Fuzzy Logic in Wireless Communications
Fuzzy Logic can be employed to model systems and situations by taking into consideration uncertainty and ambiguity. As such it can be utilized in problems for which knowledge of all factors is insufficient or impossible to obtain. Methods furnished with fuzzy logic are most commonly used in more difficult conditions such as dealing with non-linearity and time-variance.
Fuzzy logic and, more specifically, fuzzy control traditionally incorporates human expert knowledge into a rule-based framework. It may, however, be further expanded with learning algorithms to derive the fuzzy control parameters from sample data. These parameters may be obtained by combining fuzzy logic with related soft computing disciplines such as, e.g., neural networks, evolutionary computation techniques etc.
One of the directions this project has taken, is in the area of Cognitive Radio. Whereas the demand for access to the radio spectrum has grown dramatically, the spectrum bands assigned to licensed users are vastly underutilized. Cognitive Radio has as its main aim to more efficiently use the radio spectrum, which is turning into an exceedingly scarce resource, while at the same time minimizing interference to the licensed (primary) user and maximizing the quality of service to the non-licensed (secondary) users. Cognitive Radio exploits artificial intelligence in order to sense, adapt to, learn from and adjust to their surroundings. Fuzzy Logic is a means proposed to meet these challenges.
The other direction is in Fuzzy Game Theory, an area in which research has been conducted together with Elisabeth Rakus-Andersson from the department of Mathematics and Natural Sciences at BTH. In a dynamic spectrum assignment environment there will be multiple agents. In order to describe their behavior, a game theoretical approach can be employed. Since uncertainty is already present in this context, which Classical Game Theory aims to model, a fuzzy aspect can be added to blur the sharp boundaries of numerical values in the classical model to create a more accurate model.