Channel Estimation Based On Compressed Sensing Of LTE Systems

With the development of society, the demand for the properties and applicationsof mobile communication systems is increasing, such as high speed transfer rate, hightransmission quality, and species diversification of business.The raising of LTE marksthe wireless communications entering a new frontier. Channel estimation technologyas an important issue of LTE systems, have always been the hot research of wirelesscommunication systems and the technical difficulties. Traditional methods of channelestimation don’t fully take into account the related information of the channel, so thatthe pilots take up a large amount of spectrum resources. The accuracy andperformance of the algorithm is not very satisfactory. The need of development ofbroadband and high-speed communications presents a greater challenge for thevalidity and accuracy of channel estimation technology.In recent years, compressed sensing theory causes more and more people’sattention. The theory of compressed sensing broke the traditional laws of the Nyquistsampling constraints, presents a new opportunity for development of signal processing.Theory of compressed sensing pointed out that we can fully tap the sparse attribute ofthe wireless channel. Appliying the theory of compressed sensing to the channelestimation technology. Not only does it reduce the number of the using pilots, so doesimprove the spectrum utilization. But also can improve the performance of channelestimation, channel information is more efficient and accurate. The blending forcompressed sensing and channel estimation technology is of great theoreticalsignificance and application prospects.This article first describes the background and research status of framework forLTE systems. Analysing the important components of LTE downlinks physical layerand channel estimation technology. An analysis of LTE channel model. Thenintroduce the background and current status of compressed sensing. Focusing onapplying the theory of compressed sensing method for channel estimation on the timedelay domain.Through research and analysis, we find that the orthogonal matchingpursuit algorithm must know in advance the channel of the limitations of sparse, wegive a method to estimate in advance the sparsity of the channel, so that the stabilityof the algorithm, convergence and performance are further improved. Thenanalysising the time-selective channels.We discover the Doppler domain of channelmeets sparse channels, and use the compressed sensing for channel estimation methodfor time-varying channel. Simulation results show the effect of frequency distributionon performance. In addition, throughing further research, compressed sensing isapplied to channel estimation for the selected channel, which is time-selective andfrequency-selective channel. We compare the compressed sensing method for channelestimation and traditional channel estimation algorithm.Simulation results show that the channel estimation method based on compressed sensing can effectively reducesystem demand for pilots, and thus stability of the algorithm, convergence andestimated performance is further improved.The method has a very significantadvantage.