Towards 6G IoT: Tracing Mobile Sensor Nodes with Deep Learning Clustering in UAV Networks

Abstract

Unmanned aerial vehicles (UAVs) in the role of flying anchor nodes have been proposed to assist the localisation of terrestrial Internet of Things (IoT) sensors and provide relay services in the context of the upcoming 6G networks. This paper considered the objective of tracing a mobile IoT device of unknown location, using a group of UAVs that were equipped with received signal strength indicator (RSSI) sensors. The UAVs employed measurements of the target’s radio frequency (RF) signal power to approach the target as quickly as possible. A deep learning model performed clustering in the UAV network at regular intervals, based on a graph convolutional network (GCN) architecture, which utilised information about the RSSI and the UAV positions. The number of clusters was determined dynamically at each instant using a heuristic method, and the partitions were determined by optimising an RSSI loss function. The proposed algorithm retained the clusters that approached the RF source more effectively, removing the rest of the UAVs, which returned to the base. Simulation experiments demonstrated the improvement of this method compared to a previous deterministic approach, in terms of the time required to reach the target and the total distance covered by the UAVs.

A Comprehensive Review on Artificial Intelligence/Machine Learning Algorithms for Empowering the Future IoT Toward 6G Era

“…Research topic Research analysis/findings Deep Learning [148] Network access and routing algorithm Survey on DL, supervised, reinforcement and imitation learning [149], [150] Indoor localization Localization error analysis [151] CSI estimation technique CSI overhead, channel measurement and sum rate analysis [152] DoA estimation Estimation accuracy analysis with the proposed, RVNN, SVR and MUSIC approaches [154] Power allocation strategy Analysis of secrecy rate, computation time and interference leakage [155] QoE forecasting mechanism Performance analysis of the proposed scheme against SVR, MLP, LSTM-based schemes [156] Anti-jamming scheme Throughput analysis Transformer algorithm [157] Medical image classification Classification accuracy analysis [158] Traffic sign recognition Classification accuracy analysis [159] Wildfire recognition and region detection Classification and detection accuracy analysis [160] Modulation recognition Classification and detection accuracy analysis [161] Intrusion detection Detection accuracy analysis Graph neural network [162] Topology control Network lifetime enhancement [163] IoT device tracking Tracking optimization in terms of execution time and distance covered by the tracking devices [164] Sentiment classification Interpretation accuracy of the aspect of text(s) [165] Vehicular traffic data prediction Prediction accuracy of the missing data from the available dataset recognition mechanism is designed in [158] with the help of DNN consisting of CNN and transformer-based algorithm.…”

Trajectory Design for UAV-Based Data Collection Using Clustering Model in Smart Farming

“…Instead of GPS, the RSSI of the received signal is used to locate objects on the ground. Similarly, Spyridis et al [ 27 ] proposed deep-learning-based clustering scheme to trace mobile sensor nodes using UAV. Ma et al [ 28 ] studied a data collection framework using UAV in an architecture-less environment where mobility changes over time.…”

A Real-Time Zanthoxylum Target Detection Method for an Intelligent Picking Robot under a Complex Background, Based on an Improved YOLOv5s Architecture

“…However, the above models cannot quickly and efficiently provide accurate positioning for picking robots in complex orchard environments. Thus, a highly robust target detection system based on computer vision and a fully autonomous automatic detection model of UAV [34][35][36] systems is of urgent need. On the other hand, though the identification efficiency of most Zanthoxylum target recognition research models based on deep learning is high, the timeliness and accuracy of the models is insufficient in complex orchard environments with different fruit sizes and serious branch clustering.…”