Angle of arrival (AoA) estimation or direction finding (DF) has been extensively studied in the literature. Many existing AoA estimation algorithms perform a search over the entire angle space for the correct result. While such computational complexity is manageable in a number of scenarios, for future commercial small and low power wireless devices such as Bluetooth based internet of things (IoT) devices, the existing methods may not be feasible. This problem is exacerbated in 3D DF where the elevation angles also need to be computed. In this paper, a low complexity direction finding approach is presented which assumes that the transmit sequence is known. Under additive white Gaussian noise with zero mean, averaging the received sequence over multiple time instances `filters' out the noise. The task of determining the azimuth and elevation angles then reduces to simpler operations. We show that the proposed method yields similar estimation accuracy as the conventional MUSIC algorithm in both free space single path as well as indoor multipath environments while needs much less computation. The proposed method may also be used in combination with existing algorithms developed for more complicated DF scenarios.