-
Applications of Distributed Machine Learning for the Internet-of-Things: A Comprehensive Survey
Authors:
Mai Le,
Thien Huynh-The,
Tan Do-Duy,
Thai-Hoc Vu,
Won-Joo Hwang,
Quoc-Viet Pham
Abstract:
The emergence of new services and applications in emerging wireless networks (e.g., beyond 5G and 6G) has shown a growing demand for the usage of artificial intelligence (AI) in the Internet of Things (IoT). However, the proliferation of massive IoT connections and the availability of computing resources distributed across future IoT systems have strongly demanded the development of distributed AI…
▽ More
The emergence of new services and applications in emerging wireless networks (e.g., beyond 5G and 6G) has shown a growing demand for the usage of artificial intelligence (AI) in the Internet of Things (IoT). However, the proliferation of massive IoT connections and the availability of computing resources distributed across future IoT systems have strongly demanded the development of distributed AI for better IoT services and applications. Therefore, existing AI-enabled IoT systems can be enhanced by implementing distributed machine learning (aka distributed learning) approaches. This work aims to provide a comprehensive survey on distributed learning for IoT services and applications in emerging networks. In particular, we first provide a background of machine learning and present a preliminary to typical distributed learning approaches, such as federated learning, multi-agent reinforcement learning, and distributed inference. Then, we provide an extensive review of distributed learning for critical IoT services (e.g., data sharing and computation offloading, localization, mobile crowdsensing, and security and privacy) and IoT applications (e.g., smart healthcare, smart grid, autonomous vehicle, aerial IoT networks, and smart industry). From the reviewed literature, we also present critical challenges of distributed learning for IoT and propose several promising solutions and research directions in this emerging area.
△ Less
Submitted 16 October, 2023;
originally announced October 2023.
-
RIS-assisted UAV Communications for IoT with Wireless Power Transfer Using Deep Reinforcement Learning
Authors:
Khoi Khac Nguyen,
Antonino Masaracchia,
Tan Do-Duy,
H. Vincent Poor,
Trung Q. Duong
Abstract:
Many of the devices used in Internet-of-Things (IoT) applications are energy-limited, and thus supplying energy while maintaining seamless connectivity for IoT devices is of considerable importance. In this context, we propose a simultaneous wireless power transfer and information transmission scheme for IoT devices with support from reconfigurable intelligent surface (RIS)-aided unmanned aerial v…
▽ More
Many of the devices used in Internet-of-Things (IoT) applications are energy-limited, and thus supplying energy while maintaining seamless connectivity for IoT devices is of considerable importance. In this context, we propose a simultaneous wireless power transfer and information transmission scheme for IoT devices with support from reconfigurable intelligent surface (RIS)-aided unmanned aerial vehicle (UAV) communications. In particular, in a first phase, IoT devices harvest energy from the UAV through wireless power transfer; and then in a second phase, the UAV collects data from the IoT devices through information transmission. To characterise the agility of the UAV, we consider two scenarios: a hovering UAV and a mobile UAV. Aiming at maximizing the total network sum-rate, we jointly optimize the trajectory of the UAV, the energy harvesting scheduling of IoT devices, and the phaseshift matrix of the RIS. We formulate a Markov decision process and propose two deep reinforcement learning algorithms to solve the optimization problem of maximizing the total network sum-rate. Numerical results illustrate the effectiveness of the UAV's flying path optimization and the network's throughput of our proposed techniques compared with other benchmark schemes. Given the strict requirements of the RIS and UAV, the significant improvement in processing time and throughput performance demonstrates that our proposed scheme is well applicable for practical IoT applications.
△ Less
Submitted 5 August, 2021;
originally announced August 2021.
-
3D UAV Trajectory and Data Collection Optimisation via Deep Reinforcement Learning
Authors:
Khoi Khac Nguyen,
Trung Q. Duong,
Tan Do-Duy,
Holger Claussen,
and Lajos Hanzo
Abstract:
Unmanned aerial vehicles (UAVs) are now beginning to be deployed for enhancing the network performance and coverage in wireless communication. However, due to the limitation of their on-board power and flight time, it is challenging to obtain an optimal resource allocation scheme for the UAV-assisted Internet of Things (IoT). In this paper, we design a new UAV-assisted IoT systems relying on the s…
▽ More
Unmanned aerial vehicles (UAVs) are now beginning to be deployed for enhancing the network performance and coverage in wireless communication. However, due to the limitation of their on-board power and flight time, it is challenging to obtain an optimal resource allocation scheme for the UAV-assisted Internet of Things (IoT). In this paper, we design a new UAV-assisted IoT systems relying on the shortest flight path of the UAVs while maximising the amount of data collected from IoT devices. Then, a deep reinforcement learning-based technique is conceived for finding the optimal trajectory and throughput in a specific coverage area. After training, the UAV has the ability to autonomously collect all the data from user nodes at a significant total sum-rate improvement while minimising the associated resources used. Numerical results are provided to highlight how our techniques strike a balance between the throughput attained, trajectory, and the time spent. More explicitly, we characterise the attainable performance in terms of the UAV trajectory, the expected reward and the total sum-rate.
