In this article, we investigate the performances of conventional cooperative sensing (CCS)‐ and superior selective reporting (SSR)‐based cooperative sensing schemes in a multiple‐input, multiple‐output (MIMO)‐aided non‐orthogonal multiple access (NOMA)‐driven cognitive radio (CR) networks. First, we derive expressions for the probabilities of false‐alarm and signal detection for an optimal detector in the considered MIMO‐NOMA CR setup. Next, we study a constrained optimization problem that maximizes the sum rates of CCS and SSR schemes, with a constraint on the power allocation at SUs. This problem is shown to be a constrained convex optimization problem. Later, we analytically characterize and compare the sum rate performances of CCS and SSR schemes in MIMO‐NOMA CR and MIMO‐based orthogonal multiple access (MIMO‐OMA) CR networks. Extensive computer simulation results are presented which corroborate our analysis, which also quantifies the gain in performance of the proposed MIMO‐NOMA CR network employing SSR scheme when compared to conventional MIMO‐OMA CR systems. Furthermore, we present an experimental study based on RTL‐SDR setup, which supports our theory and also paves way towards a practical implementation of the proposed framework.