The document proposes a lightweight intrusion detection system for IoT networks using support vector machines (SVM). It analyzes common IoT threats like denial of service attacks and models IoT traffic using Poisson distributions. Features are extracted from training data and an SVM classifier is trained and evaluated on test data. Several experiments show the linear SVM performs best, detecting attacks with over 98% accuracy while requiring less time and resources than other classifiers. The proposed system is shown to effectively detect attacks in IoT networks while keeping low computational overhead needed for resource-constrained IoT devices.
Mehryar Emambakhsh, Alessandro Bay, and Eduard Vazquez, "Deep Recurrent Neural Network for Multi-target Filtering", 25th International Conference MultiMedia Modelling (MMM), Thessaloniki, Greece, pp. 519--531, 2018
Face recognition using gaussian mixture model & artificial neural networkeSAT Journals
Abstract
Face recognition is a non-contact and friendly biometric identification technology. It has broad application prospects in the
military, public security and economic security. In this work, we also consider illumination variable database. The images have
taken from far distance and do not consider the close view face of the individual as in most of the face databases, clear face view
has been considered. In this first we located face as region of interest and then LBP and LPQ descriptors are used which is
illuminance invariant in nature. After this GMM has been used to reduce feature set by taking negative log-likelihood from each
LBP and LPQ descripted image histograms. After this ANN consumes stayed used for organization purposes. The investigational
consequencesshow excellent correctness rates in overall testing of input data.
Keywords: Illumination invariant, face recognition, LBP, LPQs,GMM,ANN
Machine Learning Laboratory set of experiments, including ANN, Backpropagation, K-Means, Hierarchical Clustering, Linear Regression, Multivariate Regression, Fuzzy Logic.
Mehryar Emambakhsh, Alessandro Bay, and Eduard Vazquez, "Deep Recurrent Neural Network for Multi-target Filtering", 25th International Conference MultiMedia Modelling (MMM), Thessaloniki, Greece, pp. 519--531, 2018
Face recognition using gaussian mixture model & artificial neural networkeSAT Journals
Abstract
Face recognition is a non-contact and friendly biometric identification technology. It has broad application prospects in the
military, public security and economic security. In this work, we also consider illumination variable database. The images have
taken from far distance and do not consider the close view face of the individual as in most of the face databases, clear face view
has been considered. In this first we located face as region of interest and then LBP and LPQ descriptors are used which is
illuminance invariant in nature. After this GMM has been used to reduce feature set by taking negative log-likelihood from each
LBP and LPQ descripted image histograms. After this ANN consumes stayed used for organization purposes. The investigational
consequencesshow excellent correctness rates in overall testing of input data.
Keywords: Illumination invariant, face recognition, LBP, LPQs,GMM,ANN
Machine Learning Laboratory set of experiments, including ANN, Backpropagation, K-Means, Hierarchical Clustering, Linear Regression, Multivariate Regression, Fuzzy Logic.
V2.0 open power ai virtual university deep learning and ai introductionGanesan Narayanasamy
OpenPOWER AI virtual University's - focus on bringing together industry, government and academic expertise to connect and help shape the AI future .
https://www.youtube.com/channel/UCYLtbUp0AH0ZAv5mNut1Kcg
The International Journal of Engineering & Science is aimed at providing a platform for researchers, engineers, scientists, or educators to publish their original research results, to exchange new ideas, to disseminate information in innovative designs, engineering experiences and technological skills. It is also the Journal's objective to promote engineering and technology education. All papers submitted to the Journal will be blind peer-reviewed. Only original articles will be published.
The papers for publication in The International Journal of Engineering& Science are selected through rigorous peer reviews to ensure originality, timeliness, relevance, and readability.
