The impact of cell site re-homing on the performance of umts core networks

The impact of cell site re-homing on the performance of umts core networks
of 14

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  International Journal of Next Generation Network (IJNGN), Vol.2, No.1, March 2010   󰀱󰀰󰀮󰀵󰀱󰀲󰀱󰀯󰁩󰁪󰁮󰁧󰁮󰀮󰀲󰀰󰀱󰀰󰀮󰀲󰀱󰀰󰀵󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀴󰀹󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠󰀠 T HE I MPACT OF C ELL S ITE R E - HOMING ON THE P ERFORMANCE OF UMTS   C ORE N ETWORKS   Ye Ouyang and M. Hosein Fallah, Ph.D., P.E. Howe School of Technology Management Stevens Institute of Technology, Hoboken, NJ, USA 07030 youyang@stevens.edu hfallah@stevens.edu  A  BSTRACT     Mobile operators currently prefer optimizing their radio networks via re-homing or cutting over the cell sites in 2G or 3G networks. The core network, as the parental part of radio network, is inevitably impacted by the re-homing in radio domain. This paper introduces the cell site re-homing in radio network and analyzes its impact on the  performance of GSM/UMTS core network. The possible re-homing models are created and analyzed for core networks. The paper concludes that appropriate re-homing in radio domain, using correct algorithms, not only optimizes the radio network but also helps improve the QoS of the core network and saves the carriers’ OPEX and CAPEX on their core networks.  K   EYWORDS   UMTS, WCDMA, GSM, Core Network, Rehoming, Network Optimization, Network Dimension, Network Plan. 1.   I NTRODUCTION   The past few years have seen mobile operators transition to next-generation mobile networks; specifically from third-generation networks (3G) to long term evolution (LTE). Subscriber numbers and network usage are up; and forecasts point to even greater expansion for many years. Mobile operators are challenged to retain existing subscribers, acquire new ones, and manage costs for serving both. However, with increased traffic, introduction of data, rich multimedia services as well as larger service areas, the mobile operators are facing the issue of radio network congestion and confronting the demands for larger service coverage areas. Normally the solution is the network expansion, increasing the size and capacity of mobile networks by installing more network infrastructure into the existing networks, which is an expensive and human-resource intensive undertaking. Therefore, from the radio network aspect, the best approach to avoid cell congestion or channel blocking due to the subscriber increase is to enhance the radio coverage and capacity via increasing the radio infrastructure such as base stations (BTS) and Base Station Controllers (BSC) in Global System for Mobile Communications (GSM), radio domain, or Node-B and Radio Network Controller (RNC) in the Universal Mobile Telecommunication System (UMTS) radio domain.  International Journal of Next Generation Network (IJNGN), Vol.2, No.1, March 2010  󰀵󰀰 Meanwhile, a new problem that exists in the radio network expansion is that the load distribution is uneven across the RNCs or BSCs after new Node-Bs or BTSs are added into radio networks. In other words, the new cell sites (BTSs and Node B) may cause a few BSCs and RNCs to overload. The traffic should be re-distributed across all the old and new BSCs and RNCs. Mobile operators usually strive to resolve this issue by cell site re-homing; move (“re-home”) a few cell sites from BSC or RNC with heavy load to that with low load. Generally re-homing is a re-distribution and re-configuration process for traffic and routing in the radio domain. It is of pivotal importance for the mobile operators to optimize the GSM/UMTS radio domain to the extent possible before investing more to expand the infrastructure. Figure 1 displays the topology of a GSM/UMTS radio domain in which BTS and Node-B are the first reference point for end subscribers to access into the mobile networks. BSC and RNC, standing above cell sites logically, are responsible for the connection, control and management of base stations in the radio domain. In particular, the radio domain in UMTS, according to 3GPP TS 25.401 and 3GPP TS 23.002, is called UMTS Terrestrial Radio Access Network (UTRAN) which includes one or multiple Radio Network Sub-system (RNS). An RNS contains one RNC and one or more Node B. Similarly, the radio domain in GSM is called GSM RAN which includes one or more Base Station Sub-system (BSS). A BSS contains two types of Network Entities: BSC and BTS. BSC plays a similar role as RNC to control and route calls for the base stations. BTS as the cell site in GSM is to access the mobile stations (MS). The re-homing technique towards the radio domain attempts to achieve the optimization of routings, loading and throughput for both 2G and 3G RAN. Figure 1. Architecture of the GSM/UMTS network and the impact scope of the cell site re-homing Oom, Jan. et al (2004) stated that the cell site re-homing procedure for