Peyman Kabiri | Iran University of Science and Technology (IUST) (original) (raw)
Papers by Peyman Kabiri
Iran Journal of Computer Science
International Journal of Wavelets, Multiresolution and Information Processing
Image fusion is an important concept in remote sensing. Earth observation satellites provide both... more Image fusion is an important concept in remote sensing. Earth observation satellites provide both high-resolution panchromatic and low-resolution multispectral images. Pansharpening is aimed on fusion of a low-resolution multispectral image with a high-resolution panchromatic image. Because of this fusion, a multispectral image with high spatial and spectral resolution is generated. This paper reports a new method to improve spatial resolution of the final multispectral image. The reported work proposes an image fusion method using wavelet packet transform (WPT) and principal component analysis (PCA) methods based on the textures of the panchromatic image. Initially, adaptive PCA (APCA) is applied to both multispectral and panchromatic images. Consequently, WPT is used to decompose the first principal component of multispectral and panchromatic images. Using WPT, high frequency details of both panchromatic and multispectral images are extracted. In areas with similar texture, extrac...
Iran Journal of Computer Science
Virtualization is critical to the infrastructure of cloud computing environment and other online ... more Virtualization is critical to the infrastructure of cloud computing environment and other online services. Hypercall interface is provided by hypervisor to offer privileged requests by the guest domains. Attackers may use this interface to send malicious hypercalls. In the reported work, repetitive hypercall attacks and sending hypercalls within irregular sequences to Xen hypervisor were analyzed, and finally, an intrusion detection system (IDS) is proposed to detect these attacks. The proposed system is placed in the host domain (Dom0). Monitoring hypercalls traffic the system operates based on the identification of irregular behaviors in hypercalls sent from guest domains to hypervisor. Later on, the association rule algorithm is applied on the collected data within a fixed time window, and a set of thresholds for maximum number of all types of the hypercalls is extracted. The results from the implementation of the proposed system show 91% true positive rate.
Electronics
Camera tracking and the construction of a robust and accurate map in unknown environments are sti... more Camera tracking and the construction of a robust and accurate map in unknown environments are still challenging tasks in computer vision and robotic applications. Visual Simultaneous Localization and Mapping (SLAM) along with Augmented Reality (AR) are two important applications, and their performance is entirely dependent on the accuracy of the camera tracking routine. This paper presents a novel feature-based approach for the monocular SLAM problem using a hand-held camera in room-sized workspaces with a maximum scene depth of 4–5 m. In the core of the proposed method, there is a Particle Filter (PF) responsible for the estimation of extrinsic parameters of the camera. In addition, contrary to key-frame based methods, the proposed system tracks the camera frame by frame and constructs a robust and accurate map incrementally. Moreover, the proposed algorithm initially constructs a metric sparse map. To this end, a chessboard pattern with a known cell size has been placed in front o...
TURKISH JOURNAL OF ELECTRICAL ENGINEERING & COMPUTER SCIENCES
A wavelet packet transform (WPT) is a well-known technique used for data and signal-processing th... more A wavelet packet transform (WPT) is a well-known technique used for data and signal-processing that has proven to be successful in condition monitoring and fault diagnosis. In this study, using feature extraction based on wavelet transformation, sound signals emitted from automobile engines under both faulty and healthy conditions are analyzed. The intention is to categorize sound signals into both healthy and faulty classes. Sound signals are generated from 4 different automobile engines in both healthy and faulty conditions. The investigated fault is within the ignition system of the engines. In addition, there are other possible problems that may also affect the generated sound signals. In the reported study, a set of features is initially extracted from the recorded signals. The more informative features are later selected using a correlation-based feature selection (CFS) algorithm. Results prove the efficiency of wavelet-based feature extraction for the case study of the reported work.
