MIMO radar is an important category of radars that in recent years have attracted the attention of researchers. Unlike a standard phased-array radar, which transmits the scaled versions of a single waveform, a MIMO radar system can transmit via its antennas multiple probing signals that may be arbitrarily chosen. This waveform diversity causes superior capabilities compared with a standard phased array radar. Due to the presence of multiple receivers in these radar systems, a significant cost reduction will be achieved by reducing the sampling rate at the receivers. Compressive sensing method reduces the sampling rate without loss of essential signal information. This method has been employed in many signal processing applications, including processing of radar signals. Using this method, we can reduce the complexity of receivers and MIMO radar power and as a resultwe will reduce the cost substantially. It is worth noting that a condition of using this method is that the signal is sparse at their base. The previous works on compressive sensing MIMO radar, have been with the aim of the sampling rate reduction in time domain. A lot of works have been done assuming a uniform distance between arrays. Arrays with uniform spacing actually use the Nyquist rate sampling in space domain. The disadvantage of this setup is that the product of the number of transmitters and receivers should be set to have a linear relationship with the array apertureHowever, the spatial compressive sensing can also be used. In array signal processing, the resolution of angle of arrival increases, with increasing array aperture. Increasing the array aperture without increasing the number of receivers and transmitters will causea measurement error. In this thesis, we have proposed a new method which uses the sparseness of the signal and suggests a structure of random array with less elements covering a large aperture. For the first time the block coherence of the sensing matrix in spatial domain is usedas an criterion and it is proven that in spatial domain block coherence has a little value, and its relation with coherence is found. The received signal is modeled as a block sparse signal and then we use the upper bands of coherence and block coherence to find a lower band on the number of antennas required in MIMO radar. Finally, we found that by using the block compressive sensing algorithm instead of compressive sensing algorithm, less number of transceiversare required. Keywords: Spatial block compressive sensing, MIMO radar, block coherence
