Machine Learning Approach to RF Transmitter Identification

With the increasing domain and widespread use of wireless devices in recent years (mobile phones, Internet of Things, Wi-Fi), the electromagnetic spectrum has become extremely crowded. To counter security threats posed by rogue or unknown transmitters, we must identify RF transmitters not only by the data content of the transmissions but also based on the intrinsic physical characteristics of the transmitters. RF waveforms represent a particular challenge because of the extremely high data rates involved and the potentially large number of transmitters sharing a channel in a given location. These factors outline the need for rapid fingerprinting and identification methods that go beyond the traditional hand-engineered approaches. In this paper, we investigate the use of machine learning strategies to the classification and identification problem. We evaluate four different strategies: conventional deep neural nets, convolutional neural nets, support vector machines, and deep neural nets with multi-stage training. The latter was by far the most accurate, achieving 100% classification accuracy of 12 transmitters, and showing remarkable potential for scalability to large transmitter populations.