Implementation Of Timing And Frequency Synchronization Of OFDM On FPGA

Orthogonal frequency division multiplexing (OFDM) is one of the most important technologies in future broadband wireless communication system. To meet the demand of real-time high rate multimedia service grown with days, and at the same time to support data transmission in high speed mobile environment, OFDM is applied widely, for the sake of high utilization ratio of frequency spectrum, good immunities to frequency selective and multi-path wireless channel.OFDM system is very sensitive to timing errors and frequency offset because it must keep the orthogonality between the subcarriers. The main task of this research subject is to analyze the property of various algorithms, and select the algorithm with high performance to implement it in FPGA.First of all, this paper makes a brief introduction to the transmission character of wireless channel and basic theory of OFDM system, and analyses the effects of signal by timing errors and frequency offset. We contrast the data-aided synchronous algorithm with non data-aided synchronous algorithm, and decide to solve the synchronization problem of burst transmission system about IEEE 802.16-2004 with data-aided synchronous algorithm. At last, we implement and simulate this algorithm. All of the simulation of implement are carried out according the symbol and preamble of IEEE 802.16-2004 in Quartus II.In this paper, there are three innovations: First, self-correlation and cross-correlation algorithms are associated to complete frame detection and symbol timing estimation at the same time. We settle the problem of detection threshold indeterminacy effectively by correlating only with sign bit of received data, as a result, it decreases the consumption of logic resources. Second, in the fractional frequency offset estimation module, we compute the phase angle of long preamble correlation by Cordic module with pipeline organization. At last, we evaluate integral frequency offset with the means of moving window correlation, operate merely with the sign bit of frequency data, effectively resolved the disadvantage of the traditional arithmetic, simultaneously reduced the consumption of resource.