Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Globecom

Dynamic Grover Search Optimization with Deep Q-Networks for Active User Detection

Deemah H. Tashman, Soumaya Cherkaoui

Published Dec 8, 2025• 1 citation
Publisher

Abstract

Sixth-generation (6G) networks must deliver ultra-low latency and near-100 percent reliability to support massive-scale Internet of Things (IoT) deployments and Hyper-Reliable Low-Latency Communications (HRLLC). Grant-free access protocols permit devices to transmit without prior scheduling; nevertheless, this uncoordinated transmission introduces uncertainty at the receiver, necessitating Active User Detection (AUD) to ascertain which devices are active. Quantum search methods—most notably Grover’s algorithm—can accelerate AUD, yet they require knowing the optimal number of iterations, which depends on the (typically unknown and time-varying) number of valid solutions induced by the current activity pattern and channel/noise conditions. To overcome this, we formulate an optimization problem that optimizes the number of Grover iterations to maximize detection accuracy and minimize computational cost without any prior activity information. We then apply a Deep Q-Network (DQN) to learn, via deep reinforcement learning, an adaptive policy for selecting the iteration count. Simulation results verify that the DQN converges to an optimal strategy and outperforms two baseline schemes under varying fading conditions and active-user transmit powers.

DOI
10.1109/globecom59602.2025.11432642

Related papers

Quantum Minimum Searching Algorithms for Active User Detection in Wireless IoT Networks

Muhammad Idham Habibie, Claire Goursaud, Jihad Hamie

The key features of 5G, such as ultra-reliable low latency (URLLC) and massive machine-type communication (mMTC), are designed to address the need for low latency and the ability to connect a large number of devices in the IoT context. To support these constraints, mobile devices transmit information without previously establishing a connection with the base station (BS). This requires the Base Station to detect in real-time the active users (process known as Active User Detection (AUD)). With classical processors, one can employ the Maximum Likelihood (ML) method (the optimal detector, but suffers from high complexity and delay), or suboptimal ones (which are simpler, but less reliable). Meanwhile, quantum algorithms, particularly Durr and Hoyer (DHA) algorithm, addressing minimum searching problems, can significantly reduce complexity while keeping good performances. However, these algorithms were designed for generic problems, and their initialization and parameterization are blindly done. Nonetheless, we can have access to prior information on the system’s behavior. Therefore, in this paper, we aim to adapt and improve these quantum algorithms by using prior knowledge on the system for the AUD problem. We first propose a novel algorithm, the Improved Iterative Minimum Searching Algorithm (IIMSA) where we define more efficiently the parameters. Then, further enhancements of IIMSA are obtained thanks to a better initialization of the algorithms by exploiting classical preprocessing of the received signals with classical Conventional Correlation Receiver (CCR) or Zero Forcing (ZF). The obtained results show that these proposed algorithms operate more efficiently (i.e., less complexity with better accuracy).

IEEE Internet of Things Journal•Apr 1, 2024•8 citations

Enhanced BBHT Algorithm for Active User Detection in 5G

Muhammad Idham Habibie, Milad Ghadimi, Riccardo Bassoli +1 more

The key features of 5G, particularly URLLC and mMTC, are designed to achieve low latency and high scalability. To support these requirements, a proposed approach for random access scenarios, known as the GF scheme, eliminates handshakes between the BS and mobile users to reduce latency and accommodate a large number of devices. However, this introduces a new challenge known as AUD, where users send messages indicating their activity status, requiring the BS to detect and decode them. Decoding these messages, however, demands high computational complexity, which increases exponentially with the number of users. Several algorithms, such as ZF and CCR, have been proposed to mitigate this complexity, but they suffer from suboptimal performance. On the other hand, the optimal solution, known as ML, performs well but suffers from high complexity. To address this, quantum algorithms, like Grover's algorithm, have been proposed due to their ability to reduce search complexity while also keep detecting active users performance better. However, Grover's algorithm requires adaptation in this context, as the optimal number of iterations depends on the number of solutions, which is always unknown in the AUD case. To address this, the BBHT and DHA algorithms have been proposed to minimize complexity when the solution is unknown, but both still exhibit relatively high computational demands. In this paper, we propose an Enhanced BBHT that aims to reduce this complexity while maintaining detection performance. Our findings demonstrate that the Enhanced BBHT reduces computational complexity while keeping the performance stable.

