March 01, 2019
Capacity Scaling of Wireless Networks: Protocol Limited or Model Limited?
- Du J.
Given a large wireless network deployed within an area, where each node wants to transmit to a random destination node within the network at some given rate, how fast can the sum rate grow as the number of nodes scales up? Previous work have shown that, multi-hop routing provides a scaling at most as the square root of network size, whereas hierarchical cooperation protocols have the potential to support linear scaling by creating virtual MIMO transmission between clusters of nodes. In this talk, we will show some new capacity scaling bounds by accounting for the constraints of local communication. Our upper bound is significantly smaller than previous known bounds for networks of any reasonable size (e.g., N<10^40). On the other hand, we realized that the model of capacity scaling, originally posed by Gupta and Kumar and inherited by all the follow up work, has an intrinsic bottleneck. It assumes equal rate for all source-destination pairs, which effectively allocates more resource/time to pairs with worse channel quality. Could such artificial enforcement of equal message size impose the ultimate limit on capacity scaling?View Original Article