||As an emerging research topic in the last few years, MIMO radar problems have been studied by a few research groups. However, the fact that each group possesses their own interpretation on defining the 'MIMO radar' scheme not only leads to confusions but also brings inconvenience for carrying out subsequent research. This thesis aims to build up a general framework, as well as provide some new insights for MIMO radar investigation from the signal processing perspective. First of all, a versatile description of the nomenclature 'MIMO radar' is defined, followed by a corresponding signal model, which is shown to be able to characterize various problems in different radar contexts. Secondly, a brief overview of different MIMO radar problems in the literature is presented, based on our unified model. Estimation and detection problems are of the main concern, as they are two fundamental functions of radar systems. The following part of the thesis is devoted to a general detection problem, under the Gaussian noise assumption. To be more specific, we seek for the optimal detector in the Neyman-Pearson sense by deploying a Bayesian approach. Novel expressions for the detection performance are derived for arbitrary second order statistic of the noise. In addition, waveform design problems are discussed afterwards, but only restricted to the white noise case. More general waveform designing are still open problems, and they are addressed at the end of this thesis, along with some other potential future work.