Key Technology Of Cognitive Radio, Ofdm-the Tdcs System

Several key techniques on Orthogonal Frequency Division Multiplexing-Transform Domain Communication Systems (OFDM-TDCS) are elaborated in this dissertation. First of all, a lower-complexity soft-demodulation (SD) algorithm is derived from the conditional probability density function of received signals. Furthermore, a special code table is created for highly efficient implementation. Simulation shows that the performance gain of bit-error-rate (BER) is enhanced from 0.3 dB to 1.5 dB.Secondly, two kinds of channel estimation techniques on OFDM-TDCS systems are investigated respectively. On the one hand, the Least Squared (LS) algorithm is adopted for slow fading channel, of which Mean Square Error (MSE) performance is analyzed theoretically and validated by simulation. To improve the BER performance, an LS iterative channel estimation algorithm is proposed. Simulation indicates that the BER performance gap between 20% pilots-aided LS and perfect estimation is descended from 3dB to 2dB owing to the first iteration.On the other hand, the Recursive Least Squared (RLS) algorithm is employed for fast fading channel. The optimal memory factor is obtained according to the analysis of MSE performance. Subsequently, another iterative channel estimation which combines RLS with Polynomial Fitting is considered to improve the estimation accuracy. Test results demonstrate that the BER performance gap between 20% pilots-aided RLS and perfect estimation is reduced from 5dB to 2dB on account of the second iteration. Finally, a feasible synchronization scheme is proposed for OFDM-TDCS systems.Based on the duality hypothesis model, the false and miss probabilities are analyzed, and then the optimal detection threshold is acquired by the rule of Neyman-Pearson. What's more, the timing delay and fractional carrier frequency offset are estimated respectively. Simulation validates the feasibility of this scheme.