Nowadays a great number of telecommunication applications are used by military as well as civilians. Modulation and demodulation are two essential elements of information transfer by radio signal. As for modulation, one or more physical parameter of an electromagnetic wave is changed through time which denotes the information to be sent. Demodulation can be executed if the modulation type is known in advance, since different demodulations require different hardware and bandwidth arrangements. Furthermore, the methods show variant robustness to noise and data rate while being used on a communication channel.
In modern civilian applications it is an important challenge to control the data rate and the bandwith allocation. Altough the possible modulation types are known by both transmitter and receiver end, the applied method is chosen adaptively, desired communication speed and channel load are taken into consideration too. Consequently, an automatic modulation classification system seems to be a reasonable choice to deliver the data succesfully.
In military application, modulation can be seen as a form of encryption, so no one can retrieve the data without being aware of the modulation type. Modulation recognition helps spy on communications by resolving the applied type of modulation. Apart from understanding the message, jamming can be achived by signals with matching modulations.
Moreover, direction of the radio transmitter can be of utmost interest. With proper antenna array and direction estimator algorithms, the location of a transmitter can be obtained. Beside radio signal detection, direction of arrival (DOA) estimation is used in exploration, astronomy and medicine.
In the following, well-known digial modulations, feature-based modulation classification and DOA estiomation algorithms are introduced. The methods are verified by simulations and modulation recognition is presented in a practical aproach as well.