Improving Scalability of Wireless Ad Hoc Networks Using MPR

摘要: Improving Scalability Of Wireless Ad Hoc Networks Using MPR Zheng Wang, Hamid R. Sadjadpour, Department of Electrical Engineering University California Santa Cruz Cruz, CA 95064, USA Email: { wzgold,hamid@soe.ucsc.edu } Abstract— Gupta and Kumar established that per node throughput capacity wireless ad hoc networks with multi- pair unicast traffic scales as λ(n) = Θ(1/ n log n). Recent investigation network coding (NC) broadcasting has demonstrated there is no increase in the order multi-pair for networks. This paper investigates scalability which allows multipacket reception (MPR) successive inter- ference cancelation (SIC) capabilities all receiver nodes network. The result shows we can p improve random Θ( n/n) using protocol model, a gain Θ(log n) compared earlier results 1 . I. I NTRODUCTION have been studied [1]. [1] showed √ systems proportional to where total number implies cannot scale when goes infinity. work by Ahlswede, Cai, Li Yeung [2] introduced concept (NC). It proved NC achieve max-flow min-cut directed graph single source multicast applications. original concentrated on graphs while most undirected links half-duplex communication. Junning et al. [3] some similarity conjecture given [4], had session broadcast, only achieves at twice improvement graphs. Thus motivation this find another method From proof procedure [1], it be concluded constrained was supported part US Army Research Office under grants W911NF-05-1-0246 Basking Chair Computer Engineering. Opinion, interpretations, conclusions recommendations are those authors not necessarily endorsed Defense. J.J Garcia-Luna-Aceves jj@soe.ucsc.edu mutual interference concurrent transmissions among nodes. Our objective reduce increasing complexity each Therefore, assume endowed SIC. Ghez [5], [6] Tong [7] framework many-to-one communications also known MPR. In context, multiple cooperate transmit their packets simultaneously multiuser detection (MUD), antennas (DA), or input output (MIMO) techniques [8], [9]. utilizes (MUD) interfer- ence decode packets. Recently, Toumpis Goldsmith [10] shown regions significantly increased access schemes combined spatial reuse (i.e., simultaneous transmissions), multihop routing packet relaying), (SIC). main contribution high probability 2 We prove upper lower bounds converge n/n). assumptions similar except equipped SIC capabilities. model based focus 2- D analysis. First, bound maximum flow through sparsity cut square unit theorem proved. Secondly, use Chernoff achieved probability. other word, achievable equal bound. Heuristically, Θ(r(n)). With connectivity condition, transmission radius least n/n), thus scheme static Comparing [3], say our stochastic an event occurs (w.h.p.) if its tends one → ∞. Per defined bits second every w.h.p. destination.