Journal of Network and Computer Applications, 2015
ABSTRACT Virtual machine technology facilitates implementation of modern Internet services, espec... more ABSTRACT Virtual machine technology facilitates implementation of modern Internet services, especially multi-tier applications. Server virtualization aims to reduce the cost of service provisioning and improve fault tolerance, portability and security of virtualized services by sharing the resources amongst consolidated servers. Although designing applications based on multi-tier architecture brings benefits to service providers in terms of service flexibility, functionality, and reusability of software in comparison to traditional architecture, deploying multi-tier applications in virtualization environments faces many challenges. In addition to virtualization overhead and virtual machines interference, dynamic and unpredictable nature of submitted workloads makes the performance evaluation of virtualized multi-tier applications very complex task. In this article, we propose a novel analytical model based on queuing network to evaluate virtualized multi-tier applications. The efficiency of the proposed model is evaluated by using a series of in-depth experiments in different configurations to study the performance of multi-tier applications. The conducted experiments examine consolidation of RUBiS and Wikipedia tiers in different scenarios under both bursty and non-bursty workloads. We present the results based on quantitative and qualitative analyses that are useful for virtualized multi-tier applications deployment. The performance modeling and experimental results are in excellent agreement and confirm the efficiency of the proposed model to evaluate and analyze the performance of virtualized multi-tier applications.
Privacy, Intrusion Detection and Response, 2012
A great deal of research attention has been paid to data mining on data streams in recent years. ... more A great deal of research attention has been paid to data mining on data streams in recent years. In this chapter, the authors carry out a case study of anomaly detection in large and high-dimensional network connection data streams using Stream Projected Outlier deTector (SPOT) that is proposed in (Zhang et al. 2009) to detect anomalies from data streams using subspace analysis. SPOT is deployed on the 1999 KDD CUP anomaly detection application. Innovative approaches for training data generation, anomaly classification, and false positive reduction are proposed in this chapter as well. Experimental results demonstrate that SPOT is effective and efficient in detecting anomalies from network data streams and outperforms existing anomaly detection methods.
2014 Iranian Conference on Intelligent Systems (ICIS), 2014
2008 Third International Conference on Digital Information Management, 2008
... Peyman Kabiri, Hamed Shahbazi, Mohsen Soryani Department of Computer Engineering, Iran Univer... more ... Peyman Kabiri, Hamed Shahbazi, Mohsen Soryani Department of Computer Engineering, Iran University of Science and Technology Narmak, Tehran, 16844 Peyman.kabiri@iust.ac.ir ... that rotating fi j (r, θ) by an an-gle θ0 rotates Fi j (u, v) by the same angle independent of w. In ...
As Mobile Ad-hoc network (MANET) has become a very importanttechnology, research concerning its s... more As Mobile Ad-hoc network (MANET) has become a very importanttechnology, research concerning its security problem, especially,in intrusion detection has attracted many researchers. Featureselection methodology plays a central role in the data analysisprocess. The proposed features are tested in different networkoperating conditions. PCA is used to analyze the selectedfeatures. This is because, redundant and irrelevant features oftenreduce performance of the detection system. Performance reductionwill occur both in speed and predictive accuracy. This paper aimsto select and analyze the network features using principalcomponent analysis. In this paper, performing various experiments,normal and attack states are simulated and the results for theselected features are analyzed.
این مقاله روشی را برای ایجاد مجمع رده¬بندهای شبکه عصبی چندلایه ارائه می¬کند. از آنجایی که هم دقت ... more این مقاله روشی را برای ایجاد مجمع رده¬بندهای شبکه عصبی چندلایه ارائه می¬کند. از آنجایی که هم دقت تک تک دسته-بندهای پایه و هم تنوع میان آن¬ها نقش مهمی در ایجاد یک مجمع دقیق دارند بنابراین شبکه عصبی چندلایه که معمولا دقت رده¬بندی بالایی دارد به عنوان رده¬بند پایه انتخاب شد. برای ایجاد تنوع از اعمال نویز بر روی مجموعه داده آموزشی استفاده شده است. در نهایت از روش رای اکثریت برای تجمع رای رده¬بندهای پایه استفاده می¬شود. روش ارائه شده با روش-های ایجاد مجمع آدابوست، بگینگ و جنگل چرخش روی 17 مجموعه داده مقایسه شده است که نشان می¬دهد از لحاظ آزمون آماری از دو روش اول بهتر بوده و با روش سوم نیز برابری می کند.
