books of a lecture manager named Feakins, who
manages Chesterton, Barry Pain and other people. I’ll probably lecture
before Catholic organizations as well as secular ones, because I need
the money, but I don’t want to for reasons you will, I know, understand.
In the first place, I don’t want in any way to make money out of my
religion, to seem to be a “professional Catholic.” In the second place, I
have delight chiefly in talking veiled Catholicism to non-Catholics, in
humbly endeavouring to be an Apostle to Bohemia. I have no real
message to Catholics, I have Catholicism’s message to modern
pagans. So I want to lecture chiefly to Pagans.
The Prairie du Chien correspondent of the Police Gazette has
cabled that you are to read a paper at the St. Paul Educational
Conference. Congratulations! And I’ll try to help you to the extent of my
power—but you don’t need help as I do.
Yours affectionately,
Joyce.

* * * * *
 Good Samaritan Hospital,
Suffern, N. Y.
August 2, 1916.
Dear Father James:
I am sorry that you received exaggerated reports of my accident.
All that happened to me was the fracture of a couple of ribs, and I have
been very comfortable in a delightful hospital. My secretary comes out
several times a week, so I am keeping up with my work.
Remember me, if you please, to Mother Mercedes and Father
O’Reilly.
Yours affectionately,
Joyce.

* * * * *

Mahesh, N. J.
Aug. 10, 1916.
Dear Father James:
This is to say that I am rapidly recovering my health, and hope to
be in my office in a week. I had a fine time in the hospital, a
comfortable place run by particularly amiable Sisters of Charity. It may
interest you to know that I had received the Blessed Sacrament half an
hour before the train struck me, and that to this fact I attribute my
escape from death—since at the place where I was struck several men
have been killed, being thrown forward and under the wheels, instead
of, (as I was), to one side.
Affectionately,
Joyce.

* * * * *

(Aug 8, 1917).
Monday.
Dear Father James:
Your letter was opened and read by one Private Kilmer, a
hardened military cuss, unused to literary activities. This is the first
literary labor he has essayed for a month, aside from studying Moss’s
Manual of Military Training and the art of shooting craps.
We’re still in New York—at the Armory from 9 A. M. to 4.30 P. M.,
but we usually get home at night. Young Louis is in the same company
as myself. We probably will go to Spartansberg, North Carolina, about
Aug. 17, but we may be here till October, or we may go to France to-
morrow. Nobody knows. I find that there are many Catholics in the
Seventh. It’s amusing that our training camp is in North Carolina—
nothing to drink and no Jesuits! Bishop is a Benedictine mitred abbott
—it’s in Belmont Abbey.
Before mobilisation I finished editing “Dreams and Images,” an
anthology of Catholic verse written since (but including) Patmore.
Terrible rush job, but pretty good book. Do you mind having it
dedicated to you? Wanted to include some of your stuff, but didn’t have
time to get any. Have dedicated it to you, but can change it, if you
wish. Don’t hesitate to say so! Can’t hurt my feelings! Hard military
character, seriously considering acquisition of habit of chewing
tobacco.
This is an absurd note, but I find myself very stupid at writing.
Perhaps after drill becomes easier, I’ll have more of a mind left at
night. Fr. Dwight honoured me by asking me for a sonnet on St.
Ignatius for America, but I fell down on the assignment. I could not
write even a limerick on St. Ignatius in my present mental state! But it’s
a comfortable life physically—mentally too, I guess, and as for
spiritually, why, pray for yours affectionately,
Joyce.

