Low-Cost Air Quality Sensor Nodes (LCAQSN) are being widely deployed across numerous cities worldwide as a new way for assessing air quality. Despite facing challenges related to accuracy and consistency, these nodes offer valuable insights, substantially reducing the costs associated with monitoring air pollutants. A significant hurdle to address is the occurrence of missing values. In this study, we hypothesize that a network of LCAQSN within the same urban environment can effectively retain and utilize shared information to accurately impute missing values, even in cases with substantial gaps in the time-series data of individual nodes. Employing various Machine Learning techniques, our analysis reveals that a network comprising 26 LCAQSN in the Greater Thessaloniki Area, Greece, with 40.93% missing values, can achieve an imputation accuracy of 0.7 R2 on a simulated test set of 10% of missing values. These findings exhibit great promise and unveil numerous opportunities for leveraging LCAQSN networks further, including data fusion and downscaling applications.