With the increase in Internet users the number of malicious users are also growing day-by-day posing a
serious problem in distinguishing between normal and abnormal behavior of users in the network. This
has led to the research area of intrusion detection which essentially analyzes the network traffic and tries
to determine normal and abnormal patterns of behavior.In this paper, we have analyzed the standard
NSL-KDD intrusion dataset using some neural network based techniques for predicting possible
intrusions. Four most effective classification methods, namely, Radial Basis Function Network, Self-
Organizing Map, Sequential Minimal Optimization, and Projective Adaptive Resonance Theory have been
applied. In order to enhance the performance of the classifiers, three entropy based feature selection
methods have been applied as preprocessing of data. Performances of different combinations of classifiers
and attribute reduction methods have also been compared.
A methodology for full system power modeling in heterogeneous data centersRaimon Bosch
The need for energy-awareness in current data centers has encouraged the use of power modeling to estimate their power consumption. However, existing models present noticeable limitations, which make them application-dependent, platform-dependent, inaccurate, or computationally complex. In this paper, we propose a platform-and application-agnostic methodology for full-system power modeling in heterogeneous data centers that overcomes those limitations. It derives a single model per platform, which works with high accuracy for heterogeneous applications with different patterns of resource usage and energy consumption, by systematically selecting a minimum set of resource usage indicators and extracting complex relations among them that capture the impact on energy consumption of all the resources in the system. We demonstrate our methodology by generating power models for heterogeneous platforms with very different power consumption profiles. Our validation experiments with real Cloud applications show that such models provide high accuracy (around 5% of average estimation error).
https://www.bsc.es/research-and-development/publications/methodology-full-system-power-modeling-heterogeneous-data
Neural Network Algorithm for Radar Signal RecognitionIJERA Editor
Nowadays, the traditional recognition method could not match the development of radar signals. In this paper, based on fractal theory and Neural Network, a new radar signal recognition algorithm is presented. The relevant point is extracted as the input of neutral network, and then it will recognize and classify the signals. Simulation results show that, this algorithm has a distinguish effect on classification under the condition of low SNR.
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology
Design of c slotted microstrip antenna using artificial neural network modeleSAT Journals
Abstract In this paper, neural network model has been used to estimation of resonance frequency of a coaxial feed C-slotted Microstrip Antenna. The Multi-Layer Perceptron Feed forward back Propagation (MLPFFBP) and Radial basis function Artificial Neural Network (RBFANN) have been used to implement the neural network model. A relative performance analysis of the proposed neural network for different training algorithms. Number of neurons and number of hidden layer is also carried out for estimating the resonance frequency. The method of moment (MOM) based IE3D software was used to generate data dictionary for training and validation set of ANN. The results obtain using ANN are compared with simulation feeding and found quite satisfactory and also it is concluded that RBFANN network is more accurate and fast compared to MLPFFBP network algorithm. Index Terms: Artificial Neural Network, C slot, Microstrip Antenna, Multilayer Feed Forward Networks, Radial basis function Artificial Neural Network, Resonance frequency.
Overview of the fundamental roles in Hydropower generation and the components involved in wider Electrical Engineering.
This paper presents the design and construction of hydroelectric dams from the hydrologist’s survey of the valley before construction, all aspects and involved disciplines, fluid dynamics, structural engineering, generation and mains frequency regulation to the very transmission of power through the network in the United Kingdom.
Author: Robbie Edward Sayers
Collaborators and co editors: Charlie Sims and Connor Healey.
(C) 2024 Robbie E. Sayers
V2.0 open power ai virtual university deep learning and ai introductionGanesan Narayanasamy
OpenPOWER AI virtual University's - focus on bringing together industry, government and academic expertise to connect and help shape the AI future .
https://www.youtube.com/channel/UCYLtbUp0AH0ZAv5mNut1Kcg
The International Journal of Engineering & Science is aimed at providing a platform for researchers, engineers, scientists, or educators to publish their original research results, to exchange new ideas, to disseminate information in innovative designs, engineering experiences and technological skills. It is also the Journal's objective to promote engineering and technology education. All papers submitted to the Journal will be blind peer-reviewed. Only original articles will be published.
