Study On Power Control Algorithm In Cognitive Radio Networks

With the rapid development of current wireless communication technology, morespectrum resources are needed and they can not satisfy our demands. At the same time,the static spectrum management leads to amount of idle spectrum in the primary bandand spectrum resources are not effectively used. Cognitive radio (CR) technology cansense and detect the idle spectrum from space, time and frequency areamulti-dimensionally and the utilization rate of spectrum resources can be enhanced.Generally speaking, white spectrum which is filled with white noise and some grayspectrum which is filled with low radiofrequency signals can be reused. Accordinglythere are three spectrum sharing model: overlay, underlay, combination of overlay andunderlay. Power control is a key technology of cognitive radio. Effective power controlcan not only satisfy the quality of service of secondary users without disturbing thenormal communication of primary user, but also reduce the interference betweensecondary users and improve the lifetime of battery.Because of the changing external environment, secondary users have to adjust theirpower to satisfy the signal-to-interference ratio. So the power control algorithm shouldconverge fast. An improved algorithm based on non-cooperative game is proposed forthe overlay model in cognitive radio networks. A tangent cost function is designed toimprove the convergence. Simulations show that, compared with other three algorithms,this algorithm not only can fast converge, but also has higher averagesignal-to-interference ratio when the number of secondary users is small.In the underlay spectrum sharing model, secondary users get access to thespectrum being used by primary user. So secondary users need to control their powerstrictly to satisfy the primary user’s interference temperature constraint. An improvedalgorithm with logarithmic target function is proposed for the underlay spectrumsharing model. Its cost part increases with the increment of interference to primary userand increases rapidly when the interference is close to the interference temperatureconstraint. Simulations demonstrate that, compared to the Foschini-Miljanic algorithm,this algorithm encourages secondary users raise their power to get highersignal-to-interference ratio when interference to primary user is small, and urgesecondary users to control their power strictly to reduce the interference to primary userwhen there are many secondary users. This algorithm is very fair to secondary users additionally.