Arithmetic Research Of Doppler Spread Estimation In Mobile OFDM Communication System

Recently OFDM(Orthogonal Frequency Division Multiplexing) has been Widely applied in many high speed data transmission fields such as digital audio broadcasting DAB,digital video broadcasting DVB,wireless LAN IEEE 802.11a, and thus attracts great attention of researchers and is called one of the possible technologies to realize Beyond 3G mobile communication systems due to its strong anti-multipath interference ability,high spectrum efficiency,implementation simplicity and many other advantages.In mobile communication systems,time variance of the channel originated in high speed relative motion between mobile station and base station,however,leads to a loss of signal orthogonality,resulting in interference between the sub-carriers(ICI) and system performance reduction.As one key parameter reflecting channel fading rate,the maximum Doppler frequency can optimize channel tracker step size,update rate of power control algorithms and adjust interleaving lengths.Also,radio network control algorithms,such as handoff and allocation in cellular system,can utilize the Doppler information.With its wide applications in OFDM systems,the algorithm research of accurate Doppler spread estimation is indispensable for the design of OFDM mobile communication systems.In this paper,an efficient Doppler estimation algorithm design applies to Rayleigh fading environment is proposed on the basis of existing Doppler estimation algorithms by analyzing the features of OFDM systems and wireless transmission channel.The algorithm has advantages of low complexity,wide estimation range, good accuracy,moreover,the algorithm is robust to noise and can provide real-time estimation due to averaging processing.The presented algorithms in this paper is based on a channel estimation auto-correlation function,that is to say,the auto-correlation function of time domain impulse response meets the 0th order Bessel function of the first kind,and the Doppler value can be educed by the Bessel inverse function.According to the characteristics of the DTMB system,during the algorithm design,firstly time domain channel estimates are obtained by a channel estimation module,followed by the auto-correlation calculation and normalization of channel estimates over several OFDM symbols,then the averaging processing is carried out to reduce the noise interference,finally the Doppler estimates are got using the LUT implementation of Bessel inverse function.Simulation results show that the proposed Doppler spread estimator can provide relatively high estimation accuracy for a wide range of Doppler spread in a typical urban(TU) reception environment with six Rayleigh fading multi-paths and is robust to varying SNR values.Meanwhile,the DTMB receiver system with the proposed estimator also shows a significantly better SER performance than the system without it.Moreover,as the channel auto-correlation is computed using the channel estimates over the known fields of OFDM symbols,so the proposed algorithms can also be applied directly into other OFDM systems with known pilot symbols inserted within the data fields.