1.

0 INTRODUCTION

The human immune system can be used as inspiration when developing algorithms to
solve difficult computational problems. This is because it is a robust, decentralized, complex,
and error tolerant biological system; i.e.it possesses properties that make it ideal for certain
application areas, such as computer intrusion detection and pattern recognition.
Artificial Immune Systems (AIS) [20] are algorithms and systems that use the human
immune system as inspiration. The human immune system is a robust, decentralized, error
tolerant and adaptive system. Such properties are highly desirable for the development of novel
computer systems. Unlike some other bio inspired techniques, such as genetic algorithms and
neural networks, the field of AIS encompasses a spectrum of algorithms that exist because
different algorithms implement different properties of different cells. All AIS algorithms mimic
the behaviour and properties of immunological cells, specifically B cells, T cells and dendritic
cells (DCs), but the resultant algorithms exhibit differing levels of complexity and can perform a
range of tasks.
The major part of AIS work to date has been the development of three algorithms derived
from more simplified models; negative selection, clonal selection and immune networks.
However, these first generation AIS algorithms have often shown considerable limitations when
applied to realistic applications. For this reason, a second generation of AIS is emerging, using
models derived from cutting edge immunology as their basis, not simply mechanisms derived
from basic models found in text books.

1
 2.0. IMMUNE SYSTEM

The immune system is a host defense system comprising many biological structures and
processes within an organism that protects against disease. To function properly, an immune
system must detect a wide variety of agents, known as pathogens, from viruses to parasitic
worms, and distinguish them from the organism's own healthy tissue. In many species, the
immune system can be classified into subsystems, such as the innate immune system versus the
adaptive immune system, or humoral immunity versus cell-mediated immunity.

In humans, the blood–brain barrier, blood–cerebrospinal fluid barrier, and similar fluid–
brain barriers separate the peripheral immune system from the neuroimmune system, which
protects the brain.

The field of Artificial Immune Systems (AIS) is concerned with abstracting the structure and
function of the immune system to computational systems, and investigating the application of
these systems towards solving computational problems from mathematics, engineering, and
information technology.

Pathogens can rapidly evolve and adapt, and thereby avoid detection and neutralization by the
immune system; however, multiple defense mechanisms have also evolved to recognize and
neutralize pathogens. Even simple unicellular organisms such as bacteria possess a rudimentary
immune system in the form of enzymes that protect against bacteriophage infections.

2.1. Structure and Function of the Immune System

The immune system generates a large variety of cells and molecules for defensive purposes.
These cells and molecules together act in a dynamic and intricate network of interactions to
detect and eliminate antigens. It is difficult to give a concise picture of such a complex system;
moreover, many of the mechanisms are not completely understood. This Section gives an
abstract view of the immune system, yet omits many details of many specific mechanisms. The
purpose is to serve as a reference to the subsequent sections. Detailed Review of the natural
immune system and its functionalities may be found elsewhere. The immune system can be
2
envisioned as a multilayer system with defense mechanisms in several layers. A biological
immune system consists of two different response systems: innate and adaptive immune system.
The main layers include the anatomic barrier, the innate immunity and the adaptive immunity.
They are described as follows

 Anatomic Barrier: The first layer is the anatomic barrier, composed of the skin and the
surface of mucous membranes. Intact skin prevents the penetration of most pathogens
and also inhibits most bacterial growth because of its low ph. On the other hand, may
pathogens enter the body by binding or penetrating through the mucous membranes;
these membranes provide a number of nonspecific mechanisms that help to prevent such
entry. Saliva, tears, and some mucous secretions act to wash away potential invader sand
also contain antibacterial and antiviral substances.
 Innate immunity: Innate immunity which is also known as nonspecific immunity, refers
to the defense mechanism against foreign invaders that individuals are born with. Innate
immunity is mainly composed of the following mechanisms:
 Physiologic barriers: This includes mechanisms like temperature, pH, oxygen
tension, and various soluble chemicals. The purpose of these mechanisms is to
provide detrimental living conditions for foreign pathogens. For instance, the low
acidity of the gastric system acts as a barrier to infection by ingested micro-
organisms, since they cannot survive the low pH of the stomach
 Phagocytic barriers: Some specialized cells (like macrophages, neutrophils and
natural killer cells) are able to ingest specific material, including whole
pathogenic micro-organisms. This Ingestion has two purposes: to kill the antigen
and to present fragments of the invader's proteins to other immune cells and
molecules.
 Inflammatory response: Activated macrophages produce proteins called
cytokines. They work as hormone-like messengers that induce the inflammatory
response, which is characterized by vasodilation and rise in capillary
permeability. These changes allow a large number of circulating immune cells to
be recruited to the site of the infection. The cytokines are also produced by other
immune cells and non-immune cells, for example those that secrete cytokines
when damaged.
3
 3.0. ARITIFICIAL IMMUNE SYSTEM (AIS)

