Article
Computer Science, Information Systems
Claudio Badii, Angelo Difino, Paolo Nesi, Irene Paoli, Michela Paolucci
Summary: The advancement of modern mobile phones and digital transport networks has facilitated access to useful information about user's mean of transportation, leading to the development of innovative applications in sustainable mobility, smart transportation, and e-health. A new approach has been proposed to collect real-time data from mobile phones for personalized assistance messages for city users, contributing to a better understanding of travel behavior and enhancing user experience in urban environments.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Engineering, Electrical & Electronic
Yanjun Qin, Haiyong Luo, Fang Zhao, Chenxing Wang, Yuchen Fang
Summary: This paper proposes a novel fusion framework for fine-grained transportation mode recognition, incorporating NIN, Dilate Convolution, and GCN, which outperforms other baseline methods with over 22.3% higher accuracy according to extensive experimental results on the SHL dataset.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2021)
Article
Engineering, Industrial
Amir Hossein Barahimi, Alireza Eydi, Abdolah Aghaie
Summary: The research addresses the issue of increasing the link capacity in a dual-mode public transport network within urban infrastructure. By developing a mathematical model and using the PSO algorithm to solve the multi-objective bi-level model, the study was applied to a part of the urban transportation network in Tehran, Iran.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Article
Computer Science, Information Systems
Orla Brimacombe, Luis C. Gonzalez, Johan Wahlstrom
Summary: In order to reduce transportation-related greenhouse gas emissions, it is important to establish efficient systems for monitoring individuals' travel patterns and carbon footprints. This paper introduces a CO(2)e emission estimator that combines transportation mode classification with mode-specific emissions data. It not only evaluates the accuracy of the emission estimation, but also identifies and discusses the sources of errors and their relative importance. The study provides recommendations for future carbon footprint estimators and highlights the importance of evaluating transportation mode classifiers based on their ability to identify carbon emitting transportation modes.
Review
Health Care Sciences & Services
Marcin Straczkiewicz, Peter James, Jukka-Pekka Onnela
Summary: Smartphones are well-suited for health research, particularly in human activity recognition systems. Existing studies have shown various methods and practices in data acquisition, preprocessing, feature extraction, and activity classification using smartphones, with future studies focusing on improving data quality, addressing missing data, incorporating diverse participants and activities, and sharing source code.
NPJ DIGITAL MEDICINE
(2021)
Article
Multidisciplinary Sciences
Chuanlin Zhang, Kai Cao, Limeng Lu, Tao Deng
Summary: This study proposed a multi-scale feature extraction fusion model combining CNN and GRU, which can automatically extract different local features and long-term dependencies of the original data to obtain a richer feature representation. The accuracy of the proposed model is 97.18%, 96.71%, and 96.28% on the WISDM, UCI- HAR, and PAMAP2 datasets respectively.
SCIENTIFIC REPORTS
(2022)
Article
Multidisciplinary Sciences
Md Zia Uddin, Ahmet Soylu
Summary: Healthcare using body sensor data, especially in elderly care, has attracted significant research interest. This study proposes a robust activity modeling and recognition system using deep Neural Structured Learning and achieves a high recall rate. The approach outperforms conventional machine learning methods such as DBN, CNN, and RNN.
SCIENTIFIC REPORTS
(2021)
Editorial Material
Statistics & Probability
Nathan Kallus
Summary: The study reveals that randomization beyond a single partition of units is needed for precision in a two-arm controlled experiment, even when conditional means are linear. Therefore, the inference-constrained mixed-strategy optimal design is proposed as the minimax-optimal for precision among designs under sufficient uniformity constraints.
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
(2021)
Article
Computer Science, Hardware & Architecture
Qinglin Yang, Xiaofei Luo, Peng Li, Toshiaki Miyazaki, Wenfeng Shen, Weiqin Tong
Summary: This paper proposes collaborative inference among mobile devices to share computation workloads and accelerate processing speed by batching inference tasks on GPUs. An algorithm based on PSO is designed for efficient collaboration, as well as a distributed algorithm to address the challenge of collecting global network information and running centralized algorithms. Extensive simulations show that the collaborative inference scheme effectively reduces inference time for mobile deep learning applications.
MOBILE NETWORKS & APPLICATIONS
(2021)
Article
Statistics & Probability
Patrick Bajari, Brian Burdick, Guido W. Imbens, Lorenzo Masoero, James Mcqueen, Thomas S. Richardson, Ido M. Rosen
Summary: Classical Randomized Controlled Trials (RCTs) are designed to draw causal inferences, but modern experiments often involve complex interactions between units, making classical designs ineffective. In this manuscript, we review novel experimental designs, such as Multiple Randomization Designs (MRDs), that allow for the study of causal effects in the presence of interference.
STATISTICAL SCIENCE
(2023)
Article
Chemistry, Multidisciplinary
Zhong Zhang, Minho Shin
Summary: In the realm of mobile privacy, there are various attack methods to leak users' private information. Despite protection mechanisms against privilege escalation, attackers can utilize inference algorithms to derive new information or enhance data quality without violating privilege limits. A proposed detection and protection mechanism using Inference Graph and Policy Engine allows users to control their privilege policies in information escalation, showing feasibility and good usability in implementation results.
APPLIED SCIENCES-BASEL
(2021)
Article
Engineering, Electrical & Electronic
Zhenxing Yao, Yu Zhong, Qiang Liao, Jie Wu, Haode Liu, Fei Yang
Summary: Existing research on travel behavior detection based on cellphone data mainly focuses on algorithm exploration and evaluation, and existing platforms have limited functions. This paper proposes a real-time urban mobility monitoring and traffic management platform using cellular data, with a field case study conducted in Guiyang.
IEEE INTELLIGENT TRANSPORTATION SYSTEMS MAGAZINE
(2021)
Article
Computer Science, Information Systems
Jia Xu, Yuanhang Zhou, Yuqing Ding, Dejun Yang, Lijie Xu
Summary: Mobile crowdsensing is an effective method for large-scale data collection, and incentive mechanisms are crucial for its success. This article proposes an auction-based biobjective robust mobile crowdsensing system with an incentive mechanism that achieves desirable properties such as computational efficiency and individual rationality. The proposed mechanism also shows improvement in platform utility compared to existing algorithms.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Computer Science, Information Systems
Tao Chen, Longfei Shangguan, Zhenjiang Li, Kyle Jamieson
Summary: This article introduces SoundSticker, a system that enables steganographic data communication over an acoustic channel. By embedding hidden bits in audible sounds, SoundSticker achieves higher data rates and remains imperceptible to listeners. The system addresses technical challenges such as waveform artifacts, bit rate adaptation, and real-time preamble detection, and outperforms state-of-the-art techniques in terms of performance and sound quality.
ACM TRANSACTIONS ON SENSOR NETWORKS
(2023)
Article
Environmental Sciences
Maria Sanchez-Aparicio, Jose Antonio Martin-Jimenez, Enrique Gonzalez-Gonzalez, Susana Laguela
Summary: This work explores the use of mobile laser scanning to analyze and evaluate the orography of urban areas, with a focus on its importance for electrifying public transport and electric buses. The study determines that a minimum point density of 1 point/m(2), measured with aerial laser scanning, is sufficient for estimating slope. Based on this, a route design for public transport is presented, taking into consideration key transit points and maximum slope requirements. The analysis shows that implementing electric buses rather than diesel ones in cities with steep slopes (up to 7%) can significantly reduce greenhouse gas emissions (32.59%) and economic costs (18.10%).