Advanced spectrum sharing technology for future 5G and Beyond 5G (B5G) networks have the potential to achieve higher spectrum utilization efficiency by utilizing knowledge of the carrier frequency and angle-of-arrival (AOA) of the signals emitted from multiple base stations. In this paper, an algorithm to perform wideband spectrum sensing along the frequency and angular domains simultaneously utilizing sub-Nyquist analog-to-digital converters (ADC) is proposed. The proposed algorithm is developed based on the conventional multi-coset sampling (MCS) algorithm in order to support AOA estimation, and is expressed as a typical compressive sensing problem. Spectrum sensing performance is evaluated using Monte Carlo simulation, and results show that it can successfully sense the wideband two-dimensional spectrum without the need for high-rate ADCs.