Artificial Immune System is a sub-field of biologically-inspired computing, and Natural

computation, with interests in Machine Learning and belonging to the broader field of Artificial
Intelligence.

Artificial Immune Systems (AISs) are adaptive systems, inspired by theoretical

immunology and observed immune functions, principles and models, which are applied to
problem solving.

AIS is distinct from computational immunology and theoretical biology that are

concerned with simulating immunology using computational and mathematical models towards
better understanding the immune system, although such models initiated the field of AIS and
continue to provide a fertile ground for inspiration.

Finally, the field of AIS is not concerned with the investigation of the immune system as a
substrate for computation, unlike other fields such as DNA computing.

3.1. TECHNIQUES USED

The common techniques are inspired by specific immunological theories that explain the
function of the mammalian adaptive immune system.

 Clonal Selection Algorithm: A class of algorithms inspired by the clonal

selection theory of acquired immunity that explains how B and T
lymphocytes improve their response to antigens over time called affinity
maturation. These algorithms focus on the Darwinian attributes of the theory
where selection is inspired by the affinity of antigen-antibody interactions,
reproduction is inspired by cell division, and variation is inspired by somatic
hyper mutation. Clonal selection algorithms are most commonly applied to
optimization and pattern recognition domains, some of which resemble parallel
hill climbing and the genetic algorithm without the recombination operator. (de
Castro, Leandro N.; Timmis, Jonathan ,2002).

4
  Clonal Selection Principle
The clonal selection principle describes the basic features of an immune response
to an antigenic stimulus. It establishes the idea that only those cells that recognize
the antigen proliferate, thus being selected against those that do not. The main
features of the clonal selection theory are that:
1. The new cells are copies of their parents (clone) subjected to a mutation
mechanism with high rates (somatic hyper mutation).
2. Elimination of newly differentiated lymphocytes carrying self-reactive
receptors.
3. Proliferation and differentiation on contact of mature cells with antigens. ( K.
Sri Lakshmi,2014)
When an antibody strongly matches an antigen the corresponding B-cell is stimulated to
produce clones of itself that then produce more antibodies. This (hyper) mutation, is quite
rapid, often as much as “one mutation per cell division” (de Castro and Von Zuben, 1999).
This allows a very quick response to the antigens. It should be noted here that in the Artificial
Immune System literature, often no distinction is made between B-cells and the antibodies
they produce.
Both are subsumed under the word ‘antibody’ and statements such as mutation of antibodies
(rather than mutation of B-cells) are common

 Negative Selection Algorithm: Inspired by the positive and negative selection processes
that occur during the maturation of T cells in the thymus called T cell tolerance. Negative
selection refers to the identification and deletion (apoptosis) of self-reacting cells that is T
cells that may select for and attack self-tissues. This class of algorithms are typically used for
classification and pattern recognition problem domains where the problem space is modeled
in the complement of available knowledge. For example, in the case of an anomaly
detection domain the algorithm prepares a set of exemplar pattern detectors trained on
normal (non-anomalous) patterns that model and detect unseen or anomalous patterns.

