Undesired, human behavior in public environments is an increasing issue in today’s society. The overload of security operators and law enforcement addresses the need for automatic detection of anomalous behavior. The EU-project ADABTS aims to facilitate the protection of EU citizens, property and infrastructure against threats of terrorism, crime and riots, by the automatic detection of abnormal human behaviour. At the Acoustics & Sonar department of TNO Defence and Safety, part of this problem is addressed by means of acoustical detection of anomalous events. The approach is based on ‘scanning’ public environments by applying beamforming on the outputs of an acoustical sensor array and applying classification algorithms for detecting specific sources. In this Master’s thesis, an initial step is taken with the development of a real-time beamforming system that delivers required sound parameters used in sound classification. A number of different beamforming methods have been considered, differing in performance and computational complexity. Conventional methods like Delay and Sum (DAS), possibly combined with the use of static, frequency-invariant windows, lack spatial resolution at especially lower frequencies and are unable of coping with multiple interfering sources. Other methods provide an improved performance on the cost of increased complexity. The method known as Minimum Variance Distortionless Response (MVDR) beamforming maintains a high spatial resolution at lower frequencies. Two main versions of this method are frequently used: a static (non-input-based) one and a dynamic (input-based) one. Static MVDR (SMVDR) is able to maintain performance at lower frequencies, but due to its static nature it does not add any extra value in multi-source environments. Dynamic MVDR (DMVDR), on the other hand, is partly capable of filtering away undesired coherent interferers and also has an improved spatial response in single-source environments. Its computational complexity, still, is an important bottleneck. The search for less intensive beamforming methods leads to a way of adaptive beamforming. Beamformers in which static beamforming and dynamic filtering are split in two different parts, are able to alleviate complexity. However, the need for adding extra elements to account for target-signal cancellation in multi-source environments destroys the computational advantages, making it unsuitable. The developed real-time application takes into account the intensive routines of DMVDR. Since the properties of the deployment platform are not known in advance, it is supplied with a mechanism for adapting to different and changing, available hardware resources such as available CPU-time, arithmetic units and memory. In this way it will always deliver the best possible solution, based on what the user is offering. Still, an extensive implementation process has led to a relatively fast execution of the algorithm. The system is supplied with a user interface for controlling a number of parameters and for obtaining the first visual effects. Furthermore, it is provided with a user-friendly mechanism for calibrating the system for each possible deployment environment.