Читать книгу Dynamic Spectrum Access Decisions - George F. Elmasry - Страница 79

This section presents a generic construct of a DSA cognitive engine that can be followed at any of the hierarchical levels explained in the previous section. Although this book is not focused on cognitive engine design, it is appropriate to demonstrate that a common (skeleton) DSA cognitive engine concept can be followed at all entities of the hierarchy presented in this chapter in order to facilitate DSA as a set of cloud services. A DSA cognitive engine is responsible for many tasks, including spectrum sensing information fusion, offering DSA services and the propagation of policies, rule sets, and configuration parameters to the lower hierarchy entities.

Figure 5.2 shows this generic representation of the DSA cognitive engine. At the center of this DSA cognitive engine is an information repository. This information repository can differ from one entity to another based on its hierarchy and place in the hierarchical networks. The information repository can receive real‐time input from local sensors, from peer DSA engines or from other hierarchic DSA engines. The information repository can also be affected by automation inputs such as policies, rules, and configurations. The information repository is coupled with an information fusion and evolver. The DSA engine also has a resource monitor that can create recommendations and warnings to external entities and can disseminate messages to other entities as needed. The resource monitor can trigger policy updates, rules updates or configuration parameter updates that propagate from higher DSA entities to lower DSA entities. The DSA decision maker is a separate process that receives DSA services requests and responds to these requests. A request can be a local request or a dissemination request as explained above. The resource monitor can create internally triggered decisions (as a result of spectrum sensing information fusion) and pass them to the decision maker for consideration.

Figure 5.2 The generic cognitive engine representation that can be used at any network hierarchical entity offering DSA services.

As Chapter 8 explains, the decision maker can consider the impact of co‐site interference before disseminating a response to a service request.

Notice that:

 this skeleton architecture applies for all hierarchical DSA decision making entities.

 local response should be instantaneous.

 dissemination based response should be slower and can depend on the control plane state.

 fusion and information evolver is a continual process that can update the information repository. The information repository can be updated based on external messages or based on local analysis. An example of this internal update is the ROC model hypothesizing threshold explained in Chapter 3. For example, local analysis at a gateway can update this threshold value and create a message that disseminates to the network nodes to update the value of this threshold in order to adjust the tradeoff between the probability of false alarm and the probability of misdetection.

This book is not about developing cognitive engines. However, some critical threads of this cognitive engine skeleton are worth covering to show some key aspects of the decision‐making process.