Federated Meta-Learning Enhanced Acoustic Radio Cooperative Framework for Ocean of Things

  title={Federated Meta-Learning Enhanced Acoustic Radio Cooperative Framework for Ocean of Things},
  author={Hao Zhao and Fei Ji and Qiang Li and Quansheng Guan and Shuai Wang and Miaowen Wen},
  journal={IEEE Journal of Selected Topics in Signal Processing},
  • Hao Zhao, Fei Ji, Miaowen Wen
  • Published 24 May 2021
  • Computer Science
  • IEEE Journal of Selected Topics in Signal Processing
Ocean of Things, consisting of multiple buoys distributed on the sea, is an acoustic radio cooperative wireless network that aims to acquire underwater information. In this paper, a deep neural network (DNN)-based receiver with data augmentation, termed chirp (C)-DNN, is developed for a buoy that uses chirp modulation-based underwater acoustic communications. To further solve the problem that the training data at a single buoy may not be sufficient, a federated meta-learning (FML) scheme is… 


Multiagent DDPG-Based Deep Learning for Smart Ocean Federated Learning IoT Networks
A novel multiagent deep reinforcement learning-based algorithm which can realize federated learning (FL) computation with Internet-of-Underwater-Things (IoUT) devices in the ocean environment and achieves 80% and 41% performance improvements than the standard actor–critic and DDPG, respectively, in terms of the downlink throughput.
Deep Transfer Learning-Based Downlink Channel Prediction for FDD Massive MIMO Systems
This work forms the downlink channel prediction as a deep transfer learning (DTL) problem, and proposes the direct-transfer algorithm based on the fully-connected neural network architecture, where the network is trained in the manner of classical deep learning and is fine-tuned for new environments.
DeepOcean: A General Deep Learning Framework for Spatio-Temporal Ocean Sensing Data Prediction
DeepOcean, a deep learning framework for spatio-temporal ocean sensing data prediction, which consists of a generative module and a prediction module, is proposed and evaluated with Argo data, where the proposed framework outperforms fifteen state-of-art baselines in terms of accuracy.
ComNet: Combination of Deep Learning and Expert Knowledge in OFDM Receivers
The proposed model-driven DL receiver offers more accurate channel estimation compared with the linear minimum mean-squared error method and exhibits higher data recovery accuracy comparing with the existing methods and FC-DNN.
Distance-Vector-Based Opportunistic Routing for Underwater Acoustic Sensor Networks
Simulation results show that DVOR outperforms the existing routing protocols in terms of packet delivery ratio, energy-efficiency, and average end-to-end delay.
OFDM for Underwater Acoustic Communications
A blend of introductory material and advanced signal processing and communication techniques, of critical importance to underwater system and network developmentThis book, which is the first to
Doppler-Resilient Orthogonal Signal-Division Multiplexing for Underwater Acoustic Communication
D-OSDM can provide low-power and high-quality UWA communications in channels with large delay and Doppler spreads and is found to become a powerful communication tool for underwater operations.
Exploiting Spatial–Temporal Joint Sparsity for Underwater Acoustic Multiple-Input–Multiple-Output Communications
Novel interference cancellation (IC) methods for handling co-channel interference problem in the design of both channel estimation and channel equalization are proposed and show that the proposed methods obtain higher output signal-to-noise ratio, lower bit error rate, and more separated constellations compared with the traditional compressed sensing channel estimation method and the traditional CE-DFE method.
Underwater acoustic communication and the general performance evaluation criteria
Characteristics of underwater acoustic channels are first introduced and compared with terrestrial communication to demonstrate the difficulties in UAC research and three criteria are presented based on the research publications and years of experience in high-rate short- to medium-range communications.
Scheduling Policies for Federated Learning in Wireless Networks
An analytical model is developed to characterize the performance of federated learning in wireless networks and shows that running FL with PF outperforms RS and RR if the network is operating under a high signal-to-interference-plus-noise ratio (SINR) threshold, while RR is more preferable when the SINR threshold is low.