Myat Thida Mon - Academia.edu (original) (raw)

Papers by Myat Thida Mon

2019 International Conference on Advanced Information Technologies (ICAIT)

With the rapid growth of current network, the demand of resources is growing significantly. Insuf... more With the rapid growth of current network, the demand of resources is growing significantly. Insufficient bandwidth results unstable throughput in a network. Software-defined network (SDN) has been proposed to provide optimal routing decision in the presence of congestion. Equal Cost Multi-path (ECMP) routing cannot guarantee optimal resource utilization. ECMP causes the long flow collision in network because it does not consider the network parameter such as bandwidth. In this paper, the Flow Path Computing Algorithm (FPCA) is proposed to minimize the network congestion by rerouting the flows in SDN. This algorithm mainly focuses on rerouting the traffic flows over the alternative path when the network congestion is detected. If the flow demand is exceeded 10% of link bandwidth, the algorithm computes the light loaded path based on the port statistics and it shifts from the congested path to light loaded path. Simulation results are presented to show the effectiveness of the FPCA algorithm over ECMP.

Proceedings of 10th International Conference on Science and Engineering 2019, (ICSE 2019), Dec 8, 2019

2020 International Conference on Advanced Information Technologies (ICAIT), 2020

In this paper, the facial expression recognition system is described based on Eigenface. Facial e... more In this paper, the facial expression recognition system is described based on Eigenface. Facial expression recognition has become increasingly common in several applications in the fields of video conferences, surveillance, financial services, health treatment, and access control system. The performance of recognition depends mainly on several challenges. Lighting variations are also the most challenging in the recognition process. In this system, illumination invariant techniques applied for decreasing the lighting affected faces as a preprocessing stage. Eigen faces are used as feature extraction step. K-NN and Multi-SVMs classifier are used to identify the facial expression and to test the performance of the proposed system. The standard dataset of JAFFE is used in the experiments and 84.29% accuracy is achieved for facial expression recognition. The accuracy of the framework is contrasted with the state of the art methods tested on the JAFFE database.

2018 18th International Symposium on Communications and Information Technologies (ISCIT), 2018

Communication networks are growing in size at an increasing rate with the traditional network. Qu... more Communication networks are growing in size at an increasing rate with the traditional network. Quality of Service (QoS) is needed for new applications and new services on the Internet. Software Defined Network (SDN) techniques are appeared to be more flexible and programmable network. By utilizing the characteristics of SDN, this paper proposes QoS sensitive routing for available bandwidth by using segment routing (SR) to find the best path according to the QoS requirements. This paper focuses on monitoring of flow entries among switches and finding the feasible path over QoS-based routing scheme. To provide network throughput to the end hosts, the available bandwidth has to be accomplished. The routing algorithm finds the path which is feasible to meet the desired QoS of data flows. If the required QoS cannot provide to the requested flow, the controller determines how to calculate depending on the request from the switch. If the initial path will not be able to achieve the available bandwidth, the algorithm reroutes the higher bandwidth flow. This paper suggests improvements in QoS using Open Network Operating System (ONOS) controller with OpenFlow protocol and the simulation experiments with effective QoS are suitable for the future internet using SDN.

Long Term Evolution (LTE) supports several QoS (quality of services) classes and tries to guarant... more Long Term Evolution (LTE) supports several QoS (quality of services) classes and tries to guarantee their requirements. There are two main different QoS classes: Guaranteed Bit rate (GBR) such as VoIP and non-Guaranteed Bit Rate (non-GBR) such as FTP or HTTP. Having these different QoS requirements in packet domain introduces an additional challenge on the LTE MAC scheduler design. Therefore, the scheduler has to be aware of the different service requirements and satisfy them. In this paper, QoS aware proportional fair downlink scheduler (QAPF) is proposed. It can optimize the use of available resources while maintaining QoS requirements of different service classes. The result will show the performance evaluation of the proposed scheduler by comparing with others. KeywordsLTE, QoS, GBR, nonGBR, QAPF

Face Detection is the first step of any automated face and facial expression recognition system. ... more Face Detection is the first step of any automated face and facial expression recognition system. It is applied in many applications such as computer vision and biometric applications like face recognition, expression recognition and classification, security access systems, video surveillance, and intelligent human computer interaction. A false detection rate and slow detection speed remain major problems in face detection. In this paper, we present a method to avoid these problems, by combining skin detection and Viola and Jones Face detection. Skin detection is used to reject false positive. After determining all possible face candidate regions, the Viola Jones detector is applied to detect face. Our method has high detection rate and accuracy than an individual face detector. It can perform quiet well in complex background and varying lighting conditions when it is applied images with complex backgrounds. Experimental results show the effectiveness of the proposed method. Keywords...

