Random access technology and code assignment schemes for UTRA systems

This thesis is devoted to the performance evaluation of Random Access Channel (RACH) and the design of Orthogonal Variable Spreading Factor (OVSF) code assignment schemes for both the UTRA-TDD and FDD systems. As getting through the RACH is the initial and primary step for accessing the network and getting service, the performance of RACH has a strong effect on Quality of Service (QoS) and therefore needs an in-depth study. At the base station, OVSF codes are the basic resource units for bandwidth assignment when a user's access request sent on RACH is correctly received. To satisfy the IMT-2000 requirement of “bandwidth on demand”, these OVSF codes should be properly managed and assigned so as to achieve high efficiency in supporting various multi-rate traffic classes.; In both the UTRA-TDD and FDD systems, slotted ALOHA is the random access protocol used in RACH. To pave the way for analyzing the delay performance of RACH, we first derive the closed-form delay distributions of slotted ALOHA and nonpersistent CSMA protocols under large population scenario. Three retransmission backoff policies, namely Uniform Backoff (UB), Binary Exponential Backoff (BEB) and Geometric Backoff (GB), are analyzed respectively. Most interestingly, we find the general stability conditions for guaranteeing finite delay moments under the BEB policy. These results fill in the gap in the complete delay analysis of random access protocols and can be generalized to the systems with large population size and using contention-based protocols. In addition, we study the exact way of r_max (maximum number of retransmissions allowed) in trading-off the blocking and delay performance.; Based on the above analytical results, we then evaluate the throughput, blocking and delay performance of RACH in both the UTRA-TDD and FDD systems. The ideal, the AWGN and the multi-path frequency selective Rayleigh slow-fading channel models are considered in the derivation of closed-form solutions. In UTRA-FDD, the power ramping technique, whereby the mobile station can increase its transmission power after each unsuccessful random access attempt, is used to enhance the success probability of retransmission. In this work, we analyze and compare three different power ramping schemes, viz. Fixed Step Power Ramping (FSPR), Linear Step Power Ramping (LSPR) and Binary Exponential Step Power Ramping (BESPR). The results derived present a framework for studying power ramping technique on other random access protocols.; Finally, we turn to the design of OVSF code assignment schemes. After analyzing the mechanism of OVSF codes in supporting multiple transmission data rates, we define “flexibility index” to measure the capability of an assignable code set in supporting multi-rate traffic. Using this new concept as the criterion, we propose and analyze two computational efficient schemes, namely Compact Assignment (CA) and Rearrangeable Compact Assignment (RCA). They both can leave the system as flexible as possible (in accommodating various multi-rate traffic classes) after each code assignment. Analytical results, verified by simulation results, show that the proposed schemes can offer the blocking probability, throughput and fairness index very close to the theoretical bounds derived. Compared with other schemes, our CA and RCA schemes have the combined advantage of simple, efficient, stable and fair.