Research Of Resource Allocation Mechanisms In Cognitive Radio Network

Recently the contradiction between increasing wireless communication demand and vacancy of spectrum resource has become obvious.Massive test results show that current static spectrum assignment policy leads to low efficiency of spectrum usage.The main factor of spectrum vacancy is the outdated static assignment policy,not spectrum resource itself.Cognitive radio(CR)technology facilitates the concept of dynamic spectrum management(DSM).Based on DSM,the flexible dynamic spectrum access(DSA)could realize efficient radio resource sharing and scheduling between different radio resources.DSM could conquer the radio resource bottleneck of wireless communication development,thus it holds important theoretical and practical sense.This thesis studies the resource allocation problem in cognitive radio network through theoretical analysis and simulation based on DSM concept.The main contents and results of this thesis is as follows:1)Proposed an auction based spectrum allocation mechanism in overlay cognitive radio network.Unlike the existing auctions,the proposed mechanism uses tokens that represents SU's contribution to the network.SUs submit their first bids and spectrum demand,then they are encouraged to form coalitions.SUs negotiate with partners or potential partners,and form cooperation-based coalitions.The coalitions report their bid information to the FC and FC allocates the available spectrum according to a probability rule.Due to the fact that some SUs may have insufficient tokens to pay their bids,and some SUs with mass tokens long for transmission help from other SUs,these SUs may form a debit and credit relationship.The negotiations are modeled as non-cooperative games,and existence of equilibrium is analyzed.Since cooperation introduces capacity gain,cooperation yields mutual benefit for both parties.Simulations are conducted to show the mutual benefit from cooperation.Simulation results also show that the proposed auction leads to satisfactory data rate and fairness.2)Studied the power allocation problem in a relay assisted multi-band underlay cognitive radio network.Underlay cognitive radio network allows SUs and PUs to transmit on the same channel simultaneously.However,the transmission of SUs cannot cause intolerable interference to PUs.Under such constraint,a cooperative transmission mechanism is designed.In a case where two PUs,two SUs and one half-duplex relay and two available channels exist,the transmission time is divided into two identical slots.In the first time slot,SU1 transmits directly on channel I to its receiver,and transmits data via channel II to the relay,similarly SU2 performs direct transmission on channel II,and transmits to the relay via channel I.In the second time slot,SU1 continues transmission on channel I,but does not use channel II,and SU2 performs similarly.The relay transmits data to receiver of SU1 via channel II,and transmits transmits data to receiver of SU2 via channel I.In this mechanism,the total power constraint and interference contraints of two PUs are taken into account.The power allocation of two SUs is modeled as non-cooperative game,and utility functions are designed based on capacities.The game is proven to be super-modular and thus Nash equilibrium exists.Through analysis on best response functions,a distributed update algorithm is proposed to achieve equilibrium under incomplete information.This update algorithm allows each SU to adapt its strategy alternatively in repeated games.The convergence of the algorithm is proven.Simulations are conducted to analyze the best response function curves in different scenarios and different initial strategies.The system capacities in different scenarios are also investigated through simulations.3)Proposed a relay-based cooperative spectrum sharing mechanism where the network consists of one PU and multiple SUs.The concept of cooperative spectrum sharing allows SUs to use portion of the PUs' radio resource for their own data transmission,under the condition that SUs help the PUs' transmission.This cooperation further improves spectrum utilization.First,a resource allocation framework is proposed: the authorized spectrum of PU is divided into two portions.One portion is leased to the SUs as their exclusive transmission channel(secondary channel),the other portion(primary channel)is used to transmit PU data and relay PU data by SUs.This cooperation contains two steps.The first step is that PU announces the bandwidth of secondary channel(PU strategy),the second step is that all SUs determine their relay power for PU(SU strategy)according to their valuations on energy comsumption and bandwidth of secondary channel.The resource allocation is modeled as a two-stage Stackelberg game.The utility functions of PU and SUs are designed based on capacities and energy comsumption.To analyze the optimal solution of the game,backward conduction is adopted.Under the assumption that the PU strategy is fixed,the optimal power allocation among SUs is analyzed,and the existence and uniqueness of equilibrium are investigated.Further the optimal strategy of PU is analyzed and the uniqueness of this optimal strategy is proven.Considering the fact that in actual scenarios,users are unlikely to obtain complete information,and the fact that the closed-form expression of PU optimal strategy cannot obtained,a distributed update algorithm is proposed.The algorithm is proven to guarantee converge.Simulations are conducted to validate the optimal strategy of PU,power allocation and utilities of SUs under PU optimal strategy,as well as the convergence of the proposed update algorithm.