Nowadays, Internet is the main communication medium but traditional network approach has been dif... more Nowadays, Internet is the main communication medium but traditional network approach has been difficult to adapt to the fast growing Internet. The emergence of Software Defined Network (SDN) has brought to predict for the solution of this problem. Methods for measuring Quality of Service (QoS) parameters such as bandwidth utilization, bandwidth allocation and delay have been introduced not only for traditional network but also for SDN-based scenarios. Bandwidth allocation is a great significance in various areas of networking. We propose a low-complexity bandwidth allocation algorithm based segment routing for real-time and non-real-time service requests to allocate network resource to ensure the request of services. In this paper, we present available bandwidth allocation scheme based on SDN to provide Quality of Service (QoS) support for various services via experiments under our SDN testbed. Key Words-Quality of Service, Bandwidth Allocation, Software Defined Networking, Segment ...

In the last years, face and facial expression recognition have become a popular area of research ... more In the last years, face and facial expression recognition have become a popular area of research in computer vision. It is typically used in network security systems and access control systems but it is also useful in other multimedia information processing areas. Performance of the face verification system depends on many conditions. One of the most problematic is varying illumination condition. In this paper, we present new preprocessing approach to eliminate illumination effect from the human face images. The different stages include Median filtering, 2D Gabor filter is used for Edge enhancement and Histogram Equalization for Image contrast illumination. Hence an enhanced image is produced by using Hybrid filters this helps in an efficient way of preprocessing image for any further digital image processing work.

2020 International Conference on Advanced Information Technologies (ICAIT), 2020

The Software Defined Network (SDN) offers centralized traffic flow control from a global view of ... more The Software Defined Network (SDN) offers centralized traffic flow control from a global view of the underlying network, covering the route for SDN's various traffic engineering research. The continuous development of virtual machine migration, data backup, or MapReduce operations, the scale of the current network become more and more enormous. Equal-Cost Multi-Path (ECMP) is a method of multipath routing to distribute the flows using static hashing method. These flows do not reveal degrading due to static hashing for the resulting collisions. Traffic overloads due to long delay and low throughput will frequently cause connection congestion and result in less flow transmission performance. The paper focuses on the Flow based Traffic Scheduling FTS scheme that provides rerouting solution by scheduling the flow to the maximum available bandwidth path between the several paths. The experimental evaluations are used to show the efficiency of FTS in relation to ECMP and Hedera with reference to average throughput and flow completion time (FCT).

2019 International Conference on Advanced Information Technologies (ICAIT), 2019

In modern years, face recognition is becoming popular in many applications in different areas bei... more In modern years, face recognition is becoming popular in many applications in different areas being videoconferencing, security, banking, law requirement, and human-computer interaction. The performance of the face verification system depends on many challenges. The Most challenge in face recognition is illumination variations. In this approach, Median Filter, Gabor Filter, and Histogram Equalization are used to reduce the illumination effect of the face images as the preprocessing stage. The extraction of the Eigen faces features use Principal Component Analysis (PCA). After that, recognize face by using multiclass Support Vector Machines. The standard databases of ORL and Yale are used in the experiment. The results show that the system efficiently increased the accuracy of face recognition rate of the system, especially in various lighting situations.

