This is an overview of my research work, organized in a topic-by-topic basis:

  1. Game dynamics.
  2. Machine Learning.
  3. Wireless communications and signal processing.

The above topics obviously overlap; for a year-by-year list of publications, go here instead.

Game Theory

My research focuses on the application of game theory for the security problem in the Cognitive Radio (CR) networks, which is a key candidate for the future wireless communications. In such model, the network operation is under control of a central entity, referred as the network coordinator. To mitigate the influence of the misbehaving users (or refers to Attacker) to the operation of the network, we propose a surveillance process, implemented by the network coordinator (or refers to Defender). The interaction between Defender and Attacker, like the avoid fare evasion or the patrolling scenarios, can be seen as the model of leadership and commitment and formulated by the Stackelberg game. Our work then characterizes the corresponding attack and surveillance strategies of Attacker and Defender through the Strong Stackelberg Equilibrium (SSE) through the Multiple Linear Program method. A comparison to the conventional surveillance game with NE strategy shows that Defender should exploit the leader position in the game by committing to a defense strategy.

I see this work as the starting point for my long-term research goal of extending the field of application of game theory and machine learning, particularly Reinforcement Learning (RL), for security, economy, and communication networks. My future research projects include a study of the stochastic game for the security problem in the next-generation wireless communication under the attack of misbehaving users, as well as a project exploring the game theory-based collaborative framework to maintain a harmonized coexistence between communication systems. The goal is to create a new communication paradigm that communications systems can work together and individually with each other through autonomy. Also, I look forward to enriching my knowledge on game theory and machine learning by exploring the application of the Stackelberg game and the stochastic model in the security and economics scenario with single leader multiple followers and multiple leaders multiple followers.

Some selected work.

[J5] Duc-Tuyen. Ta, Nhan Nguyen-Thanh, Patrick Maillé, and Van-Tam Nguyen, “Strategic Surveillance Against Primary User Emulation Attacks in Cognitive Radio Networks,” IEEE Transactions on Cognitive Communications and Networking, minor revision.

[J4] Duc-Tuyen. Ta, and Duy H.N. Nguyen, and Nhan Nguyen-Thanh, and Van-Tam Nguyen, “Collaborative Paradigm for Next Generation Wireless Networks,” EURASIP Journal on Wireless Communications and Networking, second round of peer review.

[J3] Nhan Nguyen-Thanh, and Duc-Tuyen. Ta, and Van-Tam Nguyen, “Spoofing Attack and Surveillance Game in Geo-location Database Driven Spectrum Sharing,” IET Communications , under consideration.

[C5] Duc-Tuyen. TA, Nhan Nguyen-Thanh, Patrick. Maille, Phillipe. Ciblat, Van-Tam Nguyen, “Mitigating Primary Emulation Attacks in Multi-Channel Cognitive Radio Networks: A Surveillance Game,” in Proceedings of the IEEE Global Communications Conference (GLOBECOM), Washington, DC, USA Dec 2016.

[C4] Duc-Tuyen. TA, Nhan Nguyen-Thanh, Phillipe. Ciblat, Van-Tam Nguyen, “Extra sensing game for malicious primary user emulator attack in cognitive radio network,” in Proceedings of the European Conference on Networks and Communications (EuCNC), Paris, France Jun 2015.

Please follows my Publications for more details.

Wireless Sensor Networks

Project 1: Designing and implementing an efficient and reliable Landslide Monitoring and Early Warning (LMnE) system based on the 3G/2G mobile communication system combining with a wireless sensor network at monitoring stations.

Project 2: A novel MAC protocol for the sensor networks for marine monitoring, searching, and rescuing applications. A testbed with multiple communication nodes, which is contained a GPS, sensors, and radio communication devices, is implemented.

Some selected work.

[J2] Duc-Tuyen. Ta, Tran Duc-Tan, Do Duc Dung, “Efficient and Reliable GPS-Based Wireless Ad Hoc for Marine Search Rescue System,” Multimedia and Ubiquitous Engineering , Lecture Notes in Electrical Engineering, vol. 240, pp. 911–918, May 2013.

[C3] Duc-Tuyen. TA, Duc-Tan Tran, Do Duc Dung, Van Hoang Nguyen, Vu van Yem and Xuan Nam Tran, “GPS-Based Wireless Ad Hoc Network for Marine Monitoring, Search and Rescue (MSnR),” in Proceedings of Second International Conference on Intelligent Systems, Modelling and Simulation, Kuala Lumpur, Malaysia Jun 2011.

[C2] Duc-Tuyen. TA, Duc-Tan Tran, Do Duc Dung, Van Hoang Nguyen, Vu van Yem and Xuan Nam Tran, “Wireless ad hoc network based on Global Positioning System for marine monitoring, searching and rescuing (MSnR),” in Proceedings of Asia-Pacific Microwave Conference, Melbourne, Australia Dec 2011.

Please follows my Publications for more details.

Machine Learning

Deep convolutional neural networks (CNNs) has been developed for a wide range of applications such as image recognition, nature language processing, etc. However, the deployment of deep CNNs in home and mobile devices remains challenging due to substantial requirements for computing resources and energy needed for the computation of high-dimensional convolutions. In this paper, we propose a novel approach designed to minimize energy consumption in the computation of convolutions in deep CNNs. The proposed solution includes (i) an optimal selection method for Fast Fourier Transform (FFT) configuration associated with splitting input feature maps, (ii) a reconfigurable hardware architecture for computing high-dimensional convolutions based on 2D-FFT, and (iii) an optimal pipeline data movement scheduling. The FFT size selecting method enables us to determine the optimal length of the split input for the lowest energy consumption. The hardware architecture contains a processing engine (PE) array, whose PEs are connected to form parallel flexible-length Radix-2 single-delay feedback lines, enabling the computation of variable-size 2D-FFT. The pipeline data movement scheduling optimizes the transition between row-wise FFT and column-wise FFT in a 2D-FFT process and minimizes the required data access for the element-wise accumulation across input channels. Using simulations, we demonstrated that the proposed framework improves the energy consumption by 89.7% in the inference case.

Some selected work.

[C6] Nhan Nguyen-Thanh, Han Le-Duc, Duc-Tuyen. TA, Van-Tam Nguyen, “Energy-efficient techniques using FFT for deep convolutional neural networks,” in Proceedings of the International Conference on Advanced Technologies for Communications (ATC), Hanoi, Vietnam, Oct 2016.

Please follows my Publications for more details.