The recent COVID-19 outbreak drove the attention to methods for monitoring the flow of people between human settlements, including traffic flow. Although the remote sensing of nighttime lights is a viable option to estimate traffic flow-derived indicators, changes in radiance levels at night are not all associated with traffic. This paper presents the theoretical approach proposed on the development of an algorithm able to identify spectrally unbiased control samples for regions of interest (ROI), namely roadway sections. Firstly, an experiment is presented to put in evidence the background dependency of the DNB monthly composites (vcm) radiance levels. Then, an overview of the algorithm is presented, followed by an empirical estimation of its time complexity. The results showed that the algorithm has an O(n) time complexity and that control samples and ROIs can have similar time series features, indicating that analysis without the use of control samples can lead to biased results.