2020 IEEE Conference on Computer Applications(ICCA), 2020

With the high volume of traffic in recent network, traffic between the switches has exchanged rap... more With the high volume of traffic in recent network, traffic between the switches has exchanged rapidly among the servers that may conduct the link congestion. In a traditional network, routing protocols are used the shortest path algorithm and congestion will occur in the network links. Equal-Cost Multi-Path (ECMP) algorithm cannot fairly distribute the bandwidth of the links between the flows. These flows suffer long queuing delays and degrades throughput. Flow Collision Avoidance algorithm (FCAA) is an efficient method to alleviate the network congestion by rerouting the flows to guarantee the performance of the network. Congestion occurs when flow demand exceeds link capacity and it leads to the degradation of QoS. The algorithm evaluates the existing flow’s demand based on port statistics to handle the collision of the large flow in the network avoiding the congestion. The evaluations indicate that SDN-based FCAA algorithm enforces the required end-to-end QoS for each traffic flow over ECMP.

Provisioning quality of service (QoS) is a big deal to deliver different applications over the cu... more Provisioning quality of service (QoS) is a big deal to deliver different applications over the current internet. With the advancements of using multimedia applications, the necessity of Quality of Service (QoS) is increasing rapidly. As real-time applications increase, Software Defined Network (SDN) has emerged as a wellestablished paradigm for next generations networks. By utilizing the characteristics of SDN, this paper proposes QoS provisioning based segment routing (SR) over SDN framework to find the feasible path according to the QoS requirements. This QoS provisioning architecture includes monitoring of link states among switches and providing of flow’s QoS requirements. This QoS provisioning is the use of the available bandwidth to react to the network traffic. The routing algorithm solves the problem of inefficient bandwidth. If there is no available bandwidth path, the controller will be decided depending on the flows by using the proposed algorithm. Simulation results are ...

Long Term Evolution (LTE) supports several QoS (quality of services) classes and tries to guarant... more Long Term Evolution (LTE) supports several QoS (quality of services) classes and tries to guarantee their requirements. There are two main different QoS classes: Guaranteed Bit rate (GBR) such as VoIP and non-Guaranteed Bit Rate (non-GBR) such as FTP or HTTP. Having these different QoS requirements in packet domain introduces an additional challenge on the LTE MAC scheduler design. Therefore, the scheduler has to be aware of the different service requirements and satisfy them. In this paper, QoS aware proportional fair downlink scheduler (QAPF) is proposed. It can optimize the use of available resources while maintaining QoS requirements of different service classes. The result will show the performance evaluation of the proposed scheduler by comparing with others.

2019 International Conference on Advanced Information Technologies (ICAIT)

With the rapid growth of current network, the demand of resources is growing significantly. Insuf... more With the rapid growth of current network, the demand of resources is growing significantly. Insufficient bandwidth results unstable throughput in a network. Software-defined network (SDN) has been proposed to provide optimal routing decision in the presence of congestion. Equal Cost Multi-path (ECMP) routing cannot guarantee optimal resource utilization. ECMP causes the long flow collision in network because it does not consider the network parameter such as bandwidth. In this paper, the Flow Path Computing Algorithm (FPCA) is proposed to minimize the network congestion by rerouting the flows in SDN. This algorithm mainly focuses on rerouting the traffic flows over the alternative path when the network congestion is detected. If the flow demand is exceeded 10% of link bandwidth, the algorithm computes the light loaded path based on the port statistics and it shifts from the congested path to light loaded path. Simulation results are presented to show the effectiveness of the FPCA algorithm over ECMP.

Proceedings of 10th International Conference on Science and Engineering 2019, (ICSE 2019), Dec 8, 2019

2020 International Conference on Advanced Information Technologies (ICAIT), 2020

In this paper, the facial expression recognition system is described based on Eigenface. Facial e... more In this paper, the facial expression recognition system is described based on Eigenface. Facial expression recognition has become increasingly common in several applications in the fields of video conferences, surveillance, financial services, health treatment, and access control system. The performance of recognition depends mainly on several challenges. Lighting variations are also the most challenging in the recognition process. In this system, illumination invariant techniques applied for decreasing the lighting affected faces as a preprocessing stage. Eigen faces are used as feature extraction step. K-NN and Multi-SVMs classifier are used to identify the facial expression and to test the performance of the proposed system. The standard dataset of JAFFE is used in the experiments and 84.29% accuracy is achieved for facial expression recognition. The accuracy of the framework is contrasted with the state of the art methods tested on the JAFFE database.