radio access networks (RAN) requires many manual operation steps, so it is a labor intensive and time-consuming task requiring reconfigurations of both the radio and transport networks. Take 2G RAN as a example; a cell site re-homing requires the following operations: 1) verify the hardware configuration in a  International Journal of Next Generation Network (IJNGN), Vol.2, No.1, March 2010  󰀵󰀱 target BSC by comparing it to the hardware configuration of a source BSC; 2) check if the relevant software versions for BTS are available in the target BSC by comparing them with the registered versions in the source BSC; 3) copy the cell data from the source BSC to the target BSC; 4) copy the site data from the source BSC to the target BSC; 5) copy the neighbor cell data from the source BSC to the target BSC; 6) create new external cells data in the source BSC with state “not operating;” 7) create new external cells data in the target BSC with state “not operating;” 8) halt source cell in the source BSC; 9) block transceiver (TRX) resources in the BTS; 10) set old external cells data in the source and target BSC to state “not operating.” With the above preliminary measures completed then; 11) moving the switching connection for the BTS from the source BSC to the target BSC. The re-homing procedure continues with the following operations: 12) update the Cell Global Identifier (CGI) in the Mobile Services Switching Center (MSC); 13) set new external cells data in the source and target BSCs with state “operating;” (14) de-block TRX resources in the BTS; 15) activate target cell in the target BSC; 16) remove the cell data in the source BSC; 17) remove the site data in the source BSC; 18) remove N-cell data in the source BSC; 19) remove old external cells data in the source BSC; and 20) remove old external cells data in the target BSC. The dot dashed circled area in the Figure 1 shows the impact scope of core networks due to the re-homing in radio access networks (RAN). The core network (CN) is the heart of the current mobile communication networks. CN and RAN are closely coupled via A interface between BSC and MSC in GSM network and via Iu-CS interface between RNC and Media Gateway (MGW) in UMTS network (3GPP TS 25.413and 3GPP TS 25.415). Therefore, the changes in routing, loading and throughput resulting from the re-homing at radio side will definitely impact the performance of core network in GSM or UMTS. As per Figure 1, the cell site re-homing in GSM RAN and UTRAN impacts the area with red dot dashed line circled through A and Iu-CS interfaces. The paper will study the impact of the re-homing in RAN on the performance of CN in GSM/UMTS networks. The appropriate re-homing in radio domain, followed by corresponding regulations, not only optimizes the radio network but also helps improve the QoS of core network and reduces the need for extra OPEX and CAPEX investment in on core network. The rest of the paper proceeds as follows: Section 2 summarizes the literature in the related area and the challenges in optimizing and planning mobile core networks. Section 3 introduces the proposed re-home models in mobile core networks and how a re-homing in the radio domain impacts the performance of the Core Network (CN). Section 4 which is the core of the paper discusses the numeric analysis for the proposed rehoming models and creates the algorithms to optimize the core networks via rehoming operations. Section 5 provides a case study to illustrate application of the algorithms created in Section 4. Section 6 is the conclusion to the paper. 2.   L ITERATURE R EVIEWS   The current literature is more focused on the practical or theoretical solutions to design and plan GSM, UMTS, and Long Term Evolution (LTE) radio networks but overlooks the algorithms for planning and optimization of core networks. Also earlier studies don’t provide a unified approach to optimize the traffic and throughput for mobile core networks. Furthermore, the current literature does not address the dimensioning and optimization for mobile core  International Journal of Next Generation Network (IJNGN), Vol.2, No.1, March 2010  󰀵󰀲 networks. This can be due to: 1) The mobile core network in either logical or physical structure is more complicated than radio access network, 2) The internal traffic and throughput may vary depending on the vendors’ network entities (NE). Oom, Jan. et al (2004) in their patent introduce a resource sharing method through rehoming work in wireless radio networks. The patent interprets the detailed rehoming procedure and achieves the optimization of GSM radio networks via re-balancing and re-distributing the traffic going through the network. Their method is a good starting point for extending the rehoming to UMTS core networks. Shalak, R. et al (2004) make a qualitative study of the performance of UMTS core network, in which equipment of multiple vendors of UMTS CN are compared. Harmatos, J. (2002) proposes a model to