الفهرس 
Chapter 1 : Introduction
Problem definitionOnly 14 pages are availabe for public view
 |  | from | 119 |  |  |
| | | from | 119 | | |
Abstract
In this thesis, a new strategy is presented called Mixed Mobility Prediction (MMP).This strategy is used to solve handoff problem in personal communications service networks by predicting the next cell for a Mobile Terminal (MT). When a MT moving among cells during the active call, this active call has to be switched to new base station of the next cell in order to keep the active call continued. Therefore the next cell of the MT has to be predicted in order to be able to reserve a resource for MT’s active call to prevent it from dropping down. MMP consists of two predictors: the first one called Association Rule Predictor (ARP) which is used for users having sufficient history of paths stored in the network. The second predictor called Weighted Ant Colony Predictor (WACP), which used for users have no history of paths stored in the network. ARP and WACP are merged together in the context of MMP for users having insufficient history of paths stored in the network. ARP predicts the next cell of MT by using calculations of support and confidence taking into consideration time of entering current cell and time of calling inside current cell. WACP predicts the next cell for MTs having no histories by using Ant Colony algorithm in addition to giving weights to the famous places found in each cell and roads leads to each neighboring cell. MMP predicts the next cell of MTs having no histories stored in the network by merging the decisions of both predictors (ARP and WACP). The prediction of MMP is calculated by multiplying a specific weight to the summation of prediction decisions of both ARP and WACP in order to get an accurate prediction decision. The proposed approach is compared with other methods of mobility prediction and its noticed that Mixed Mobility Prediction is outperformed than the compared the state-of-arts methods in terms of; Prediction Accuracy (PA), and Quality of Measure (QM).