The papers for publication in The International Journal of Engineering& Science are selected through rigorous peer reviews to ensure originality, timeliness, relevance, and readability.
With the increase in Internet users the number of malicious users are also growing day-by-day posing a
serious problem in distinguishing between normal and abnormal behavior of users in the network. This
has led to the research area of intrusion detection which essentially analyzes the network traffic and tries
to determine normal and abnormal patterns of behavior.In this paper, we have analyzed the standard
NSL-KDD intrusion dataset using some neural network based techniques for predicting possible
intrusions. Four most effective classification methods, namely, Radial Basis Function Network, Self-
Organizing Map, Sequential Minimal Optimization, and Projective Adaptive Resonance Theory have been
applied. In order to enhance the performance of the classifiers, three entropy based feature selection
methods have been applied as preprocessing of data. Performances of different combinations of classifiers
and attribute reduction methods have also been compared.
A methodology for full system power modeling in heterogeneous data centersRaimon Bosch
The need for energy-awareness in current data centers has encouraged the use of power modeling to estimate their power consumption. However, existing models present noticeable limitations, which make them application-dependent, platform-dependent, inaccurate, or computationally complex. In this paper, we propose a platform-and application-agnostic methodology for full-system power modeling in heterogeneous data centers that overcomes those limitations. It derives a single model per platform, which works with high accuracy for heterogeneous applications with different patterns of resource usage and energy consumption, by systematically selecting a minimum set of resource usage indicators and extracting complex relations among them that capture the impact on energy consumption of all the resources in the system. We demonstrate our methodology by generating power models for heterogeneous platforms with very different power consumption profiles. Our validation experiments with real Cloud applications show that such models provide high accuracy (around 5% of average estimation error).
https://www.bsc.es/research-and-development/publications/methodology-full-system-power-modeling-heterogeneous-data
Neural Network Algorithm for Radar Signal RecognitionIJERA Editor
Nowadays, the traditional recognition method could not match the development of radar signals. In this paper, based on fractal theory and Neural Network, a new radar signal recognition algorithm is presented. The relevant point is extracted as the input of neutral network, and then it will recognize and classify the signals. Simulation results show that, this algorithm has a distinguish effect on classification under the condition of low SNR.
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology
Design of c slotted microstrip antenna using artificial neural network modeleSAT Journals
Abstract In this paper, neural network model has been used to estimation of resonance frequency of a coaxial feed C-slotted Microstrip Antenna. The Multi-Layer Perceptron Feed forward back Propagation (MLPFFBP) and Radial basis function Artificial Neural Network (RBFANN) have been used to implement the neural network model. A relative performance analysis of the proposed neural network for different training algorithms. Number of neurons and number of hidden layer is also carried out for estimating the resonance frequency. The method of moment (MOM) based IE3D software was used to generate data dictionary for training and validation set of ANN. The results obtain using ANN are compared with simulation feeding and found quite satisfactory and also it is concluded that RBFANN network is more accurate and fast compared to MLPFFBP network algorithm. Index Terms: Artificial Neural Network, C slot, Microstrip Antenna, Multilayer Feed Forward Networks, Radial basis function Artificial Neural Network, Resonance frequency.
Overview of the fundamental roles in Hydropower generation and the components involved in wider Electrical Engineering.
This paper presents the design and construction of hydroelectric dams from the hydrologist’s survey of the valley before construction, all aspects and involved disciplines, fluid dynamics, structural engineering, generation and mains frequency regulation to the very transmission of power through the network in the United Kingdom.
Author: Robbie Edward Sayers
Collaborators and co editors: Charlie Sims and Connor Healey.
(C) 2024 Robbie E. Sayers
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...Dr.Costas Sachpazis
Terzaghi's soil bearing capacity theory, developed by Karl Terzaghi, is a fundamental principle in geotechnical engineering used to determine the bearing capacity of shallow foundations. This theory provides a method to calculate the ultimate bearing capacity of soil, which is the maximum load per unit area that the soil can support without undergoing shear failure. The Calculation HTML Code included.
