Routing Based on Security

Routing Based on Security

I. A. Almerhag (University of Tripoli, Libya)
DOI: 10.4018/978-1-4666-5808-0.ch015


Even though it is an essential requirement of any computer system, there is not yet a standard method to measure data security, especially when sending information over a network. However, the most common technique used to achieve the three goals of security is encryption. Three security metrics are derived from important issues of network security in this chapter. Each metric demonstrates the level of achievement in preserving one of the security goals. Routing algorithms based on these metrics are implemented to test the proposed solution. Computational effort and blocking probability are used to assess the behavior and the performance of these routing algorithms. Results show that the algorithms are able to find feasible paths between communicating parties and make reasonable savings in the computational effort needed to find an acceptable path. Consequently, higher blocking probabilities are encountered, which is the price to be paid for such savings.
Chapter Preview

Routing Metrics

A routing algorithm and a metric are the basic building blocks for packets routing in WANs. Existing routing protocols are using either a single metric, or multiple metrics. Generally speaking, there is a direct relationship between the number of metrics in use and both the performance and complexity of the routing algorithm. Examples of conventional metrics in use are: hop count, cost, delay, jitter, bandwidth and reliability.

Quality of service routing is a well-known problem and much research has been done in this area. It is concerned with finding a path across a network for a message to follow starting at a source node till it reaches its final destination. This process relies on a routing algorithm that uses topological data, also called metrics or weights, collected in the initial phase. In the case of routing, each forwarding decision is associated with a particular service response, so that a “best-effort” path to a particular destination address may differ from a “low-latency” path, which in turn may differ from a “high-bandwidth” path, and so on.

Basically, QoS is a collection of technologies that allow applications to request and receive predictable service levels in terms of data throughput capacity (bandwidth), latency variations (jitter), cost, reliability and delay (Cisco Systems, 2001).The main goal of QoS is to provide different services to different network traffic over various technologies. Emerging networks, such as Asynchronous Transfer Mode (ATM), can provide QoS guarantees on bandwidth and delay for the transfer of continuous media data (Eberle & Oertli, 1998).

Using bandwidth is the simplest way to describe QoS. However, it is not good enough to define QoS using a single metric only. Practically, a number of metrics are used to express the concept of QoS. The computation of path metric value depends on the individual metric performance. Three main types of metrics were defined; these are: additive, concave and multiplicative. Typical examples of these metric types are delay, bandwidth and packet loss respectively (Kenyon, 2002).

Complete Chapter List

Search this Book: