The use of Global Navigation Satellite System (GNSS) receivers, such as the Global Positioning System (GPS), is wide spread in the society today. GNSS receivers are used for navigation in many safety critical sectors, such as maritime, aviation and first responder applications, but also for timing synchronization in important infrastructural applications such as the power grid, mobile telephony systems and financial investment systems. With the wide spread usage and high reliance of GNSS, also comes vulnerability to outages due to interference, jamming and spoofing. The first step towards mitigating the effects of jamming is to detect that there is a jammer and if the delivered GNSS position and time solution is unreliable. The main goal of this work is to evaluate different approaches for detecting interference signals in the GNSS frequency bands. The focus of the evaluation has been on detectors that makes no assumption on the jammer signal characteristics. The evaluations in this work are based on measurements of authentic GPS and jamming signals. It has been shown that detectors based on received energy, automatic gain control (AGC) levels, and receiver carrier-to-noise (C/N0) estimates are able to detect many different types of signals ranging from narrow band (continuous wave) signals to wide band (>20 MHz) signals. The performance of the different methods depends on the application and its requirements, as well as on the hardware and the possibility to change parameters such as the decision threshold and detection time. Energy detection is a conceptually simple detector that often performs very well. However, to compute the energy, one essentially needs to have access to the raw IF samples. These are generally not available in an off-the-shelf product. AGC values could be available and therefore an AGC based detector could be a good alternative to the energy detector, with similar properties. Of course, the AGC gain is necessarily quantized, and as such some detection performance is lost as compared to dealing with raw IF data. C/N0-based detectors in general are not suitable for applications where the received satellite signal strength normally varies, such as dynamic scenarios in urban environments. The reason is that these detectors cannot distinguish between an increased noise-plus-jammer-power and a decreased desired signal power.