Font Size: a A A

Simulation Research On Receiver Of Downlink Non-orthogonal Multiple Access For 5G

Posted on:2018-02-14Degree:MasterType:Thesis
Country:ChinaCandidate:Y J LiuFull Text:PDF
GTID:2348330515971259Subject:Information and Communication Engineering
Abstract/Summary:PDF Full Text Request
With the commercial deployment of the forth generation(4G)mobile communication system,investigations on the fifth generation(5G)is becoming a hot topic.Faced with the requirements of 5G network on more massive connection,higher spectrum efficiency and larger capacity,the Non-orthogonal Multiple Access(NOMA)has been focused on considerably.NOMA achieves multiplexing through introduction of the power domain and successive interference cancellation(SIC)is used at the receiver in order to distinguish signals of users.Since there are multiple users at the transmission side,interference is caused.Thus,how to detect signals of user is an important issue.This thesis mainly studies the detection strategies of downlink NOMA system.First of all,the thesis introduces basic theory of NOMA and takes the two-user model as an example to verify that NOMA can get better performance of system compared with orthogonal frequency division multiplexing access(OFDMA).As the core technology of NOMA,power allocation algorithm and detection algorithm is introduced as well.Besides,the basic theory of MIMO is presented.Secondly,the link-level simulation platform of NOMA downlink is built through modularized thought,based on which,the thesis simulates the performance of Ideal SIC receiver,Symbol-level SIC receiver and Codeword-level SIC receiver in the context of single-and multiple-antenna respectively.The results show that by allocating power reasonably,the BER of Codeword-level SIC coincides with that of Ideal SIC and outperforms that of Symbol-level SIC.Meanwhile,with the increase of power gap between the cell center and edge users,the BER of all three receivers improves.And with the increase of modulation order,the performance of receivers is worse.In addition,power allocation algorithm is studied in single antenna scenario,which regards maximize the weighted sum rate as the principle of power allocation.The simulation result shows that power allocation factor is not only related with SNR difference between uses but also with user's own SNR.Further,the thesis pays attention on receiver performance in diversity and multiplex scenario.The results show that Symbol-level SIC is deeply influenced by inter-user interference and inter-stream interference,therefore Symbol-level SIC is more sensitive to error propagation which will result in much worse BER performance.The simulation on EP(Error Propagation)is conducted as well,which shows that EP will reduce system capacity.Eventually,join detection algorithm which removes the reconstruction of interference signal by detecting user signal through join constellation at receiver terminal is simulated.The result shows that BER performance of the algorithm is far superior to Symbol-level SIC.And compared with Codeword-level SIC,the complexity decreases 50%approximately.
Keywords/Search Tags:NOMA, 5G, SIC, Join Detection
PDF Full Text Request
Related items