Optimized heteronuclear-ion quantum demodulator with inverse-quartic and inverse-quadratic temporal scalings

Efficient and precise measurement of electromagnetic signals is crucial for both fundamental science and practical applications. Demodulation is a widely used technique for receiving and recovering modulated signals. Here we show a quantum demodulator using two heteronuclear ions to simultaneously detect the amplitudes and frequencies of unknown periodic signals, overcoming the constraint of double-parameter optimization. With reinforcement learning, the quantum Fisher information scales as t4 for frequency estimation and as t2 for amplitude estimation, achieving inverse-quartic and inverse-quadratic temporal scaling, respectively. This work demonstrates the potential of heteronuclear ions in efficient quantum sensing and highlights the role of reinforcement learning in advancing quantum control. Efficient and precise measurement of electromagnetic signals is critical for both fundamental science and practical applications, yet traditional techniques often face inherent constraints in multi-parameter optimization. Here, the authors experimentally realize a quantum demodulator using two heteronuclear ions, achieving inverse-quartic and inverse-quadratic temporal scaling for frequency and amplitude estimation, respectively, thereby highlighting the power of reinforcement learning in advancing quantum sensing technologies.