Self-optimization of random access channel in 3rd Generation Partnership Project Long Term Evolution

摘要

The long term evolution (LTE) is being designed to enhance third generation radio access technologies, and it represents a major step towards high-performance mobile data networks that are fully optimized for packet-switched connections. In an initial access, the LTE user equipment executes the random access (RA) procedure while searching for serving base station and initiating services. For a good cell coverage to be obtained, feasible uplink data rate and low delays in, for example, call setup, the optimal random access channel (RACH) performance is essential. Although manual tuning of procedures can be very costly, it is important to design self-organizing network (SON) algorithms that optimize procedures such as RA. In this paper, a mathematical framework for the RACH modeling in LTE technology is presented on the basis of probability theory and statistics. A self-optimization algorithm for RACH preamble sequence allocation and preamble split is then examined using the deduced mathematical framework and system-level simulations. Performance analysis is semi-analytical: First, RACH performance results are produced using analytical model, and second, outputs are fed into a dynamic LTE network simulator. Results show that the proposed optimization approach clearly improves uplink performance and ensures a certain success rate at the first RA attempt so that shorter handover durations and call setup delays can be achieved.