Title:
Designing orthogonal chirp division multiplexing waveforms for next generation wireless communications
Designing orthogonal chirp division multiplexing waveforms for next generation wireless communications
Author(s)
Omar, Muhammad Shahmeer
Advisor(s)
Ma, Xiaoli
Editor(s)
Collections
Supplementary to
Permanent Link
Abstract
The objective of this dissertation is to analyze the performance of orthogonal chirp division multiplexing (OCDM) as a waveform for wireless communications, identify issues to its adoption in future networks, and propose solutions to these problems. OCDM is an extension of the chirp spread spectrum (CSS) as it employs linear chirps to carry data. However, in contrast with CSS, OCDM can achieve the Nyquist signaling rate and therefore, it has the same spectral efficiency as orthogonal frequency division multiplexing (OFDM). The first part of this dissertation analyzed the performance of OCDM affected by typical wireless channel impairments such as multipath fading, carrier frequency offset (CFO) and interference. Results showed that OCDM has better performance in multipath fading channels without channel coding than OFDM and is more robust to interference than both single carrier block transmissions and OFDM. This makes it an attractive alternative to OFDM in future networks. On the other hand, the analysis highlighted key issues that prevent its adoption in practical systems. First, it suffers from inter-symbol interference (ISI) in frequency-selective channels and hence does not lend itself easily to uplink multiuser transmissions and pilot-based channel estimation. Second, there is no way of digitally introducing guard bands in OCDM without altering hardware sampling rates. Third, OCDM has a very high peak-to-average power ratio (PAPR). Solutions were subsequently proposed for each of the identified issues. To control the signal bandwidth, a low complexity OCDM transmitter was proposed to digitally alter the spectral shape. This made it compatible with digital filtering techniques that were originally proposed for OFDM to further reduce out-of-band emissions (OOBE). A machine learning-based transmitter was proposed to lower the PAPR of both OFDM and OCDM signals while minimizing distortion and maintaining spectral mask compliance. Several pilot multiplexing techniques were proposed which allow for pilot and data symbols to exist in the same block without causing interference. A complex precoding technique was also proposed to enable full diversity and orthogonal multiuser transmissions in the uplink.
Sponsor
Date Issued
2022-05-03
Extent
Resource Type
Text
Resource Subtype
Dissertation