2021 29th Signal Processing and Communications Applications Conference (SIU)最新文献

{"title":"Performance Comparison of Target Tracking Filters in Underwater Multipath Environments","authors":"Ahmet Güneş, A. Güllü","doi":"10.1109/SIU53274.2021.9477906","DOIUrl":"https://doi.org/10.1109/SIU53274.2021.9477906","url":null,"abstract":"Using time-of-arrivals or time-difference-of-arrivals for the localization and tracking of underwater vehicles is a common approach. Although these two approaches are claimed to have similar performances, the analytical derivations oversee the biases in the delays. Another problem in the underwater tracking is to choose from multiple arrivals which occur due to underwater multipath effects. In this study, ray tracing is used for simulating the complex underwater environment. Nearest neighborhood and probabilistic data association filters are compared for timeofarrival and time-difference-of-arrival measurement models.","PeriodicalId":160693,"journal":{"name":"2021 29th Signal Processing and Communications Applications Conference (SIU)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127451970","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}

{"title":"Determination Of The Impact Distance Applied On Long Fuel Distribution Pipes Using Wireless Sensors And Lora Radio","authors":"Cem Sisman, Selim Sagir, I. Kaya","doi":"10.1109/SIU53274.2021.9478053","DOIUrl":"https://doi.org/10.1109/SIU53274.2021.9478053","url":null,"abstract":"This study is aimed to determine the source of a sound occurred on solid objects. Such that; It is aimed to detect the position of the sound caused by sudden impact on objects such as water or fuel pipes reaching long distances by using sensors placed at certain intervals on these objects. At the same time, separately synchronized LoRa radios are used for communication between the sensors and the processing center where the data is processed. While a significant part of the work is the positioning of the sound source in solid objects, another noteworthy part has been the work done for the synchronization of LoRa radios and the evaluation of the received signals together.","PeriodicalId":160693,"journal":{"name":"2021 29th Signal Processing and Communications Applications Conference (SIU)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127168210","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}

{"title":"Partially Observable Markov Model Based Multi-Armed Bandits Structured Radar Emission Management Algorithm","authors":"Nuri Baran Ayana, Duygu Acar İçer, Hasan İhsan Turhan, M. Demirekler","doi":"10.1109/SIU53274.2021.9478025","DOIUrl":"https://doi.org/10.1109/SIU53274.2021.9478025","url":null,"abstract":"In this study, an EMCON aimed scheduling is studied for a network of radars with overlapping coverage. The objective of scheduling is to minimize the probability of detection of the radar positions by the enemy. This probability increases due to the radiation of the radar. The method used in this study is adapted from [8]. The proposed algorithm uses Multi Armed Bandits (MAB) structure. For the computation of the Gittins index, a Partially Observable Markov Decision Process (POMPD) model is developed and solved. The proposed sensor management system is implemented in the MATLAB environment and tested on an example.","PeriodicalId":160693,"journal":{"name":"2021 29th Signal Processing and Communications Applications Conference (SIU)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132953488","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}

{"title":"Channel Estimation Using Optical MIMO OFDM Based LED Index Modulation","authors":"Furkan Batuhan Okumus, E. Panayirci","doi":"10.1109/SIU53274.2021.9478042","DOIUrl":"https://doi.org/10.1109/SIU53274.2021.9478042","url":null,"abstract":"In this paper, a new MIMO channel estimation technique is proposed for generalized LED index modulation (GLIM-OFDM) developed for multi-input multi-output (MIMO) orthogonal frequency division multiplexing (OFDM) based VLC systems. The mean square error (MSE) of the channel estimation algorithm, the bit error rate (BER) and the Cramer-Rao (CR) lower-bound of the system were obtained analytically and the performance of the algorithm according to the signal-to-noise ratio was investigated by computer simulations. From the obtained MSE, BER and CR curves it is concluded that the MSE performance of the proposed channel estimation algorithm is very high and the BER performance of the system obtained with the estimated channel information is very close to the BER performance in the case of perfect knowledge of the channel.","PeriodicalId":160693,"journal":{"name":"2021 29th Signal Processing and Communications Applications Conference (SIU)","volume":"685 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132183026","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}