Automobile industry is an important industry in many countries. Monitoring the operating conditio... more Automobile industry is an important industry in many countries. Monitoring the operating conditions in automobiles is a big challenge not only when the product leaves the factory but also during the after sale maintenance programs. This paper proposes a new approach in acoustic emission based condition monitoring and fault detection that uses wavelet method. The fault that was investigated in this study is related to ignition system of the automobile, more specifically speaking, operation of the sparks. The sampled dataset includes sound samples collected in repair shops. Noise was a major problem for this work. Recording the sound signals in repair shop will include unwanted noise. Human voice, sound of the wind and the sound made by other operating automobile engines are also recorded in the sampled sound record, where, some of them may include arbitrary and random noise. Energy, RMS, Kurtosis, Skewness, Marse, Crest Factor, Zero count Minimum count and Maximum count features are ...
Nowadays, honeypots are widely used to divert attackers from the original target and keep them bu... more Nowadays, honeypots are widely used to divert attackers from the original target and keep them busy within a decoy environment. DeMilitarized Zone (DMZ) is an important zone for network administrators, because many of the services to the public network is provided at this zone. Many of the security tools such as firewalls, intrusion detection systems and several other security systems can be used to secure DMZ. But honeypots are supplementary devices used to discover attacks and capture forensics against the attackers. The most important solution to secure the DMZ is to detect attacks against servers of this zone and void these intrusions by leading them to honeypots and capturing enough forensics against the attackers. This research work is focused on providing a solution for problem areas such as response to intrusion attempts and redirection of the intruders to honeypots. The proposed system detects malicious activities and redirects them to a decoy system to capture forensics. H...
جو زمين يک سيستم آشوبناک است و پديدههايی که در آن رخ میدهند، از پيچيدگی بالايی برخوردار میباشن... more جو زمين يک سيستم آشوبناک است و پديدههايی که در آن رخ میدهند، از پيچيدگی بالايی برخوردار میباشند. روزانه حجم انبوهی از دادههای آبوهوايی در صدها ايستگاه همديدی کشور ثبت میشوند. در اين مقاله، امکان بکارگيری اين دادهها برای پيشبينی کوتاه مدت وضع هوا مورد بررسی قرار گرفته است. برای پيشبينی بر مبنای اين دادهها، چارچوبی پيشنهاد شده است که مبتنی بر بکارگيری مجموعهای از نگاشتهای وضعيت کنونی به وضعيت 3 ساعت آينده است. در بخشی از مقاله امکان پيشبينی وضعيت 3 ساعت آينده ايستگاه همديدی همدان بر پايه وضعيت جاری اين ايستگاه و 9 ايستگاه پيرامونش مورد بررسی قرار گرفته است. برای نگاشت از وضعيت کنونی به وضعيت آينده، نخست پارامترهای ورودی مهمتر در توصيف وضعيت کنونی انتخاب و سپس از يک رگرسيون کننده برای نگاشت به وضعيت آينده استفاده شده است. نتايج تجربی نشان میدهند که پيشبينی دقيق تنها با تکيه بر دادههای ايستگاههای همديدی ميسر نيست.
Intrusion detection in wireless networks has been a challenging research problem in network secur... more Intrusion detection in wireless networks has been a challenging research problem in network security for more than half a century. This paper presents a novel intrusion detection method based on Received Signal Strength Acceptance Test (RSSAT) to improve the IDS capabilities in anomaly-based Host-based Intrusion Detection Systems (HIDS). The new system can identify suspicious behaviors detecting anomalies in the received signal strength from the access points. Several scenarios are implemented in Omnet++ environment to evaluate the performance of the proposed scheme. A test criterion is used to improve accuracy in detecting forged signal powers and at the same time to reduce number of the false positives i.e. number of false attack alerts resulted from the legitimate signal powers.
