Two-dimensional MMSE channel estimation for single-carrier frequency equalization
Jianhua Lü
Abstract
Jianhua Lü
Abstract
Cyclic-prefix-based block processing is adopted in single-carrier frequency equalization(SC-FDE) system,which makes discrete channel estimation and equalization possible.A two-dimensional minimum mean square error(MMSE) channel estimation algorithm was developed for the SC-FDE system in this paper.With separation property of mobile wireless channels,the two-dimensional estimation can be simplified as an MMSE interpolation in the frequency domain cascaded by an MMSE filter in the time domain.The simplified algorithm is suitable to SC-FDE with unique word(UW),i.e.UW-SC-FDE,as well as with pilot,i.e.Pilot-SC-FDE.Simulation results indicate that the algorithm is much superior to the one-dimensional channel estimation algorithm,for both UW-SC-FDE and Pilot-SC-FDE systems.The performance of the proposed algorithm gets close to that of ideal channel estimation.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Cyclic-prefix-based block processing is adopted in single-carrier frequency equalization(SC-FDE) system,which makes discrete channel estimation and equalization possible.A two-dimensional minimum mean square error(MMSE) channel estimation algorithm was developed for the SC-FDE system in this paper.With separation property of mobile wireless channels,the two-dimensional estimation can be simplified as an MMSE interpolation in the frequency domain cascaded by an MMSE filter in the time domain.The simplified algorithm is suitable to SC-FDE with unique word(UW),i.e.UW-SC-FDE,as well as with pilot,i.e.Pilot-SC-FDE.Simulation results indicate that the algorithm is much superior to the one-dimensional channel estimation algorithm,for both UW-SC-FDE and Pilot-SC-FDE systems.The performance of the proposed algorithm gets close to that of ideal channel estimation.
Key concepts: Cyclic prefix, SC-FDE, Equalization (audio), Algorithm, Channel (broadcasting), Computer science, Frequency domain, Minimum mean square error