COLLEGE BUS MANAGEMENT SYSTEM PROJECT REPORT.pdfKamal Acharya
The College Bus Management system is completely developed by Visual Basic .NET Version. The application is connect with most secured database language MS SQL Server. The application is develop by using best combination of front-end and back-end languages. The application is totally design like flat user interface. This flat user interface is more attractive user interface in 2017. The application is gives more important to the system functionality. The application is to manage the student’s details, driver’s details, bus details, bus route details, bus fees details and more. The application has only one unit for admin. The admin can manage the entire application. The admin can login into the application by using username and password of the admin. The application is develop for big and small colleges. It is more user friendly for non-computer person. Even they can easily learn how to manage the application within hours. The application is more secure by the admin. The system will give an effective output for the VB.Net and SQL Server given as input to the system. The compiled java program given as input to the system, after scanning the program will generate different reports. The application generates the report for users. The admin can view and download the report of the data. The application deliver the excel format reports. Because, excel formatted reports is very easy to understand the income and expense of the college bus. This application is mainly develop for windows operating system users. In 2017, 73% of people enterprises are using windows operating system. So the application will easily install for all the windows operating system users. The application-developed size is very low. The application consumes very low space in disk. Therefore, the user can allocate very minimum local disk space for this application.
Saudi Arabia stands as a titan in the global energy landscape, renowned for its abundant oil and gas resources. It's the largest exporter of petroleum and holds some of the world's most significant reserves. Let's delve into the top 10 oil and gas projects shaping Saudi Arabia's energy future in 2024.
Courier management system project report.pdfKamal Acharya
It is now-a-days very important for the people to send or receive articles like imported furniture, electronic items, gifts, business goods and the like. People depend vastly on different transport systems which mostly use the manual way of receiving and delivering the articles. There is no way to track the articles till they are received and there is no way to let the customer know what happened in transit, once he booked some articles. In such a situation, we need a system which completely computerizes the cargo activities including time to time tracking of the articles sent. This need is fulfilled by Courier Management System software which is online software for the cargo management people that enables them to receive the goods from a source and send them to a required destination and track their status from time to time.
Cosmetic shop management system project report.pdfKamal Acharya
Buying new cosmetic products is difficult. It can even be scary for those who have sensitive skin and are prone to skin trouble. The information needed to alleviate this problem is on the back of each product, but it's thought to interpret those ingredient lists unless you have a background in chemistry.
Instead of buying and hoping for the best, we can use data science to help us predict which products may be good fits for us. It includes various function programs to do the above mentioned tasks.
Data file handling has been effectively used in the program.
The automated cosmetic shop management system should deal with the automation of general workflow and administration process of the shop. The main processes of the system focus on customer's request where the system is able to search the most appropriate products and deliver it to the customers. It should help the employees to quickly identify the list of cosmetic product that have reached the minimum quantity and also keep a track of expired date for each cosmetic product. It should help the employees to find the rack number in which the product is placed.It is also Faster and more efficient way.
TECHNICAL TRAINING MANUAL GENERAL FAMILIARIZATION COURSEDuvanRamosGarzon1
AIRCRAFT GENERAL
The Single Aisle is the most advanced family aircraft in service today, with fly-by-wire flight controls.
The A318, A319, A320 and A321 are twin-engine subsonic medium range aircraft.
