Enhancing Quality of Service in an eUTRAN LTE Network using the RK4 Differential Equation: A Deterministic Versus a Stochastic Approach

• Pacôme

  English

  Author

• Ghislain

  English

  Author

• Bodjré Aka Hugues

  English

  Author

As user traffic and radio channel interference are subject to constant change, it is crucial to enhance the
Quality of Service (QoS) in an eUTRAN LTE network to ensure optimum performance. As such, this study
investigates the application of 4th-order Runge-Kutta differential equations (RK4) to the problem of radio resource
management, contrasting two methods: one that relies on the assumption of constant and predictable traffic, and
the other that takes network fluctuations into account at random. The main issue is that it is not easy to predict and
handle network congestion well, which impacts transmission latency, bandwidth, and resource allocation. With the
use of a 24-hour simulation, we can observe that the deterministic method provides reliable but restricted
management, with peak-time resource allocations of up to 90%, bandwidth occupancy rates of 80 Mbps, and
maximum latency levels of 120 ms. On the other hand, it is better reflected by the stochastic method, which shows
that resource allocation may fluctuate by ±5%, bandwidth can vary by ±10 Mbps, and transmission delays can
reach 130-140 ms. These findings show that determinism provides a controlled picture of the network, but
stochasticity allows for more effective dynamic adaptation, which is crucial for real-time resource adjustment and
congestion prediction.

• pdf