Signal theory based encryption of faster-than-Nyquist signals for fiber and wireless transmission

New applications such as the Internet of Things, autonomous driving, Industry X.0 and many more will transmit sensitive information via fibers and over the air with envisioned data rates beyond terabits per second. Therefore, the encryption has to be simple, fast and spectrally efficient, so that the power consumption and latency are low and the scarce bandwidth is not wasted. Various encryption schemes, based on mathematical algorithms, quantum theory, chaos communication or spectral spreading below the noise level have been explored. Besides power, spectral efficiency and latency, most of these approaches face additional challenges such as limited data rates, compatibility issues with communication standards and integration. Here, we propose a signal theory based method that enables the encryption of super-signals with bandwidths of hundreds of gigahertz without any additional bandwidth. In proof-of-concept experiments we demonstrate the encryption of a 270 GBd faster than Nyquist super-signal in a 252.4 GHz bandwidth. The encryption is simple, fast and power efficient, and offers a solution for secure data transmission in existing and future communication networks.