Liusheng Huang

Liusheng Huang

Auto-generated profile, based on publicly available information

Liusheng Huang

School of Computer Science and Technology, University of Science and Technology of China, Hefei, China

Research Focus & Metrics

Academic Career Length

27Years (2000-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
23 April 2018 in IEEE Transactions on Mobile Computing
Verified
Journal Article
Closed access
GGGuoju Gao
Guoju Gao
Mingjun XiaoMingjun Xiao
Mingjun Xiao
JWJie Wu
Jie Wu
LHLiusheng Huang
Liusheng Huang
CHChang Hu
Chang Hu
In this paper, we focus on the incentive mechanism design for a vehicle-based, nondeterministic crowdsensing system. In this crowdsensing system, vehicles move along their trajectories and perform corresponding sensing tasks with different probabilities. Each task may be performed by multiple vehicles jointly so as to ensure a high probability of success. Designing an incentive mechanism for such a crowdsensing system is challenging since it contains a non-trivial set cover problem. To solve this problem, we propose a truthful, reverse-auction-based incentive mechanism that includes an approximation algorithm to select winning bids with a nearly minimum social cost and a payment algorithm to determine payments for all participants. Moreover, we extend the problem to a more complex case in which the Quality of sensing Data (QoD) of each vehicle is taken into consideration. For this problem, we propose a QoD-aware incentive mechanism, which consists of a QoD-aware winning-bid selection algorithm and a QoD-aware payment determination algorithm. We prove that the proposed incentive mechanisms have truthfulness, individual rationality and computational efficiency. Moreover, we analyze the approximation ratios of the winning-bid selection algorithms. The simulations, based on a real vehicle trace, also demonstrate the significant performances of our incentive mechanisms.

23 April 2021 in IEEE Transactions on Mobile Computing
Verified
Journal Article
Closed access
ZMZhenguo Ma
Zhenguo Ma
YXYang Xu
Yang Xu
HXHongli Xu
Hongli Xu
ZMZeyu Meng
Zeyu Meng
LHLiusheng Huang
Liusheng Huang
YXYinxing Xue
Yinxing Xue
The emerging Federated Learning (FL) enables IoT devices to collaboratively learn a shared model based on their local datasets. However, due to end devices’ heterogeneity, it will magnify the inherent synchronization barrier issue of FL and result in non-negligible waiting time when local models are trained with the identical batch size. Moreover, the useless waiting time will further lead to a great strain on devices’ limited battery life. Herein, we aim to alleviate the negative impact of synchronization barrier through adaptive batch size during model training. When using different batch sizes, stability and convergence of the global model should be enforced by assigning appropriate learning rates on different devices. Therefore, we first study the relationship between batch size and learning rate, and formulate a scaling rule to guide the setting of learning rate in terms of batch size. Then we theoretically analyze the convergence rate of global model and obtain a convergence upper bound. On these bases, we propose an efficient algorithm that adaptively adjusts batch size with scaled learning rate for heterogeneous devices to reduce the waiting time and save battery life. We conduct extensive simulations and testbed experiments, and the experimental results demonstrate the effectiveness of our method.

16 November 2017 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
YWYang Wang
Yang Wang
WDWenjian Ding
Wenjian Ding
LHLiusheng Huang
Liusheng Huang
ZWZheng Wei
Zheng Wei
HLHengchang Liu
Hengchang Liu
JSJohn A. Stankovic
John A. Stankovic
Despite the clear benefits of electric vehicles (EVs) in terms of reducing greenhouse gas emissions and traditional energy consumptions, the popularization of EVs remains a challenge in the short run. There are currently two major ways of refueling electric taxicabs (ETs): recharging and battery swapping. ETs usually choose battery swapping so as not to waste precious time during their daily work. While previous studies focused on planning battery swapping stations for private EVs, we investigate ways of supporting the upgrade of an entire urban taxi system, with demands differing both in scale and nature. Furthermore, most ETs are fully charged in the early morning and need to swap battery in approximately the same time period of the day, which results in a bottleneck for battery swapping and hurts the QoS of the taxicab fleet. With this insight, we model the ET fleet as a mobile sensor network, analyze the historical sensing data of taxi routes, and evaluate the battery swapping demand profile and the power consumption of individual taxis, as well as the driving time between positions in the road network. Based on these inputs, we propose a method to calculate an optimized battery swapping station scheme, then describe a real-time algorithm to schedule a subset of the unoccupied taxicabs to swap batteries early by giving them allowances to avoid congestion. Our strategies are then evaluated via a real world 366-day, 3,976-taxi dataset. The results demonstrate that compared to uniform deployment, our planning scheme reduces the average time-cost by 67.2%; and based on our deployment, our scheduling strategy decreases the in-station queuing time by 45.63%, 52.26% and 29.57%, and the average driving time-cost to the stations by 42.3%, 43.69% and 28.77% compared to actual taxicab refueling, random scheduling, and a static heuristic strategy, respectively. Furthermore, our approach reduces the number of battery swapping that last more than 30 minutes by 40.8%, which is a significant improvement on the QoS of the whole system.

