Pub Date : 2020-11-09DOI: 10.1109/ISTT50966.2020.9279376
Karthiyayeni Govindasamy, A. Mahmud, Siva Priya Thiagarajah, A. Aziz, M. Roslee
We investigate new performance analysis of multi-channel Scalable Video Coding (SVC) in multi-frequency system using Area Expected Distortion (AED). We begin by outlining the obtained explicit expression of AED in multi-frequency system namely fractional frequency reuse (FFR), for single layer and two-layer SVC. We start with mathematical model for single-layer and two-layered SVC, where expected distortion of the SVC that depends on the trade-off between the probability of outage and quantization accuracy is derived. The novel performance measurement of AED is then used to measure the relationship between the average distortion rate, outage probability and channel allocation per area in multi-frequency systems. Analysis of AED for FFR shows that the two-layer SVC outperforms the single layer transmission with 60 percent lesser expected distortion at 20dB, and we deduced a system parameter for two layer SVC, that given minimum AED when the threshold distance is at 0.86km and 0.78km at 20dB and 0dB respectively. The analytical results are verified using the Monte Carlo simulations.
{"title":"On the Area Expected Distortion of Scalable Videos in Multi-Frequency System","authors":"Karthiyayeni Govindasamy, A. Mahmud, Siva Priya Thiagarajah, A. Aziz, M. Roslee","doi":"10.1109/ISTT50966.2020.9279376","DOIUrl":"https://doi.org/10.1109/ISTT50966.2020.9279376","url":null,"abstract":"We investigate new performance analysis of multi-channel Scalable Video Coding (SVC) in multi-frequency system using Area Expected Distortion (AED). We begin by outlining the obtained explicit expression of AED in multi-frequency system namely fractional frequency reuse (FFR), for single layer and two-layer SVC. We start with mathematical model for single-layer and two-layered SVC, where expected distortion of the SVC that depends on the trade-off between the probability of outage and quantization accuracy is derived. The novel performance measurement of AED is then used to measure the relationship between the average distortion rate, outage probability and channel allocation per area in multi-frequency systems. Analysis of AED for FFR shows that the two-layer SVC outperforms the single layer transmission with 60 percent lesser expected distortion at 20dB, and we deduced a system parameter for two layer SVC, that given minimum AED when the threshold distance is at 0.86km and 0.78km at 20dB and 0dB respectively. The analytical results are verified using the Monte Carlo simulations.","PeriodicalId":345344,"journal":{"name":"2020 IEEE 5th International Symposium on Telecommunication Technologies (ISTT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129570316","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-09DOI: 10.1109/ISTT50966.2020.9279379
S. Saleh, W. Ismail, I. Z. Abidin, M. Jamaluddin, M. Bataineh, Asem Alzoubi
Simple 2x1 and 3x1 UWB linear microstrip patch antenna arrays are analyzed and designed in this paper. The used feeding networks are based on UWB compact 2-way equal split and unequal split Tapper Transmission Lines (TTL) and Nonuniform Transmission Line (NTL) Wilkinson Power Divider (WPD) and UWB equal split 3-way TTL Bagley Power Divider (BPD). The simulated S11 is less than −10 dB for the designed five antenna arrays through the UWB frequency band. The obtained peak gain is between 4 and 6.5 dBi which can be improved using a high gain antenna. The simulation in this work is carried out uses High-Frequency Structure Simulator (HFSS) software based on the Finite Element Method (FEM).