 Immune Network Algorithm: Algorithms inspired by the idiotypic network theory

proposed by Niels Kaj Jerne that describes the regulation of the immune system by anti-
idiotypic antibodies (antibodies that select for other antibodies). This class of algorithms

5
focus on the network graph structures involved where antibodies (or antibody producing
cells) represent the nodes and the training algorithm involves growing or pruning edges
between the nodes based on affinity (similarity in the problems representation space).
Immune network algorithms have been used in clustering, data visualization, control, and
optimization domains, and share properties with artificial neural network.

 Dendritic Cell Algorithms: The Dendritic Cell Algorithm (DCA) is an example of an

immune inspired algorithm developed using a multi-scale approach. This algorithm is
based on an abstract model of dendritic cells (DCs). The DCA is abstracted and
implemented through a process of examining and modeling various aspects of DC
function, from the molecular networks present within the cell to the behavior exhibited by
a population of cells as a whole. Within the DCA information is granulated at different
layers, achieved through multi-scale processing.

4.0. NEGATIVE SELECTION ALGORITHM (NSA)

Negative Selection Algorithm is one of the most popular Artificial Immune System
models that have attracted much attention from researchers. Forrest et al. proposed Negative
Selection Algorithm, which is based on the concept of self/non-self-discrimination behavior in
the immune system. It is motivated from the fact of negative selection of T cells in the thymus
and worked upon the immune system philosophy to recognize unknown antigens or non-self-cells
without reacting to self-cells. It produces a set of self-patterns (strings) that define the normal
network patterns (strings). This set can easily identify non-self-patterns and marked them as non-
self or anomalous. If any random pattern matches with any self-pattern then it is removed so that
it cannot becomes a detector. A detector is a pattern or a set of patterns that only recognizes the
complement of self-patterns. Those patterns that do not get matched with self-patterns become
detectors and signify non-self. Afterward, these detectors are used to identify non-self or
anomalies. These detectors examine incoming patterns and if any new pattern matches with the
detector, then this represent the detection of an anomaly. The NSA was motivated by the negative
selection process occurring within the Natural Immune System (NIS).

6
 The classical NSAs consist of two stages Firstly, the NSAs generate a detectors set in the
non-self-space. A candidate detector is generated in the whole space. If the detector does not
match with the known self-states, it becomes a mature detector and is added to the detector set. In
the second stage, the unknown states are tested with the detector set. If an unknown state is
matched by any mature detector, the NSAs assert that an anomaly occurs. Fig. 1 illustrates the
core idea of constant-sized and variable-sized detectors in 2-dimensional space. The grey area
represents the self-region, which is usually given through the training data (self-samples). The
white circles are the possible detectors covering the non-self-region. “Holes” are illustrated in
black. Fig. 1.a presents constant-sized NSA while Fig. 1.b illustrates the principles of variable-
sized detectors.

(Source: Chikh Ramdane et al,2017)

4.1. APPLICATIONS OF NEGATIVE SELECTION ALGORITHM

Since his emergence, NSA has attracted the attention of many researchers and has been applied in
numerous real world applications. Its development and application domains generally similar of
those are of computational intelligence approaches such as artificial neural networks,
evolutionary algorithms and fuzzy systems. In the following, a list of recent applications of NSA
is presented: computer security, anomaly detection, data mining, optimization and so on.

4.2. ADVANTAGES OF NEGATIVE SELECTION ALGORITHM

1. The network or computer is constantly monitored for any invasion or attack.

7
 2. The system can be modified and changed according to needs of specific client and can
help outside as well as inner threats to the system and network.
3. It effectively prevents any damage to the network.
4. It provides user friendly interface which allows easy security management systems.

4.3 DISADVANTAGES OF NEGATIVE SELECTION ALGORITHM

The disadvantage of Negative Selection Algorithm is they cannot detect the
source of the attack and in any case of attack, they just lock the whole network.