The family offers a choice of engines
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdffxintegritypublishin
Advancements in technology unveil a myriad of electrical and electronic breakthroughs geared towards efficiently harnessing limited resources to meet human energy demands. The optimization of hybrid solar PV panels and pumped hydro energy supply systems plays a pivotal role in utilizing natural resources effectively. This initiative not only benefits humanity but also fosters environmental sustainability. The study investigated the design optimization of these hybrid systems, focusing on understanding solar radiation patterns, identifying geographical influences on solar radiation, formulating a mathematical model for system optimization, and determining the optimal configuration of PV panels and pumped hydro storage. Through a comparative analysis approach and eight weeks of data collection, the study addressed key research questions related to solar radiation patterns and optimal system design. The findings highlighted regions with heightened solar radiation levels, showcasing substantial potential for power generation and emphasizing the system's efficiency. Optimizing system design significantly boosted power generation, promoted renewable energy utilization, and enhanced energy storage capacity. The study underscored the benefits of optimizing hybrid solar PV panels and pumped hydro energy supply systems for sustainable energy usage. Optimizing the design of solar PV panels and pumped hydro energy supply systems as examined across diverse climatic conditions in a developing country, not only enhances power generation but also improves the integration of renewable energy sources and boosts energy storage capacities, particularly beneficial for less economically prosperous regions. Additionally, the study provides valuable insights for advancing energy research in economically viable areas. Recommendations included conducting site-specific assessments, utilizing advanced modeling tools, implementing regular maintenance protocols, and enhancing communication among system components.
Immunizing Image Classifiers Against Localized Adversary Attacksgerogepatton
This paper addresses the vulnerability of deep learning models, particularly convolutional neural networks
(CNN)s, to adversarial attacks and presents a proactive training technique designed to counter them. We
introduce a novel volumization algorithm, which transforms 2D images into 3D volumetric representations.
When combined with 3D convolution and deep curriculum learning optimization (CLO), itsignificantly improves
the immunity of models against localized universal attacks by up to 40%. We evaluate our proposed approach
using contemporary CNN architectures and the modified Canadian Institute for Advanced Research (CIFAR-10
and CIFAR-100) and ImageNet Large Scale Visual Recognition Challenge (ILSVRC12) datasets, showcasing
accuracy improvements over previous techniques. The results indicate that the combination of the volumetric
input and curriculum learning holds significant promise for mitigating adversarial attacks without necessitating
adversary training.
Student information management system project report ii.pdfKamal Acharya
Our project explains about the student management. This project mainly explains the various actions related to student details. This project shows some ease in adding, editing and deleting the student details. It also provides a less time consuming process for viewing, adding, editing and deleting the marks of the students.
Democratizing Fuzzing at Scale by Abhishek Aryaabh.arya
Presented at NUS: Fuzzing and Software Security Summer School 2024
This keynote talks about the democratization of fuzzing at scale, highlighting the collaboration between open source communities, academia, and industry to advance the field of fuzzing. It delves into the history of fuzzing, the development of scalable fuzzing platforms, and the empowerment of community-driven research. The talk will further discuss recent advancements leveraging AI/ML and offer insights into the future evolution of the fuzzing landscape.
1. Towards a Lightweight Intrusion Detection
System for the Internet of Things
1
Rashila Shrestha
M.Tech IT 2021
Nov 30, 2022
SANA ULLAH JAN, SAEED AHMED, VLADIMIR SHAKHOV, AND INSOO KOO
University of Ulsan
IEEE Access
2. Outlines
● Introduction
● Related Works
● Motivations
● Analysis of IoT threats
● Proposed Intrusion Detection System
● Experiments (1-7)
● Discussion
● Research Questions
● Conclusion and Future Scope
2
4. Introduction
● Integration of the internet into the entities of the different domains of human society is emerging.
● However, wide-range IoT networks make them prone to cyber attacks.
● One of the main types of attack is denial of service (DoS), where the attacker floods the network with
a large volume of data to prevent nodes from using the services.
● An intrusion detection mechanism is considered a chief source of protection for information and
communications technology.
● However, conventional intrusion detection methods need to be modified and improved for
application to the Internet of Things
○ certain limitations, like resource-constrained devices, the limited memory and battery capacity of
nodes, and specific protocol stacks.
4
6. Related Works
● Farahnakian and Heikkonen proposed a deep autoencoder (DAE)-based
classification model to detect and classify intrusions in the network.
