Research Interest

My research interests are in the areas of wireless networks and their applications, with current focuses on ML/AI in wireless networks, edge computing, and network security.

Research Direction

Federated Learning in Wireless Edge Networks

 

As an emerging paradigm of ML, federated learning (FL) trains a global ML model by aggregating local ML models computed by distributed devices based on their local data. Along a different avenue, continuing innovations of wireless technologies enable the future Internet of Things (IoTs) for many applications, including connected and autonomous vehicles, collaborative robots, cyber-physical systems, etc. In this research, we explore FL in wireless edge networks to achieve collaborative intelligence in wireless networked systems, while taking into account salient features of such wireless FL, including heterogeneous local data as well as computation and communication capabilities of wireless devices, wireless interference, and time-varying wireless channels.

D. Li*, X. Gong, “Anarchic federated learning with delayed gradient averaging”, ACM MobiHoc’23, Washington DC, USA, Oct. 23-26, 2023.

D. Li*, X. Gong, “Anarchic convex federated learning”, IEEE INFOCOM’23 Workshop on Distributed Machine Learning and Fog Networks (FOGML), online, May 20, 2023.

Y. Zhao*, X. Gong, S. Mao, “Truthful incentive mechanism for federated learning with crowdsourced data labeling”, IEEE INFOCOM’23, New York area, USA, May 17-20, 2023.

Y. Zhu*, X. Gong, “Distributed policy gradient with heterogeneous computation for federated reinforcement learning”, CISS’23, Baltimore, MD, USA, Mar. 22-24, 2023.

D. Li*, Y. Zhao*, X. Gong, “Quality-aware distributed computation and communication scheduling for fast convergent wireless federated learning”, WiOpt’21, Oct. 18-21, online, 2021.

Y. Zhao*, X. Gong, “Quality-aware distributed computation for cost-effective non-convex and asynchronous wireless federated learning”, WiOpt’21, Oct. 18-21, online, 2021.

Y. Zhao*, X. Gong, “Quality-aware distributed computation and user selection for cost-effective federated learning”, IEEE INFOCOM’21 Workshop on Distributed Machine Learning and Fog Networks (FOGML), May 10, 2021.

(my students are marked by *)

Distributed Computation Offloading in Wireless Edge Networks

 

Edge computing exploits devices at or near end users (e.g., smartphone, laptop, desktop) to perform a substantial amount of networking, computing, control, and management. It has various emerging applications, including ML/AI, mixed reality, Internet of Things, autonomous vehicle, UAV, and robotics. Distributed computing makes use of distributed devices to perform computation in a collaborative manner. In this research, we explore edge devices connected wirelessly to carry out computation offloading in a distributed manner.

M. Chen, X. Gong, Y. Cao, “Delay-optimal distributed edge computation offloading with correlated computation and communication workloads”, IEEE Transactions on Mobile Computing, to appear, 2022.

X. Gong, “Delay-optimal distributed edge computing in wireless edge networks”, IEEE INFOCOM’20, July 6-9, online, 2020.

(my students are marked by *)

Quality-Aware Data Crowdsourcing

 

Data crowdsourcing leverages the “wisdom” of a potentially large crowd of “workers” (e.g., mobile users) by collecting data from them. It has a wide range of applications, including spectrum sensing, environmental monitoring, human annotation. The value and usefulness of the data collected in crowdsourcing rely on the quality of participating workers’ data, which generally varies for different workers. To exploit the potential of crowdsourcing, this research explores quality-aware crowdsourcing that makes use of the quality information of workers’ data (e.g., for tasks allocation or data aggregation), while addressing the challenges of unknown data quality.

Y. Zhao*, X. Gong, F. Lin, X. Chen, “Data poisoning attacks and defenses in dynamic crowdsourcing with online data quality learning”, IEEE Trans. on Mobile Computing, to appear, 2021.

Z. Shi, G. Yang, X. Gong, S. He, J. Chen, “Quality-aware incentive mechanisms under social influences in data crowdsourcing”, IEEE/ACM Trans. Networking, to appear, 2021.

Y. Zhao*, X. Gong, X. Chen, “Privacy-preserving incentive mechanisms for truthful data quality in data crowdsourcing”, IEEE Trans. on Mobile Computing, to appear, 2021.

X. Gong, N. B. Shroff, “Truthful data quality elicitation for quality-aware data crowdsourcing”, IEEE Trans. Control of Network Systems, vol. 7, no. 1, pp. 326-337, Mar. 2020.

X. Gong, N. B. Shroff, “Truthful mobile crowdsensing for strategic users with private data quality”, IEEE/ACM Trans. Networking, vol. 27, no. 5, pp. 1959-1972, Oct. 2019.

Y. Zhao*, X. Gong, “Truthful quality-aware data crowdsensing for machine learning”, IEEE SECON’19, Boston, MA, USA, June 10-13, 2019.

X. Zhang*, X. Gong, “Online data quality learning for quality-aware crowdsensing”, IEEE SECON’19, Boston, MA, USA, June 10-13, 2019.

X. Gong, “Incentivizing quality-based data crowdsourcing,” IJCAI-ECAI Workshop on Game-Theoretic Mechanisms for Data and Information, Stockholm, Sweden, July 14, 2018.

X. Gong, N. B. Shroff, “Incentivizing truthful data quality in quality-aware mobile data crowdsourcing,” ACM MobiHoc’18, Los Angeles, CA, USA, June 25-28, 2018. [slides] [presentation video]

X. Gong, N. B. Shroff, “Truthful mobile crowdsensing for strategic users with private qualities,” WiOpt’17, Paris, France, May 15-19, 2017. [slides]

(my students are marked by *)

Socially-Aware Mobile Networking and Computing

Social relationships play an increasingly important role in people's interactions with each other. As mobile devices are carried and operated by human beings, it is natural to explore social aspects of human networks as a new dimension for mobile networking and computing. In this research, we explore how mobile users’ social relationships impact their interactions for mobile networking and computing, and how to leverage this impact to improve the performance of mobile networks.

X. Gong, L. Duan, and X. Chen, “When network effect meets congestion effect: Leveraging social services for wireless services,” ACM MobiHoc’15, Hangzhou, China, June 22- June 25, 2015. [slides]

X. Gong, X. Chen, K. Xing, D. Shin, M. Zhang, and J. Zhang, “Personalized location privacy in mobile networks: A social group utility approach,” IEEE INFOCOM’15, Hong Kong, China, Apr.26- May 1, 2015. [slides]

X. Gong, X. Chen, J. Zhang, and H. V. Poor “From social trust assisted reciprocity (STAR) to utility-optimal mobile crowdsensing,” IEEE GlobalSIP’14, Atlanta, GA, USA, Dec. 3-5, 2014. [poster]

X. Chen, X. Gong, L. Yang, and J. Zhang, “A social group utility maximization framework with applications in database assisted spectrum access,” IEEE INFOCOM’14, Toronto, Canada, Apr.27- May 2, 2014. [slides] (Runner-up Best Paper Award)

X. Gong, X. Chen, and J. Zhang, “Social group utility maximization in mobile networks: From altruistic to malicious behavior,” IEEE CISS’14, Princeton, NJ, USA, Mar. 19-21, 2014.

X. Gong, X. Chen, and J. Zhang, “Social group utility maximization game with applications in mobile social networks,” Allerton’13, Monticello, IL, USA, Oct. 2-4, 2013.

X. Chen, B. Proulx, X. Gong, and J. Zhang, “Social trust and social reciprocity based cooperative D2D communications,” ACM MobiHoc’13, Bangalore, India, July 29- Aug. 1, 2013.