| In this thesis, a number of coded Gaussian minimum-shift keying (GMSK) transceiver systems using the Laurent representation and iterative decoding are proposed and evaluated. First, a 4-state reduced-complexity linear demodulator is shown to perform close to the optimal 16-state demodulator for BT = 0.3. This reduced-complexity receiver is used with a coded GMSK system, and a model for considering the code and modulation memory together as a serially concatenated code is proposed. Transceiver systems based on this model have no bandwidth expansion above the original coded GMSK, and are shown to have significantly better performance. The effect of SISO parameters on performance are examined. Results are presented for both additive white Gaussian noise (AWGN) and Rayleigh fading channels. The causes of performance loss in a Rayleigh fading environment are identified, and methods for making the system more robust to fading using an adaptive matched filter and coding are shown. Simulation results are presented demonstrating that the performance of the proposed adaptive receiver is superior to the non-adaptive receiver in a Rayleigh channel. |