○ Obtained classifier by combining the set of auto-encoder(AE)
○ Output of AE at (n-1)th layer is the input to AE at nth layer.
○ Better intrusion detection compared to deep belief network (DBN) and combination of AE+DBN
● Shone et al. proposes a nonsymmetric deep auto-encoder (NDAE) to learn features
in an unsupervised manner.
○ A set of NDAEs is stacked to perform the learning and classification tasks.
6
7. Related Work contd..
● Oh et al. proposed a malicious pattern detection.
○ Auxiliary Shifting Method and early detection scheme
○ Reduce memory usage in pattern matching process
○ Efficient early detection of malicious pattern
○ Limitation:
■ Failed to detect and classify other attacks, such as denial of service (DoS), false data injection,
etc.
● Ali et al. proposed a fast learning network (FLN) with particle swarm optimization
(PSO) applied for convergence of classifier parameters.
○ Limitations:
■ Complexity of the system was too high to be applied to sensor nodes due to their low
computation and energy-storage capabilities.
7
8. Related Work contd..
● Moukhafi et al. combined a hybrid genetic algorithm (GA) and an SVM with PSO for
feature subset selection in their proposed intrusion detection system.
○ Succeeded in differentiating DoS attack from other attacks with 100% accuracy
○ Limitation:
■ Could not discriminate normal class signals from other types of attacks with reasonable
accuracy.
● Kabir et al. proposed an optimum allocation–based least square SVM (OA-LS-SVM)
for intrusion detection systems.
○ This technique first combines the training and testing datasets. Then, an optimal allocation (OA)
scheme determines the volume of training and testing sets.
○ Later, it selects representative samples directly from training and testing datasets for the classifier.
8
9. Motivation
● All those proposed methods were able to perform efficiently in detecting intrusions
in the network.
● These techniques rely on resource-intensive computing and may be too heavy for
the low-capacity nodes of IoT networks.
● Thus, it was necessary to design an IDS requiring low computational costs, minimal
energy, and little memory in the network nodes.
9
11. IoT Traffic Modeling
● To analyze network behavior, a mathematical model of the traffic is used.
● Poisson process has been used for traffic modeling in IoT system.
● Also generates training and testing samples for SVM performance analysis.
11
12. Proposed Intrusion Detection System
● Two main phases of the system:
○ Training and evaluation phase
● In training phase, labeled data samples are obtained
● In the first stage, features are extracted to obtain feature pool.
● The resulting feature pool along with a vector of labels is then used to train the
classifier.
● After a trained classifier is obtained, it is then presented to classify the unobserved
samples from the test dataset.
● To evaluate the performance of the classifier, similar features used in the training
phase are extracted from the test samples in the test dataset.
● These unlabeled test samples are then given input to the classifier to obtain the
predicted output.
12
14. Performance Evaluation Parameters
● A true positive is an outcome where the model correctly predicts the positive class.
● Similarly, a true negative is an outcome where the model correctly predicts the
negative class.
● A false positive is an outcome where the model incorrectly predicts the positive
class.
● A false negative is an outcome where the model incorrectly predicts the negative
class.
14
15. 15
Performance Evaluation Parameters
● True Positive Rate: The ratio of true positives to the number of observations
predicted as positive. Also known as sensitivity, recall, or hit rate:
● False Positive Rate: The ratio of false positives to the sum of false positives and true
negatives. Also known as fall-out.
16. Performance Evaluation Parameters contd..
● False Detection Rate: Defined as (1 – accuracy) and calculated as follows:
● Accuracy (ACC): The ratio of true values to total observations, calculated as follow:
16
17. Experiment 1
● SVM-based classifier is trained with three kernels
○ linear, polynomial, and radial-basis function
● The four feature combinations:
○ mean and maximum (max) values
○ mean and median values
○ max and median values
○ mean, max, and median values.
● Datasets with 100 observations
○ 40 observations for training the classifier
○ 60 observations for testing
17
20. Analysis
● Linear-SVM outperformed the counter poly SVM and rbf-SVM in terms of accurately
classifying the input signal using all four combinations of features.
