XTC: a practical topology control algorithm for ad-hoc networks

Abstract

The XTC ad-hoc network topology control algorithm introduced in this paper shows three main advantages over previously proposed algorithms. First, it is extremely simple and strictly local. Second, it does not assume the network graph to be a Unit Disk Graph; XTC proves correct also on general weighted network graphs. Third, the algorithm does not require availability of node position information. Instead, XTC operates with a general notion of order over the neighbors' link qualities. In the special case of the network graph being a Unit Disk Graph, the resulting topology proves to have bounded degree, to be a planar graph, and-on average-case graphs-to be a good spanner.

Topology Control with Limited Geometric Information

“…As shown in [11], if neighborhood orderings are distance-based (as in (1)), then G XT C is connected. Note that according to this definition, each node u orders its neighbors in increasing order of distance, breaking ties using ids.…”

“…It is shown in [11] that G XT C is symmetric provided G is and this is independent of ≺. It is also shown that if G is a Euclidean graph and ≺ u is defined as…”

“…More recently, Wattenhofer and Zollinger [11] have proposed a topology control protocol called XTC that does not rely on specific location information for each node, but rather requires each node to only know the distance to each of its neighbors. Although this is an improvement over location-based topology control algorithms, XTC still suffers from lack of robustness to errors in distance information.…”

“…This is a significant problem because in general it seems hard to accurately estimate pairwise node distances. For example, [11] mentions the use of the strength of the received signal (RSSI) as a way to estimate distances. While this technique is relatively cheap and does not need additional hardware, it is known to have low accuracy.…”

“…As mentioned in [11], the XTC protocol (shown above) consists of three main steps: (i) neighbor ordering (Line 1), (ii) neighbor order exchange (Line 2), and (iii) edge selection (Lines 3-11). In the edge selection step a vertex u decides to drop v from its set of neighbors if there is a vertex w that u and v both agree is mutually better.…”