Research On Power Control In Cognitive Radio Systems

In recent years, the contradiction between explosive increasing demand for wirelesscommunication services and limited spectrum resource has become increasingly prominent.How to effectively improve spectrum utilization efficiency has become an important topicin the field of wireless communication. For the traditional fixed way of allocation spectrumresources and the access methods, cognitive radio (CR) is one of an nintelligent wirelesscommunication technology which has been attracted the attention of many people. It canadaptively adjust parameters of the system and rapidly fill the spectrum holes or unusedspectrum to solve the problem of the spectrum underutilization. In cognitive radio systems,the transmit power of second user can realize spectrum sharing as one of the coretechnology, and also is the only variable that can be manipulated by second users, as wellas cause interference to other users which is the main reason. Threrfore, the power controlproblem in cognitive radio system naturally attracted widespread attention by researchers.Currently, the power control problem of the single target or multi-target in onetimeslot has been solved by the majority of the researchers at home and abroad, whichcombined different mathematical optimization algorithms based on game theory,cooperative manner, intelligent optimization, as well as robust optimization etc, andachieved gratifying results. However, for these problems of the highly costs of centralizednetwork, the poor ability of adapting dynamic communication environment, and theconsidering of uncertainties, system performances can not be the most accurate descriptionthrough the above power control methods. Based on the aforementioned consideration, inthis paper new theories and mathematical tools were adopted to achieve new power controlstrategies and the corresponding solutions in cognitive radio system. The main researchresults as follows:1）In cognitive radio system, most of the traditional power control schemes areproposed based on centralized networks which took a high cost. In this paper, a novelpower control approach with the objective to minimize total power consumption of SUs isproposed based on chaos particle swarm optimization (CPSO) under the cognitive radionetwork on the underlay scenario for the research platform of the distributed cognitiveradio network in the underlay scenario. The approach guarantees the communicationquality of secondary users (SUs)and the service quality of primary usres (PUs). The utilityfunctions and constraints of original optimization problem are transformed into a fitnessfunction by PSO and penalty function theory. Simulation results indicate that the proposedCPSO algorithm can reduce total transmit power consumption, obtain faster convergencerate and improve searching quality compared with the standard particle swarm optimization (PSO) and the adaptation particle swarm optimization (APSO) algorithmsrespectively.2）In cognitive radio networks, due to mobility of the user and the randomness ofspectrum holes cause rapidly changing on the radio network environment, it is particularlyimportant to stable and reliable transmission. In this paper, in order to adapt dynamiccommunication environment, a dynamic particle swarm optimization (DPSO) powercontrol algorithm based on underlay cognitive radio networks (CRNs) of PUs and SUsspectrum sharing presented to minimized total transmit power of SUs in consideration ofthe same constraints at the last chapter. According to the precise penalty function theory,the initial optimization model is converted into a fitness function. Moreover, the changes inthe environment detected by sensitive particle in DPSO, and transmit power updated byordinary particle in real time. The comparison of simulation result among DPSO algorithm,particle swarm optimization (PSO) and chaos particle swarm optimization (CPSO)algorithms is provided to show that DPSO can get minimum transmit power of SUs, fastconvergence rate, and stability to cognitive radio networks under dynamical environment.3）In order to make primary users (PUs) receive minimum interference generated fromall secondary user (SUs) in cognitive radio networks (CRNs), while ensure the quality ofservices (QoS) of SUs, and adopt the principle of the artificial fish algorithm (AFSA), animproved artificial fish swarm algorithm (IAFSA) of the survival mechanism is alsopresented based on a underlay model to solve the problem of power control, whichconsiders interference plus noise ratio (SINR) of each SU under the minimum thresholdand the transmit power of each SU below the maximum permitted power. Simulationresults show that, in comparison with the Particle Swarm Optimization (PSO) and ChaosParticle Swarm Optimization (CPSO) algorithms, both AFSA and IAFSA can lead SUs totransmit less power in order to reduce the interference to PUs, and simultaneously providefast global convergence, stability, robustness to CRNs, and better communicationperformance.4）The distributed power control algorithm can reduce the exchange information ofusers, and system overhead. A basic distributed power control algorithm (DPCA) is putsforward based on the minimum interference power to PUs in this paper. In order to reduceinterference to the PUs, an improved distributed power control algorithm (IDPCA) isproposed. Simulation results show that the performance of two proposed algorithms aresuperior to the traditional iterative water filling algorithm (IWFA) in perfect and imperfectchannel environment, and the improved algorithm is better than the standard distributedalgorithm.5）In cognitive radio system, the constraint of the interference to PUs is usuallyconsidered, while the second user communication quality is not always strictly guaranteed in great many of power control scheme. Power control is an effective means to savespectrum resources and improve spectrum utilization in cognitive radio networks (CRNs).The focus of this chapter derives a standard distributed optimal power control strategy inthe high signal to interference plus noise ratio (SINR) case based on maximization of thedata rate for each second user (SU) under the two constraints of maximum transmit powerof each SU at all subcarriers and the maximum tolerated interference power for primaryusers (PUs) at each subcarrier, considering an orthogonal frequency-division multiplexing(OFDM) framework. Due to the nonconvexity of objective function, geometricprogramming (GP) is introduced to transform it into a convex optimization problem.Furthermore, we apply the Lagrange relaxation of the coupling constraints method andsubgradient iterative algorithm in a distributed way in order to solve the dual problem. Toenhance the quality of service (QoS) requirements for SUs at each subcarrier, an improvedalgorithm is presented. Numerical simulation results show that the standard algorithm andimproved algorithm are all superior to traditional IWFA in enhancing the data rate of eachSU, convergence speed and computational complexity.6）In the actual communication environment, the performance of the system isaffected the uncertainty of parameters. The focus of this paper is to find a robust powercontrol strategy with uncertain noise plus interference (NI) in CRNs under orthogonalfrequency-division multiplexing (OFDM) framework. Our object is to maximize data ratefor each second user (SU) at all subcarriers under three constraints, i.e. the transmit powerrange of each SU, the tolerable interference power for primary users (PUs) no more thangiven thresholds, as well as the satisfied probabilities of the transmit rate of each SU lessthan desire transmit rate for the quantized NI error. In consideration of the feedback errorsfrom the quantization following uniform distribution, the probabilistic constraint istransformed into closed deterministic forms. By using Lagrange relaxation of the couplingconstraints method and subgradient iterative algorithm in a distributed way, we solve thisdual problem. Numerical simulation results show that our proposed algorithm is superior tothe robust power control scheme based on interference gain worst case approach andnon-robust algorithm without quantization error in perfect channels in the improvement ofdata rate of each SU, convergence speed and computational complexity.