This paper proposes a new method that uses multivariate analysis of the variance algorithm for in... more This paper proposes a new method that uses multivariate analysis of the variance algorithm for intrusion detection. The proposed methodology manipulates the original features to produce a new feature set with two major advantages. The first advantage is that while reducing the similarity between different classes this method will increase the similarity between the instances of each class. As a result of this feature the resulted classes will have a more convex shape. As a second advantage this algorithm will produce new orthogonal features each holding a larger amount of information than their ancestor. The nearest neighbour algorithm was used to detect intruders. Although this algorithm is fast and less complex than the other classification algorithm in this content promising results were achieved.
In the recent years, machine learning has become very attractive for control engineers. This appr... more In the recent years, machine learning has become very attractive for control engineers. This approach looks more interesting as complexity of systems increases. A Novel Neural Fuzzy based adaptive compensation algorithm is developed. In this approach there is no need to create a transfer function to model the behaviour of the system. This paper reports an ongoing research in adaptive control at the Nottingham Trent University, in collaboration with industry. A Neural Fuzzy Engine (NFE) is used for learning the system behaviour. The trained NFE determines the amount of compensation for each location in the defined working area. At the final stage results are analysed, and they are found to show that this approach can be used to develop control systems which suffer from errors due to flexibility.
This paper presents a black box approach to the system parameters estimation using machine learni... more This paper presents a black box approach to the system parameters estimation using machine learning methods. In mechanical systems not all of the system parameters are presentable by the equations. The nature of some of them is not quiet clear (e.g. backlash errors). However implementing the black box approach using a machine learning method some of these parameters are simulated. The simulated data is used to predict the error at each position during the displacement of the manipulator.
This paper reports an attempt for solving the position inaccuracy and flexibility problems in a v... more This paper reports an attempt for solving the position inaccuracy and flexibility problems in a very inaccurate robot arm by means of machine learning methods. Error sources in the system are studied and problems are pointed out. Utilising a real time monitoring system, positioning data for the end-effector are collected from a touch sensitive sensor. Collected data is converted into appropriate error maps. Using a novel machine learning method, the controller utilises error maps to predict system’s error at each position and compensates for them [3][4]. The machine learning engine generalises the data for the points between the sample points. The employed machine learning method has no training stage.
Iran Journal of Computer Science
International Journal of Wavelets, Multiresolution and Information Processing
Image fusion is an important concept in remote sensing. Earth observation satellites provide both... more Image fusion is an important concept in remote sensing. Earth observation satellites provide both high-resolution panchromatic and low-resolution multispectral images. Pansharpening is aimed on fusion of a low-resolution multispectral image with a high-resolution panchromatic image. Because of this fusion, a multispectral image with high spatial and spectral resolution is generated. This paper reports a new method to improve spatial resolution of the final multispectral image. The reported work proposes an image fusion method using wavelet packet transform (WPT) and principal component analysis (PCA) methods based on the textures of the panchromatic image. Initially, adaptive PCA (APCA) is applied to both multispectral and panchromatic images. Consequently, WPT is used to decompose the first principal component of multispectral and panchromatic images. Using WPT, high frequency details of both panchromatic and multispectral images are extracted. In areas with similar texture, extrac...
Iran Journal of Computer Science
Virtualization is critical to the infrastructure of cloud computing environment and other online ... more Virtualization is critical to the infrastructure of cloud computing environment and other online services. Hypercall interface is provided by hypervisor to offer privileged requests by the guest domains. Attackers may use this interface to send malicious hypercalls. In the reported work, repetitive hypercall attacks and sending hypercalls within irregular sequences to Xen hypervisor were analyzed, and finally, an intrusion detection system (IDS) is proposed to detect these attacks. The proposed system is placed in the host domain (Dom0). Monitoring hypercalls traffic the system operates based on the identification of irregular behaviors in hypercalls sent from guest domains to hypervisor. Later on, the association rule algorithm is applied on the collected data within a fixed time window, and a set of thresholds for maximum number of all types of the hypercalls is extracted. The results from the implementation of the proposed system show 91% true positive rate.
