CFAR Outlier Detection With Forward Methods
IEEE Transactions on Signal Processing
| Hi-index | 0.00 |
The detection threshold setting and noise uncertainty are known to be critical aspects for energy detectors. Adaptive methods for threshold setting outperform non-adaptive methods due to flexibility and robustness. There exists several adaptive threshold setting methods of which the forward consecutive mean excision (FCME) algorithm is among the most attractive ones since it is blind, computationally simple and efficient. However, in some situations, it may give a too large threshold. We propose to apply median filtering with the FCME. Real-life real-time measurement results show that proposal enables more stable thresholds even in the situations when there are no signal-free reference samples for the initial threshold computing.