Development of Multi-agent Deep Reinforcement Learning with Prioritized Experience Replay for Communication and Sensing Integrated Network in 5G mmWave System
Jul 2026· 2026 7th International Conference on Smart Systems and Inventive Technology (ICSSIT)· pp. 1042-1050· 0 citations· 19 references
Abstract
The ability of various isolated devices to sense their surroundings can be improved by 5G millimetre wave (mmWave) communication technology. By jointly supporting data transmission and sensing tasks, the framework improves overall spectrum efficiency in wireless networks. Among them, the Integrated Sensing and Communication (ISAC) has become the standard in wireless communications. Specifically, mmWave technology is highly effective for bandwidth-intensive communication services and delivers improved spatial and temporal accuracy through its large spectrum availability and directional beamforming characteristics. To meet the requirements, a multi-agent-based deep learning technique is proposed for better development. Over this sensing network of 5G mmWave, the resource allocation process is handled by Multi-agent Deep Reinforcement Learning with Prioritized Experience Replay (MDRL-PER), whereas the system is provided based on allocated resource for better communication. Finally, the performance of the system is assessed through distinct evaluation metrics and compared with existing methodologies. Hence, the superior results are obtained to ensure the efficacy of the communication network.
The Millimeter-wave massive Multiple-Input Multiple-Output (MIMO) is a core mechanism for the Sixth-Generation (6G) wireless communication networks. By including numerous antennas in the compact model of advanced smartphones, the MIMO enhances the network capacity and the Spectral Efficiency (SE). The growth of 6G technology is significant for the future and provided evolutionary and revolutionary solutions. The resource allocation in the MIMO-based wireless networks is selected for various users, aiming to optimize the network resource distribution. But the high increase in the antennas and users poses complexities for the resource allocation and interference suppression for the MIMO systems. In this research, an advanced Deep Reinforcement Learning (DRL)-based approach is proposed for efficient dynamic spectrum allocation in 6G MIMO systems. To perform spectrum allocation in 6G MIMO systems, an Adaptive Deep Multiagent Reinforcement Learning with Co-ordinate Attention (ADMRL-CA) model is developed. The DMRL mechanism is capable of handling varying traffic and channel conditions. The incorporation of the CA mechanism enhances the policy learning process for accurate spectrum allocation. The parameters of the ADMRL-CA are fine-tuned using the Flying workers phase Modified Termite Queen Algorithm (FMTQA). Finally, the model performance is analyzed with various existing models. The SE of the recommended FMTQA-ADMRL-CA is increased by 4.44% of DRL, 6.66% of SAC, 2.77% of DDPG and 10.88% of DMRL-CA when system uses as 32
nd
batch size. Hence, it is guaranteed that the recommended FMTQA-ADMRL-CA can allocate the spectrum efficiently and robustly in 6G MIMO systems than the existing methods.
Asha Aiyappan, Jafar A. Alzubi, M. P. Rajakumar et al.· Scientific Reports· 0 citations
A Deep Reinforcement Learning (DRL)-based framework for dynamic spectrum access in 6G heterogeneous Cognitive Radio Networks (Het-CRNs), wherein secondary users learn optimal channel selection policies through direct interaction with the radio environment, without requiring explicit statistical channel models is proposed.
Naadir Kamal, R. Kumar· Global Journal of Engineerin...· 0 citations
The development of 5G and higher wireless communications systems has posed a great need on intelligent transmission techniques capable of supporting ultra-high data rates, low latency, high spectral efficiency, and dependable connectivity in dynamic networked environments. Beamforming is one of the emerging technologies that is critical in improving signal directionality and reducing interference, especially in millimeter-wave and massive multiple-input multiple-output (MIMO) communication systems. The traditional methods of beamforming are however usually challenged by issues associated with a high computational complexity, high energy usage and low adaptability to the fast-varying channel conditions. Through the research, a Hybrid Beaming Technique to enhance the aspect of performance in 5G and above communication networks using Machine Learning, combining Artificial Neural Networks (ANN) and Deep Convolutional Neural Networks (DNN) to optimize the beam selection, channel estimation, and adaptive resource allocation in the 5G and above communication networks. Within the framework proposed, ANN is used as a predictive analysis of channel state information and intelligent beam weight optimization, and CNN is used as a feature extraction of complex spatial channel patterns and interference mapping, in order to make accurate decisions of beam steering in real time. Incorporating the energy efficiency of analog beam control with the flexibility and precision of digital beam processing, the hybrid beamforming architecture forms a robust and scalable communication model. The results of the simulations have shown that the proposed ANN-DNN-based hybrid beaming technique is significantly better in terms of throughput, signal-to-interference-plus-noise ratio (SINR), spectral efficiency, and coverage reliability than traditional beamforming techniques. The framework also exhibits less beam misalignment and greater flexibility in case of user mobility and dense deployment. The proposed intelligent hybrid beaming model presents a bright solution to next-generation wireless systems, such as 5G networks, to support high-capacity, low-latency, and energy-efficient communication infrastructures to future smart and connected environments.
B Jaya, Ette Hari Krishna· International journal of com...· 0 citations
Reconfigurable intelligent surface (RIS) has emerged as a promising technology for next-generation wireless networks due to its ability to intelligently manipulate the propagation environment. In RIS-assisted millimeter-wave multiantenna MIMO communication networks, the joint optimization of RIS phase configuration and resource allocation under heterogeneous user priorities remains challenging. This paper proposes a deep learning-based framework that incorporates user priority weights into both channel estimation and resource allocation through and unsupervised learning. We formulate the joint optimization problem of RIS phase shifts, base station beamforming, and user priority scheduling under α-fairness criteria. A neural network architecture is designed to learn the mapping from channel state information and user weights to optimal resource allocation policies. Simulation results demonstrate that the proposed approach achieves significant performance improvements of 6.5–13.8% in throughput compared to baseline schemes across multiple metrics. The devised framework attains enhanced performance metrics with lower computational burden, which renders it far more expandable than iterative optimization approaches.
Chao-Qun Pei, Gewei Tan· 2026 8th International Confe...· 0 citations
With the advent of seventh-generation (7G) wireless systems, the spectrum environment is extremely dynamic and heterogeneous, and traditional methods of sensing do not offer reliable and efficient performance. This paper introduces a self-evolving spectrum sensing system to enable adaptive and intelligent spectrum access based on a deep reinforcement-based learning paradigm. The framework depicts the sensing process as a sequence decision problem where an autonomous agent is able to continually refine its policy as it engages with the environment. The multi-objective reward formulation is designed to maximize the combination of the detection accuracy, false alarms, energy usage, and using the spectrum. Moreover, an adaptive representation of state mechanism is also introduced to indicate the temporal changes and short-term changes in the spectrum occupancy. The self-evolution strategy proposed adjusts learning parameters and decision policies in a dynamic manner that gives a robust operating in a non-stationary environment. The overall analysis of the experiment shows that the structure achieves detection probability of 97.1, false alarm rate reduced to minimum of 3.8, spectral usage maximized to above 92 and convergence rate is quicker as compared to the existing techniques. These results confirm the appropriateness of the proposed method in overcoming the issues of the next-generation wireless systems.
A.L Sriram, H. N. Divya, S. Sabarinathan et al.· International Conference on...· 0 citations
Simulation results position D3QN-PER as a strong candidate for deployment as a near-RT RIC xApp within the O-RAN architecture, advancing the vision of AI-native mobility management for 6G.
Kalpesh Popat, Divyakant T. Meva· Telecommunications Systems· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
