Ad-hoc networks, a promising trend in wireless technology, fail to work properly in a global setting. In most cases, self-organization and cost-free local communication cannot compensate the need for being connected, gathering urgent information just-in-time. Equipping mobile devices additionally with GSM or UMTS adapters in order to communicate with arbitrary remote devices or even a fixed network infrastructure provides an opportunity. Devices that operate as intermediate nodes between the ad-hoc network and a reliable backbone network are potential injection points. They allow disseminating received information within the local neighborhood. The effectiveness of different devices to serve as injection point differs substantially. For practical reasons the determination of injection points should be done locally, within the ad-hoc network partitions. We analyze different localized algorithms using at most 2-hop neighboring information. Results show that devices selected this way spread information more efficiently through the ad-hoc network. Our results can also be applied in order to support the election process for clusterheads in the field of clustering mechanisms.
Deep Dive into Localized Support for Injection Point Election in Hybrid Networks.
Ad-hoc networks, a promising trend in wireless technology, fail to work properly in a global setting. In most cases, self-organization and cost-free local communication cannot compensate the need for being connected, gathering urgent information just-in-time. Equipping mobile devices additionally with GSM or UMTS adapters in order to communicate with arbitrary remote devices or even a fixed network infrastructure provides an opportunity. Devices that operate as intermediate nodes between the ad-hoc network and a reliable backbone network are potential injection points. They allow disseminating received information within the local neighborhood. The effectiveness of different devices to serve as injection point differs substantially. For practical reasons the determination of injection points should be done locally, within the ad-hoc network partitions. We analyze different localized algorithms using at most 2-hop neighboring information. Results show that devices selected this way spre
Mobile devices such as cellular phones, notebooks, MP3-players, digital cameras, and alike become more commonplace in our daily life [1].
Multi-hop ad-hoc networks are composed of a set of these devices that communicate with each other over a wireless medium [2] using for instance Wi-Fi and Bluetooth.
Such networks can be established spontaneously whenever devices are in transmission range. In addition to being mobile, the desire of being connected with other devices or, in particular, the Internet, arises immediately [3].
The goal has been redefined to overcome the limitations of pure ad-hoc networks by augmenting them with instant Internet access. Technological a d v a n c e s l i k e U M T S a n d G S M f o r l i n k i n g t o a backbone network drive considerable progress in this respect.
Devices offering both ad-hoc as well as backbone connectivity can operate as intermediate nodes between the different network types. We denote them using the term injection points [4]. The resulting network is of a hybrid nature.
Injection points serve two different purposes: a point where information dissemination starts [5]and where services are being placed [6].
In the first case, the injection point is of essential importance at the moment of receiving information and passing this information to the neighborhood.
The injection point might represent a bottleneck, depending on the amount of data passing through. Injection points become particularly attractive when offering a service. Information dissemination can be seen as such a service that is usable by devices in the injection point’s surroundings.
However, it is necessary to pay particular attention on the efficiency of those injection points. Consequently, the dynamic and self-organizing augmentation of networks by additional links plays an important role. The costs of such augmentation should usually be made as small as possible.
Multiple different device properties have significant impact on the suitability of serving as injection point. Those include available power, technical equipment, load balancing issues, and for instance also the time a device is expected to remain available [7].
Aside from that, the effectiveness of injection points in terms of information spreading is crucial as well. The problem is to choose nodes that can disseminate information (in a region) more efficiently, i.e. with a minimized number of hops or messages.
For practical reasons the determination of injection points should be done locally, within the ad-hoc network partitions.
Thus, our contribution in this paper is to analyze several localized approaches to determine potential injection points. We apply this analysis to different network densities. We finally can offer guidelines on selecting the proper approach, depending on the network density at hand.
The approaches introduced use at most 2-hop topological information only, rather than relying on geographical positions.
The remainder of this paper is organized as follows. The underlying system model as well as the problem statement is illustrated in Section 2. In Section 3 localized approaches for the determination of injection points are introduced. We present an empirical study of these approaches in Section 4 and finally give a conclusion in Section 5.
A wireless network consists of a set of computers connected by wireless network links. We assume that technologies like Bluetooth and Wi-Fi can be employed to create ad hoc communication links within the transmission range at no charge. All nodes have the same communication range with bidirectional communication links.
Additional cellular network links such as GSM/UMTS might be employed by appropriately equipped devices to establish supplementary communication links to a network backbone. These links, however, will induce costs.
Our working assumption considers that the backbone network does not keep track of participating mobile devices. Thus, the request for information injection has to be initiated by a mobile device, the injection point candidate.
Note that we do not need to consider the details of the MAC and network layer for our investigations.
Furthermore we assume that each node knows its current one-hop neighbors and even the list of two-hop neighbors. Geographical positions of the nodes are not used at all.
We define the injection point candidate problem as finding nodes that serve as more appropriate injection points according to some criteria than other nodes.
The objective is to classify nodes in order to indicate nodes that are more appropriate with high probability than the remaining nodes. These nodes are called injection point candidates.
Bridge nodes connect two or more groups of nodes as illustrated in Fig. 1a. Thus, bridge nodes present a potential single-point-of-failure in terms of partitioning. These problematic situations appear when a node v has a high number of neighbors and the neighbors are grouped. To illustrate this term s
…(Full text truncated)…
This content is AI-processed based on ArXiv data.
