Spectrum Allocation for Network Slices with Inter-Numerology Interference using Deep Reinforcement Learning

Abstract

Network slicing and mixed-numerology schemes are essential technologies to efficiently accommodate different services in 5G radio access networks (RAN). To fully take advantage of these techniques, the design of spectrum slicing policies needs to account for the limited availability of the radio resources as well as the inter-numerology interference generated by slices employing different numerologies. In this context, we formulate a binary non-convex problem that maximizes the aggregate capacity of multiple network slices. The resulting spectrum allocation minimizes the inter-numerology interference under the frequent channel fluctuations characterizing the various users. To address the computational complexity of the designed objective function, we leverage deep reinforcement learning (DRL) to design a model-free solution computation. In detail, the trained centralized DRL agent exploits the channel fading statistic in order to provide a spectrum allocation that minimizes the internumerology interference. Results reveal that the proposed DRL scheme achieves performance that is comparable to the optimal one. It also outperforms a baseline scheme that statically allocate the radio resources.

Applications of Machine Learning in Resource Management for RAN-Slicing in 5G and Beyond Networks: A Survey

“…The simulation results show that the proposed method performs better in terms of the mentioned QoE factors compared to the fixed or dynamic (considering the priority and traffic of each slice) methods. In [88], the issue of radio and power allocation is modeled as a binary non-convex problem in the mixed-numerology interference environment taking into account channel state information, aiming to increase the service capacity for different slices. Then, a DQN-based method is used to overcome the computational complexity of the problem.…”

Intelligent multi-branch allocation of spectrum slices for inter-numerology interference minimization

“…Finally, our previous work [6] proposes a DRL agent that multiplexes spectrum slices of different numerologies employing an optimal INI-aware reward function formulation. However, this approach is unpractical when the number of resources and numerologies increases due to the fact that the agent requires the enumeration of every feasible allocation to approximate the optimal policy.…”

“…• Single branch allocation: the subchannel allocation is computed using the DRL agent proposed in our previous work [6], which is based on DQN. We will refer to this agent as "single branch resource allocation (SBRA) agent".…”

“…In detail, we design a DRL agent that infers the relationship between the INI power and the wireless channel fading dynamic to maximize the aggregated network slice throughput. Note that, differently from our previous work in [6], we propose an agent based on BDQ networks that efficiently scales with the number of spectrum resources and network slices. More precisely, the main contribution of this work are:…”

“…Results show that the agent successfully converges under different combinations of network configurations and it provides a good approximation of the optimal solution in most scenarios. In addition, it provides better gains than the single-branch agent counterpart when the most complex system scenarios are assessed [6].…”