In this work, we propose the use of energy harvesting and semantic communication for Internet of Things (IoT) systems. The system allows IoT devices to harvest energy from a base station and then uses the harvested energy for extracting and transmitting semantic information, i.e., scene graphs, to the base station. The proposed network thus copes with the energy and network resource constraints of the IoT devices. To maximize the total image data or scene graph transmitted to the base station, we formulate a problem that optimizes the energy harvesting duration, the selection of original image or portions of scene graphs, transmit power, and channel allocation to the IoT devices. Under the high dynamics and uncertainty of the context and size of the collected images as well as the wireless channels and computing resources, we propose two advanced deep reinforcement learning (DRL) algorithms, i.e., advantage Actor-Critic (A2C) and proximal policy optimization (PPO), to solve the problem. Simulation results are implemented on the real dataset clearly showing that the performance achieved by the proposed algorithms is much higher than that achieved by the baseline scheme. This implies that more original images or triplets are transmitted.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Joint Energy Harvesting, Semantic Transmission Selection, Channel Allocation and Power Control for Resource-Constrained IoT Networks


    Contributors:


    Publication date :

    2024-06-24


    Size :

    337760 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Relay Selection and Resource Allocation for Energy Harvesting Cooperative Networks

    Deep, M. Prudhvi / Jain, Shubham / Ubaidulla, P. | IEEE | 2017


    Joint Computing Resource and Bandwidth Allocation for Semantic Communication Networks

    Zhao, Fangzhou / Bagwe, Gaurav / Mohammed, Ezedin et al. | IEEE | 2023



    OFDMA Constrained Resource Allocation with imperfect Channel Knowledge

    Brah, Felix / Vandendorpe, Luc / Louveaux, Jerome | IEEE | 2007


    Resource Allocation Strategy Based on RF Energy Harvesting in Heterogeneous Networks

    Gao, Jincheng / Zhao, Yisheng / Chen, Mengjia et al. | IEEE | 2019