Quantum Sensing
EVER since the second quantum revolution, quantum sensing has emerged as a key enabler for numerous applications. Especially, quantum wireless sensing is envisioned to bring a new paradigm shift to wireless sensing. Unlike the traditional wireless sensing that exploits radio frequency (RF) components, the quantum wireless sensing exploits an atomic receiver for the electromagnetic (EM) wave reception. As shown in Fig. 1, the atomic receiver leverages the Rydberg atoms for detecting the RF signal’s amplitude, frequency, and phase. Replacing the dipole antenna of the classical receiver with the Rydberg atom offers several benefits, such as enhanced sensing granularity, long-range communications, and broadband tunability. To fully exploit the aforementioned potential, recent research on the atomic receiver has mainly focused on its integration with wireless communications. The atomic receiver is incorporated into the multiple-input-multiple-output (MIMO) for the signal detection of multi-user. Here, a biased Gerchberg-Saxton (GS) algorithm was proposed to solve the non-linear biased phase retrieval (PR) problem induced by the magnitude-only measurement of the atomic receiver.
Fig 1. Wireless signal processing based on an atomic receiver. When the EM wave is absorbed or emitted, the energy level of the electron is transferred to another energy level, which is known as atomic level transition
Despite these emerging attentions on the atomic receiver, its integration to the wireless sensing is still at the nascent stage. For example, authors employed the simple geometric relationship between a phase difference and angle-of-arrival (AoA) of a single target, which is not applicable for multi-user environments. Although several hardware demonstrations have validated the effectiveness of quantum wireless sensing, AoA estimation on multi-user is still challenging due to the absence of a signal processing algorithm for the atomic receiver. Motivated by this observation, in this letter, we propose a Quantum-MUSIC, multiple signal classification (MUSIC) for quantum wireless sensing. To the best of our knowledge, the proposed algorithm is the first signal processing algorithm for the quantum wireless sensing of multi-user. Here, the proposed algorithm first modifies the biased GS algorithm to retrieve the channel information from the magnitude-only measurement. Thereafter, applying the MUSIC algorithm enables accurate quantum wireless sensing in the multi-user environment. In this letter, we compare the sensing performance of traditional RF-based MUSIC with that of proposed Quantum-MUSIC and unveil the potential of the atomic receiver.
Fig 2. AoA RMSE versus (left) the number of users K, and (right) the number of antennas M
the AoA RMSE according to the number of users is presented in Fig. 2 (left), showing better performance compared to the MUSIC regardless of the number of users. Eventually, the impact of the number of antennas for quantum wireless sensing is shown in Fig 2 (right). The RMSE of both MUSIC. and Quantum-MUSIC decreases as M increases, but the Quantum-MUSIC outperforms the MUSIC.
we propose the Quantum-MUSIC for the quantum wireless sensing of multi-user. The proposed algorithm first recovers the channel from the magnitude-only measurement by the modified biased-GS algorithm. Thereafter, the traditional MUSIC is adopted for multi-user sensing. Through a comprehensive comparison, the Quantum-MUSIC shows the superiority compared to the classical MUSIC, revealing the potential of quantum wireless sensing and the atomic receiver. Future work is expected to be developing a signal processing algorithm for multi-band quantum wireless sensing, where multi-user transmit an extremely broad range of frequencies from megahertz to terahertz.
Multi-band wireless network (MBN) is envisioned to be a key enabler for next-generation communications and sensing. By utilizing multiple spectrums, MBN is expected to bring several advantages, such as higher capacity, reduced latency, and enhanced sensing performance. Motivated by these visions, multi-band wireless sensing has recently attracted extensive attention. However, the classical receiver is governed by the fundamental limitations induced by the multiple radio frequency (RF) components and dipole antennas, such as high implementation cost, band-limited processing, and mutual coupling effect. To overcome these limitations, quantum wireless sensing is a promising solution for multi-band systems. The core of quantum wireless sensing is the Rydberg atomic receiver, which leverages the ’Rydberg atom’ as atomic antennas.
Fig 3. Illustrations of Rydberg atom structure and energy level diagram for multi-band EM waves reception.
Rydberg atom structure and energy level diagram for multi-band electromagnetic (EM) waves reception is illustrated as Fig.1. The energy level of Rydberg atoms excites from the Rydberg state |3⟩ to other Rydberg states |4⟩, |5⟩ as the electromagnetic (EM) wave reception, where the transition intensity can be read out by the photodetector (PD) and optical-readout mechanisms.
Fig 4. Schematic diagram of the multi-band atomic receiver.
To fully leverage the Rydberg atom in Fig.3 for multi-band wireless sensing, we proposed multi-band atomic receiver as Fig.4. We assume that each single user utilizes different frequencies the atomic receiver comprises M vapor cells with an equivalent number of photodetectors (PDs) and lasers. Here, we deploy a number of B local oscillators (LOs) to transmit known reference signals, where the carrier frequencies of LOs are same as that of users. Thereafter, the transition intensity induced by the interaction between impinging EM waves and the Rydberg atoms, which is known as Rabi frequency can be read out by multiple PDs.
Utilizing Rydberg atomic receiver Fig.4., the goal of multi-band quantum wireless sensing is to estimate the AoAs of multi-user from the measurement. To achieve this objective, the proposed algorithm first recovers the channel matrix from the intensity-only measurement acquired by Rydberg atomic receivers. Thereafter, correlations between the retrieved channel matrix and the multi-band steering vectors are calculated to estimate the AoAs.
Fig 5. AoA RMSE versus the BBR.
We define the band-to-band ratio (BBR) in x-axis which is the ratio between low and high frequencies in the multi-band frequency set. The root mean square error (RMSE) is defined with estimated AoA and true AoA of each multiple users. We can observe that the proposed algorithm is superior across all BBR values compared to the other existing quantum and RF wireless sensing algorithms.
For future research, an extremely broad range of multi-band scenarios, such as MHz to THz, will be considered.