Distributed Simulation of Large-Scale PCS Networks

There has been rapid growth in demand for mobile communications over the past few years that has led to intensive research and development of complex PCS (personal communication service) networks. Capacity planning and performance modeling is necessary to maintain a high quality of service to the mobile subscriber while minimizing cost. Simulation is widely used in such studies, however, because these models are extremely time consuming to execute, only small-scale PCS networks have previously been simulated. In this paper, we examine the use of the Time Warp distributed simulation mechanism in simulating large scale (1024 or more cells) PCS networks. An object-oriented, distributed discrete event simula-tor using Time Warp has been developed, and initial performance measurements completed. Speedups in the range of 2.8 to 7.8 using 8 Unix workstations have been obtained, enabling simulation runs that require 20 hours on a single workstation to be completed in only 3.5 hours. A personal communication service (PCS) network 3] is a wireless communication network, which provides communication services for portables or PCS subscribers. The service area of a PCS network is partitioned into several sub-areas or cells, and radio channels are assigned to the cells. When a subscriber makes a phone call, the system tries to allocate a radio channel for the connection. If all channels are busy, then the phone call is blocked. It is important to engineer the system so that the blocking probability is low (e.g., less than 1%). Discrete event simulation is a popular technique for designing and modeling large high-capacity personal communication services (PCS) networks. In a PCS simulation, the network is usually represented by hexagonal or square cells. Ideally, PCS simulation experiments should model large networks containing thousands of cells. However, due to limited computing resources, most studies only examine small-scale networks containing fewer than 50 cells 7, 13], and output statistics of the boundary cells are usually discarded to avoid the boundary eeect. Lin and Mak 9] showed that this approach may lead to biased output statistics, and suggest using a large network with a wrap-around topology to achieve reliable simulation results. To illustrate the impact of network size, consider the following PCS simulation. Assume that each cell is assigned a xed number of radio channels, referred to as the xed channel assignment scheme. A portable (or mobile phone) resides at a cell for a period of time, then moves to another cell. When a …