作者: R. Narasimhan , D.C. Cox
DOI: 10.1109/PIMRC.1998.733570
关键词:
摘要: A new handoff algorithm based on recognition of patterns in the received signal power is presented. The uses constancy large scale variation with respect to a base station improve performance comparison conventional hysteresis rule. probabilistic neural network (PNN) used for pattern classifier. For given environment, training run performed. set consists averaged samples (in dB) from nearby stations within adjacent spatial windows. probability failure strength defined and determines possible serving station(s) each class. performed if class sequence classes requires handoff. Performance this rule made environments four five stations. Simulation results indicate that fixed failure, recognition-based considerably fewer handoffs This reduction unnecessary decreases signalling load.