Hawaii International Conference on System Sciences

All too often a seemingly insurmountable divide between theory and practice can be witnessed. In this paper we try to contribute to narrowing this gap in the field of ad-hoc routing. In particular we consider two aspects: We propose a new geometric routing algorithm which is outstandingly efficient on practical average-case networks, however is also in theory asymptotically worst-case optimal. On the other hand we are able to drop the formerly necessary assumption that the distance between network nodes may not fall below a constant value, an assumption that cannot be maintained for practical networks. Abandoning this assumption we identify from a theoretical point of view two fundamentamentally different classes of cost metrics for routing in ad-hoc networks.

/pdf/geometric-ad-hoc-routing-of-theory-and-practice-21lmlozaab.pdf

Geometric ad-hoc routing: of theory and practice

Topology control in ad-hoc networks tries to lower node energy consumption by reducing transmission power and by confining interference, collisions and consequently retransmissions. Commonly low interference is claimed to be a consequence to sparseness of the resulting topology. In this paper we disprove this implication. In contrast to most of the related work claiming to solve the interference issue by graph sparseness without providing clear argumentation or proofs, we provide a concise and intuitive definition of interference. Based on this definition we show that most currently proposed topology control algorithms do not effectively constrain interference. Furthermore we propose connectivity-preserving an spanner constructions that are interference-minimal.

/pdf/does-topology-control-reduce-interference-2zl06qrf7p.pdf

Does topology control reduce interference

In this paper we present GOAFR, a new geometric ad-hoc routing algorithm combining greedy and face routing. We evaluate this algorithm by both rigorous analysis and comprehensive simulation. GOAFR is the first ad-hoc algorithm to be both asymptotically optimal and average-case efficient. For our simulations we identify a network density range critical for any routing algorithm. We study a dozen of routing algorithms and show that GOAFR outperforms other prominent algorithms, such as GPSR or AFR.

/pdf/worst-case-optimal-and-average-case-efficient-geometric-ad-3q87txrpcp.pdf

Worst-Case optimal and average-case efficient geometric ad-hoc routing

Efficient routing among a set of mobile hosts is one of the most important functions in ad hoc wireless networks. Routing based on a connected dominating set is a promising approach, where the search space for a route is reduced to the hosts in the set. A set is dominating if all the hosts in the system are either in the set or neighbors of hosts in the set. The efficiency of dominating-set-based routing mainly depends on the overhead introduced in the formation of the dominating set and the size of the dominating set. In this paper, we first review a localized formation of a connected dominating set called marking process and dominating-set-based routing. Then, we propose a dominant pruning rule to reduce the size of the dominating set. This dominant pruning rule (called Rule k) is a generalization of two existing rules (called Rule 1 and Rule 2, respectively). We prove that the vertex set derived by applying Rule k is still a connected dominating set. Rule k is more effective in reducing the dominating set derived from the marking process than the combination of Rules 1 and 2 and, surprisingly, in a restricted implementation with local neighborhood information, Rule k has the same communication complexity and less computation complexity. Simulation results confirm that Rule k outperforms Rules 1 and 2, especially in networks with relatively high vertex degree and high percentage of unidirectional links. We also prove that an upper bound exists on the average size of the dominating set derived from Rule k in its restricted implementation.

/pdf/an-extended-localized-algorithm-for-connected-dominating-set-nznskre8jb.pdf

An extended localized algorithm for connected dominating set formation in ad hoc wireless networks

We propose a new geometric spanner for static wireless ad hoc networks, which can be constructed efficiently in a localized manner. It integrates the connected dominating set and the local Delaunay graph to form a backbone of the wireless network. Priori arts showed that both structures can be constructed locally with bounded communication costs. This new spanner has these following attractive properties: 1) the backbone is a planar graph, 2) the node degree of the backbone is bounded from above by a positive constant, 3) it is a spanner for both hops and length, 4) it can be constructed locally and is easy to maintain when the nodes move around, and 5) moreover, the communication cost of each node is bounded by a constant. Simulation results are also presented for studying its practical performance.