5.0. Classification of Negative Selection Algorithm

There are a diverse family of NSAs has been developed, the essential characteristics of the original NSA
introduced by Forrest are still remaining. However the first NSA has large time cost complexity and space
complexity. According to data representation there are two types of NSA: the binary NSAs (BNSAs) and
the Real-Valued NSAs (RNSAs). Table1 presents a simple taxonomy of NSA.
Table 1: Negative Selection Algorithm classification

Real

NSA types Binary NSA Constant-sized detectors variable-sized detectors

Criteria's
Abbreviation BNSA CRNSA VRNSA(V-detectors )
Data
representation Binary value(strings) Real value
Radius Constant constant variable
Matching rules r-contiguous bits(rcb),r-chunks,
landscape-affinity matching, Euclidian distances and its derivations
Hamming distance
- Suitable for discrete space

Advantages (representation and search)
-Implementation simplicity
The binary representation has some
limitations for the real world
problems
Drawbacks
High level representation, expressiveness and scalability.
- - Small number of detector - Best coverage of
non self
- The lack of continuous adaptability -Low detection rate and high
false positive rate - Large time cost and space complexity
- Great number of detectors to cover the - Presence of the holes -
non-self-space some overlapping
-Great-Overlapping between detector between detectors

8
 REFERENCES

 B.Santos Kumar, T.Chandra Sekhara Phani Raju, M.Ratnakar, Sk.Dawood Baba,

N.Sudhakar (2013). Intrusion Detection System- Types and Prevention, (IJCSIT)
International Journal of Computer Science and Information Technologies, Vol. 4
(1)
 Chikh Ramdane and Salim Chikhi, Sétif University, Sétif, Algeria, MISC Laboratory,
Constantine University, Constantine, Algeria (2017). Negative Selection Algorithm:
Recent Improvements and Its Application in Intrusion Detection System. International
Journal of Computing Academic Research (IJCAR) ISSN 2305-9184, Volume 6, Number
2

 Dr. Jamie Twycross, Dr. Uwe Aickelin. (2007). An Immune-Inspired Approach to.
Anomaly Detection.
 De Castro, Leandro N.; Timmis, Jonathan (2002). Artificial Immune Systems: A New
Computational Intelligence Approach. Springer. pp. 57–58. ISBN 978-1-85233-
594-6.
 De Castro, L. and J. Timmis, 2002. Artificial immune systems: a novel paradigm for
pattern recognition. L. Alonso, J. Corchado and C. Fyfe (Eds.), Artificial Neural
Networks in Pattern Recognition, University of Paisley, pp: 67-84

 Fabio Gonzalez (2003). A Study of Artificial Immune Systems Applied to Anomaly

Detection.

 H. Bersini, F.J. Varela, (1990)Hints for adaptive problem solving gleaned from immune
networks. Parallel Problem Solving from Nature, First Workshop PPSW 1,
Dortmund,FRG, October,
 Julie Greensmith, Amanda Whitbrook and Uwe Aickelin.(2008). Artificial Immune
Systems.

9
 K. Sri Lakshm(2014).Implementation of Artificial Immune System Algorithms.
International Journal of Application or Innovation in Engineering &
Management(IJAIEM)

 Neyole Misiko Jacob & Muchelule Yusuf Wanjal (2017). A Review of Intrusion
Detection Systems, Global Journal of Computer Science and Technology:
Software & Data Engineering, Volume 17, Issue 3,Version 1.0
 Sanjay Sharma1 , R.K. Gupta2. A Model for Intrusion Detection based on Negative Selection
Algorithm and J48 Decision Tree International Journal for Research in Applied Science
& Engineering Technology (IJRASET) ISSN: 2321-9653; IC Value: 45.98; SJ Impact
Factor :6.887 Volume 5 Issue XII
 V. Cutello, G. Nicosia, M. Pavone, J. Timmis (2007) An Immune Algorithm for Protein
Structure Prediction on Lattice Models, IEEE Transactions on Evolutionary
Computation,vol.11no.1,pp. 101–117.
http://www.dmi.unict.it/nicosia/papers/journals/Nicosia-IEEE-TEVC07.pdf
 www.wikipedia.com. Immune System. Accessed on the 24th of Dec 2018
 www.wikipedia.com. Artificial Immune System. Accessed on the 25th of Dec 2018


10