1 November 2000 in Journal of Computer Science and Technology
Verified
Journal Article
Closed access
PDF file
LHLiusheng Huang
Liusheng Huang
HCHuaping Chen
Huaping Chen
XWXun Wang
Xun Wang
GCGuoliang Chen
Guoliang Chen

29 May 2018 in IEEE Transactions on Knowledge and Data Engineering
Verified
Journal Article
Closed access
YNYiwen Nie
Yiwen Nie
WYWei Yang
Wei Yang
LHLiusheng Huang
Liusheng Huang
XXXike Xie
Xike Xie
ZZZhenhua Zhao
Zhenhua Zhao
SWShaowei Wang
Shaowei Wang
Recently, local differential privacy (LDP), as a strong and practical notion, has been applied to deal with privacy issues in data collection. However, existing LDP-based strategies mainly focus on utility optimization at a single privacy level while ignoring various privacy preferences of data providers and multilevel privacy demands for statistics. In this paper, we for the first time propose a framework to optimize the utility of histogram estimation with these two privacy requirements. To clarify the goal of privacy protection, we personalize the traditional definition of LDP. We design two independent approaches to minimize the utility loss: Advanced Combination, which composes multilevel results for utility optimisation, and Data Recycle with Personalized Privacy, which enlarges sample size for an estimation. We demonstrate their effectiveness on privacy and utility, respectively. Moreover, we embed these approaches within a Recycle and Combination Framework and prove that the framework stably achieves the optimal utility by quantifying its error bounds. On real-world datasets, our approaches are experimentally validated and remarkably outperform baseline methods.

19 February 2024 in IEEE Transactions on Mobile Computing
Verified
Journal Article
Closed access
QMQianpiao Ma
Qianpiao Ma
YXYang Xu
Yang Xu
HXHongli Xu
Hongli Xu
JLJianchun Liu
Jianchun Liu
LHLiusheng Huang
Liusheng Huang
Federated learning (FL) is an effective approach to train models collaboratively among distributed edge nodes (i.e., workers) while facing three crucial challenges, edge heterogeneity, resource constraint, and Non-IID data. Under the parameter server (PS) architecture, a single parameter server may become the system bottleneck and cannot well deal with the edge heterogeneity, while the peer-to-peer (P2P) architecture causes significant communication consumption to achieve satisfactory training performance. To this end, hierarchical aggregation (HA) architecture is proposed to cluster workers to tackle the edge heterogeneity and reduce communication consumption for FL. However, the existing researches on HA architecture cannot provide a unified clustering approach for various inter-cluster aggregation patterns (e.g., centralized or decentralized structure, synchronous or asynchronous mode). In this paper, we explore the quantitative relationship between the convergence bounds of different inter-cluster patterns and several factors, e.g., data distribution, frequency of clusters participating in inter-cluster aggregation (for asynchronous modes), and inter-cluster topology (for decentralized structures). Based on the convergence bounds, we design a unified clustering algorithm FedUC to organize workers for different patterns. Experimental results on classical models and datasets show that FedUC can greatly accelerate the model training of different patterns by 1.79-7.39× compared with the state-of-the-art clustering methods.