Electronics
Camera tracking and the construction of a robust and accurate map in unknown environments are sti... more Camera tracking and the construction of a robust and accurate map in unknown environments are still challenging tasks in computer vision and robotic applications. Visual Simultaneous Localization and Mapping (SLAM) along with Augmented Reality (AR) are two important applications, and their performance is entirely dependent on the accuracy of the camera tracking routine. This paper presents a novel feature-based approach for the monocular SLAM problem using a hand-held camera in room-sized workspaces with a maximum scene depth of 4–5 m. In the core of the proposed method, there is a Particle Filter (PF) responsible for the estimation of extrinsic parameters of the camera. In addition, contrary to key-frame based methods, the proposed system tracks the camera frame by frame and constructs a robust and accurate map incrementally. Moreover, the proposed algorithm initially constructs a metric sparse map. To this end, a chessboard pattern with a known cell size has been placed in front o...
TURKISH JOURNAL OF ELECTRICAL ENGINEERING & COMPUTER SCIENCES
A wavelet packet transform (WPT) is a well-known technique used for data and signal-processing th... more A wavelet packet transform (WPT) is a well-known technique used for data and signal-processing that has proven to be successful in condition monitoring and fault diagnosis. In this study, using feature extraction based on wavelet transformation, sound signals emitted from automobile engines under both faulty and healthy conditions are analyzed. The intention is to categorize sound signals into both healthy and faulty classes. Sound signals are generated from 4 different automobile engines in both healthy and faulty conditions. The investigated fault is within the ignition system of the engines. In addition, there are other possible problems that may also affect the generated sound signals. In the reported study, a set of features is initially extracted from the recorded signals. The more informative features are later selected using a correlation-based feature selection (CFS) algorithm. Results prove the efficiency of wavelet-based feature extraction for the case study of the reported work.
Journal of Network and Computer Applications, 2015
ABSTRACT Virtual machine technology facilitates implementation of modern Internet services, espec... more ABSTRACT Virtual machine technology facilitates implementation of modern Internet services, especially multi-tier applications. Server virtualization aims to reduce the cost of service provisioning and improve fault tolerance, portability and security of virtualized services by sharing the resources amongst consolidated servers. Although designing applications based on multi-tier architecture brings benefits to service providers in terms of service flexibility, functionality, and reusability of software in comparison to traditional architecture, deploying multi-tier applications in virtualization environments faces many challenges. In addition to virtualization overhead and virtual machines interference, dynamic and unpredictable nature of submitted workloads makes the performance evaluation of virtualized multi-tier applications very complex task. In this article, we propose a novel analytical model based on queuing network to evaluate virtualized multi-tier applications. The efficiency of the proposed model is evaluated by using a series of in-depth experiments in different configurations to study the performance of multi-tier applications. The conducted experiments examine consolidation of RUBiS and Wikipedia tiers in different scenarios under both bursty and non-bursty workloads. We present the results based on quantitative and qualitative analyses that are useful for virtualized multi-tier applications deployment. The performance modeling and experimental results are in excellent agreement and confirm the efficiency of the proposed model to evaluate and analyze the performance of virtualized multi-tier applications.
Privacy, Intrusion Detection and Response, 2012
A great deal of research attention has been paid to data mining on data streams in recent years. ... more A great deal of research attention has been paid to data mining on data streams in recent years. In this chapter, the authors carry out a case study of anomaly detection in large and high-dimensional network connection data streams using Stream Projected Outlier deTector (SPOT) that is proposed in (Zhang et al. 2009) to detect anomalies from data streams using subspace analysis. SPOT is deployed on the 1999 KDD CUP anomaly detection application. Innovative approaches for training data generation, anomaly classification, and false positive reduction are proposed in this chapter as well. Experimental results demonstrate that SPOT is effective and efficient in detecting anomalies from network data streams and outperforms existing anomaly detection methods.
2014 Iranian Conference on Intelligent Systems (ICIS), 2014
2008 Third International Conference on Digital Information Management, 2008
... Peyman Kabiri, Hamed Shahbazi, Mohsen Soryani Department of Computer Engineering, Iran Univer... more ... Peyman Kabiri, Hamed Shahbazi, Mohsen Soryani Department of Computer Engineering, Iran University of Science and Technology Narmak, Tehran, 16844 Peyman.kabiri@iust.ac.ir ... that rotating fi j (r, θ) by an an-gle θ0 rotates Fi j (u, v) by the same angle independent of w. In ...
