The millimeterwave radar has made possible high resolution tracking, activity classification, and vital signs detection, at higher precisions than is possible with most other wireless approaches. However, detecting multiple moving targets is a challenge, as dynamic scene with a lot of motion leads to clutter and noise, which interfere with the responses of targets of interest. We present a digital beamforming approach using the MIMO radar, with a range resolution of 6.4 cm and a Doppler resolution of 0.18 m/s, which reduces interference between closely neighboring targets. Thus, measurements of individual target micro-Doppler signatures are possible, even in the presence of multiple other moving targets, and the signatures are, thereby, used to train a Deep Neural Network (DNN) for activity classification. The DNN has been applied to recognize six exercise-based classes, correctly predicting with over 95% classification accuracy for all classes, but that is extendable to fall detection and other activities.