Cognitive radio (CR) has been classified as a promising technology which can significantly mitigate the impact of spectrum underutilization and spectrum scarceness arising from massive expansion of wireless applications. That is as a result of the CR capability to improve both the spectral and energy management efficiently. Spectrum sensing is the most important technological requirement for accomplishment of cognitive radio systems. For a known primary user (PU) signal, matched filtering is the common technique used for signal detection over a specific band of spectrum, therefore, detection performance enhancement of the matched filter detector has gained great attention. In this paper, a new multiple antenna elements (MAE) and matched filtering (MF) based spectrum sensing technique named as MAE/MF is proposed for detection capability enhancement. Multiple antenna elements utilization improves the received signal to noise ratio (SNR) in proportional to the achieved antenna array gain. The likelihood ratio test is used to decide the presence/absence of PU signal. Simulation results showed that the proposed MAE/MF technique outperforms the single antenna element based matched filter (SAE/MF) and other existing techniques in the state-of-the-art, especially at extremely low signal to noise ratios and limited number of received signal samples.