Performance of cooperative spectrum sensing in a cognitive radio (CR) network is investigated where each CR node uses an improved energy detector (IED) to sense the primary user (PU), and makes a local decision regarding the presence of PU using double thresholds (DTHs). The local hard decisions are combined at fusion centre (FC) to obtain the global decision. The advantage of a DTH-based system over a single threshold based one is, a CR node can opt for no decision when a decision variable lies in the fuzzy zone between two thresholds. Such censoring reduces the transmission overhead between CR and FC without significantly affecting the receiver operating characteristics. In this study, the performance of the abovementioned CR network has been assessed in terms of the average number of normalised transmitted sensing bits (knor), the total error probability (Pe,n) and optimal number of CR users that ensures minimum Pe,n. It was observed that knor increases as the signal power raise factor of IED increases or failed sensing probability decreases. Further, the agility of the network improves as the PU death rate increases. Impact of reporting channel on the sensing performance has also been indicated.