△ Less
Submitted 6 June, 2021;
originally announced June 2021.
-
Finite-length performance comparison of network codes using random vs Pascal matrices
Authors:
Tan Do-Duy,
M. Angeles Vazquez-Castro
Abstract:
In this letter, we evaluate the finite-length performance of network coding when using either random or structured encoding matrices. First, we present our novel construction of structured network codes over Fq (q = 2^m) using Pascal matrices. We present their encoding, re-encoding and decoding in matrix notation and derive their packet loss rate. Second, we propose a novel methodology to compute…
▽ More
In this letter, we evaluate the finite-length performance of network coding when using either random or structured encoding matrices. First, we present our novel construction of structured network codes over Fq (q = 2^m) using Pascal matrices. We present their encoding, re-encoding and decoding in matrix notation and derive their packet loss rate. Second, we propose a novel methodology to compute the optimal finite-length coding rate for representative and realistic traffic applications. Finally, our method allows to compare the performance of our codes with the performance of popular random codes. We show that our constructions always have better throughput and minimal overhead, which is more significant for short code lengths. Further, their larger decoding delay fulfils the delay constraints of realistic scenarios (e.g. 5G multihop networks).
△ Less
Submitted 27 November, 2019;
originally announced November 2019.
-
Optimal Finite Length Coding Rate of Random Linear Network Coding Schemes
Authors:
Tan Do-Duy,
M. Ángeles Vázquez-Castro
Abstract:
In this paper, we propose a methodology to compute the optimal finite-length coding rate for random linear network coding schemes over a line network. To do so, we first model the encoding, reencoding, and decoding process of different coding schemes in matrix notation and corresponding error probabilities. Specifically, we model the finite-length performance for random linear capacity-achieving s…
▽ More
In this paper, we propose a methodology to compute the optimal finite-length coding rate for random linear network coding schemes over a line network. To do so, we first model the encoding, reencoding, and decoding process of different coding schemes in matrix notation and corresponding error probabilities. Specifically, we model the finite-length performance for random linear capacity-achieving schemes: non-systematic (RLNC) and systematic (SNC) and non-capacity achieving schemes: SNC with packet scheduling (SNC-S) or sliding window (SWNC). Then, we propose a binary searching algorithm to identify optimal coding rate for given target packet loss rate. We use our proposed method to obtain the region of exponential increase of optimal coding rate and corresponding slopes for representative types of traffic and erasure rates. Our results show the tradeoff for capacity-achieving codes vs non-capacity achieving schemes, since the latter trade throughput with delay, which is reflected in the decrease of the exponential slope with the blocklength. We also show the effect of the number of re-encoding times, which further decreases the slope.
△ Less
Submitted 15 May, 2018;
originally announced May 2018.
-
Geo-Network Coding Function Virtualization for Reliable Communication over Satellite
Authors:
Tan Do-Duy,
M. Angeles Vazquez Castro
Abstract:
In this paper, we propose a design solution for the implementation of Virtualized Network Coding Functionality (VNCF) over a service coverage area. Network Function Virtualization (NFV) and Network Coding (NC) architectural designs are integrated as a toolbox of NC design domains so that NC can be implemented over different underlying physical networks including satellite or hybrid networks.
The…
▽ More
In this paper, we propose a design solution for the implementation of Virtualized Network Coding Functionality (VNCF) over a service coverage area. Network Function Virtualization (NFV) and Network Coding (NC) architectural designs are integrated as a toolbox of NC design domains so that NC can be implemented over different underlying physical networks including satellite or hybrid networks.
The design includes identifying theoretical limits of NC over wireless networks in terms of achievable rate region and optimizing coding rates for nodes that implement VNCF. The overall design target is to achieve a given multicast transmission target reliability at receiver sides. In addition, the optimization problem uses databases with geo-tagged link statistics and geo-location information of network nodes in the deployment area for some computational complexity/energy constraints.
