Beyond Proportional Fairness: A Resource Biasing Framework for Shaping Throughput Profiles in Multihop Wireless Networks

Throughput performance of multihop wireless networks is governed by how the network's transport capacity (in bit-meters per second) is partitioned among different network flows. Max-min fair allocation leads to poor throughput performance for all flows because connections traversing a large number of hops consume a disproportionate share of resources. While proportional fair allocation provides a significant improvement, we point out here that there is a much richer space of resource allocation strategies for introducing a controlled bias against resource-intensive long connections in order to significantly improve the performance of shorter connections. We present an analytical model that gives insight into the impact of a particular resource allocation strategy on network performance, in a manner that captures the effect of finite network size and spatial traffic patterns. Our simulation results demonstrate that it is possible to provide significantly better performance to shorter connections than max-min fair or proportional fair resource allocations, with minimal impact on the performance of long connections, using mixed bias strategies blending "fair" allocations with a strong bias against long connections.