As Mobile Ad-hoc network (MANET) has become a very importanttechnology, research concerning its s... more As Mobile Ad-hoc network (MANET) has become a very importanttechnology, research concerning its security problem, especially,in intrusion detection has attracted many researchers. Featureselection methodology plays a central role in the data analysisprocess. The proposed features are tested in different networkoperating conditions. PCA is used to analyze the selectedfeatures. This is because, redundant and irrelevant features oftenreduce performance of the detection system. Performance reductionwill occur both in speed and predictive accuracy. This paper aimsto select and analyze the network features using principalcomponent analysis. In this paper, performing various experiments,normal and attack states are simulated and the results for theselected features are analyzed.
این مقاله روشی را برای ایجاد مجمع رده¬بندهای شبکه عصبی چندلایه ارائه می¬کند. از آنجایی که هم دقت ... more این مقاله روشی را برای ایجاد مجمع رده¬بندهای شبکه عصبی چندلایه ارائه می¬کند. از آنجایی که هم دقت تک تک دسته-بندهای پایه و هم تنوع میان آن¬ها نقش مهمی در ایجاد یک مجمع دقیق دارند بنابراین شبکه عصبی چندلایه که معمولا دقت رده¬بندی بالایی دارد به عنوان رده¬بند پایه انتخاب شد. برای ایجاد تنوع از اعمال نویز بر روی مجموعه داده آموزشی استفاده شده است. در نهایت از روش رای اکثریت برای تجمع رای رده¬بندهای پایه استفاده می¬شود. روش ارائه شده با روش-های ایجاد مجمع آدابوست، بگینگ و جنگل چرخش روی 17 مجموعه داده مقایسه شده است که نشان می¬دهد از لحاظ آزمون آماری از دو روش اول بهتر بوده و با روش سوم نیز برابری می کند.
Automobile industry is an important industry in many countries. Monitoring the operating conditio... more Automobile industry is an important industry in many countries. Monitoring the operating conditions in automobiles is a big challenge not only when the product leaves the factory but also during the after sale maintenance programs. This paper proposes a new approach in acoustic emission based condition monitoring and fault detection that uses wavelet method. The fault that was investigated in this study is related to ignition system of the automobile, more specifically speaking, operation of the sparks. The sampled dataset includes sound samples collected in repair shops. Noise was a major problem for this work. Recording the sound signals in repair shop will include unwanted noise. Human voice, sound of the wind and the sound made by other operating automobile engines are also recorded in the sampled sound record, where, some of them may include arbitrary and random noise. Energy, RMS, Kurtosis, Skewness, Marse, Crest Factor, Zero count Minimum count and Maximum count features are ...
Nowadays, honeypots are widely used to divert attackers from the original target and keep them bu... more Nowadays, honeypots are widely used to divert attackers from the original target and keep them busy within a decoy environment. DeMilitarized Zone (DMZ) is an important zone for network administrators, because many of the services to the public network is provided at this zone. Many of the security tools such as firewalls, intrusion detection systems and several other security systems can be used to secure DMZ. But honeypots are supplementary devices used to discover attacks and capture forensics against the attackers. The most important solution to secure the DMZ is to detect attacks against servers of this zone and void these intrusions by leading them to honeypots and capturing enough forensics against the attackers. This research work is focused on providing a solution for problem areas such as response to intrusion attempts and redirection of the intruders to honeypots. The proposed system detects malicious activities and redirects them to a decoy system to capture forensics. H...
