Blockchain-Enabled Task Offloading With Energy Harvesting in Multi-UAV-Assisted IoT Networks: A Multi-Agent DRL Approach

Abstract

The integration of unmanned aerial vehicles (UAVs) with mobile edge computing (MEC) and Internet of Things (IoT) technology in smart farms is pivotal for efficient resource management and enhanced agricultural productivity sustainably. This paper addresses the critical need for optimizing task offloading in secure UAV-assisted smart farm networks, aiming to reduce total delay and energy consumption while maintaining robust security in data communications. We propose a multiagent deep reinforcement learning (DRL)-based approach using a deep double Q-network (DDQN) with an action mask (AM), designed to manage task offloading dynamically and efficiently. The simulation results demonstrate the superior performance of our method in managing task offloading, highlighting significant improvements in operational efficiency by reducing delay and energy consumption. This aligns with the goal of developing sustainable and energy-efficient solutions for next-generation network infrastructures, making our approach an advanced solution for achieving both performance and sustainability in smart farming applications.

A comprehensive review on internet of things task offloading in multi-access edge computing

“…The goal of resource allocation is addressed in approximately 10 % of the reviewed literature [ 99 , 103 , 105 , 111 , 116 , 121 ]. In addition, 48 % of the research work focuses on both task allocation and resource allocation issues [ 76 , 80 , 83 , 87 , 89 , 141 , 142 , 144 , 146 , 92 , 95 , 101 , 102 , 107 , [112] , [113] , [114] , [115] , [117] , [118] , [119] , [120] , [122] , [123] , [124] , [126] , [127] , [128] , [129] , [132] , [133] , [134] , [135] , [136] , 148 , 149 ]. The Distribution and overlay of IoT task offloading problems addressed in MEC are shown in Fig.…”

“…Device-to-edge (D2E) task offloading in a Multi-Access Edge Computing (MEC) network involves offloading a task from a user's device to an edge server or cloudlet located at the network's edge. This approach allows tasks to be processed closer to the source of data, reducing latency and improving the overall performance of applications [ 73 , 74 , 76 , 77 , 79 , 80 , 83 , 84 , [86] , [87] , [88] , [89] , [144] , [145] , [146] , [147] , 91 , 94 , 97 , 100 , [102] , [103] , [104] , 106 , 107 ], [ 113 , 114 , [116] , [117] , [118] , 120 , 123 , [126] , [127] , [128] , 130 , 132 , 135 , 137 , 148 , 160 , 161 , 167 ].…”

“…In computational offloading within a MEC network, a single server as a task offloading destination refers to the scenario where a task from a user's device is offloaded and executed on a single edge server or cloudlet. This means that the entire computational load of the task is handled by a single server [ 73 , 80 , 84 , 88 , 89 , 143 , 145 , 146 , 100 , 103 , 104 , 106 , 110 , 113 , 116 , 120 , 130 , 135 , [137] , [148] , [149] ].…”

“…[ 73 , 85 , 138 , 140 , 146 , 147 , 92 , 96 , [99] , [100] , [101] , [102] , 112 , 116 , 119 , 121 , 127 , 131 , 148 ]. In addition, Pytorch is introduced in some implementations to support machine learning algorithms involved in task offloading mechanisms [ 103 , 114 , 124 , 129 , 132 , [135] , [136] , [137] ].…”

MLRS-RL: An Energy-Efficient Multilevel Routing Strategy Based on Reinforcement Learning in Multimodal UWSNs

“…The computation rate maximization problems in a UAVenabled MEC wireless powered system are investigated under both partial and binary computation offloading modes, subject to the energy-harvesting causal constraint and the UAV's speed constraint. In [16], authors considered an online dynamic offloading and resource scheduling algorithm to address the stochastic optimization problem of minimizing energy and computing resource consumption of energy harvesting devices while meeting the quality of service requirements of IoT devices. However, while energy harvesting can extend the device's uptime, it does not mitigate the energy consumption of edge AI.…”