{"title":"Simple UWB Linear Antenna Array Based On Compact Wilkinson And Bagley Power Dividers","authors":"S. Saleh, W. Ismail, I. Z. Abidin, M. Jamaluddin, M. Bataineh, Asem Alzoubi","doi":"10.1109/ISTT50966.2020.9279379","DOIUrl":"https://doi.org/10.1109/ISTT50966.2020.9279379","url":null,"abstract":"Simple 2x1 and 3x1 UWB linear microstrip patch antenna arrays are analyzed and designed in this paper. The used feeding networks are based on UWB compact 2-way equal split and unequal split Tapper Transmission Lines (TTL) and Nonuniform Transmission Line (NTL) Wilkinson Power Divider (WPD) and UWB equal split 3-way TTL Bagley Power Divider (BPD). The simulated S11 is less than −10 dB for the designed five antenna arrays through the UWB frequency band. The obtained peak gain is between 4 and 6.5 dBi which can be improved using a high gain antenna. The simulation in this work is carried out uses High-Frequency Structure Simulator (HFSS) software based on the Finite Element Method (FEM).","PeriodicalId":345344,"journal":{"name":"2020 IEEE 5th International Symposium on Telecommunication Technologies (ISTT)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133563646","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-09DOI: 10.1109/ISTT50966.2020.9279388
Nur Suraya binti Mandiar, N. A. Malik, N. A. A. Latiff, S. Yusof, S. Hamzah
Smart grid is a backbone for power supply system. Thus, an eco-friendly smart microgrid is developed which able to benefit the community. Smart microgrid increase integration of usage of renewable energy where it leans towards greener energy and healthier environment. The most common renewable energy source used in this smart microgrid is solar but since it is weather dependent sources, it will not be used during rainy weather. Therefore, rainwater harvesting system is implemented in this project by using micro turbine hydro-generator. A decision making algorithm is developed in this system using C++ so that it can effectively decide selected optimum power supply to the smart microgrid. Users also able to access their load usage data and power supplied data from application in their smartphones. This is by applying Internet of Things (IoT) communication where it is leading to a new level of lifestyle and environment. In this paper, a total eco-friendly condition is achieved in condition 1: hot weather at case 2: 2pm and condition 2: rainy weather at case 2: 2pm.
{"title":"Eco-Friendly Smart Renewable Microgrid System","authors":"Nur Suraya binti Mandiar, N. A. Malik, N. A. A. Latiff, S. Yusof, S. Hamzah","doi":"10.1109/ISTT50966.2020.9279388","DOIUrl":"https://doi.org/10.1109/ISTT50966.2020.9279388","url":null,"abstract":"Smart grid is a backbone for power supply system. Thus, an eco-friendly smart microgrid is developed which able to benefit the community. Smart microgrid increase integration of usage of renewable energy where it leans towards greener energy and healthier environment. The most common renewable energy source used in this smart microgrid is solar but since it is weather dependent sources, it will not be used during rainy weather. Therefore, rainwater harvesting system is implemented in this project by using micro turbine hydro-generator. A decision making algorithm is developed in this system using C++ so that it can effectively decide selected optimum power supply to the smart microgrid. Users also able to access their load usage data and power supplied data from application in their smartphones. This is by applying Internet of Things (IoT) communication where it is leading to a new level of lifestyle and environment. In this paper, a total eco-friendly condition is achieved in condition 1: hot weather at case 2: 2pm and condition 2: rainy weather at case 2: 2pm.","PeriodicalId":345344,"journal":{"name":"2020 IEEE 5th International Symposium on Telecommunication Technologies (ISTT)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129482953","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-09DOI: 10.1109/ISTT50966.2020.9279385
W. Maqbool, S. K. Syed-Yusof, N. A. A. Latiff, Bushra Naeem, Bilal Shabbir, N. N. N. A. Malik
In Cognitive Radio Wireless Mesh Networks (CR-WMNs) the end-to-end path selection is more challenging due to its heterogeneous nature. To utilize the network resources efficiently, a cognitive user must find the best available path from the source node to the end node. The factors which may affect the path in CR-WMNs are mobility, bandwidth, throughput and number of users. In this paper, all these parameters have been considered in fuzzy logic based mechanism to select an end-to-end optimal path. In this technique, the mentioned parameters of all available paths are mapped in a fuzzy logic system to go through fuzzy rules to find the optimal path. The results emphasized on improvement of the path analyzing process and path selection. The proposed algorithm is simulated in Matlab. The simulation results demonstrate the efficiency of the fuzzy system.
