Real-World Applications - Immunological Computation: Theory and Applications - page 183

PROCEDURE InitializationOfStatus()
While (size of AB < a threshold)
If (detector expresses NOSYMTOMS in network)
create new antibody (ab) from log file
add newly generated antibody (ab) to AB
For (ab AB)

clone and mutate ab to maximize affinity with ab
add best n clones to AB

PROCEDURE Continuous_adaptation()
antigen_count ← 0
LOOP
receive incoming network packets
ag ← preprocess packets into antigen
antigen_count antigen_count +1

IF (antigen_count = K )
AG ← last K antigens
Update_Population(AG)
antigen_count ← 0

←

compute degree of danger()
WHILE(danger is high)
compute temporal danger zone

AG ← all packets in the danger zone

FOREACH (ag AG)

FOREACH (ab AB)

compute affinity (ab,ag)
high_aff ← highest affinity value
IF (high_aff > a threshold)
block the scanned port

Figure 7.8 The pseudocode for detecting scan probes.

Research by Carter (2000) attempted to create an inductive computation algorithm
based on metaphors taken from immunology. h ey describe an evolutionary search
algorithm based on the model of immune network dynamics.

7.4 Application in Robotics and Control

Robot control works by Ishiguro et al. (1996, 1998), Wantanabe et al . (1998, 1999),
and Lee et al. (1999) focused on the development of a dynamic decentralized
consensus-making mechanism based on the “immune network theory.” h ey
attempted to create a mechanism by which a single, self-su cient autonomous
robot, called the immunoid, could perform the task of collecting various amounts of
garbage from a constantly changing environment. h e authors used the metaphors of
antibodies, which were potential behaviors of the immunoid; antigens corresponded
to environmental inputs such as existence of garbage, wall, and home bases (Figure 7.9).
For the immunoid to make the best decision, it detects antigens and matches the
content of the antigen with a selection of all the antibodies that it possesses. h eir
model included the concepts of “dynamics,” responsible for the variation of the