× close
Queuing system. credit: future internet (2024). DOI: 10.3390/fi16010018
One of the characteristics of 5G networks is the segmentation of the network, so-called slicing. Physically the network is the same, but logically it is divided into slices depending on the current request. This approach guarantees a certain level of signal quality. Resources are dynamically allocated to specific segments. If some resources are not currently in use, they can be redirected to another segment.
It is important to optimize this process so that slicing yields good results. Mathematicians at RUDN University have figured out the best way to implement resource distribution when two services share flexible traffic, for example when a browser is running and data is being transferred at the same time.Their research will be published in the journal future internet.
“Network slicing allows you to deploy independent network resources on the same infrastructure. This mechanism allows providers to allocate logically separated network segments to users, with each segment having a specific Designed and optimized to suit your requirements.
“For example, one for cellular communications and one for the Internet of Things. Therefore, studying the question of how to organize the redistribution of resources when slicing a 5G network That’s important,” said Dr. Irina Kochetkova. , Associate Professor, Computer Science and Telecommunications Research Institute, RUDN University.
Mathematicians at RUDN University have built a mathematical model based on queue theory and Markov decision-making processes. Resource allocation is based on his three principles: maximum matching for even resource division, maximum share of signals resulting in reallocation of resources, and maximum utilization of resources. Based on this, mathematicians built iterative algorithms to achieve optimal distributions and conducted numerical experiments.
Numerical experiments showed fast convergence, i.e., finding the optimal solution quickly, in just three iterations. This demonstrates the effectiveness of the proposed approach. The new algorithm converges faster than so-called brute force, or exhaustive search.
From numerical experiments, mathematicians were able to conclude on what exactly the optimal allocation of resources depends. These factors turned out to be the ratio of the current state of the system and the weight of the reward function, that is, the “importance” of the individual parameters in the behavior of the algorithm.
“To efficiently compute optimal resource scheduling policies, we developed a sequential algorithm, starting with maximizing resource utilization as a starting point. Using a numerical demonstration, we We presented optimal solutions for two services for bulk data transfer.
“The algorithm converges quickly in three iterations. It is effective due to a balanced approach based on three principles,” said Dr. Anastasiya Vlaskina, assistant professor at the Institute of Computer Science and Telecommunications at RUDN University. Masu.
For more information:
Irina Kochetkova et al., Controllable Queuing System with Elastic Traffic and Signaling for Resource Capacity Planning in 5G Network Slicing, future internet (2024). DOI: 10.3390/fi16010018
