基于BP神经网络的用户感知体系优化仿真

研究提高网络质量评估准确度问题。用户感知是网络质量评估的标尺,而现有无线网络优化中影响用户感知的有众多因素,和网络性能指标之间没有明显的映射关系,采用传统的数学方法很难对复杂的非线性问题定量分析,难以从网络性能指标中实现用户感知评估导致网络质量评估不准。为解决上述问题,提出一种基于BP神经网络的用户感知体系。采用具有非线性映射功能的神经网络,建立用户感知体系模型,实现用户感知的评估。通过比较验证,对用户感知度的评估达到了合理误差,并能够提升网络优化主动智能性,对网络质量评估更准确,对网络优化工作提供了设计依据。

Simulation of User-Perceived System Optimization Based on BP Neural Network

The problem of improving the network quality evaluation accuracy was studied.The user perception is the ruler of network quality.For many factors affect the user perception,it has no obvious mapping relation between user perception and network performance indicators.The complicated non-linear empirical problem is difficult to be solved by traditional mathematical methods.It is difficult to realize evaluation of user perception from network performance indicators which leads unaccuracy evaluation of network quality.To solve the above problems,an architecture of user-perceived wireless network was introduced,which adopted the neural network with the non-linear mapping function to evaluate the user perception.With comparison and authentication,reasonable error in the evaluation of user perception is achieved,the initiative of network optimization can be increased,and the method can be used for guiding the network optimization.