Numerical results provide validation of our design solution on how network conditions and system constraints impact the design and implementation of NC and how VNCF allows reliable communication over wireless networks with reliability and connectivity up to theoretical limits.
△ Less
Submitted 12 March, 2018;
originally announced March 2018.
-
Network Coding Function Virtualization
Authors:
Tan Do-Duy,
M. Angeles Vazquez Castro
Abstract:
Network Functions Virtualization (NFV) and Network Coding (NC) have attracted much attention in recent years as key concepts for providing 5G networks with flexibility and differentiated reliability, respectively. In this paper, we present the integration of NC architectural design and NFV. In order to do so we first describe what we call a virtualization process upon our proposed architectural de…
▽ More
Network Functions Virtualization (NFV) and Network Coding (NC) have attracted much attention in recent years as key concepts for providing 5G networks with flexibility and differentiated reliability, respectively. In this paper, we present the integration of NC architectural design and NFV. In order to do so we first describe what we call a virtualization process upon our proposed architectural design of NC that should help to offer the reliability functionality to a network. The process consists of identifying the required functional entities of NC and analyzing when the functionality should be activated towards complexity/energy efficiency. The relevance of our proposed NC function virtualization is its applicability to any underlying physical network, satellite or hybrid thus enabling softwarization, and rapid innovative deployment. Finally, we validate our framework to a study case of geo-control of network reliability that is based on device's geographical location-based signal/network information.
△ Less
Submitted 12 March, 2018;
originally announced March 2018.
-
Design of Virtualized Network Coding Functionality for Reliability Control of Communication Services over Satellite
Authors:
Tan Do-Duy,
M. A. Vazquez-Castro
Abstract:
Network coding (NC) is a novel coding technology that can be seen as a generalization of classic point-to-point coding. As with classic coding, both information theoretical and algebraic views bring different and complementary insights of NC benefits and corresponding tradeoffs. However, the multi-user nature of NC and its inherent applicability across all layers of the protocol stack, call for no…
▽ More
Network coding (NC) is a novel coding technology that can be seen as a generalization of classic point-to-point coding. As with classic coding, both information theoretical and algebraic views bring different and complementary insights of NC benefits and corresponding tradeoffs. However, the multi-user nature of NC and its inherent applicability across all layers of the protocol stack, call for novel design approaches towards efficient practical implementation of this technology.
In this paper, we present a possible way forward to the design of NC as a virtual network functionality offered to the communication service designer. Specifically, we propose the integration of NC and Network Function Virtualization (NFV) architectural designs. The integration is realized as a toolbox of NC design domains that the service designer can use for flow engineering. Our proposed design framework combines network protocol-driven design and system modular-driven design approaches. In particular, the adaptive choice of the network codes and its use for a specific service can then be tailored and optimized depending on the ultimate service intent and underlying (virtualized) system or network.
We work out a complete use case where we design geo-network coding, an application of NC for which coding rate is optimized using databases of geo-location information towards an energy-efficient use of resources. Our numerical results highlight the benefits of both the proposed NC design framework and the specific application.
△ Less
Submitted 12 March, 2018;
originally announced March 2018.
-
Efficient Communication over Cellular Networks with Network Coding in Emergency Scenarios
Authors:
Tan Do-Duy,
M. Angeles Vazquez Castro
Abstract:
Emergency communications requires reliability and flexibility for disaster recovery and relief operation. Based upon existing commercial portable devices (e.g., smartphones, tablets, laptops), we propose a network architecture that uses cellular networks and WiFi connections to deliver large files in emergency scenarios under the impairments of wireless channel such as packet losses and intermitte…
▽ More
Emergency communications requires reliability and flexibility for disaster recovery and relief operation. Based upon existing commercial portable devices (e.g., smartphones, tablets, laptops), we propose a network architecture that uses cellular networks and WiFi connections to deliver large files in emergency scenarios under the impairments of wireless channel such as packet losses and intermittent connection issues. Network coding (NC) is exploited to improve the delivery probability. We first review the state-of-the-art of NC for emergency communications. Then, we present the proposed network architecture which utilizes multiple radio interfaces of portable devices to support data delivery. A random linear NC scheme is exploited at source to enhance the reliability for content delivery against packet losses. Besides, an analytical model for the successful decoding probability in linear NC is derived. Finally, we evaluate the effectiveness of the proposed architecture with NC in terms of the delivery ratio of content for intermittent connectivity scenarios.
△ Less
Submitted 12 March, 2018;
originally announced March 2018.
