Applying Case-Based Reasoning to Tactical Cognitive Sensor Networks for Dynamic Frequency Allocation

Abstract

This paper proposes a cognitive radio engine platform for making exploitation of available frequency channels usable for a tactical wireless sensor network in presence of incumbent communication devices known as the primary user (PU) required to be protected from undesired harmful interference. In the field of tactical communication networks, it is desperate to find available frequencies for opportunistic and dynamic access to channels in which PU is in active. This paper introduces a cognitive engine plaform for determining available channels on the basis of case-based reasoning technique deployable as core functionality on cognitive radio engine to enable dynamic spectrum access (DSA) with high fidelity. Towards this, this paper introduces a plausible learning engine to characterize channel usage pattern to extract best channel candiates for the tactical cognitive radio node (TCRN). Performance of the proposed cognitive engine is verified by conducting simulation tests which confirm the reliability in functional aspect of the proposed cognitive engine covering the learning engine as well as the case-based reasoning engine with showing how well TCRN can avoid the collision against the PU operation considered as the etiquette secondary user (SU) should have.

Intelligent Dynamic Spectrum Resource Management Based on Sensing Data in Space-Time and Frequency Domain

“…This list is defined in advance so that users can switch to an idle channel to avoid having an interference effect on incumbent users. As for the channel selection method for generating such a backup channel list, occupancy probability and state transition probability based channel selection methods have been studied based on history data representing spectrum statistics [ 7 , 8 , 9 ].…”

“…Therefore, it should be dynamically optimized according to time varying parameters such as channel state information. To this end, in previous studies [ 7 , 8 , 9 ], the usage patterns of incumbent users were learned using the occupancy probability and state transition probability based on history data representing spectrum statistics to infer idle channels in a dynamic spectrum environment. In addition, in [ 10 , 11 , 12 , 13 ], in order to enable continuous communication without interference to incumbent users, a reactive handoff method based on spectrum sensing and a proactive handoff method using a backup channel list are defined.…”