Sub-optimal all-sky detection of periodic gravitational waves

The design of a (roughly speaking optimal) GLRT-based detector for continuous gravitational waves (GWs) is a simple task and results in a bank-of-filter structure. Unfortunately, the computational burden of such a GLRT is unaffordable with the current computational powers: resorting to some kind of sub-optimal detectors seems unavoidable. We propose three new detection algorithms that can be implemented with a reasonable computational complexity, and analyze their behaviors both in terms of error probabilities and in terms of computational costs. With respect to these performance figures the improvements over the classical GLRT are impressive.