中文摘要：

为消除EMI现场测试中环境常驻窄带信号的影响,通过完善改进的差分法,提出一种能够保留特征小信号的常驻窄带信号剔除算法。先对环境噪声信号进行幅偏和频偏修正,将修正后的环境噪声与混合信号进行对比：对于混合信号小于环境噪声的部分,将仪器底噪存为设备辐射;对于环境噪声淹没在仪器底噪中,或者混合信号大于环境噪声超过6 d B的部分,将混合信号当成设备辐射信号;对于环境噪声大于仪器底噪,且混合信号大于环境噪声不超过6 d B的部分,先将环境噪声中的底噪剔除,然后用混合信号减去差值,进而得出设备辐射信号。实验证明该方法能有效剔除环境常驻窄带噪声信号,差分结果与标准暗室测试结果相关系数达到0.974 4。

英文摘要：

In order to eliminate the impact of the environment permanent narrowband signal in EMI field tests, a permanent narrowband signal culling algorithm by perfecting the improved differential method is proposed, which can retain the characteristic small signals. The operational procedure of this method is that first revise amplitude de- viation and frequency deviation of ambient noise signal, then compare the revised ambient noise with mixed signal. When the mixed-signal is less than the ambient noise, the instrument＇ s background noise will be stored as equip- ment radiation signal. When the ambient noise is drowned in the instrument＇ s background noise, or the mixed sig- nal is greater than the ambient noise exceeding 6 dB, the mixed signal will be regarded as equipment radiation sig- nal. When the ambient noise is greater than the instrument＇ s background noise and the former is less than the mixed signal, but not more than 6 dB, the concrete steps of figuring out the equipment＇ s radiation signal are as fol- lows： first get a value by removing the instrument＇ s background noise from ambient noise; next use mixed noise minus the value, thus we＇ ll get the final value. The experiments demonstrate that this method can eliminate the en- vironment permanent narrowband noise signal effectively, and the correlation coefficient between the differential re- suits and the anechoic test results can reach 0.974 4.