Underutilized radio frequencies are the chief apprehension in advance radio communication. The radio recourses are sparse and costly and their efficient allocation has become a challenge. Cognitive radio networks are the ray of hope. Cognitive radio networks use dynamic spectrum access technique to opportunistically retrieve and share the licensed spectrum. The licensed users are called primary users and the users that opportunistically access the licensed spectrum all called secondary users. The proposed system is a feedback system that work on demand and supply concept, in which secondary receivers senses the vacant spectrum and shares the information with the secondary transmitters. The secondary transmitters adjust their transmission parameters of transmit power and data rate in such a way that date rate is maximized. Two methods of spectrum access using frequency division multiple access (FDMA) and Time division multiple access (TDMA) are discussed. Interference temperature limit and maximum achievable capacity are the constraints that regulate the entire technique. The aim of the technique is to control the transmitter power according to the data requirements of each secondary user and optimizing the resources like bandwidth, transmit power using machine learning and feed forward back propagation deep neural networks making full use of the network capacity without hampering the operation of primary network.
Transmit power control and data rate enhancement in cognitive radio network using computational intelligence
