JavaScript is disabled for your browser. Some features of this site may not work without it.
| dc.contributor.author | GETAHUN, GEBRIE | |
| dc.date.accessioned | 2022-11-18T07:03:11Z | |
| dc.date.available | 2022-11-18T07:03:11Z | |
| dc.date.issued | 2022-08-25 | |
| dc.identifier.uri | http://ir.bdu.edu.et/handle/123456789/14462 | |
| dc.description.abstract | Wireless systems, because of the environment in which they operate, channels are prone to mul-tipath interferences. This causes various kinds of multipath fading and disturbances affecting the sent information. In order to maintain the BER within the defined bounds in time-varying channels, the receiver uses an adaptive equalization based on traditional equalizer algorithms. Most of these techniques are proved to be much complex and not good in dealing with non-linear and time varying channels. Since most wireless communication channels are susceptible to channel interferences due to multipath propagations, we need to model and use an efficient adaptive channel equalization techniques in order to reduce such interferences. In this thesis the impact of interference at receiving end are considered and analyzed in different structures of an Adaptive Neuro Fuzzy Inference System (ANFIS) with different training data and membership functions. Realization of equalizers based on neuro-fuzzy techniques seems to be most appro-priate option for the mobile cellular channel especially for non-invertible channels. This thesis presents the performance of ANFIS as equalizers for time varying channels with hybrid learning algorithms, taking into account interferences from multipath propagations and the occurance of co-channels interference (CCI) in cellular wireless network operation. The MATLAB simulation result shows the performances of ANFIS equalizers using more member-ship functions and using more train data samples have good performance in case of BER, and also the performances of ANFIS equalizers are better than RBF NN equalizers on the impacts of interference expressed by BER versus SIR and SINR also greatly affects the overall system performance for different number of membership functions and train data samples forming dif-ferent ANFIS structures and RBF neural networks in addition to the conventional AWGN noise expressed by SNR for different structures of ANFIS and RBF based channel equalization tech-niques. Key Words: ANFIS, CCI, ISI, RBF NN, ANFIS EQALIZER | en_US |
| dc.language.iso | en_US | en_US |
| dc.subject | Faculty of Electrical and Computer Engineering | en_US |
| dc.title | PERFORMANCE ANALYSIS OF ADAPATIVE NEURO- F UZZY I NFERENCE SYSTEM (ANFIS) E QUALIZER FOR DISTORTED WIRELESS TIME V ARYING CHANNEL | en_US |
| dc.type | Thesis | en_US |