● Detection time for all types of SVM was similar for respective combinations of
features.
● Mean and max took the longest time to classify 120 observations in the test set.
“Mean, max, and median was the combination of features that required the shortest time
from among all combinations of features to obtain final results”
20
23. Experiment 2
● Test the performance of classifier:
○ Signal S, of length 1000 elements obtained using bigram technique(N-gram with N=2)
○ S = {s1, s2, ..., sr} r: random number between 1-1000
○ Snorm = {s1, s2, ..., sr} was obtained through Poisson distribution with parameter λnorm,
○ Signal Sintr = {Sr+1, Sr+2, ..., S1000} was generated with a Poisson distribution from parameter λintr.
○ Finally, the two vectors, Snorm and Sint were concatenated to obtain one signal, S = [Snorm,Sintr], of
1000 elements
● λnorm = 2 and λintr =4 are used to obtain the training and testing signals.
23
24. 24
Analysis
● Linear kernel-based classifier outperformed the other two classifiers in terms of
accuracy.
● The polynomial kernel-based classifier reported the worst accuracy for features sets
1, 2 and 3.
● In feature set 4, the RBF kernel-based classifier showed the lowest accuracy.
● The RBF kernel-based classifier outperformed the linear and polynomial kernel-
based classifiers in terms of TPR.
● The worst results were reported by the polynomial kernel based classifier
26. Experiment 3
● Scenario:
○ similar λnorm, λintr = 6
● Performance of Kernel-based classifier improved compared to linear and RBF
kernel-based classifiers
● Lower window size for feature set 1:
○ increase the accuracy of TPR of polynomial kernel-based classifier.
● Further, decrease in w :
○ classifier’s performance degraded to lower in linear and RBF kernel-based classifiers.
● Using feature sets 2,3,and 4:
○ polynomial kernel-based classifier showed comparable performance to remaining
classifiers.
○ Improved performance in FPR, FDR and τ.
● Increase in λintr increases the efficiency of all classifiers for all sets of features
26
28. Experiment 4
● Scenario:
○ λintr = 1.5 and λnorm = 2
● Analysis
○ performance of the polynomial kernel–based classifier was
worse compared to the other two classifiers in many cases.
○ performance of the linear and RBF kernel–based classifiers
are comparable to each other.
28
30. Experiment 5
● Scenario:
○ λintr was further decreased to 1 to analyze the effect on the
performance of the classifiers.
● Analysis:
○ performance by all classifiers improved for all cases and classifier types
○ linear kernel–based classifier achieved the best results among the
three types of classifiers
● Results strengthen the statement that an increase in the
difference between λnorm and λintr increases the performance
of the classifiers.
30
32. Experiment 6
● Compared proposed SVM algorithm with NN, KNN and DT algorithm
● SVM was implemented using three kernels.
○ linear, polynomial and radial-basis function - SVM(L), SVM(P) and SVM(R)
● Selected parameters and algorithm:
○ Neural Network, a feed forward network with input, hidden and output
layers comprised of 2 or 3 and 2 and 1 nodes respectively.
○ Levenberg Marquardt (LM) in training phase for backpropagation
○ KNN classifier was used to classify an input sample by obtaining the
Euclidean distance from 3 nearest neighbor
○ Decision Tree, a classification decision tree was fitted in training phase to
perform classification of test samples
32
34. Experiment 7
● Efficiency of proposed algorithm was analysed based on accuracy and CPU time with
the algorithm in GA-SVM, A-IDS, and WFS-IDS.
● Measure of accuracy shows how efficiently an algorithm classify an input signal as
normal or intruded.
● To measure the lightweightness of an algorithm, the two parameters including
consumed energy and elapsed time were used.
34
35. Analysis
● Datasets - 10,000 samples where the 60% samples are reserved for training and 40%
for testing purpose
○ three features; mean, maximum and median measures of the packet arrival rate
● Generate a parallel synchronized dataset to the CICIDS2017 for proving the
efficiency of proposed algorithm.
