In this paper, the design of a subspace compressive GLRT (SSC-GLRT) detector has been pursued for MIMO airborne radar system when the secondary sets of data are available in addition to the primary one. It is observed that with very reduced amount of measurement data at sub-Nyquist rate that is equal to the number of virtual bi-static radar in the presence of a single target, SSC-GLRT provides the same performance as the conventional GLRT with much larger length of the data, i.e., an order of magnitude higher given by the product of the number of elements in the received antenna array and coherent pulse interval (CPI). Therefore, SSC-GLRT is much energy efficient and useful in the scenario where the computational complexity becomes a burden. Also, this paper illustrates an elegant way of mapping the angular and Doppler domain to an one dimensional discrete vector for the formulation of compressive sampling (CS) based ℓ1 reconstruction of the signal subspace when it is unknown so that it can successively be used in SSC-GLRT. The simulation example further corroborates the effectiveness of the proposed SSC-GLRT detector.