International Conference on Communications•Jun 8, 2025•1 citation

Grover's Search-Inspired Quantum Reinforcement Learning for Massive MIMO User Scheduling

Ruining Fan, Xingyu Huang, Mouli Chakraborty +2 more

The efficient user scheduling policy in the massive Multiple Input Multiple Output (mMIMO) system remains a significant challenge in the field of 5 G and Beyond 5 G (B5G) due to its high computational complexity, scalability, and Channel State Information (CSI) overhead. This paper proposes a novel Grover's search-inspired Quantum Reinforcement Learning (QRL) framework for mMIMO user scheduling. The QRL agent can explore the exponentially large scheduling space effectively by applying Grover's search to the reinforcement learning process. The model is implemented using our designed quantum-gate-based circuit, which imitates the layered architecture of reinforcement learning, where quantum operations act as policy updates and decision-making units. Moreover, the simulation results demonstrate that the proposed method achieves proper convergence and significantly outperforms classical Convolutional Neural Networks (CNN) and Quantum Deep Learning (QDL) benchmarks by 51 % and 43 %, respectively.

arXiv:2601.20688•IEEE Wireless Communications and Networking Conference•Apr 13, 2026•0 citations

Grover-Accelerated Variational Quantum Learning for Time-Critical Task Offloading in 6G Satellite-Terrestrial Integrated Networks

Chenfei Gu, Yuchao Chen, Wei Lu

In 6G Satellite-Terrestrial Integrated Networks (STIN), Low Earth Orbit (LEO) satellites serve as edge nodes to enable time-critical task offloading for ground users. However, the high mobility and transient visibility of satellites impose stringent constraints on decision latency. While Deep Reinforcement Learning (DRL) is widely adopted to handle such dynamic environments, classical approaches suffer from inefficient state representation and serial action search, often failing to complete decisions within the fast-fading channel coherence time. To address this, we propose Grover-Accelerated Variational Quantum Learning (GVQL). By integrating heterogeneous quantum encoding for high-dimensional feature mapping, Variational Quantum Circuits (VQC) for value approximation and Grover’s algorithm for quadratic search speedup, GVQL effectively overcomes these computational bottlenecks. Extensive simulations demonstrate that GVQL significantly outperforms representative baselines in decision timeliness and reliability under strict timing constraints.

International Wireless Communications and Mobile Computing Conference (IWCMC)•Jun 1, 2026•0 citations

Grover-Based PLS: AUD and Beamforming With Artificial Noise in CD-NOMA

Deemah H. Tashman, Soumaya Cherkaoui

Sixth-Generation (6G) networks will require massive connectivity, ultra-low latency, and robust security, making reliable Active User Detection (AUD) essential for interference control and physical layer protection. This letter proposes a Grover-based physical layer security (PLS) framework for a code-domain non-orthogonal multiple access (CD-NOMA) network, where the base station employs artificial-noise (AN)-assisted beamforming and identifies the active set via Grover’s quantum search algorithm. We consider two threat models: passive eavesdroppers formed by detected inactive users, and active eavesdroppers selected as the top$f\%$most frequent transmitters among detected active users. By aligning beams and AN with the Grover-based AUD output, the proposed scheme enlarges the main–wiretap rate gap and significantly improves the average secrecy rate compared with compressive sensing and classical correlation receiver baselines, while approaching maximum-likelihood detection performance with a quadratic reduction in search complexity. The impact of the information/AN power split, the base station transmit power, and the fraction of highly active users treated as eavesdroppers on secrecy is characterized through numerical simulations, and design insights are extracted for 6G PLS under both passive and active eavesdropping.

arXiv:2607.05429•IEEE Communications Letters•Jan 1, 2026•0 citations

A Survey of Quantum Deep Reinforcement Learning for Resource Allocation in Future Wireless Networks

Ni Made Erma Pratiwi Astiti, Byung Moo Lee

The significant rise in wireless communication, driven by sixth-generation (6G) networks, underscores the urgent need for innovative resource allocation methods to achieve efficiency, scalability, and reliability. This survey explores the integration of Quantum Deep Reinforcement Learning (QDRL) as a strategy to address the challenges posed by dynamic and complex wireless environments. By leveraging quantum computing features such as superposition and entanglement, QDRL enhances computational efficiency and mitigates the scalability limitations of classical and deep reinforcement learning (DRL) in high-dimensional optimization. The study reviews recent QDRL approaches for key resource allocation tasks, including power distribution, user scheduling, and channel assignment, and examines quantum algorithms such as the Quantum Approximate Optimization Algorithm (QAOA) and Quantum Neural Networks (QNN) for improved decision-making in wireless systems. Furthermore, it emphasizes the connection between quantum computing and DRL, highlighting hybrid architectures that balance computational cost, energy efficiency, and scalability. Finally, the survey identifies research gaps and outlines future directions, paving the way for robust QDRL strategies to support next-generation wireless networks.

IEEE Communications Surveys & Tutorials•Nov 28, 2025•16 citations
Sixth-generation (6G) networks must deliver ultra-low latency and near-100 percent reliability to support massive-scale Internet of Things (IoT) deployments and Hyper-Reliable Low-Latency Communications (HRLLC). Grant-free access protocols permit devices to transmit without prior scheduling; nevertheless, this uncoordinated transmission introduces uncertainty at the receiver, necessitating Active User Detection (AUD) to ascertain which devices are active. Quantum search methods—most notably Grover’s algorithm—can accelerate AUD, yet they require knowing the optimal number of iterations, which depends on the (typically unknown and time-varying) number of valid solutions induced by the current activity pattern and channel/noise conditions. To overcome this, we formulate an optimization problem that optimizes the number of Grover iterations to maximize detection accuracy and minimize computational cost without any prior activity information. We then apply a Deep Q-Network (DQN) to learn, via deep reinforcement learning, an adaptive policy for selecting the iteration count. Simulation results verify that the DQN converges to an optimal strategy and outperforms two baseline schemes under varying fading conditions and active-user transmit powers.
The key features of 5G, such as ultra-reliable low latency (URLLC) and massive machine-type communication (mMTC), are designed to address the need for low latency and the ability to connect a large number of devices in the IoT context. To support these constraints, mobile devices transmit information without previously establishing a connection with the base station (BS). This requires the Base Station to detect in real-time the active users (process known as Active User Detection (AUD)). With classical processors, one can employ the Maximum Likelihood (ML) method (the optimal detector, but suffers from high complexity and delay), or suboptimal ones (which are simpler, but less reliable). Meanwhile, quantum algorithms, particularly Durr and Hoyer (DHA) algorithm, addressing minimum searching problems, can significantly reduce complexity while keeping good performances. However, these algorithms were designed for generic problems, and their initialization and parameterization are blindly done. Nonetheless, we can have access to prior information on the system’s behavior. Therefore, in this paper, we aim to adapt and improve these quantum algorithms by using prior knowledge on the system for the AUD problem. We first propose a novel algorithm, the Improved Iterative Minimum Searching Algorithm (IIMSA) where we define more efficiently the parameters. Then, further enhancements of IIMSA are obtained thanks to a better initialization of the algorithms by exploiting classical preprocessing of the received signals with classical Conventional Correlation Receiver (CCR) or Zero Forcing (ZF). The obtained results show that these proposed algorithms operate more efficiently (i.e., less complexity with better accuracy).
The key features of 5G, particularly URLLC and mMTC, are designed to achieve low latency and high scalability. To support these requirements, a proposed approach for random access scenarios, known as the GF scheme, eliminates handshakes between the BS and mobile users to reduce latency and accommodate a large number of devices. However, this introduces a new challenge known as AUD, where users send messages indicating their activity status, requiring the BS to detect and decode them. Decoding these messages, however, demands high computational complexity, which increases exponentially with the number of users. Several algorithms, such as ZF and CCR, have been proposed to mitigate this complexity, but they suffer from suboptimal performance. On the other hand, the optimal solution, known as ML, performs well but suffers from high complexity. To address this, quantum algorithms, like Grover's algorithm, have been proposed due to their ability to reduce search complexity while also keep detecting active users performance better. However, Grover's algorithm requires adaptation in this context, as the optimal number of iterations depends on the number of solutions, which is always unknown in the AUD case. To address this, the BBHT and DHA algorithms have been proposed to minimize complexity when the solution is unknown, but both still exhibit relatively high computational demands. In this paper, we propose an Enhanced BBHT that aims to reduce this complexity while maintaining detection performance. Our findings demonstrate that the Enhanced BBHT reduces computational complexity while keeping the performance stable.
