YC
Yang Chen
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
5 records found
1
Energy-Aware Collaborative Perception in HetVNets
Balancing Accuracy and Sustainability
Journal article
(2026)
-
Mengying Zhou, Shaobin Wang, Qiang Duan, Aaron Yi Ding, Xin Wang, Yang Chen
Environmental perception is essential for autonomous vehicles. Collaborative perception, supported by vehicular communication technologies like C-V2X, aggregates data from nearby sensors to extend sensing range and improve decision accuracy. However, it raises communication overhead and computational energy because vehicles process more data than their own sensor input. The issue is exacerbated in heterogeneous vehicular networks (HetVNets). Our measurements show that, under equal workloads, low-capability vehicles incur 10% higher computational latency and energy, resulting in a lower energy-to-accuracy gain ratio and a higher marginal cost than high-capability peers. In reality, users prioritize driving range over marginal accuracy gains. We therefore advocate that collaborative perception should move beyond pure accuracy maximization to include user-centric sustainability goals. We present GreenFusion, an energy-aware collaborative perception scheme that embeds explicit sustainability and fairness metrics. GreenFusion adapts each vehicle's engagement and role according to information value and capability, enabling selective sharing of noteworthy data. In evaluations, GreenFusion maintains perception performance while reducing energy consumption for low-capability vehicles by 81.0% and 31.5% on average compared with fully connected and information-adaptive centralized baselines, respectively. In a typical driving scenario, these savings correspond to a 65.6% increase in driving range, demonstrating practical sustainability benefits without sacrificing perception. GreenFusion reframes collaborative perception from an accuracy-only objective to a balanced accuracy-energy strategy, fostering a more sustainable, practical vehicular networking framework that improves resilience, longevity, and user experience.
...
Environmental perception is essential for autonomous vehicles. Collaborative perception, supported by vehicular communication technologies like C-V2X, aggregates data from nearby sensors to extend sensing range and improve decision accuracy. However, it raises communication overhead and computational energy because vehicles process more data than their own sensor input. The issue is exacerbated in heterogeneous vehicular networks (HetVNets). Our measurements show that, under equal workloads, low-capability vehicles incur 10% higher computational latency and energy, resulting in a lower energy-to-accuracy gain ratio and a higher marginal cost than high-capability peers. In reality, users prioritize driving range over marginal accuracy gains. We therefore advocate that collaborative perception should move beyond pure accuracy maximization to include user-centric sustainability goals. We present GreenFusion, an energy-aware collaborative perception scheme that embeds explicit sustainability and fairness metrics. GreenFusion adapts each vehicle's engagement and role according to information value and capability, enabling selective sharing of noteworthy data. In evaluations, GreenFusion maintains perception performance while reducing energy consumption for low-capability vehicles by 81.0% and 31.5% on average compared with fully connected and information-adaptive centralized baselines, respectively. In a typical driving scenario, these savings correspond to a 65.6% increase in driving range, demonstrating practical sustainability benefits without sacrificing perception. GreenFusion reframes collaborative perception from an accuracy-only objective to a balanced accuracy-energy strategy, fostering a more sustainable, practical vehicular networking framework that improves resilience, longevity, and user experience.
Journal article
(2024)
-
Yiliang Liu, Michel Attoui, Rima Baalbaki, Runlong Cai, George Biskos, Yang Chen, Juha Kangasluoma
Sub-10 nm metal-based nanoparticles have garnered immense interest due to their unique properties and versatile applications. In this study, we created sub-10 nm Ag-based particles with a spark discharge generator and explored the parameters impacting their size distribution and charging properties, including carrier gas flow rates, spark discharge voltage, electrode gap distances, and capacitance. Our findings illuminate that there is a comparable influence of different factors on both self-charged and neutral particles. Among the different factors, carrier gas flow rates emerging as a paramount determinant in particle size. While increasing spark discharge voltage and capacitance within the spark circuit increases particle concentrations, the associated changes in particle size prove to be less straightforward. Significant differences between the concentration of positive and negative self-charged particles manifest when the carrier gas flow rate surpasses 5.0 L min−1, with positive particles ranging from 0.8 to 1.2 nm and negative particles spanning 0.8 to 3.0 nm. Self-charged particles close to 1 nm tend to exhibit positive charges, whereas those larger than 2 nm tend to acquire negative charges, which suggests the growth of negative particles is faster than positive ones in the spark chamber. Nevertheless, these disparities between bipolar particles diminish with the increase of residence time, leading to the observation of similar particle size distributions. Positive particles consistently bear a single charge, while some negative particles exceeding 3 nm exhibit multiple charges, primarily under carrier gas flow rates exceeding 7.5 L min−1. This study provides insights into the control of properties of nano-sized metal particles, which are crucial for their practical utilization.
...
