This paper addresses the problem of optimal cooperative spectrum sensing in a cognitive-enabled sensor network where cognitive sensors can cooperate in the sensing of the spectrum. Such sensor networks are assumed to be power resource constrained. With a given threshold for the accuracy of the spectrum detection, we find the optimal number of cognitive sensors participating in the cooperative spectrum sensing and the optimal sensing interval that minimize the total energy consumption of the cooperative sensing. First, the mathematical lower bound and upper bound for the number of cooperative cognitive sensors are found. Then the optimization problem to minimize the total energy consumed by a group of sensors is presented. Finally, an efficient approximate solution to the optimization problem is proposed. Numerical calculations validate the accuracy and the performance of the proposed scheme. The impact of the noise uncertainty, the choice of the energy detection threshold, and the spectrum bandwidth on the detection accuracy and the minimum total energy consumption is also studied.