Geometric spanners for wireless ad hoc networks

A maximal independent set (MIS) S for a graph G is an independent set and no proper superset of S is also independent. A set S is dominating if each node in the graph is either in S or adjacent to one of the nodes in S. The subgraph weakly induced by S is the graph G′ such that each edge in G′ has at least one end point in S. A set S is a weakly-connected dominating set (WCDS) of G if S is dominating and G′ is connected. G′ is a sparse spanner if it has linear edges. The nodes of WCDS have been proposed in the literature as clusterheads for clustered wireless ad hoc networks. In this paper, we present two distributed algorithms for constructing a WCDS for wireless ad hoc networks in linear time. The first algorithm has an approximation ratio of 5, and requires O(n log n) messages, while the second algorithm has a larger approximation ratio, and requires only O(n) messages. Both of these algorithms are used to obtain sparse spanners. The spanner obtained by the second algorithm has a topological dilation of 3, and a geometric dilation of 6. Both of these algorithms are based on the construction of a MIS. The first algorithm requires the construction of a spanning tree. The second algorithm is fully localized, and does not depend on the spanning tree, which makes the maintenance of the WCDS simpler, and guarantees the maintenance of the same approximation ratio.

Maximal independent set, weakly-connected dominating set, and induced spanners in wireless ad hoc networks

A set S is dominating if each node in the graph G = (V, E) is either in S or adjacent to at least one of the nodes in S. The subgraph weakly induced by S is the graph G' = (V, E') such that each edge in E' has at least one end point in S. The set S is a weakly-connected dominating set (WCDS) of G if S is dominating and G' is connected G' is a sparse spanner if it has linear edges. In this paper, we present two distributed algorithms for finding a WCDS in O(n) time. The first algorithm has an approximation ratio of 5, and requires O(n log n) messages. The second algorithm has a larger approximation ratio, but it requires only O(n) messages. The graph G' generated by the second algorithm forms a sparse spanner with a topological dilation of 3, and a geometric dilation of 6.

/pdf/weakly-connected-dominating-sets-and-sparse-spanners-in-1egz80dkj5.pdf

Weakly-connected dominating sets and sparse spanners in wireless ad hoc networks

An algorithm for efficiently creating a message efficient virtual backbone in a wireless ad hoc network utilizes three phases to establish an efficient network among the independent transceivers of a wireless ad hoc network. Independent transceivers within the transmission range of each other are neighbors. A leader election and tree construction phase constructs a tree of neighboring transceivers with one transceiver being designated the root and with each transceiver establishing and recording its location in the tree structure and the identifiers of its neighbors; and reporting when the tree is established. A level calculation phase determines the level of each transceiver away from the root transceiver, with each transceiver recording the level of its neighbors. Precedence for each transceiver is established with consideration of each transceiver's tree level and identifier, with tree level being paramount in deciding precedence; and reporting when the levels of the tree are established. A backbone construction phase establishes all transceivers as a dominator or a dominatee, with the dominators forming the network backbone and the dominatees all being neighbors to a dominator. Within the network each transceiver only needs to know the information of its neighboring transceivers.

Technique for establishing a virtual backbone in an ad hoc wireless network

Let α2(G), γ(G) and γc(G) be the 2-independence number, the domination number, and the connected domination number of a graph G respectively. Then α2(G) ≤ γ (G) ≤ γc(G). In this paper , we present a simple heuristic for Minimum Connected Dominating Set in graphs. When running on a graph G excluding Km (the complete graph of order m) as a minor, the heuristic produces a connected dominating set of cardinality at most 7α2(G) - 4 if m = 3, or at most if m ≥ 4. In particular, if running on a planar graph G, the heuristic outputs a connected dominating set of cardinality at most 15α2(G) - 5.

Khaled M. Alzoubi

Papers

Geometric spanners for wireless ad hoc networks

Maximal independent set, weakly-connected dominating set, and induced spanners in wireless ad hoc networks

Weakly-connected dominating sets and sparse spanners in wireless ad hoc networks

Technique for establishing a virtual backbone in an ad hoc wireless network

A simple heuristic for minimum connected dominating set in graphs