Cognitive radio now-a-days has opened a new horizon for researchers to mitigate the up-growing demand for RF spectrum within our limited resources. The threshold plays very important role for sensing process in energy based detectors as it is widely used in CR network due to its fast performance and ability to perform without any prior knowledge or information. Conventional energy detectors has fixed threshold which performs very poor in low SNR conditions for achieving targeted performance. In this paper, we propose an adaptive threshold method consists of two control parameters to adapt the requirements utilizing targeted probability of detection and false alarm. To prove the effectiveness of our proposed method, we first determine thresholds based on different principles like Constant Detection Rate (CDR), Constant False Alarm Rate (CFAR) and Minimizing spectrum sensing error (MSSE) from Berkeley model for low SNR and then compare it with these thresholds and exhibit promising improvements. Comparative results from simulation show the variety of thresholds can be utilized by our method to cope up with the requirement efficiently which leads to improve the throughput by decreasing the number of samples which means the sensing time while protecting the primary user and avoiding the collision probability.