2018 18th International Symposium on Communications and Information Technologies (ISCIT), 2018

Communication networks are growing in size at an increasing rate with the traditional network. Qu... more Communication networks are growing in size at an increasing rate with the traditional network. Quality of Service (QoS) is needed for new applications and new services on the Internet. Software Defined Network (SDN) techniques are appeared to be more flexible and programmable network. By utilizing the characteristics of SDN, this paper proposes QoS sensitive routing for available bandwidth by using segment routing (SR) to find the best path according to the QoS requirements. This paper focuses on monitoring of flow entries among switches and finding the feasible path over QoS-based routing scheme. To provide network throughput to the end hosts, the available bandwidth has to be accomplished. The routing algorithm finds the path which is feasible to meet the desired QoS of data flows. If the required QoS cannot provide to the requested flow, the controller determines how to calculate depending on the request from the switch. If the initial path will not be able to achieve the available bandwidth, the algorithm reroutes the higher bandwidth flow. This paper suggests improvements in QoS using Open Network Operating System (ONOS) controller with OpenFlow protocol and the simulation experiments with effective QoS are suitable for the future internet using SDN.

Long Term Evolution (LTE) supports several QoS (quality of services) classes and tries to guarant... more Long Term Evolution (LTE) supports several QoS (quality of services) classes and tries to guarantee their requirements. There are two main different QoS classes: Guaranteed Bit rate (GBR) such as VoIP and non-Guaranteed Bit Rate (non-GBR) such as FTP or HTTP. Having these different QoS requirements in packet domain introduces an additional challenge on the LTE MAC scheduler design. Therefore, the scheduler has to be aware of the different service requirements and satisfy them. In this paper, QoS aware proportional fair downlink scheduler (QAPF) is proposed. It can optimize the use of available resources while maintaining QoS requirements of different service classes. The result will show the performance evaluation of the proposed scheduler by comparing with others. KeywordsLTE, QoS, GBR, nonGBR, QAPF

Face Detection is the first step of any automated face and facial expression recognition system. ... more Face Detection is the first step of any automated face and facial expression recognition system. It is applied in many applications such as computer vision and biometric applications like face recognition, expression recognition and classification, security access systems, video surveillance, and intelligent human computer interaction. A false detection rate and slow detection speed remain major problems in face detection. In this paper, we present a method to avoid these problems, by combining skin detection and Viola and Jones Face detection. Skin detection is used to reject false positive. After determining all possible face candidate regions, the Viola Jones detector is applied to detect face. Our method has high detection rate and accuracy than an individual face detector. It can perform quiet well in complex background and varying lighting conditions when it is applied images with complex backgrounds. Experimental results show the effectiveness of the proposed method. Keywords...

Nowadays, Internet is the main communication medium but traditional network approach has been dif... more Nowadays, Internet is the main communication medium but traditional network approach has been difficult to adapt to the fast growing Internet. The emergence of Software Defined Network (SDN) has brought to predict for the solution of this problem. Methods for measuring Quality of Service (QoS) parameters such as bandwidth utilization, bandwidth allocation and delay have been introduced not only for traditional network but also for SDN-based scenarios. Bandwidth allocation is a great significance in various areas of networking. We propose a low-complexity bandwidth allocation algorithm based segment routing for real-time and non-real-time service requests to allocate network resource to ensure the request of services. In this paper, we present available bandwidth allocation scheme based on SDN to provide Quality of Service (QoS) support for various services via experiments under our SDN testbed. Key Words-Quality of Service, Bandwidth Allocation, Software Defined Networking, Segment ...

In the last years, face and facial expression recognition have become a popular area of research ... more In the last years, face and facial expression recognition have become a popular area of research in computer vision. It is typically used in network security systems and access control systems but it is also useful in other multimedia information processing areas. Performance of the face verification system depends on many conditions. One of the most problematic is varying illumination condition. In this paper, we present new preprocessing approach to eliminate illumination effect from the human face images. The different stages include Median filtering, 2D Gabor filter is used for Edge enhancement and Histogram Equalization for Image contrast illumination. Hence an enhanced image is produced by using Hybrid filters this helps in an efficient way of preprocessing image for any further digital image processing work.