{"title":"Optimal End-to-End Path Selection Mechanism for CR-WMNs Based on Fuzzy Logic System","authors":"W. Maqbool, S. K. Syed-Yusof, N. A. A. Latiff, Bushra Naeem, Bilal Shabbir, N. N. N. A. Malik","doi":"10.1109/ISTT50966.2020.9279385","DOIUrl":"https://doi.org/10.1109/ISTT50966.2020.9279385","url":null,"abstract":"In Cognitive Radio Wireless Mesh Networks (CR-WMNs) the end-to-end path selection is more challenging due to its heterogeneous nature. To utilize the network resources efficiently, a cognitive user must find the best available path from the source node to the end node. The factors which may affect the path in CR-WMNs are mobility, bandwidth, throughput and number of users. In this paper, all these parameters have been considered in fuzzy logic based mechanism to select an end-to-end optimal path. In this technique, the mentioned parameters of all available paths are mapped in a fuzzy logic system to go through fuzzy rules to find the optimal path. The results emphasized on improvement of the path analyzing process and path selection. The proposed algorithm is simulated in Matlab. The simulation results demonstrate the efficiency of the fuzzy system.","PeriodicalId":345344,"journal":{"name":"2020 IEEE 5th International Symposium on Telecommunication Technologies (ISTT)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126445659","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-09DOI: 10.1109/ISTT50966.2020.9279368
Ivan Ku, Hung Lee Vei, Ayman A. El-Saleh, T. Le, M. Y. Alias
Multiuser multiple-input multiple-output transmission is identified as one of the promising techniques to vastly increase the transmission rate especially in densely deployed networks where inter-user interference is a major issue. To achieve this, channel state information (CSI) is needed to ensure accurate precoder design at the transmitter in order to effectively mitigate inter-user interference. However, the CSI is usually estimated at the receiver and, after quantization, is sent back to the transmitter. The feedback of quantized CSI incurs additional overhead which then leads to trade-off performances between transmission rate and energy efficiency. We will analyze this trade-off and propose the economy efficiency metric as a potential tool to determine the best operating point when the network is jointly optimized using transmission rate and energy efficiency.
{"title":"Trade-Off Performances in Multiuser MIMO Networks with Quantized CSI Feedback","authors":"Ivan Ku, Hung Lee Vei, Ayman A. El-Saleh, T. Le, M. Y. Alias","doi":"10.1109/ISTT50966.2020.9279368","DOIUrl":"https://doi.org/10.1109/ISTT50966.2020.9279368","url":null,"abstract":"Multiuser multiple-input multiple-output transmission is identified as one of the promising techniques to vastly increase the transmission rate especially in densely deployed networks where inter-user interference is a major issue. To achieve this, channel state information (CSI) is needed to ensure accurate precoder design at the transmitter in order to effectively mitigate inter-user interference. However, the CSI is usually estimated at the receiver and, after quantization, is sent back to the transmitter. The feedback of quantized CSI incurs additional overhead which then leads to trade-off performances between transmission rate and energy efficiency. We will analyze this trade-off and propose the economy efficiency metric as a potential tool to determine the best operating point when the network is jointly optimized using transmission rate and energy efficiency.","PeriodicalId":345344,"journal":{"name":"2020 IEEE 5th International Symposium on Telecommunication Technologies (ISTT)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131890649","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-09DOI: 10.1109/ISTT50966.2020.9279369
Arlind Billa, Ibraheem Shayea, A. Alhammadi, Qazwan Abdullah (غزوان عبد الله محمد طربوش), M. Roslee
Localization system is a process of monitoring any area or object using any known information gathered from different sources such as wired or wireless networks. Internet of Things (IoT) technology plays a more significant role, which directly affects our lives. Navigation IoT based system is becoming a very appealing research topic nowadays since it is used in various context-aware and localization-aware applications that cover many fields such as tracking, healthcare, or security. Moreover, this field has taken so much attention lately since there is continuous progress in sensing technologies and computing capabilities. This paper provides a brief review of the indoor localization technologies used for IoT-based navigation systems. This review gives fundamental guidelines for designing an intelligent indoor localization system.
{"title":"An Overview of Indoor Localization Technologies: Toward IoT Navigation Services","authors":"Arlind Billa, Ibraheem Shayea, A. Alhammadi, Qazwan Abdullah (غزوان عبد الله محمد طربوش), M. Roslee","doi":"10.1109/ISTT50966.2020.9279369","DOIUrl":"https://doi.org/10.1109/ISTT50966.2020.9279369","url":null,"abstract":"Localization system is a process of monitoring any area or object using any known information gathered from different sources such as wired or wireless networks. Internet of Things (IoT) technology plays a more significant role, which directly affects our lives. Navigation IoT based system is becoming a very appealing research topic nowadays since it is used in various context-aware and localization-aware applications that cover many fields such as tracking, healthcare, or security. Moreover, this field has taken so much attention lately since there is continuous progress in sensing technologies and computing capabilities. This paper provides a brief review of the indoor localization technologies used for IoT-based navigation systems. This review gives fundamental guidelines for designing an intelligent indoor localization system.","PeriodicalId":345344,"journal":{"name":"2020 IEEE 5th International Symposium on Telecommunication Technologies (ISTT)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116973681","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-09DOI: 10.1109/ISTT50966.2020.9279389
A. Omar, Mosab M. Qwakneh, Khaled A. AlHammami, Yanal S. Faouri
A metamaterial with frequency reconfigurability over a wide range of frequencies is presented in this paper. The proposed unit-cell has a circular geometrical shape with a triple split-ring resonator (SRR) and stepped-impedance resonator (SIR) technique applied to provide better miniaturization factor. The operating frequency is controlled by the presence/absence of a perfect conductor switches. Employing nine switches, the frequency is controlled over the range (1–7) GHz providing a various range of resonance frequencies, frequency bands, and different bandwidths. The suggested metamaterial is simulated using Ansys Electronics Desktop simulator (HFSS), and MATLAB software is used to analyze the characteristics of the unit-cell in terms of permittivity, permeability, and the refractive index. The reflection and transmission coefficients are extracted to observe the behavior of the metamaterial. The results show a negative near-zero refractive index at a certain frequency range. The permeability and permittivity curves exhibit a single negative (SNG) and a double negative (DNG) operating region.
