Cognitive Radio Terminal Selection via Reinforcement Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Determining which terminals should perform cognitive radio communication in an environment with multiple IoT terminals is complex due to the need for efficient frequency management and interference avoidance, which existing methods have not adequately addressed.
Innovation Solution
A method and device that utilize a VNF transfer graph selection process within an NFV MANO system to determine suitable terminals by obtaining inter-network and intra-network delay values from a virtualized infrastructure manager, calculating roundtrip times, and applying round-robin scheduling to ensure network QoS requirements are met, enabling efficient terminal selection for cognitive radio communication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple IoT terminals participate in cognitive radio communication, then frequency use efficiency is improved, but terminal determination complexity increases
Solution Approach 1:
The patent transforms the complex terminal determination problem into a structured parameter-based selection process. It defines specific parameters including communication success probability (based on spectrum sensing results), position information, and access probabilities. By changing the determination criteria from arbitrary to parameter-driven, the system efficiently selects terminals while maintaining frequency use efficiency in multi-terminal IoT environments.
2Reliability
If spectrum sensing is performed to detect surrounding environment, then communication success probability is improved, but detection difficulty and time consumption increase
Solution Approach 1:
The patent performs spectrum sensing as a preliminary action before terminal determination and cognitive radio communication. The base station collects spectrum sensing results from multiple terminals in advance, uses this information to calculate communication success probabilities, and then makes determination decisions. This preliminary detection approach ensures reliable communication selection while streamlining the overall process.
3Object-affected harmful factors
If position information is utilized for terminal selection, then interference avoidance is improved, but information processing requirements increase
Solution Approach 1:
The patent applies position information to identify local spatial characteristics of terminals. By analyzing the spatial distribution and positions of terminals relative to primary users and each other, the base station can determine which terminals are in favorable locations with lower interference risk. This local quality assessment based on position enables effective interference avoidance while processing only essential spatial information.
Data Source
AI summary
Disclosed is a control device for determining cognitive radio terminals. The device comprises: a spectrum sensing information storage unit for receiving and storing interference amount information for each frequency from spectrum sensing devices; a communication success probability calculation unit for receiving position information from secondary user terminals within a network and calculating a communication success probability of each secondary user terminal; a reinforced learning unit for setting the communication success probability as an initial access probability for each of the secondary user terminals and enabling learning such that an access probability, a state function, and a utility function are updated for each of the secondary user terminals at each iteration; and a cognitive radio terminal determining unit for, when the leaning is completed in the reinforced learning unit, selecting a secondary user terminal to execute cognitive radio communication on the basis of a final access probability.