Capacity scaling in delay tolerant networks with heterogeneous mobile nodes

Abstract

We provide a general framework for the analysis of the capacity scaling properties in mobile ad-hoc networks with heterogeneous nodes and spatial inhomogeneities. Existing analytical studies strongly rely on the assumption that nodes are identical and uniformly visit the entire network space. Experimental data, however, have shown that the mobility pattern of individual nodes is typically restricted over the area, while the overall node density is often largely inhomogeneous, due to prevailing clustering behavior resulting from hot-spots. Such ubiquitous features of realistic mobility processes demand to reconsider the scaling laws for the peruser throughput achievable by the store-carry-forward communication paradigm which provides the foundation of many promising applications of delay tolerant networking. We show how the analysis of the asymptotic capacity of dense mobile ad-hoc networks can be transformed, under mild assumptions, into a Maximum Concurrent Flow (MCF) problem over an associated Generalized Random Geometric Graph (GRGG). Our methodology allows to identify the scaling laws for a general class of mobile wireless networks, and to precisely determine under which conditions the mobility of nodes can indeed be exploited to increase the per-node throughput. At last we propose a simple, asymptotically optimal, scheduling and routing scheme that achieves the maximum transport capacity of the network.

Worst-case latency of broadcast in intermittently connected networks

“…The multi-commodity flow problem has indeed been used to find the capacity of wireless (and other) networks (e.g. (Garetto et al, 2007)). However, our reason for using this concept is not the capacity itself, but rather the connection with graph expansion (Hoory et al, 2006), as will be shown in the next section.…”

On Delay Constrained Multicast Capacity of Large-Scale Mobile Ad-Hoc Networks

“…The following lemma is presented in [18]. Inequality (16) holds since the total number of bits transmitted or received in T time slots cannot exceed n s pW T. Inequality (17) holds since the total number of bits transmitted to relay nodes cannot exceed n s (p+1)W T. Inequality (18) holds since each successful target delivery associates an exclusion region which is a disk with radius ∆α B /2. Note that β B > κ(1 + pγ 2 ) log(n s p) implies that H(d B , 2γ, t) ≥ κ(1 + pγ 2 ) log(n s p)…”

“…Over the past few years, there have been a lot of interest in characterizing the capacity of MANETs under a range of mobility models [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20]. Most of these work assumes unicast traffic flows and studies the unicast capacity.…”

Multicast Capacity of Wireless Ad Hoc Networks Under Gaussian Channel Model

“…Bhandari and Vaidya [1] studied the unicast capacity of multi-channel wireless networks with random (c, f ) assignment. Garetto et al [4] studied the capacity scaling in delay tolerant networks with heterogeneous mobile devices. Their methodology allows to identify the scaling laws for a general class of mobile wireless networks, and to precisely determine under which conditions the mobility of nodes can indeed be exploited to increase the per-node throughput.…”