{"title":"Triple Split Ring Resonators Metamaterial for Frequency Agility","authors":"A. Omar, Mosab M. Qwakneh, Khaled A. AlHammami, Yanal S. Faouri","doi":"10.1109/ISTT50966.2020.9279389","DOIUrl":"https://doi.org/10.1109/ISTT50966.2020.9279389","url":null,"abstract":"A metamaterial with frequency reconfigurability over a wide range of frequencies is presented in this paper. The proposed unit-cell has a circular geometrical shape with a triple split-ring resonator (SRR) and stepped-impedance resonator (SIR) technique applied to provide better miniaturization factor. The operating frequency is controlled by the presence/absence of a perfect conductor switches. Employing nine switches, the frequency is controlled over the range (1–7) GHz providing a various range of resonance frequencies, frequency bands, and different bandwidths. The suggested metamaterial is simulated using Ansys Electronics Desktop simulator (HFSS), and MATLAB software is used to analyze the characteristics of the unit-cell in terms of permittivity, permeability, and the refractive index. The reflection and transmission coefficients are extracted to observe the behavior of the metamaterial. The results show a negative near-zero refractive index at a certain frequency range. The permeability and permittivity curves exhibit a single negative (SNG) and a double negative (DNG) operating region.","PeriodicalId":345344,"journal":{"name":"2020 IEEE 5th International Symposium on Telecommunication Technologies (ISTT)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129942563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-09DOI: 10.1109/ISTT50966.2020.9279373
Shan Jaffry, S. F. Hasan
Autonomous network traffic prediction will be a key feature in beyond 5G networks. In the past, researchers have used statistical methods such as Auto Regressive Integrated Moving Average (ARIMA) to provide traffic prediction. However ARIMA based models fail to provide accurate predictions in highly dynamic cellular environment. Hence, researchers are exploring deep learning techniques such as Recurrent Neural Networks (RNN) and Long-Short-Term-Memory (LSTM) to develop autonomous cellular traffic prediction models.This paper proposes a LSTM based cellular traffic prediction model using real world call data record. We have compared the LSTM based prediction with ARIMA model and vanilla Feed-Forward Neural Network (FFNN). The results show that LSTM and FFNN can accurately predict cellular traffic. However, it has been found that LSTM models converged more quickly in terms of training the model for prediction.
{"title":"Cellular Traffic Prediction using Recurrent Neural Networks","authors":"Shan Jaffry, S. F. Hasan","doi":"10.1109/ISTT50966.2020.9279373","DOIUrl":"https://doi.org/10.1109/ISTT50966.2020.9279373","url":null,"abstract":"Autonomous network traffic prediction will be a key feature in beyond 5G networks. In the past, researchers have used statistical methods such as Auto Regressive Integrated Moving Average (ARIMA) to provide traffic prediction. However ARIMA based models fail to provide accurate predictions in highly dynamic cellular environment. Hence, researchers are exploring deep learning techniques such as Recurrent Neural Networks (RNN) and Long-Short-Term-Memory (LSTM) to develop autonomous cellular traffic prediction models.This paper proposes a LSTM based cellular traffic prediction model using real world call data record. We have compared the LSTM based prediction with ARIMA model and vanilla Feed-Forward Neural Network (FFNN). The results show that LSTM and FFNN can accurately predict cellular traffic. However, it has been found that LSTM models converged more quickly in terms of training the model for prediction.","PeriodicalId":345344,"journal":{"name":"2020 IEEE 5th International Symposium on Telecommunication Technologies (ISTT)","volume":"346 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116457986","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-09DOI: 10.1109/ISTT50966.2020.9279342
T. Murray, S. F. Hasan
Natural and man-made disasters often trap victims inside rubble, who must be rescued within tight time-constraints. A number of technologies are available to track the victims, each having their own benefits and trade offs. This paper presents a review of literature on the technologies that are used for victim localization, and highlights some of the problems associated with their use in post disaster environments.
