Conventional two-dimensional phased arrays face two major shortcomings: the presence of ambiguities in direction of arrival measurements and beam broadening endfire effects. The literature provides methods for addressing and minimizing these problems on conventional planar phased array structures, but there has been no investigation into solving these issues with three-dimensional geometries. In this thesis, the design and performance of a cubic phased array that can eliminate endfire effects and dramatically improve direction of arrival ambiguity resolution is investigated. Both beamforming and direction of arrival simulations are performed in MATLAB and 4nec2 simulation environments for cubic phased arrays of various sizes and at different frequencies and demonstrate that the endfire effects are eliminated and direction of arrival ambiguity resolution is dramatically improved. These findings are expected to lead to new designs of high fidelity three-dimensional phased arrays. / Master of Science / Conventional two-dimensional, flat, plane antenna arrays have revolutionized how sensing and detection systems perform. These systems, however, face two major shortcomings due to their "flat" geometry. The computation that determines the direction from which an object is approaching or a signal has been transmitted will have two solutions that are opposite each other in the same way that the polynomial expression x2 = +2 or -2 has two solutions that are opposite each other. This is known as the ambiguity problem and presents major uncertainty in direction finding or direction of arrival measurements. The second major shortcoming has to do with transmitting a signal at different directions. The antenna elements in the array are stationary, but the beams that each element transmits can be aimed in specific directions by controlling the phase of the voltage sources for each respective antenna. This is why it is called a phased array. When every element is transmitting directly forward, it is known as broadside. As the voltage sources for the elements are shifted, or steered, away from this direction, it is known as beam steering. When the beam is steered 90 degrees from the broadside direction, the beams of one column of elements are actually transmitting into the next column of elements, effectively transmitting out of a one-dimensional line array. This is known as endfire and has significant negative effects that are often desired to be avoided. Current scientific literature provides methods for addressing and minimizing these problems on conventional two-dimensional planar phased array structures, but there has been no investigation into solving these issues with three-dimensional geometries. In this thesis, the design and performance of a cubic phased array is presented. The cubic phased array eliminates endfire effects entirely because each face of the cube is identical; when transmitting at 90 degrees off broadside, the transmit area of the cube is identical to that of the broadside direction. The cubic geometry also dramatically improves the direction-finding process. By introducing a third dimension, the mathematics can more precisely determine the direction from which the object or the signal is coming, thus dramatically decreasing the ambiguity simply as a function of geometry. Both beam steering and direction of arrival simulations are performed in MATLAB and 4nec2 simulation environments for cubic phased arrays of various sizes and at different frequencies. This demonstrates that the endfire effects are eliminated and direction of arrival performance is dramatically improved. These findings are expected to lead to new designs of high fidelity three-dimensional phased arrays for a multitude of applications, especially for space applications where the three-dimensional geometry has the added benefit of resolving the requirements for compensation for the tumbling motion of objects in orbit.