2020 International Conference on Advanced Information Technologies (ICAIT), 2020

The Software Defined Network (SDN) offers centralized traffic flow control from a global view of ... more The Software Defined Network (SDN) offers centralized traffic flow control from a global view of the underlying network, covering the route for SDN's various traffic engineering research. The continuous development of virtual machine migration, data backup, or MapReduce operations, the scale of the current network become more and more enormous. Equal-Cost Multi-Path (ECMP) is a method of multipath routing to distribute the flows using static hashing method. These flows do not reveal degrading due to static hashing for the resulting collisions. Traffic overloads due to long delay and low throughput will frequently cause connection congestion and result in less flow transmission performance. The paper focuses on the Flow based Traffic Scheduling FTS scheme that provides rerouting solution by scheduling the flow to the maximum available bandwidth path between the several paths. The experimental evaluations are used to show the efficiency of FTS in relation to ECMP and Hedera with reference to average throughput and flow completion time (FCT).

2019 International Conference on Advanced Information Technologies (ICAIT), 2019

In modern years, face recognition is becoming popular in many applications in different areas bei... more In modern years, face recognition is becoming popular in many applications in different areas being videoconferencing, security, banking, law requirement, and human-computer interaction. The performance of the face verification system depends on many challenges. The Most challenge in face recognition is illumination variations. In this approach, Median Filter, Gabor Filter, and Histogram Equalization are used to reduce the illumination effect of the face images as the preprocessing stage. The extraction of the Eigen faces features use Principal Component Analysis (PCA). After that, recognize face by using multiclass Support Vector Machines. The standard databases of ORL and Yale are used in the experiment. The results show that the system efficiently increased the accuracy of face recognition rate of the system, especially in various lighting situations.

2020 IEEE Conference on Computer Applications(ICCA), 2020

With the high volume of traffic in recent network, traffic between the switches has exchanged rap... more With the high volume of traffic in recent network, traffic between the switches has exchanged rapidly among the servers that may conduct the link congestion. In a traditional network, routing protocols are used the shortest path algorithm and congestion will occur in the network links. Equal-Cost Multi-Path (ECMP) algorithm cannot fairly distribute the bandwidth of the links between the flows. These flows suffer long queuing delays and degrades throughput. Flow Collision Avoidance algorithm (FCAA) is an efficient method to alleviate the network congestion by rerouting the flows to guarantee the performance of the network. Congestion occurs when flow demand exceeds link capacity and it leads to the degradation of QoS. The algorithm evaluates the existing flow’s demand based on port statistics to handle the collision of the large flow in the network avoiding the congestion. The evaluations indicate that SDN-based FCAA algorithm enforces the required end-to-end QoS for each traffic flow over ECMP.

Provisioning quality of service (QoS) is a big deal to deliver different applications over the cu... more Provisioning quality of service (QoS) is a big deal to deliver different applications over the current internet. With the advancements of using multimedia applications, the necessity of Quality of Service (QoS) is increasing rapidly. As real-time applications increase, Software Defined Network (SDN) has emerged as a wellestablished paradigm for next generations networks. By utilizing the characteristics of SDN, this paper proposes QoS provisioning based segment routing (SR) over SDN framework to find the feasible path according to the QoS requirements. This QoS provisioning architecture includes monitoring of link states among switches and providing of flow’s QoS requirements. This QoS provisioning is the use of the available bandwidth to react to the network traffic. The routing algorithm solves the problem of inefficient bandwidth. If there is no available bandwidth path, the controller will be decided depending on the flows by using the proposed algorithm. Simulation results are ...

Long Term Evolution (LTE) supports several QoS (quality of services) classes and tries to guarant... more Long Term Evolution (LTE) supports several QoS (quality of services) classes and tries to guarantee their requirements. There are two main different QoS classes: Guaranteed Bit rate (GBR) such as VoIP and non-Guaranteed Bit Rate (non-GBR) such as FTP or HTTP. Having these different QoS requirements in packet domain introduces an additional challenge on the LTE MAC scheduler design. Therefore, the scheduler has to be aware of the different service requirements and satisfy them. In this paper, QoS aware proportional fair downlink scheduler (QAPF) is proposed. It can optimize the use of available resources while maintaining QoS requirements of different service classes. The result will show the performance evaluation of the proposed scheduler by comparing with others.