7.5 Cognitive Adaptation Engines 189 individuals who interact with the system. The knowledge base is the representation of the expertise. Each piece of expertise is typically termed a rule and represented using an IF/THEN format. For example, one rule in our system may be, IF frequency band of interest is currently in use, THEN alter frequency. The expertise or the rules are created by the domain expert. The domain expert would say specifically what frequency is the optimal to use. Typically, as shown in Figure 7.2, a knowledge engineer is used to encode the expert's knowledge into a form that can be used by the expert system. For some projects the domain expert and the knowledge engineer may be the same person, or in our case the information produced by the domain expert actually comes from an equation. The second major component in the system is the working storage. This com- ponent holds the problem-specific information for the problem currently being solved. This information is typically termed the facts. For example, a fact may exist stating that frequency band 5.325 GHz is currently in use. This information can be added anytime the system is running. In our case, the facts are brought in from sensors on the wireless system. Information such as the battery life, channel noise figure, and received signal-to-noise ratio can all be represented as facts. The inference engine is also a primary system component in an expert system. This component includes the code that combines the information from the working