جو زمين يک سيستم آشوبناک است و پديدههايی که در آن رخ میدهند، از پيچيدگی بالايی برخوردار میباشن... more جو زمين يک سيستم آشوبناک است و پديدههايی که در آن رخ میدهند، از پيچيدگی بالايی برخوردار میباشند. روزانه حجم انبوهی از دادههای آبوهوايی در صدها ايستگاه همديدی کشور ثبت میشوند. در اين مقاله، امکان بکارگيری اين دادهها برای پيشبينی کوتاه مدت وضع هوا مورد بررسی قرار گرفته است. برای پيشبينی بر مبنای اين دادهها، چارچوبی پيشنهاد شده است که مبتنی بر بکارگيری مجموعهای از نگاشتهای وضعيت کنونی به وضعيت 3 ساعت آينده است. در بخشی از مقاله امکان پيشبينی وضعيت 3 ساعت آينده ايستگاه همديدی همدان بر پايه وضعيت جاری اين ايستگاه و 9 ايستگاه پيرامونش مورد بررسی قرار گرفته است. برای نگاشت از وضعيت کنونی به وضعيت آينده، نخست پارامترهای ورودی مهمتر در توصيف وضعيت کنونی انتخاب و سپس از يک رگرسيون کننده برای نگاشت به وضعيت آينده استفاده شده است. نتايج تجربی نشان میدهند که پيشبينی دقيق تنها با تکيه بر دادههای ايستگاههای همديدی ميسر نيست.
Intrusion detection in wireless networks has been a challenging research problem in network secur... more Intrusion detection in wireless networks has been a challenging research problem in network security for more than half a century. This paper presents a novel intrusion detection method based on Received Signal Strength Acceptance Test (RSSAT) to improve the IDS capabilities in anomaly-based Host-based Intrusion Detection Systems (HIDS). The new system can identify suspicious behaviors detecting anomalies in the received signal strength from the access points. Several scenarios are implemented in Omnet++ environment to evaluate the performance of the proposed scheme. A test criterion is used to improve accuracy in detecting forged signal powers and at the same time to reduce number of the false positives i.e. number of false attack alerts resulted from the legitimate signal powers.
This paper proposes a new method that uses multivariate analysis of the variance algorithm for in... more This paper proposes a new method that uses multivariate analysis of the variance algorithm for intrusion detection. The proposed methodology manipulates the original features to produce a new feature set with two major advantages. The first advantage is that while reducing the similarity between different classes this method will increase the similarity between the instances of each class. As a result of this feature the resulted classes will have a more convex shape. As a second advantage this algorithm will produce new orthogonal features each holding a larger amount of information than their ancestor. The nearest neighbour algorithm was used to detect intruders. Although this algorithm is fast and less complex than the other classification algorithm in this content promising results were achieved.
In the recent years, machine learning has become very attractive for control engineers. This appr... more In the recent years, machine learning has become very attractive for control engineers. This approach looks more interesting as complexity of systems increases. A Novel Neural Fuzzy based adaptive compensation algorithm is developed. In this approach there is no need to create a transfer function to model the behaviour of the system. This paper reports an ongoing research in adaptive control at the Nottingham Trent University, in collaboration with industry. A Neural Fuzzy Engine (NFE) is used for learning the system behaviour. The trained NFE determines the amount of compensation for each location in the defined working area. At the final stage results are analysed, and they are found to show that this approach can be used to develop control systems which suffer from errors due to flexibility.
This paper presents a black box approach to the system parameters estimation using machine learni... more This paper presents a black box approach to the system parameters estimation using machine learning methods. In mechanical systems not all of the system parameters are presentable by the equations. The nature of some of them is not quiet clear (e.g. backlash errors). However implementing the black box approach using a machine learning method some of these parameters are simulated. The simulated data is used to predict the error at each position during the displacement of the manipulator.
This paper reports an attempt for solving the position inaccuracy and flexibility problems in a v... more This paper reports an attempt for solving the position inaccuracy and flexibility problems in a very inaccurate robot arm by means of machine learning methods. Error sources in the system are studied and problems are pointed out. Utilising a real time monitoring system, positioning data for the end-effector are collected from a touch sensitive sensor. Collected data is converted into appropriate error maps. Using a novel machine learning method, the controller utilises error maps to predict system’s error at each position and compensates for them [3][4]. The machine learning engine generalises the data for the points between the sample points. The employed machine learning method has no training stage.