This thesis is on trellis coded modulation (TCM) schemes with asymmetrical constellation. This modulation technique can be easily adapted to the intelligent systems like Cognitive Radio and Software Defined Radio (SDR) with improved error performance. Different types of asymmetric constellation methods for QPSK, 8-PSK and 16-PSK are used, which result in better performance compared to the case of standard symmetrical constellation assignment. In a cognitive radio environment, the channel conditions change frequently. Thus it suggests the requirement for a non-linear adaptive modulation technique with variable parameters. This requirement can be met by using trellis coded modulation with asymmetrical constellation [15,16].
The approach to this thesis is to use digital modulation techniques namely M-ary Phase Shift Keying (MPSK), in combination with convolutional codes of specific code rate, and then combining these convolutional codes to a TCM mapper, and finally carrying out TCM encoding and decoding. Towards this, the following steps are performed; designing the convolutional encoder based on Ungerboeck‟s design scheme, using a signal mapper to map the modulation scheme, making use of Ungerboeck's set partitioning technique to obtain the maximum distance between signals [2], obtaining the trellis which maximizes the performance, and carrying out decoding using the Viterbi algorithm [18]. The method used in this thesis yields an improvement in error, minimum distance and coding gain as compared to those of conventional trellis coded modulation technique.