For fixed routing, a single, permanent route is configured for each source- destination pair of nodes in the network. It determines routes using a least cost algorithm. The routes are fixed, or at least only change when there is a change in the topology of the network. A central routing matrix is created for implementing fixed routing which is shown below:

: First Generation adaptive algorithm

In this algorithm, each node maintains two vectors:

Where

Di = delay vector for node i

dij = current estimate of minimum delay from node i to node j (dij = 0)

N = number of nodes in network

Si = successor node vector for node i

sij = the next node in the current minimum delay route from i to j

Periodically each node exchanges its delay vector with all of its neighbors. On the basis of all incoming delay vectors, a node k updates both of its vectors as follows:

Delay for going k to j node, dkj_{ }= min [d _{ij} + l _{ki }]

i € A

s _{kj }= i using i that minimizes the preceding expression

where

A = set of neighbor nodes for k

l _{k}i = current estimate of delay from k to i

Figure below provides an example of the original ARPANET algorithm:

• For network above, snapshot for node 1 shown

• Later, link costs change in the network

• Update comes in from its neighbors (2, 3, 4)