We consider the problem of simultaneously estimating three parameters of multiple microseimic events, i.e., the hypocenter, moment tensor, and origin time. This problem is of great interest because its solution could provide a better understanding of reservoir behavior and can help to optimize the hydraulic fracturing process. The existing approaches employing spatial source sparsity have advantages over traditional full-wave inversion-based schemes; however, their validity and accuracy depend on the knowledge of the source time-function, which is lacking in practical applications. This becomes even more challenging when multiple microseimic sources appear simultaneously. To cope with this shortcoming, we propose to approach the problem from a frequency-domain perspective and develop a novel sparsity-aware framework that is blind to the source time-function. Through our simulation results with synthetic data, we illustrate that our proposed approach can handle multiple microseismic sources and can estimate their hypocenters with an acceptable accuracy. The results also show that our approach can estimate the normalized amplitude of the moment tensors as a by-product, which can provide worthwhile information about the nature of the sources.