频谱检测中基于FCM的自适应门限选择机制

在认知无线电中,能量检测法是次用户实现频谱检测的重要方法之一。其中,检测参数的设置尤为重要。然而,当无线网络环境改变时,一些重要检测参数(如检测器门限)将会随之改变,因而及时准确地获取检测参数就显得十分必要。本文首先在理想高斯白噪声信道条件下推导了能量检测的最佳门限表达式。为了在变化的网络环境中快速自适应地获得最佳门限并降低检错率,提出了一种基于模糊C均值(fuzzy C means,FCM)的自适应门限选择机制。该机制只需根据接收到的能量样本的相似性和差异性进行聚类,选取隶属度值差异最小的能量样本作为最佳门限值,而无需信噪比、初始门限等先验信息,因此在能量检测中具有更强的自适应性。Matlab仿真结果证明,新机制下获取的最佳门限与本文中推导的高斯白噪声信道下的最佳检测器门限相比,具有很好的拟合度。

FCM based adaptive threshold selection mechanism in spectrum detection

Energy detection is an important method in cognitive radio for secondary users to achieve spectrum detection, where detecting parameter setting is a key problem. However, as the network environment changes, some crucial detection parameters, such as detector threshold, will change as well. Thus it is necessary to obtain detection parameters accurately and timely. To solve this problem, an optimal threshold in energy detection over the additive white Gaussian noise channel is deduced and then an adaptive method is proposed to find the optimal threshold based on fuzzy C means(FCM). Priori information about signal to noise ratio and the initial threshold is not required in this method. Only clustering according to the similarities and differences of the received energy samples needs to be achieved, and then select the energy samples with the minimum degrees of membership differences as the optimal threshold. Matlab simulation results show that the proposed mechanism has a good degree of fitting with the deduced optimal detector threshold over the additive white Gaussian noise channel.