CAJ | 학술논문

为有效管理嵌入式系统，尤其是减少移动终端的电源功耗，设计一种更加精确的动态电源管理方案。在Linux平台上运行，基于API行为特点，利用BP神经网络算法进行应用类型预测，通过对应用类型的预测，提前对系统状态进行调整。实验结果表明，在不影响系统性能的前提下，该方案可有效降低功耗，实现对嵌入式设备电源的实时、动态管理。
위유효관리감입식계통，우기시감소이동종단적전원공모，설계일충경가정학적동태전원관리방안。재Linux평태상운행，기우API행위특점，이용BP신경망락산법진행응용류형예측，통과대응용류형적예측，제전대계통상태진행조정。실험결과표명，재불영향계통성능적전제하，해방안가유효강저공모，실현대감입식설비전원적실시、동태관리。
In order to effectively manage embedded systems, especially reduce power consumption of the mobile terminal,this paper proposes a power management scheme,which is based on the design of a more refined dynamic power management scheme. It is based on Application Program Interface ( API ) behavioral characteristics, using BP neural network algorithm to predict the type of application,through the effective prediction of application types. It can adjust the system state in advance,without affecting system performance,effectively reducing power consumption,realize the power of real-time embedded devices and dynamic management.