An approach for just-in-time radio access network capacity planning in CDMA networks

This paper describes a holistic capacity planning methodology for radio access networks (RANs) and presents a case study focusing an the third generation (3G) code division multiple access (CDMA) technology. It addresses the complex problem of how a RAN should be optimally and just-in-time equipped to accommodate fluctuations in traffic patterns over time, while fulfilling the quality of service (QoS) requirements of a wireless network operator (WNO) and without over-provisioning. The just-in-time RAN capacity planning methodology provides a software engine that automatically and iteratively computes various RAN configurations (which may correspond to various "what-if" deployment scenarios) based on multiple inputs that include network configuration, traffic volume measurements, and traffic forecast. It uses a bottom-up approach that sequentially evaluates the capacity needs over the air interface, at the base stations, and over the transport (backhaul) domain. Material in excess/deficit is identified at each RAN level and may be redistributed to avoid ordering of new material and reduce overall cost. This methodology provides the rapid and repeatable dimensioning which is required to implement just-in-time provisioning processes, and to support the network cost modeling an operator requires to quickly model quality of service and call plan strategy. Furthermore, thanks to its high level of automation, it allows the WNOs to complete a multi-year capacity planning process quickly and efficiently. The approach uses a generic process enabling its applicability to any CDMA-based network. It has been intensively tested with real field data and has been recently adopted by several WNOs. (C) 2008 Alcatel-Lucent.