Background & Objective: This paper proposes a modified eigenvalue-based Generalized
Likelihood Ratio Test (GLRT). The proposed model is appropriate for cognitive wireless radio system
under colored noise and unknown primary user's power. Herein, we introduce an optimization problem,
which maximizes the detection probability under the constraints of false alarm, sensing time, the
weighting of eigenvalues, and Energy Efficiency (EE). In addition, we improve the detection probability
by achieving the proper number of cooperative secondary users using the power method.
Conclusion: Simulation results show that the computational complexity of the proposed method is
lower than that for traditional methods, Energy Detection (ED) and Eigenvalue-Based (EVB).