Research On Channel Estimation Methods For Internet Of Vehicles

In recent years,to further improve the safety,efficiency and sustainability for transportation,the Intelligent Transportation System(ITS)centered on Internet of Vehicles(IoV)technology has been proposed to meet these needs,the IoV has attracted researchers' so much attention that it has become a hot topic soon.In 2010,after a few years' revisions,the communication standard IEEE 802.11 p protocol for a IoV communication was officially published.In the communication system of the IoV,the channel parameters are complicated and variant,hence a reliable channel estimation technique is critical to the performance of the communication system.Orthogonal frequency division multiplexing(OFDM)technology is a key technique adopted by IEEE 802.11 p.Based on these,the thesis mainly studies the channel estimation method of OFDM system under IEEE 802.11 p protocol.Firstly,the thesis analyzes the implementation process of Physical Layer(PHY)and the fundamental working mechanism of Media Access Control Layer(MAC)under IEEE 802.11 p.The characteristics of wireless channels under IOV environment are studied.Several conventional pilot-based OFDM channel estimation algorithms are introduced in detail.The performance comparison is given according to simulation experiments,and a simulation platform based on IEEE 802.11 p has also been developed.The platform can be used to verify the performance of each channel estimation algorithm in IEEE 802.11 p.Aiming at the shortcomings of conventional channel estimation algorithms and double selective channel characteristics with sparsity in time-frequency domain for IoV environment,the thesis introduces a channel estimation method based on Compressed Sensing(CS)technique.The four pilots in every OFDM symbols of IEEE 802.11 p frames are exploited to perform channel estimation with CS.Since the channel estimation is performed between each OFDM symbol,a time-varying channel that can be better tracked with thi s method.Furthermore,the pilot is too sparse in OFDM symbols in the original pilot structure.The thesis proposes an improved 8-pilot structure for the problem of too sparse pilots in OFDM symbols.The improved structure uses virtual null subcarriers that do not transmit valid data as pilot subcarriers hence it can improve the estimation accuracy without modifying too much original structure of IEEE 802.11 p.Aiming at the traditional time-domain sampling method,the channel recovery dictionary is not refined enough,and it cannot accurately reflect the characteristics of the transmission channel path.This thesis uses multipath sparse fractional delay channel model to simulate the wireless multipath channel of OFDM system.A time domain oversampling method at the receiver is proposed to refine the channel recovery dictionary to improve channel estimation accuracy.Meanwhile,for the problem that the complexity of the algorithm increases due to the expansion of the CS measurement matrix caused by oversampling,a Reduced-Set Matching Pursuit(RMP)algorithm is proposed for channel estimation,furthermore,it proposes a Doubly Reduced Match Pursuit(DRMP)algorithm to reduce the computational complexity while maintaining a high estimation accuracy.The simulation experiment results verify the reliability and the validity for the improved 8-pilot structure proposed in this thesis and the DRMP channel estimation method with time domain oversampling.