Sub-10 nm metal-based nanoparticles have garnered immense interest due to their unique properties and versatile applications. In this study, we created sub-10 nm Ag-based particles with a spark discharge generator and explored the parameters impacting their size distribution and charging properties, including carrier gas flow rates, spark discharge voltage, electrode gap distances, and capacitance. Our findings illuminate that there is a comparable influence of different factors on both self-charged and neutral particles. Among the different factors, carrier gas flow rates emerging as a paramount determinant in particle size. While increasing spark discharge voltage and capacitance within the spark circuit increases particle concentrations, the associated changes in particle size prove to be less straightforward. Significant differences between the concentration of positive and negative self-charged particles manifest when the carrier gas flow rate surpasses 5.0 L min−1, with positive particles ranging from 0.8 to 1.2 nm and negative particles spanning 0.8 to 3.0 nm. Self-charged particles close to 1 nm tend to exhibit positive charges, whereas those larger than 2 nm tend to acquire negative charges, which suggests the growth of negative particles is faster than positive ones in the spark chamber. Nevertheless, these disparities between bipolar particles diminish with the increase of residence time, leading to the observation of similar particle size distributions. Positive particles consistently bear a single charge, while some negative particles exceeding 3 nm exhibit multiple charges, primarily under carrier gas flow rates exceeding 7.5 L min−1. This study provides insights into the control of properties of nano-sized metal particles, which are crucial for their practical utilization.
Model partitioning is a promising solution to reduce the high computation load and transmission of high-volume data. Within the scope of Edge AI, the fundamentals of model partitioning involve splitting the model for local computing at the edge and offloading heavy computation tasks to the cloud or server. This approach benefits scenarios with limited computing and battery capacity with low latency requirements, such as connected autonomous vehicles. However, while model partitioning offers advantages in reducing the onboard computation, memory requirements and inference time, it also introduces challenges such as increased energy consumption for partitioned computations and overhead for transferring partitioned data/model. In this work, we explore hybrid model partitioning to optimize computational and communication energy consumption. Our results provide an initial analysis of the tradeoff between energy and accuracy, focusing on the energy-aware model partitioning for future Edge AI applications.
...
Model partitioning is a promising solution to reduce the high computation load and transmission of high-volume data. Within the scope of Edge AI, the fundamentals of model partitioning involve splitting the model for local computing at the edge and offloading heavy computation tasks to the cloud or server. This approach benefits scenarios with limited computing and battery capacity with low latency requirements, such as connected autonomous vehicles. However, while model partitioning offers advantages in reducing the onboard computation, memory requirements and inference time, it also introduces challenges such as increased energy consumption for partitioned computations and overhead for transferring partitioned data/model. In this work, we explore hybrid model partitioning to optimize computational and communication energy consumption. Our results provide an initial analysis of the tradeoff between energy and accuracy, focusing on the energy-aware model partitioning for future Edge AI applications.
Journal article
(2021)
-
Zihang Lin, Yuwei Zhang, Qingyuan Gong, Yang Chen, Atte Oksanen, Aaron Yi Ding
Social networks now connect billions of people around the world, where individuals occupying different positions often represent different social roles and show different characteristics in their behaviors. The structural hole (SH) theory demonstrates that users occupying the bridging positions between different communities have advantages since they control the key information diffusion paths. Users of this type, known as SH spanners, are important when it comes to assimilating social network structures and user behaviors. In this article, we review the use of SHs theory in social network analysis, where SH spanners take advantage of both information and control benefits. We investigate the existing algorithms of SH spanner detection and classify them into information flow-based algorithms and network centrality-based algorithms. For practitioners, we further illustrate the applications of SH theory in various practical scenarios, including enterprise settings, information diffusion in social networks, software development, mobile applications, and machine learning (ML)-based social prediction. Our review provides a comprehensive discussion on the foundation, detection, and practical applications of SHs. The insights can facilitate researchers and service providers to better apply the theory and derive value-added tools with advanced ML techniques. To inspire follow-up research, we identify potential research trends in this area, especially on the dynamics of networks.
...
Social networks now connect billions of people around the world, where individuals occupying different positions often represent different social roles and show different characteristics in their behaviors. The structural hole (SH) theory demonstrates that users occupying the bridging positions between different communities have advantages since they control the key information diffusion paths. Users of this type, known as SH spanners, are important when it comes to assimilating social network structures and user behaviors. In this article, we review the use of SHs theory in social network analysis, where SH spanners take advantage of both information and control benefits. We investigate the existing algorithms of SH spanner detection and classify them into information flow-based algorithms and network centrality-based algorithms. For practitioners, we further illustrate the applications of SH theory in various practical scenarios, including enterprise settings, information diffusion in social networks, software development, mobile applications, and machine learning (ML)-based social prediction. Our review provides a comprehensive discussion on the foundation, detection, and practical applications of SHs. The insights can facilitate researchers and service providers to better apply the theory and derive value-added tools with advanced ML techniques. To inspire follow-up research, we identify potential research trends in this area, especially on the dynamics of networks.
Journal article
(2020)
-
Jiayun Zhang, Yang Chen, Qingyuan Gong, Aaron Yi Ding, Yu Xiao, Xin Wang, Pan Hui
We identified three temporal patterns shown in commit activities among Chinese and American companies and found that Chinese businesses are more likely to follow long work hours than American ones. We also conducted a survey on the trends of, reasons for, and results of overtime work. Our study could provide references for developers to choose workplaces and for companies to make regulations.
...
We identified three temporal patterns shown in commit activities among Chinese and American companies and found that Chinese businesses are more likely to follow long work hours than American ones. We also conducted a survey on the trends of, reasons for, and results of overtime work. Our study could provide references for developers to choose workplaces and for companies to make regulations.