Modal space-time channel decomposition for dynamic resource allocation in wireless networks

In mobile wireless networks, dynamic allocation of power and bandwidth must respond to changes in the quality of the wireless link in addition to changes in traffic conditions. A key stumbling block in accounting for the quality of the wireless link is accurate yet tractable modeling of the time-varying channel characteristics. We introduce a framework for dynamic resource allocation in code-division multiple access (CDMA) systems based on the notion of canonical space-time modes (CSTMs). The CSTMs are defined by fixed space-time basis waveforms that correspond to a fundamental sampling of multipath delays, Doppler shifts, and spatial angles encountered during propagation. The CSTMs decompose the dispersive wireless channel into nondispersive virtual channels, each providing a fraction of the overall bandwidth. We consider rate and bit error rate as the Quality-of-Service (QoS) parameters. The QoS requirements are accommodated by distributing the CSTMs between different streams in proportion to their rates and adjusting the powers in the CSTMs according to the required bit error rates. The performance of this simple resource allocation strategy is compared for various CSTM-based CDMA transceiver structures.