Dual Degree Dissertation Delay Limited Capacity with Coding

The thesis determines the optimal policy (out of successive decoding and FDMA) that minimizes the sum power of the users in the presence of coding gap while meeting their minimum rate requirements in a time-varying channel for the uplink setup. It is shown that, in the presence of coding gap, successive decoding is no longer an optimal policy in all channel states and that there are certain channel states where FDMA outperforms successive decoding. For these channel states, the thesis determines an optimal bandwidth allocation policy which minimizes the sum power of the users while providing them with their minimum required rates. Further, this optimal bandwidth allocation policy is shown to be a function of the channel state vector. From this we bring out the benefit of the dynamic allocation of bandwidth to the users in the presence of coding gap (dynamic FDMA). Further, for the channel states where successive decoding is optimal and the minimum rate requirement of each user is greater than a particular quantity , it is shown that the optimal decoding order is to decode the users in the decreasing order of their channel state values. The thesis also develops some insights on how the minimum sum power for successive decoding and FDMA depend on the channel state vector, coding gap and the individual channel state values of the users.