* * * * *

(Sept., 1917).
Thursday.
Dear Fr. James:
I never expected to write to you on lined paper or Y. M. C. A. paper
—uses of adversity! The Y. M. C. A. gives us poor “dough boys” the
only place we have to sit down and read and write—so far the Knights
of Columbus have not been in evidence in this highly papistical camp. I
am, as you know, a member of the 165th Infantry, U. S., formerly the
69th New York. I have recently been transferred from Co. H. to
Headquarters Co., and exchanged my 8 hours a day of violent physical
exercise (most deadening to the brain, a useful anodyne for me,
coming as it did after my grief) for exacting but interesting statistical
work. I am called Senior Regimental Statistician, but in spite of all
these syllables still rank as a private. My work is under the direction of
the Regimental Chaplain, Fr. Francis Patrick Duffy. The people I like
best here are the wild Irish—boys of 18 or 20, who left Ireland a few
years ago, some of them to escape threatened conscription, and
travelled about the country in gangs, generally working on the
railroads. They have delightful songs that have never been written
down, but sung in vagabonds’ camps and country jails. I have got
some of the songs down and hope to get more—“The Boston
Burglar”—“Sitting in My Cell All Alone”—they are fine, a veritable Irish-
American folk-lore. Before I was transferred to Headquarters Co., I
slept in a tent with a number of these entertaining youths, and enjoyed
it tremendously. We sang every night from 9 to 9.30. Now I am in more
sophisticated but less amusing company—ambitious youths, young
office men, less simple and genial than my other friends.
I get to New York for part of Saturday and Sunday. Aline is staying
with her mother in New York. It is good that I am not with the 7th
Regiment in Spartanburg—as it is I can telephone to Aline every night.
I will send you a picture of Rose soon. Her death was a piercing blow,
but beautiful. It happened at the best time. Aline was there and I and
our parish priest. Rose was happy but did not want to get well. “I’ll
drink it in another house,” she said, when the nurse coaxed her to take
some broth. Perhaps she meant the new house in Larchmont which
we move to in October. And perhaps not. There was a Mission in our
parish-church, just a couple of blocks from the house, and while Rose
died the voices of the Sisters singing “O Salutaris Hostia” could be
heard in the room. I thank you for your letter, beautifully interpretative.
Certainly Rose makes Heaven dearer to us.
Yours affectionately,
Joyce.

* * * * *

Headquarters Company, 165th U. S. Infantry,

Camp Albert E. Mills, New York.
October 28, 1917.
Dear Father James:
I hope you don’t mind being typewritten to by an unskillful typist. I
have learned all I know of this art since I became a soldier. It is
ridiculous for a newspaperman to learn typewriting in the army, but I
never had the time to learn it before—I found it better to have an
expert typist for a secretary and dictate my stuff to him. I used to turn
out four or five thousand words an hour when Watts was working for
me. I haven’t much speed as yet but I ought to be pretty good at it
when I get back. There are two things I always wanted to learn—how
to typewrite and how to serve Mass. I’m learning the one and I’m going
to get Fr. Duffy to let me pinch-hit for his orderly at Mass some
mornings. So I’ll be an accomplished cuss when I come back from the
Wars—I’ll know how to typewrite and to serve Mass and to sing the
Boston Burglar.
Numerous philanthropic citizens are sending cigarettes to soldiers,
and pipe-tobacco (in which last bounty I often share). But some
enlightened person should start a nation-wide campaign for sending
cigars to Senior Regimental Statistical Representatives. I am Senior
Statistical Representative of the 165th U. S. Infantry, I am!
Yours affectionately,
Joyce.

* * * * *
 Headquarters Company,
165th Infantry,
A. E. F., France.
April 5, 1918.
Dear Father James:
When I next visit Campion, I will teach you (in addition to “The
Boston Burglar”) an admirable song called “Down in the Heart of the
Gas-House District.” I sing it beautifully. Its climax is, “If you would
meet good fellows face to face, Just hang around Jimmy
Cunningham’s place, For it’s down in the heart of the Gas-house
district, In Old New York.”
I received two most welcome letters from you recently. One had to
do with the case of ——. It contains excellent advice, and I am carrying
it in my pocket, intending to clip your name and the name of the
college from it and to give it to him. I am deeply grateful, my dear
Father, for your great and characteristic kindness in this matter. I think
we’ll yet see him practicing and teaching philosophy. You see, in his
veins is a dangerous mixture of bloods—like my own—German and
Irish, perhaps that accounts for his troubles. Gaelic impetuosity and
Teutonic romanticism—no, sentimentality—no, self-pity—no, what is it
that makes Germans German? I mean Sturm und Drang, Sorrows of
Werther and that sort of thing. Well, it doesn’t matter what it’s called.
But all people with any German blood have a bit of it. And highly
diluted it’s rather nice. Which reminds me that Father Duffy says I’m
half German and half human.
I wrote a poem called “Rouge Bouquet.” I’ll ask Aline to send you a
copy—it’s long—and you remind her. It will tell you of an incident which
will indicate to you that handling statistics in Regimental Headquarters
isn’t the dry task you imagine it. The statistics aren’t dry—they’re wet—
and red. Also, I’m attached to the Regimental Intelligence Section,
which is exciting. I had a very thrilling week not long ago—I’m not
allowed to write about it yet. I’ll tell you about it some day. By the way,
it was at that time—Feb. 26, that you wrote to me. I wasn’t in the office
then, I was out working as an observer—finest job in the army! By the
way, I’m a Sergeant now. I’ll never be anything higher. To get a
commission I’d have to go away for three months to a school, and then
—whether or not I was made an officer—I’d be sent to some outfit
other than this. And I don’t want to leave this crowd. I’d rather be a
Sergeant in the 69th than a Lieutenant in any other regiment in the
world. Get Aline to send you a copy of “Holy Ireland”—it will explain,
somewhat.
Please give my love to anybody who will accept it. Remember me
with special fervour to Sister Eugene and Sister Stanislaus. I enjoy
reading the Campion, and am glad to see the newspaper has become
a reality.
You speak of French anti-clericals in your letter. They all must be in
Paris. I haven’t found any in my rather extensive wanderings around
France. As a common soldier, I have the privilege of intimacy with the
French peasants—and I find them edifyingly good Catholics. Pray for
me, dear Father James.
Affectionately,
Joyce.