{"title":"Present State of the Art in Post Disaster Victim Localization","authors":"T. Murray, S. F. Hasan","doi":"10.1109/ISTT50966.2020.9279342","DOIUrl":"https://doi.org/10.1109/ISTT50966.2020.9279342","url":null,"abstract":"Natural and man-made disasters often trap victims inside rubble, who must be rescued within tight time-constraints. A number of technologies are available to track the victims, each having their own benefits and trade offs. This paper presents a review of literature on the technologies that are used for victim localization, and highlights some of the problems associated with their use in post disaster environments.","PeriodicalId":345344,"journal":{"name":"2020 IEEE 5th International Symposium on Telecommunication Technologies (ISTT)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114694879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-09DOI: 10.1109/ISTT50966.2020.9279341
Siti Raihanah Abdani, M. A. Zulkifley, Muhammad Nazir Siham, Nurshafiza Zanal Abiddin, N. A. Aziz
One of the initiatives by the Malaysian government to reduce foreign dependency on staple food stock is by giving subsidies to the rice farmers. The amount received by them directly correlates with the cultivated paddy areas that include subsistence in fertilizers, seeds, and machinery. Hence, it is important for the Malaysian government to identify the exact areas that have been cultivated so that the subsidies will reach the targeted groups correctly. Currently, the surveying process is done manually by filed observer, which is a costly and tedious process. Hence, a remote sensing approach is proposed for an automated surveying system that semantically segments the satellite images of the paddy fields according to the intended class. A deep learning approach is adopted where a fully convolutional network with spatial pyramid pooling (SPP) module is designed to segment the images into four types of class, which are cultivated areas, uncultivated areas, backgrounds, and others. The encoder backbone of the network is based on VGG16, where the SPP module is comprised of four parallel branches of multiscale feature maps. The up-sample process is done through two layers of transposed convolution, where the output will be resized back according to the input image. The results show that the proposed network with SPP kernel set of 4x4, 5x5, 6x6, and 7x7 returns the best performance with a mean accuracy of 0.9869 and Jaccard index of 0.8326. The model faced its biggest training challenge when the clouds obstructed the surface information, which makes the areas uninformative. In the future, the network can be further improved by adding feed-forward layers and residual skip connections that help in reducing the zero gradient diminishing problem.
{"title":"Paddy Fields Segmentation using Fully Convolutional Network with Pyramid Pooling Module","authors":"Siti Raihanah Abdani, M. A. Zulkifley, Muhammad Nazir Siham, Nurshafiza Zanal Abiddin, N. A. Aziz","doi":"10.1109/ISTT50966.2020.9279341","DOIUrl":"https://doi.org/10.1109/ISTT50966.2020.9279341","url":null,"abstract":"One of the initiatives by the Malaysian government to reduce foreign dependency on staple food stock is by giving subsidies to the rice farmers. The amount received by them directly correlates with the cultivated paddy areas that include subsistence in fertilizers, seeds, and machinery. Hence, it is important for the Malaysian government to identify the exact areas that have been cultivated so that the subsidies will reach the targeted groups correctly. Currently, the surveying process is done manually by filed observer, which is a costly and tedious process. Hence, a remote sensing approach is proposed for an automated surveying system that semantically segments the satellite images of the paddy fields according to the intended class. A deep learning approach is adopted where a fully convolutional network with spatial pyramid pooling (SPP) module is designed to segment the images into four types of class, which are cultivated areas, uncultivated areas, backgrounds, and others. The encoder backbone of the network is based on VGG16, where the SPP module is comprised of four parallel branches of multiscale feature maps. The up-sample process is done through two layers of transposed convolution, where the output will be resized back according to the input image. The results show that the proposed network with SPP kernel set of 4x4, 5x5, 6x6, and 7x7 returns the best performance with a mean accuracy of 0.9869 and Jaccard index of 0.8326. The model faced its biggest training challenge when the clouds obstructed the surface information, which makes the areas uninformative. In the future, the network can be further improved by adding feed-forward layers and residual skip connections that help in reducing the zero gradient diminishing problem.","PeriodicalId":345344,"journal":{"name":"2020 IEEE 5th International Symposium on Telecommunication Technologies (ISTT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126523561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：