On the use of feedback in multiple antenna common information broadcasting

Common information broadcasting, or multicasting, wireless systems send the same message to multiple users simultaneously. The problem of multiple antenna multicasting using limited feedback is considered, where a transmitter with M_T antennas transmits common information to K users, each equipped with a single antenna. We propose a novel optimization criterion that leads to a higher overall throughput than the traditional max-min criterion can achieve. This is made possible as we leverage channel quality fluctuations among the users by coding the common information over multiple coherence times. Doing so frees ourselves from constraining our coding rate to accommodate the worst user at every coherence time. Generally the user with the poorest channel conditions dictates the multicast throughput. Hence quantizing the channel vector´s magnitude is arguably as important as quantizing the channel vector´s direction, unlike the downlink broadcast channel where directional quantization is paramount to reduce interference while magnitude quantization is less important. We use a simple linear combination of separate magnitude and directional distance functions to create a minimum distance codebook. In our simulations, we show that when K is large and when coding over many coherence periods, our proposed scheme appreciably outperforms traditional Max-Min criterion.