* * * * *

April 8, 1918.
My dear Father James:
I am sending you with this letter some verses that will, I think,
interest you. They are my first attempt at versification in a dug-out—a
dug-out not far from the event of which the verses tell. I have sent a
copy to Aline, but since the trans-Atlantic mails are so uncertain, will
you be so kind as to write to her asking if she has received it?
Somewhere within a radius of five miles from where I write now
slumbers (for it’s long after bedtime for soldiers who keep regular
hours) one Drury Sheehan. Every day he meets someone from my
outfit and asks after me and says he’s coming to see me. And some
days ago I—riding on a truck—passed and was passed several times
by his outfit. And I paged him, so to speak, unavailingly. Also numerous
letters from the States tell me that young Father Charles O’Donnell is
somewhere near. We’ll have a great reunion one of these days. I wish
you’d be in it! But one of these nights I’ll jump off a freight at Prairie du
Chien, break a pane of glass in a basement window and go to sleep in
the Bishop’s room. The Brother will be aghast next morning at
discovering a ragged mustachioed shaven headed soldier in such a
place—I’ll try to explain to him that I’m an adopted alumnus of
Campion—I’ll probably use French in my excitement—and he will
promptly shoot me. Then I’ll spend two months in the Campion
Infirmary, which is, I am told, a very nice place.
I am having a delightful time, but it won’t break my heart for the
war to end. I feel much comforted and strengthened by many prayers
—those of my wife and little children, those of many dear friends,
mostly priests and Sisters. I value highly my share in your Masses.
Always pray hard, my dear Father, for
Your affectionate friend,
Joyce.

* * * * *

May 15, 1918.

Dear Father James:
I thought you’d like to see how I looked after six months of
soldiering. I’m having a lot of fun (I’m a Sergeant in the Regimental
Intelligence Section now, you remember) but I haven’t been to Mass
for three weeks! It so happened that I was away from churches (except
the ruins thereof) and from Fr. Duffy. So you’ll have to be praying
doubly hard for your affectionate friend,
Joyce.

* * * * *

To Mrs. Kilburn Kilmer

 To My Mother, on Her Birthday, 1914,
With a Book of Poems

Gentlest of critics, does your memory hold

(I know it does) a record of the days
When I, a schoolboy, earned your generous praise
For halting verse and stories crudely told?
Over those boyish scrawls the years have rolled,
They might not bear the world’s unfriendly gaze,
But still your smile shines down familiar ways,
Touches my words and turns their dross to gold.

Dearer to-day than in that happy time

Comes your high praise to make me proud and strong.
In my poor notes you hear Love’s splendid chime
So unto you does this, my work, belong.
Take, then, this little book of fragile rhyme;
Your heart will change it to authentic song.

* * * * *

To A. K. K.
August the Fourth
Nineteen Hundred and Eleven

Now the English larks are singing,

And the English meadows flinging
Scarlet flags of blazing poppies to the fragrant summer air,
And from every tower and steeple
All the wondering English people
Hear a chime of fairy music, though no bell-ringers are there.