Rate Allocation for the Multi-Source Downlink Channel with Minimax Optimization

We consider the downlink transmission with multiple sources. Each source transmits over an orthogonal channel by using cooperative broadcasting. Each destination is required to receive and decode a given amount of information. A destination can receive different partitions of information from multiple sources. The sources cooperate and allocate the flow rates along each link to minimize the maximum individual transmit power of each source. To simplify the computations involved in the optimization process, we suggest a sub-optimal algorithm. Simulation results indicate that our optimal scheme gives improvements between 4.2 dB and 5.6 dB over the single-source scheme, where a destination selects only one source for transmission, in our considered networks and rate requirements. Our sub-optimal algorithm suffers a loss of below 0.4 dB compared with the optimal one. The expressions involved in the sub-optimal algorithm are less complicated and it requires M iterations, where M is the number of destinations, and a bisection search in each iteration.