论文部分内容阅读
提出一种基于神经网络预测信道质量指示(CQI)的长期演进下行调度方法,设计了基于神经网络的CQI预测模型,使用系统中已获得的CQI历史值来预测调度时刻的当前CQI值,调度器使用此预测值进行调度.仿真结果表明,该方法对由CQI延迟所导致的系统性能下降有明显改善.
A long-term evolution downlink scheduling method based on neural network prediction channel quality indicator (CQI) is proposed. A CQI prediction model based on neural network is designed. The CQI historical value obtained in the system is used to predict the current CQI value at the scheduling time. The scheduler This prediction is used for scheduling.The simulation results show that this method has obvious improvement on the system performance degradation caused by the CQI delay.