.png){kind=icon}
.png){kind=icon}
Return
Novel Cooperative Automatic Modulation Classification Using Unmanned Aerial Vehicles
.png){kind=icon}
PRE
AI
DOI：10.1109/JSEN.2021.3123048.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Automatic modulation classification (AMC) has been intriguing many researchers as it has many civil and military applications. Recently, cooperative AMC (CAMC) using a dynamic or ad hoc sensor network becomes appealing and challenging. As the unmanned aerial vehicles (UAVs) can facilitate three-dimensional communication/sensor network, we propose a novel CAMC approach based on a dynamic (ad hoc) UAV network. In our proposed new CAMC approach, the local classification decisions, which are made by spatially distributed nodes (UAVs) using our previously proposed graph-based modulation classifier, are gathered to reach an overall decision by a new weighted voting mechanism pertinent to individual received signal qualities. Note that the fusion center does not have to be a fixed UAV and it can be dynamically reassigned to any UAV within the same network in each sensing interval. The corresponding weights to individual UAVs are to be determined according to their cumulative states and the temporal discount factor. As a result, our proposed new CAMC approach can be fully distributed as no control center (or hub) is necessary. Besides, our new CAMC scheme can accommodate realistic ad hoc network variations to allow the existing UAVs to depart and/or the new UAVs to join in any sensing interval. Monte Carlo simulation results demonstrate that our proposed new CAMC scheme is quite robust and outperforms the existing CAMC method.
Keywords:
Sensors
Modulation
Unmanned aerial vehicles
Vehicle dynamics
Synchronization
Monte Carlo methods
Feature extraction
Cooperative automatic modulation classification (CAMC)
decision-level fusion
graph-based automatic modulation classification
temporary fusion center (TFC)
unmanned aerial vehicle (UAV)
weighted voting mechanism
Journal
IF:
4.5
Papers:
2.1W
Citations:
7.3W