plan UMTS core network based on the requirements from radio access network. The model also considers the premise of planning work in cost minimization, which helps mobile operators minimize Capital Expenditure (CAPEX).Their solutions are more based on techno-economic aspect to achieve the maximization of core network performance and minimization of Total Cost of Ownership (TCO) through network planning. Many other articles discussed the planning and dimensioning methods for GSM and UMTS network (Konstantinopoulou, C.N., 2000; Mishra, A , 2003; Britvic, V. et al, 2004; Vrabel, A. et al , 2007; and Szekely, I. et al, 2008). These articles are high level perspectives and interpretation of the on network transition and the evolution of the core network architecture. Ouyang, Y. and Fallah, M. H. (2009) make a further study to extend the mobile network planning and dimensioning into the level of network entities. That is, to study the throughput generated and absorbed in each interfaces of network entities. The algorithms created for the interface level provide a guideline for mobile operators to dimension their UMTS core networks. We believe the area that most of the current literature has overlooked is the core network optimization. Therefore, the objective of this paper is to analyze the impact of radio network rehoming on the core network domain, to create a unified algorithm to address the traffic/throughput re-distribution and re-configuration in UMTS core networks, and to achieve the throughput/ traffic optimization in core network domain via the rehoming operations between RNC and MGW. P ROPOSED R E -H OMING M ODELS   This section will analyze how a re-homing in the Radio Domain impacts the performance of the Core Network. As per the twenty re-homing steps in section 1, the routing configuration, traffic loading, and data throughput of involved RNC or BSC are changed after a re-homing of related cell sites (Node-B or BTS). The changes in RNC or BSC due to re-homing will result in the new changes in the traffic load and routings in 3G MGW via Iu-CS interface or 2G MSC via A interface. Let’s consider a possible scenario: from the radio domain aspect, if a re-homing evenly re-distributes the traffic across all involved RNC and reduces the loading of all involved RNC to a reasonable value (normally below 70% or 80% of the loading threshold of a RNC), it can be recognized as a successful re-homing in the radio domain. However, the re-distributed traffic in a RNC flowing into its paired MGW via Iu-CS interface sometimes may be over the tolerable threshold of the MGW in traffic loading. Hence, this re-homing is not executable from the core network aspect. One simple approach to resolve this issue is to add new MGW to share  International Journal of Next Generation Network (IJNGN), Vol.2, No.1, March 2010  󰀵󰀳 the load. That will require CAPEX. An alternative is “re-rehoming,” to re-distribute the traffic across the involved MGW based on the re-homing for RNC. The intention of the second re-homing is to 1) Ensure the successful implementation of the re-homing in the radio domain; 2) optimize the core networks by re-distributing the traffic across all involved 3G MGW, 3G MSC Server (MSS) or 2G MSC. The effectiveness of re-homing depends on optimization of both the radio and the core networks. The potential re-homing models for the core network are described below. 2.1   M ODEL 1:   F ROM S INGLE 3G   MGW   T O S INGLE 3G   MGW Figure 2 illustrates the re-homing process for mode 1: Two single and neighboring MGWs: MGW A reports to MSS A and MGW B reports to MSS B belong to the same market A. Both MGW A and B are the only media gateway under control by their paired MSS. MSS A and B must belong to the same geographical service area or market. In this model, a RNC or BSC used to connect to MGW A is re-homed to MGW B. For simplicity of Operation and Maintenance (O&M), the scope of a re-homing is normally in the same service area or market, which means that in model 1 the re-homing is only allowed in Market A but not to extend to other markets. The scope of the re-homing is applied to all the following models in the paper. Sometimes there are multiple MGWs under control of the same MSS in a service market. From the aspect of re-homing, it is theoretically expected that the traffic distribution across these MGWs will always be even and equal. Therefore, two other principles for re-homing in the core network side should be defined: 1) a re-homing is not allowed to execute from one MGW to another if this source and target MGW belong to the same MSS. 2) If a RNC or BSC is re-homed into a target MSS who controls multiple target MGWs, the re-homed traffic by the RNC/BSC should be evenly distributed into all the target MGWs under the target MSS. The principles are applied to all the re-homing modes. Figure 2. Re-homing model 1, 2 and 3
Related Documents
View more
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks