Adaptive Communications and Trajectory Design for UAV-assisted Wireless Networks: a Multi-Scale Decision Framework

Prof. Nicolò Michelusi, Arizona State University

Website: https://faculty.engineering.asu.edu/michelusi/

Email: nicolo.michelusi@asu.edu

The demand for wireless broadband is growing in the United States and across the world. Unmanned aerial vehicles (UAVs) are envisioned as key components of 5G wireless technology and beyond: thanks to their low cost, improved line-of-sight over terrestrial base stations, and controllable mobility, they will enable low-cost wireless broadband access. Nonetheless, UAVs’ integration into wireless networks poses unique challenges on the network and physical layers, due to the intricate coupling between trajectory design and communication resources to be jointly optimized, and uncertain air-to-ground channel propagation conditions. Furthermore, UAVs need to seamlessly operate under sources of randomness and uncertainty typical of wireless networks. This project aims to design techniques to enable real-time physical-layer adaptation of the communication resources, and adaptive trajectory designs to optimize communication performance and energy-efficiency of the system. This research addresses the global industrial and societal need for ubiquitous wireless broadband access by enabling a cost-effective integration of UAVs into wireless networks. This research integrates an educational and outreach program designed to foster research interests and participation of underrepresented students in electrical engineering, through activities created in collaboration with programs at ASU and local high schools.

This project develops a novel decision-making framework to address the critical need for adaptation in UAV-assisted wireless networks operating under uncertainty. Adaptive techniques are developed that leverage the high mobility of UAVs to optimize communication metrics such as latency, throughput, outage probability, area spectral efficiency, energy efficiency, by focusing on the interplay between network-level optimization and physical-layer communication, trajectory design, and control. A key novelty is a multi-scale decision framework to achieve scalable design. The framework leverages multiple spatio-temporal scales induced by the coupling between trajectory and channel propagation conditions to centralize slow timescale trajectory decisions and decentralize fast timescale communications decisions. The design aspect leverages unique features of single- and multi-antennas, operating at sub-6GHz or millimeter-wave frequencies, and provides adaptation to uncertain and dynamic channel conditions. The second goal consists of designing adaptive multi-UAV wireless systems, including UAV selection, user association, resource allocation, optimal charging schedules to enable uninterrupted operation, and contention-based access schemes to improve coverage and grant-free access. The research results are tested experimentally on NSF PAWR AERPAW by designing a software-defined-radio implementation. The experimental results are integrated into theoretical models for continuous improvement and testing.

This award reflects NSF’s statutory mission and has been deemed worthy of support through evaluation using the Foundation’s intellectual merit and broader impacts review criteria.

The demand for connectivity is growing across the world. Access to high-rate wireless is becoming critical for a stable society due to its key role in education, economic and government activities, political discourse, and basic utilities. This research addresses the global need for ubiquitous wireless broadband access, by enabling a cost-effective integration of UAVs into wireless networks. This research will have a technological and economic impact by allowing high throughput, low-cost wireless broadband access to be deployed worldwide without expensive infrastructure. An integrated education and outreach program is designed to foster the interest in research and to increase the participation of underrepresented students in electrical engineering through activities created in collaboration with programs at ASU and high schools in the Phoenix area. These activities include: (1) integration of research material into undergraduate and graduate course offerings and high-school curricula; (2) research training for students at the graduate and undergraduate levels.

Advanced wireless communications are already impacting nearly every human being. Some areas that have not historically been impacted by communication and networking advances are starting to become influenced. These include the automotive and agricultural industries. Our work on UAV-aided communications will have a transformational impact on the next generation of wireless communications that will likely indirectly influence a variety of industries, including the automotive and agricultural domains.

The ultimate goal of this research is to provide technical solutions to meet the demanding challenges of 5G (1000x increase in capacity, 10x reduction in latency, 100x increase in density of mobile devices). Thus, our research is likely to contribute to the success of 5G and will benefit society through the plethora of applications that 5G promises -- Internet of things to improve automation of transportation systems, home management for instance; supporting the increase in demand for mobile high throughput services such as virtual reality; lower the cost of deployment of communication systems, and thus provide benefits to rural or underdeveloped areas with limited access to the internet.

JOURNAL PUBLICATIONS

• 1. N. Michelusi, "Non-Coherent Over-the-Air Decentralized Gradient Descent," in IEEE Transactions on Signal Processing, vol. 72, pp. 4618-4634, 2024, doi: 10.1109/TSP.2024.3460690.

This paper presents a scalable Decentralized Gradient Descent (DGD) algorithm for wireless systems that eliminates the need for scheduling, topology, and CSI, using a Non-Coherent Over-The-Air (NCOTA) consensus scheme. Nodes transmit simultaneously, and average channel gains are used for unbiased consensus estimation, achieving O(1/√k) convergence. Numerical results show faster convergence than existing methods, especially in dense networks, with extensions for various fading models. 

• 2. B. Keshavamurthy, Y. Zhang, C. R. Anderson, N. Michelusi, D. J. Love, and J. V. Krogmeier,  "Statistical Characterization of 28GHz V2X Channels via Autonomous Beam-Steered Measurements," submitted to the IEEE Transactions on Antennas and Propagation, under submission

This paper performs a comprehensive analysis of millimeter wave propagations in V2X applications across a variety of deployment sites (urban suburban, foliage), based on experimental measurements collected on the NSF POWDER testbed using an autonomous beam-steered measurement platform.

• 3. F.P.C. Lin, S. Hosseinalipour, N. Michelusi and C. G. Brinton, "Delay-Aware Hierarchical Federated Learning," in IEEE Transactions on Cognitive Communications and Networking, vol. 10, no. 2, pp. 674-688, April 2024, doi: 10.1109/TCCN.2023.3329024.

This paper presents delay-aware hierarchical federated learning (DFL), which improves distributed ML by addressing communication delays and enhancing learning efficiency. DFL uses local stochastic gradient descent iterations and periodic model aggregation through edge servers, achieving O(1/k) convergence. Numerical results show faster convergence, reduced resource consumption, and robustness to delays compared to traditional federated learning.

• 4. B. Keshavamurthy, M. Bliss, N. Michelusi, “MAESTRO-X: Distributed Orchestration of Rotary-Wing UAV Relay Swarms“, in IEEE Transactions on Cognitive Communications and Networking, vol. 9, no. 3, pp. 794-810, June 2023, doi: 10.1109/TCCN.2023.3248859. 

This paper develops a framework to orchestrate a swarm of UAVs serving as wireless relays for ground users. It aims to minimize the service delay subject to average power constraints of UAVs.

• 5. T.H. Chou, N. Michelusi, D. J. Love, J. V. Krogmeier, "Compressed Training for Dual-Wideband Time-Varying Sub-Terahertz Massive MIMO", in IEEE Transactions on Communications, vol. 71, no. 6, pp. 3559-3575, June 2023, doi: 10.1109/TCOMM.2023.3247789. 

The paper develops pilot-efficient channel estimation strategies to enable Sub-Terahertz communications with Massive MIMO.

• 6. B. Keshavamurthy, N. Michelusi, "Learning-Based Spectrum Sensing and Access in Cognitive Radios via Approximate POMDPs," in IEEE Transactions on Cognitive Communications and Networking, vol. 8, no. 2, pp. 514-528, June 2022, doi: 10.1109/TCCN.2021.3129802

The paper develops efficient spectrum sensing and access strategies, to enable efficient use of spectrum resources in crowded wireless systems. It could enable a swarm of UAVs to communicate reliably while generating minimal interference to nearby systems. We study a use case involving a military deployment in the "Alleys of Austin" scenario of DARPA SC2, involving communications between 9 guardsmen and one UAV.

• 7. N. Michelusi, G. Scutari, C.S. Lee, “Finite-Bit Quantization For Distributed Algorithms With Linear Convergence“, in IEEE Transactions on Information Theory, vol. 68, no. 11, pp. 7254-7280, Nov. 2022, doi: 10.1109/TIT.2022.3176253.

This paper develops a black-box quantization scheme to enable the execution of distributed optimization schemes, in which distributed agents communicate over finite-capacity wireless channels. This scheme could enable a swarm of UAVs to efficiently coordinate their operations and make real-time control decisions.

• 8. S. Hosseinalipour, S. S. Azam, C. G. Brinton, N. Michelusi, V. Aggarwal, D. J. Love, and H. Dai, "Multi-Stage Hybrid Federated Learning Over Large-Scale D2D-Enabled Fog Networks," in IEEE/ACM Transactions on Networking, vol. 30, no. 4, pp. 1569-1584, Aug. 2022, doi: 10.1109/TNET.2022.3143495.

This paper develops multi-stage hybrid federated learning, a hybrid of intra- and inter-layer model learning that considers the network as a multi-layer cluster-based structure. The design solutions developed in this paper could foster the integration of mobile networks into the learning architecture, including UAVs collecting data, such as aerial photos and videos used to solve a machine learning task.

• 9. M. Hussain, N. Michelusi, "Learning and Adaptation for Millimeter-Wave Beam Tracking and Training: a Dual Timescale Variational Framework", in IEEE Journal on Selected Areas in Communications, vol. 40, no. 1, pp. 37-53, Jan. 2022, doi: 10.1109/JSAC.2021.3126086

This paper develops efficient beam-alignment schemes in millimeter-wave systems, by learning the mobility pattern of beam dynamics. The solutions developed in this paper could be used to communicate reliably at mm-wave frequencies with highly-mobile UAVs.

• 10. F.P.C. Lin, S Hosseinalipour, S.S. Azam, C. G. Brinton, N. Michelusi, "Semi-Decentralized Federated Learning with Cooperative D2D Local Model Aggregations," in IEEE Journal on Selected Areas in Communications, vol. 39, no. 12, pp. 3851-3869, Dec. 2021, doi: 10.1109/JSAC.2021.3118344

This paper develops a semi-decentralized federated learning architecture. The design solutions developed in this paper could foster the integration of mobile networks into the learning architecture, including UAVs collecting data, such as aerial photos and videos used to solve a machine learning task. 

CONFERENCE PUBLICATIONS

1. B. Keshavamurthy, and N. Michelusi, "Adaptive Control of MIMO UAVs for Prioritized Harvesting via Hierarchical Reinforcement Learning," Asilomar 2024 (invited paper).

This paper presents a hierarchical reinforcement learning-based approach for optimizing the control and coordination of a fleet of MIMO-capable UAVs to efficiently harvest prioritized traffic from heterogeneous users while maximizing rewards subject to mobility and power constraints.

2. N. Michelusi, "CSI-Free Over-The-Air Decentralized Learning Over Frequency Selective Channels," ICASSP 2024 - IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Seoul, Korea, Republic of, 2024, pp. 13076-13080, doi: 10.1109/ICASSP48485.2024.10447686.

This paper introduces a decentralized learning scheme for serverless systems operating over frequency-selective channels, utilizing waveform superposition in OFDM to enable CSI-free operation and unbiased estimation of signals for decentralized gradient descent.

3. M. F. U. Abrar and N. Michelusi, "Biased Over-the-Air Federated Learning under Wireless Heterogeneity," IEEE ICC Workshops, Denver, CO, USA, 2024, pp. 111-116, doi: 10.1109/ICCWorkshops59551.2024.10615488.

This paper investigates the design of OTA device pre-scalers for federated learning in heterogeneous wireless environments, optimizing the trade-off between bias and variance in model updates to improve convergence and outperform existing schemes.

4. B. Keshavamurthy and N. Michelusi, "Orchestrating UAVs for Prioritized Data Harvesting: A Cross-Layer Optimization Perspective," 2024 IEEE International Conference on Communications Workshops (ICC Workshops), Denver, CO, USA, 2024, pp. 1268-1273, doi: 10.1109/ICCWorkshops59551.2024.10615846.

This paper develops a framework to optimize the operation of MIMO-enabled multi-UAV systems to accommodate diverse demands and quality of service requirements. The problem is formulated as a cumulative fleet-wide reward maximization problem, considering a 3D mobility power consumption constraint on each UAV, and optimized via cross-layer optimization techniques.

5. A. S. Bondre, C. D. Richmond and N. Michelusi, "Delay-Doppler Parameter Estimation for DFRC-OTFS Using 2D Root-MUSIC," 2024 IEEE Radar Conference (RadarConf24), Denver, CO, USA, 2024, pp. 1-6, doi: 10.1109/RadarConf2458775.2024.10548156.