The efficient user scheduling policy in the massive Multiple Input Multiple Output (mMIMO) system remains a significant challenge in the field of 5 G and Beyond 5 G (B5G) due to its high computational complexity, scalability, and Channel State Information (CSI) overhead. This paper proposes a novel Grover's search-inspired Quantum Reinforcement Learning (QRL) framework for mMIMO user scheduling. The QRL agent can explore the exponentially large scheduling space effectively by applying Grover's search to the reinforcement learning process. The model is implemented using our designed quantum-gate-based circuit, which imitates the layered architecture of reinforcement learning, where quantum operations act as policy updates and decision-making units. Moreover, the simulation results demonstrate that the proposed method achieves proper convergence and significantly outperforms classical Convolutional Neural Networks (CNN) and Quantum Deep Learning (QDL) benchmarks by 51 % and 43 %, respectively.
In 6G Satellite-Terrestrial Integrated Networks (STIN), Low Earth Orbit (LEO) satellites serve as edge nodes to enable time-critical task offloading for ground users. However, the high mobility and transient visibility of satellites impose stringent constraints on decision latency. While Deep Reinforcement Learning (DRL) is widely adopted to handle such dynamic environments, classical approaches suffer from inefficient state representation and serial action search, often failing to complete decisions within the fast-fading channel coherence time. To address this, we propose Grover-Accelerated Variational Quantum Learning (GVQL). By integrating heterogeneous quantum encoding for high-dimensional feature mapping, Variational Quantum Circuits (VQC) for value approximation and Grover’s algorithm for quadratic search speedup, GVQL effectively overcomes these computational bottlenecks. Extensive simulations demonstrate that GVQL significantly outperforms representative baselines in decision timeliness and reliability under strict timing constraints.
Sixth-Generation (6G) networks will require massive connectivity, ultra-low latency, and robust security, making reliable Active User Detection (AUD) essential for interference control and physical layer protection. This letter proposes a Grover-based physical layer security (PLS) framework for a code-domain non-orthogonal multiple access (CD-NOMA) network, where the base station employs artificial-noise (AN)-assisted beamforming and identifies the active set via Grover’s quantum search algorithm. We consider two threat models: passive eavesdroppers formed by detected inactive users, and active eavesdroppers selected as the topf%f\%f%most frequent transmitters among detected active users. By aligning beams and AN with the Grover-based AUD output, the proposed scheme enlarges the main–wiretap rate gap and significantly improves the average secrecy rate compared with compressive sensing and classical correlation receiver baselines, while approaching maximum-likelihood detection performance with a quadratic reduction in search complexity. The impact of the information/AN power split, the base station transmit power, and the fraction of highly active users treated as eavesdroppers on secrecy is characterized through numerical simulations, and design insights are extracted for 6G PLS under both passive and active eavesdropping.
The significant rise in wireless communication, driven by sixth-generation (6G) networks, underscores the urgent need for innovative resource allocation methods to achieve efficiency, scalability, and reliability. This survey explores the integration of Quantum Deep Reinforcement Learning (QDRL) as a strategy to address the challenges posed by dynamic and complex wireless environments. By leveraging quantum computing features such as superposition and entanglement, QDRL enhances computational efficiency and mitigates the scalability limitations of classical and deep reinforcement learning (DRL) in high-dimensional optimization. The study reviews recent QDRL approaches for key resource allocation tasks, including power distribution, user scheduling, and channel assignment, and examines quantum algorithms such as the Quantum Approximate Optimization Algorithm (QAOA) and Quantum Neural Networks (QNN) for improved decision-making in wireless systems. Furthermore, it emphasizes the connection between quantum computing and DRL, highlighting hybrid architectures that balance computational cost, energy efficiency, and scalability. Finally, the survey identifies research gaps and outlines future directions, paving the way for robust QDRL strategies to support next-generation wireless networks.