TL;DR
This paper investigates how re-homing cell sites in the radio access network (RAN) impacts the performance of the core network (CN) in GSM/UMTS systems. By analyzing various re-homing models and their effects on traffic distribution and resource utilization, the authors demonstrate that strategic re-homing can optimize both the RAN and CN, leading to improved Quality of Service (QoS) and significant cost savings (OPEX and CAPEX) for mobile operators.
What Problem Does This Paper Address?
Mobile operators are constantly striving to optimize their networks to handle increasing subscriber numbers and data demands while minimizing operational and capital expenditures. A common practice for optimizing the radio access network (RAN) is "cell site re-homing," which involves reassigning cell sites (like Base Transceiver Stations - BTS, or Node-Bs) to different controllers (Base Station Controllers - BSCs, or Radio Network Controllers - RNCs) to balance load and improve coverage. However, the core network (CN), which is intricately linked to the RAN, can be negatively impacted by these changes if not managed carefully. This paper addresses the crucial question of how these RAN re-homing operations affect the performance of the GSM/UMTS core network and proposes a framework for optimizing this interaction.
Methods
The paper introduces a systematic approach to analyzing the impact of cell site re-homing on the core network. The authors first detail the extensive manual steps involved in a typical cell site re-homing process in the RAN. They then propose nine distinct re-homing models that categorize various scenarios of traffic redistribution between different network entities in the core network, such as Media Gateways (MGWs) and Mobile Switching Centers (MSCs). These models consider re-homing between 2G and 3G network components and involve single or multiple source/target entities.
A significant portion of the paper is dedicated to numerical analysis. This involves forecasting key performance indicators (KPIs) for switches (MSC, MGW, MSS) both before and after a re-homing event. Parameters like Busy Hour Calling Attempt (BHCA) capacity, T1/E1 port utilization, and SS7 signaling utilization are modeled. The authors also develop formulas to estimate the cost savings associated with re-homing by comparing the required infrastructure (trunks, switches) for a re-homing scenario versus a traditional network expansion.
Datasets
Not discussed in the paper. The paper focuses on theoretical modeling and numerical analysis rather than empirical data from live networks.
Key Insights and Ideas
- Interdependence of RAN and CN: The paper strongly emphasizes that the RAN and CN are not independent entities. Changes in the RAN, like cell site re-homing, have direct and significant consequences on the CN's performance and resource utilization.
- Re-homing as a Core Network Optimization Tool: While traditionally viewed as a RAN optimization technique, the paper demonstrates that re-homing, when properly planned and executed with core network considerations, can also optimize the CN.
- Categorization of Re-homing Scenarios: The proposed nine re-homing models provide a comprehensive taxonomy for understanding and analyzing the diverse ways traffic can be redistributed within the core network due to RAN re-homing. This structured approach is valuable for systematic planning.
- Cost Savings Potential: The core insight is that strategic re-homing can significantly reduce the need for new infrastructure investments (CAPEX) and ongoing operational costs (OPEX) by optimizing the utilization of existing core network resources.
- Balancing QoS and Cost: The paper highlights the critical trade-off mobile operators face between maintaining high Quality of Service (QoS) and managing costs. Re-homing, when done correctly, helps achieve this balance.
Key Results
The paper provides a framework for analyzing the impact of re-homing on core network performance metrics like BHCA capacity, T1/E1 port utilization, and SS7 signaling. While specific quantitative results are presented through numerical examples and formulas, the paper's primary "result" is the demonstration that:
- Re-homing can lead to a reduction in the number of required trunks and potentially eliminate the need for new switch purchases, as illustrated in the case study where saved trunks are quantified.
- The proposed models and numerical analysis allow operators to forecast the impact of re-homing on resource needs and estimate cost savings compared to traditional network expansion.
- The case study shows a reduction in trunks from 1120 to 920 after re-homing, directly translating to saved resources and avoided costs.
Limitations and Future Work
- Manual Re-homing Process: The paper acknowledges that current cell site re-homing procedures are labor-intensive and time-consuming, requiring many manual steps.
- Focus on GSM/UMTS: The analysis is primarily focused on GSM and UMTS core networks. The evolving landscape towards 4G (LTE) and beyond might require further research.
- Vendor Dependency: The paper notes that internal traffic and throughput can vary depending on vendors' network entities, which is a factor in core network planning.
- Future Work: The authors explicitly state their continuing research will focus on developing a tool to achieve intelligent and automatic re-homing between switches, aiming to replace the current manual and labor-intensive operations. This suggests a move towards more automated and intelligent network management.
Summary updated