This paper develops a 2D MUSIC channel estimation algorithm for OTFS, accounting for multiple pilot sequences. This introduced an added level of flexibility in designing pilot waveforms for channel estimation, resembling radar techniques. We introduced the use of linear frequency modulated (LFM) chirp in an OTFS frame, demonstrating superior channel estimation accuracy compared to a single-pilot OTFS frame.

6. A. S. Bondre, C. D. Richmond, A. Alkhateeb, and N. Michelusi, "Sparse Delay Doppler Channel Estimation for OTFS Modulation using 2D MUSIC," IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Rhodes Island, Greece, 2023, pp. 1-5, doi: 10.1109/ICASSP49357.2023.10095505.

This paper develops channel estimation techniques for OTFS modulation in highly mobile networks, such as UAVs.

7. N. Michelusi, "Decentralized Federated Learning via Non-Coherent Over-the-Air Consensus," ICC 2023 - IEEE International Conference on Communications, Rome, Italy, 2023, pp. 3102-3107, doi: 10.1109/ICC45041.2023.10279097.

This paper develops a scheme that enables to solve decentralized machine learning and sensing tasks in a wireless ecosystem. The scheme developed relies on non-coherent over-the-air consensus to enable diffusion of information across the network: it does not require channel estimation at transmitters or receivers, and it relies on the waveform superposition property of the wireless channel, overcoming the need to schedule the transmission of users. It is therefore suitable to enable coordination in highly mobile UAV swarms.

8. B. Keshavamurthy, Y. Zhang, C. R. Anderson, N. Michelusi, D. J. Love, J. V. Krogmeier, "Propagation Measurements and Analyses at 28 GHz via an Autonomous Beam-Steering Platform," IEEE ICC 2023 (accepted)

This paper presents the analysis of channel propagation measurements performed on the NSF POWDER testbed. The results will improve our understanding of mm-wave wireless channels, and will aid the design of beam-tracking algorithms in highly mobile wireless systems.

9. B. Keshavamurthy, Y. Zhang, C. R. Anderson, N. Michelusi, D. J. Love and J. V. Krogmeier, "Propagation Measurements and Analyses at 28 GHz via an Autonomous Beam-Steering Platform," ICC 2023 - IEEE International Conference on Communications, Rome, Italy, 2023, pp. 5042-5047, doi: 10.1109/ICC45041.2023.10279397.

This paper develops a framework to orchestrate a swarm of UAVs serving as wireless relays for ground users. It aims to minimize the service delay subject to average power constraints of UAVs.

10. M. Hussain, N. Michelusi, "Adaptive Beam Alignment in Mm-Wave Networks: A Deep Variational Autoencoder Architecture," IEEE Global Communications Conference (GLOBECOM), 2021, pp. 1-6, doi: 10.1109/GLOBECOM46510.2021.9685969

This paper develops efficient beam-alignment schemes in millimeter-wave systems, by learning the mobility pattern of beam dynamics. The solutions developed in this paper could be used to communicate reliably at mm-wave frequencies with highly-mobile UAVs.

11. F.P.C. Lin, S. Hosseinalipour, S.S. Azam, C. G. Brinton, N. Michelusi, "Federated Learning Beyond the Star: Local D2D Model Consensus with Global Cluster Sampling," IEEE GLOBECOM 2021, pp. 1-6, doi: 10.1109/GLOBECOM46510.2021.9685456

This paper develops a semi-decentralized federated learning architecture. The design solutions developed in this paper could foster the integration of mobile networks into the learning architecture, including UAVs collecting data, such as aerial photos and videos used to solve a machine learning task. 

12. T.H. Chou, N. Michelusi, D. J. Love, J. V. Krogmeier, "Wideband Millimeter-Wave Massive MIMO Channel Training via Compressed Sensing," IEEE GLOBECOM 2021, pp. 1-6, doi: 10.1109/GLOBE- COM46510.2021.9685901

The paper develops pilot-efficient channel estimation strategies to enable Sub-Terahertz communications with Massive MIMO.

PHD DISSERTATIONS

Bharath Keshavamurthy, "Modeling and Optimization for Non-Terrestrial Networks", 2024

https://fullcircle.asu.edu/faculty/uav-supported-wireless-broadband-to-increase-remote-access/