The Research On PAR Reduction For OFDM System Base On Conjugate-gradient Method Algorithm

Orthogonal frequency division multiplexing (OFDM) is an attractive technique forhigh-bit-rate transmission in a radio environment. By dividing the total bandwidth into manyorthogonal narrow sub channels, the effects of multipath delay spread can be minimized. So ithas been widely used in several areas such as Asymmetrical Digital Subscriber Line (ADSL),Digital Audio Broadcasting (DAB), High Definition TV (HDTV) and Wireless Local-AreaNetworks (WLAN). Unfortunately, a central disadvantage of OFDM is potentially highpeak-to-average power ratio (PAR) values, which can result in significant distortion whentransmitted through a nonlinear device, such as a transmitter power amplifier. This weaknessmakes a high request for the linearity of power amplifier in transmitter and receiver, andincreases the complexity of A/D and D/A convertor. People have already put forward a lot ofmethods to reduce the kinds of PAR currently. There are 3 most familiarly methods: (1)Clipping;(2)The method of Coding; (3)The method of probability. And the method of probability canimprove the PAR function of the OFDM signal consumedly under the situation that not producesthe signal mutation. There are 3 common method of probability: Optimum combination of partialtransmits sequence; SLM; PTS. Among them, the SLM method and the PTS method are twokinds of the most typical method. Although the PTS and the SLM calculate ways don't lose truecalculate way, they needed to carry on additional IFFT of many times an operation, and thereforetheir amount of calculation is very big. And all need in the receiver on carrying to understand toshoot the assistance that the machine adopt to add the power information by the square, so all ofall theses will increase the system complications.Based on introducing the basic OFDM principle, we analyze the reason and influence ofhigh PAR in OFDM systems, give out mathematic definition of the PAR and depict regularitiesof distribution for statistic characteristics. And combined to introduce in brief to lower a PARmethod traditionally. This paper presents the Conjugate-gradient Method algorithm to solveglobally optimization problem of PAR. Based on characteristic of the Conjugate-gradientmethod algorithm and condition limit of the PAR problem, the new method can set upcorresponsive barrier function and avert the algorithm complexity from the customizedinterior-point method (IPM) which should have matrix inverse. It can assure the astringency of the algorithm. The simulation results verify that the algorithm can not only solve high PAReffectively, but also reduce the algorithm complexity.