On-Off Keying Signal Detection Based on Hidden Markov Model for Rydberg Atomic Sensor

Rydberg atomic sensors have been seen as novel radio frequency (RF) measurements and the high sensitivity to a large range of frequencies makes it attractive for communications reception. In high-speed symbol transmission scenarios, the effect of rising and falling edges as symbol switching should not be ignored. In this work, we adopt a mixed Gaussian distribution to characterize the output distribution under symbol switching, especially under higher signal power and shorter symbol duration according to experimental measurements. Under high symbol rate, unequal rising and falling edges on signal detection make the system nonlinear. Based on the experimental measurement, we characterize such nonlinear effects via a state transition model, and adopt Hidden Markov Model (HMM) to characterize the received signal. We propose a Monte-Carlo method to compute the achievable transmission rate and bit error rate (BER) performance. In real experiments, a lower BER can be achieved by the Viterbi algorithm compared with single symbol detection (SSD). Moreover, the performance of Viterbi decoding is better with higher symbol rate compared with SSD.