Self-interference in OFDM-Based Systems: Identification and Separation

• Main Author: Guvenkaya, Ertugrul
• Format: Others
• Published: Scholar Commons 2013
• Subjects: Electrical and Computer Engineering
• Online Access: http://scholarcommons.usf.edu/etd/4499; http://scholarcommons.usf.edu/cgi/viewcontent.cgi?article=5696&context=etd

Orthogonal frequency-division multiplexing (OFDM) is communication technique that is robust against multipath spread, efficient in equalization implementation, and flexible in dynamic spectrum usage. With the need of wideband signals in high data rate communications, OFDM has gained widespread attention and usage as an efficient way of implementing multicarrier communication. Despite its prominent advantages, received signal quality degrades because of inter-symbol interference (IS) and inter-carrier interference (ICI) due to insufficient guard time in multipath channels and time variance as well as hardware impairments in OFDM systems. When several of the impairments occurs simultaneously, interferences combine at the receiver side. Thus, interference awareness becomes important concept for OFDM-based systems that are excepted to be highly adaptive for coping with existing impairment. By identifying the type and the source of the interference, mitigation of the impacts with corresponding adaptation in the system increases the performance of the overall system. In this thesis, self-interference identification, visualization, and separation of various type of interferences in OFDM systems have been studied. An interference spectrogram is developed in order to be able to investigate several interferences with one diagram solely. Identification of different sources of interference that result in same consequences at the receiver, such as ISI due to insufficient guard time and ICI due to time variance in the channel, is performed. Also, powers of corresponding interferences are estimated separately with the proposed technique. Computer simulations and measurements also confirmed the effectiveness of the interference separation method. Information symbols are represented in various domains of electro space such as time, frequency, and code. Channel impairments have different effects on the symbols in different domains. For instance, multipath spread causes ISI in single carrier systems while interference between symbols occurs when time variance exists in OFDM systems. Visualization and measurement of distortions in various domains need to be addressed for better understanding of multidimensional communication systems and their performance under distortion. Therefore, eye diagram in frequency domain for OFDM-based signals is introduced for visualization and identification of time varying impairments such as frequency spread and carrier frequency offset. Real channel measurements are performed to illustrate the effectiveness of the method in practical cases. The study conducted in this thesis can be utilized for interference awareness in OFDM-based wireless communication systems. Also, methods and diagrams that are developed can be used for testing the wireless systems, interference measurements, and educational purposes.