● The DDoS attack was implemented in the network at random points in time.
● This subset contained a total of 225,745 samples attributed in 78 features.
● Performance comparison among the algorithms GA-SVM, A-IDS, WFS-IDS and the
proposed algorithm by utilizing the generated dataset using Poisson distribution.
35
37. Analysis
● The accuracy of the proposed algorithm is the highest at 98.03%. The worst
performance is shown by the A-IDS algorithm with 89.76% accuracy.
● CPU execution time of the proposed scheme is the lowest with 208.9063, and 2.281
seconds for training and testing phases, respectively.
● These results prove the lightweightness property of the proposed algorithm with the
given set.
37
38. Analysis contd..
● The results of proposed, GA-SVM and WFS-IDS were comparable.
● The A-IDS shows comparatively poor performance.
● The proposed IDS successfully achieve highest TPs, TNs and lowest FNs instances.
38
39. Discussion
● Three features are used for feature selection and reduction:
○ system processing time can be reduced due to low time consumption of single attribute acquisition
from input data instead of multiple attributes.
○ extracting just 2 or 3 features from that single attribute takes lower time as compared to extraction of
up to 40 features from the multiple attributes.
○ the complexity of SVM is also reduced because of utilizing a much lower number of dimensions
(features) of input samples.
● The packet arrival rate was considered
○ follow the Poisson distribution, of the traffic intensity to the node.
● The Poisson distribution is just used for performance evaluation.
39
40. Research Questions
Why to use Packet Arrival Rate Attribute?
● The intrusions affects the traffic intensity.
● Either the data rate is increased(blackhole attack, warmhole attack), or
decreased(jamming, packet flooding)
● Thus, by monitoring the traffic intensity or packet arrival rate, intrusion in the
network can be detected.
40
41. B. Why SVM?
● Efficient Performance in ML algorithm
● Performance comparison in the project confirm the claim
C. How do we get less training and testing times?
● The major reason which reduces the computational time of the proposed algorithm
is the elimination of the feature selection which is the part of training phase only.
41
42. Conclusions
● Design lightweight IDS for anomaly detection in the IoT
● A common type of attack, known as DDoS, is the target.
● The proposed IDS focuses on two major issues
○ the attribute of the receiving data used to classify the signal
○ the machine learning based classifier
● Comparative analysis of SVM-based classifier with other machine learning-based
classifiers
● The results show that an SVM-based IDS can perform satisfactorily in detection of
attacks.
42
43. Future Work
● The effect of changing traffic intensity is not clearly pronounced or masked by
intruders.
● The algorithms which consider the complex 40 features may be able to detect more
sophisticated intrusions.
43
44. References
[1] Bruno Bogaz Zarpelão, Rodrigo Sanches Miani, Cláudio Toshio Kawakani, and Sean Carlisto de Alvarenga. A
survey of intrusion detection in internet of things. Journal of Network and Computer Applications, 84:25–37, 2017.
[2] Sabrina Sicari, Alessandra Rizzardi, Luigi Alfredo Grieco, and Alberto Coen-Porisini. Security, privacy and trust in
internet of things: The road ahead. Computer networks, 76:146–164, 2015.
[3] Dhananjay Singh, Gaurav Tripathi, and Antonio J Jara. A survey of internet-of-things: Future vision, architecture,
challenges and services. In Internet of things (WF-IoT), 2014 IEEE world forum on, pages 287–292. IEEE, 2014.
[4] Aref Meddeb. Internet of things standards: who stands out from the crowd? IEEE Communications Magazine,
54(7):40–47, 2016.
[5] Hichem Sedjelmaci, Sidi Mohamed Senouci, and Tarik Taleb. An accurate security game for low-resource iot
devices. IEEE Transactions on Vehicular Technology, 66(10):9381–9393, 2017.
44