Dynamic spectrum access employing cognitive radios has been proposed, in order to opportunistically use underutilized spectrum portions of a heavily licensed electromagnetic spectrum. Cognitive radios opportunistically share the spectrum, while avoiding any harmful interference to the primary licensed users. One major category of cognitive radios consists of is interweave cognitive radios. In this category, cognitive radios employ spectrum sensing to detect the empty bands of the radio spectrum, also known as spectrum holes. Upon detection of such a spectrum hole, cognitive radios dynamically share this empty band. However, as soon as the primary user appears in the corresponding band, cognitive radios have to vacate the band and look for a new spectrum hole. This way, reliable spectrum sensing becomes a key functionality of a cognitive radio network. The hidden terminal problem and fading effects have been shown to limit the reliability of spectrum sensing. Distributed cooperative detection has therefore been proposed to improve the detection performance of a cognitive radio network. In this thesis, a distributed detection scheme based on hard fusion of local results is considered. Each cognitive radio senses the spectrum and sends the result to the fusion center, and there the final decision is made about the presence or absence of the primary user. Note that, in general, cognitive radios are low-power sensors and thus energy consumption becomes a critical issue. In this thesis, several energy-efficient approaches are proposed, in order to minimize the maximum average energy consumption per sensor, while satisfying the sensing reliability of the cognitive radio network. The sensing reliability is defined by a lower bound on the probability of detection and an upper bound on the probability of false alarm. This way, the primary user is protected from the cognitive radio transmitter’s interference and also the chance of losing spectrum access through erroneous detection of the primary user in an empty band is constrained. First, a censoring scheme is considered where cognitive radios send their results to the fusion center only if they are deemed to be informative. Second, a combined censoring and truncated sequential sensing scheme is depicted which is shown to be more energy-efficient than the former case due to the sensing energy reduction. And third, a combined censoring and sleeping scheme is discussed where on top of censoring, each cognitive radio switches off its sensing module with a specific sleeping rate, in order to save energy both on transmission and sensing. It is shown that all the proposed schemes, particularly combined censoring and sleeping as well as censored truncated sequential sensing delivers significant energy savings. Further, we conclude that when a cognitive radio system is appropriately well-designed in terms of energy efficiency, increasing the number of cooperative cognitive sensors, not only improves the detection performance, but also reduces the average energy consumption of individual cognitive radios. Finally, an optimal fusion strategy for energy-constrained hard-fusion based cognitive radio networks is presented, which optimizes the network throughput subject to a constraint on the average energy consumption of individual radios and a constraint on the amount of interference to the primary user. It is shown that the majority rule is either optimal or close to optimal in terms of the network throughput.