Modulated signals are known to be spectrally redundant and this redundancy can be exploited in the design of optimum receiver. This is the main topic of this thesis. It is shown that the structure of such a receiver is system-dependant and related to: channel spacing, signal excess bandwidth, number of available receiver antennas, modulation scheme, and users' symbol rates.

In terms of cyclostationarity, it is shown that modulation schemes can be classified into three categories. Binary Phase-Shift Keying (BPSK), Quadrature Phase-Shift Keying (QPSK), and Off-set QPSK (OQPSK) are representative of these three categories. An optimum receiver can be designed and analyzed for each of these three categories.

A complete analysis of decision feedback equalizer (DFE) in cyclostationary environment is presented. The analysis is based on minimum time averaged mean square error (MTAMSE) criterion. It is shown that the conventional DFE is not an optimum receiver for BPSK and OQPSK signals but is optimum for QPSK signals if and only if all users transmit at an identical symbol rate. Two different structures are suggested for DFE. The first is based on harmonic series representation (HSR) of a cyclostationary random process (CRP) and is called Frequency-shift DFE (FRESH-DFE). The second is based on time series representation (TSR) of a CRP and is called generalized DFE (GDFE).It is shown that FRESH-DFE and GDFE are identical in performance, and in the case of "single symbol rate", their structures are also identical. We shall refer to FRESH-DFE (or GDFE) in single symbol rate systems as optimum DFE (ODFE). The results of analysis and simulation presented in this thesis show that the ODFE has a better performance than the conventional DFE, specially in high interference environment.

An optimum multi-antenna receiver is also considered, and it is shown that the bit error rate (BER) can be reduced by increasing the number of antennas. This reduction of BER can be converted into additional capacity if moderate BER is acceptable.

The indoor cellular mobile environment in the presence of high interference and multipath fading is chosen to evaluate the structure, complexity and performance of the optimum receiver. In this evaluation, the co-channel and adjacent channel interference (CCI and ACI) signals are treated separately and it is shown that the effect of ACI can be removed despite its considerable spectral overlaps with the signal of interest (SOI). It is also demonstrated that the ODFE with the same complexity as the conventional DFE can perform better in multipath fading environment.

The developed concepts were applied to cellular environment. To this end, the thesis studies signal to noise ratio (SNR), signal to interference ratio (SIR) and the capacity of cellular systems. It is shown that the capacity can be increased if moderate BER is chosen as a criterion.

Computer simulation results are presented to support the theoretical results. It is shown that for the same level of complexity, the performance of ODFE is superior to that of conventional DFE.  

Back to the list of graduate students