Data recordings are often corrupted by noise, and it can be difficult to isolate clean data of interest. For example, mobile electroencephalography is commonly corrupted by motion artifact, which limits its use in real-world settings. Here, we describe a novel noise-canceling algorithm that uses canonical correlation analysis to find and remove subspaces of corrupted data recordings that are most strongly correlated with subspaces of reference noise recordings. The algorithm, termed iCanClean, is computationally efficient, which may be useful for real-time applications, such as brain computer interfaces. In future work, we will quantify the algorithm's performance and compare it with alternative cleaning methods.