{"title":"Clustering Surface EMG Measurements with Similar Timing Patterns","authors":"Duru Berfin Su, Bora Büyüksaraç, B. Tunç","doi":"10.1109/SIU53274.2021.9477771","DOIUrl":"https://doi.org/10.1109/SIU53274.2021.9477771","url":null,"abstract":"Gait analysis is a process that qualified health specialists investigate a subject's posture differences in an environment where this person walks. Gait analysis can be enhanced when it is recorded with the presence of an sEMG system. EMG is a diagnostic medical device that can demonstrate muscle activations and deactivations. The purpose of this study is to display how muscle activation patterns of a single person alter during his/her locomotive movements in gait. Even healthy individuals exhibit different muscle activation patterns in their gait data. When the subject of the study is a person who has disabilities or problems related to his/her musculoskeletal system, the mentioned fluctuations in the activation patterns are considered to be even higher. The presented method is to gather data through gait analysis and the measuring of muscle movements with the guidance of an sEMG. When these sEMG signals are collected and appropriate signal processing steps are implemented, a trained health professional can recognize the fluctuations of muscle activation patterns of individuals much more easily. The kinematics and Spatio-temporal data of gait is obtained through Xsens wireless 3D motion tracking technology system. Simultaneously, sEMG data is recorded to indicate muscle activations in every single gait. EMG signals are uploaded into the computer and necessary digital signal processing steps are carried out on the raw data recordings. With the help of kinematics data, every gait cycle is determined and EMG data is categorized according to these cycles. In order to normalize EMG signals with respect to maximum contractions, every signal gait is examined separately and the maximum contraction value of each gait cycle is calculated individually. The normalization step is completed by implementing these values to gait cycles. The normalization process is followed by setting a threshold value to understand when the muscle in the examination is activated or deactivated. 25% threshold setting is displayed in this study. The results of the clustering algorithm show 3 types of clusters created by MATLAB regarding how similar or dissimilar the activation patterns of the gait cycles in the uploaded data.","PeriodicalId":160693,"journal":{"name":"2021 29th Signal Processing and Communications Applications Conference (SIU)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130971701","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}

{"title":"Siu 2021 [Front cover]","authors":"","doi":"10.1109/siu53274.2021.9477793","DOIUrl":"https://doi.org/10.1109/siu53274.2021.9477793","url":null,"abstract":"","PeriodicalId":160693,"journal":{"name":"2021 29th Signal Processing and Communications Applications Conference (SIU)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130234619","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}

{"title":"Neural Network Based Real Time Evaluation for Portable Apnea Device with Three Channels","authors":"Onur Peker, A. Yilmaz","doi":"10.1109/SIU53274.2021.9477854","DOIUrl":"https://doi.org/10.1109/SIU53274.2021.9477854","url":null,"abstract":"In this paper, the results of processing airflow area, oxygen saturation, heart rate variability and Neural Network based real time evaluation for a previously designed portable apnea device have been demonstrated. This results has been presented by a specific interface of “Artificial Neural Networks Based Sleep Apnea Detection and Analysis” and network performances were presented in comparative manner. Apart from having real time apnea detection, desired artificial neural networks can be generated via apnea user interface by using feature extracted signals in accordance with the selected activation function and training with selected target vector. The attributes obtained from each signal were evaluated separately or together and the detection characteristics of the models were examined. The Time Delayed Neural Networks (TDNNs) with hybrid data performs better than the other networks trained with the set obtained from single channel. It is easy to perform and compare networks with extracting new features and training new networks by this dedicated interface.","PeriodicalId":160693,"journal":{"name":"2021 29th Signal Processing and Communications Applications Conference (SIU)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126299048","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}

{"title":"A Comparative Study on Different Labelling Schemes and Cross-Corpus Experiments in Speech Emotion Recognition","authors":"P. Baki, Berna Erden, Serkan öncül","doi":"10.1109/SIU53274.2021.9477924","DOIUrl":"https://doi.org/10.1109/SIU53274.2021.9477924","url":null,"abstract":"Performance of the speech emotion recognition systems depends on many factors such as quality of the speech data, environment, cultural differences, language, emotion categorization scheme, etc. In this work, we create a baseline speech emotion recognition model based on convolutional neural networks using the RAVDESS dataset. First, we compare the performance of the model with different labeling schemes. Then, we perform cross-corpus experiments on datasets recorded in different languages. The results show that emotion groups with common arousal or valence categories are often confused and using multiple corpora in training improves the generalization capacity of the model.","PeriodicalId":160693,"journal":{"name":"2021 29th Signal Processing and Communications Applications Conference (SIU)","volume":"323 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115869125","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}

{"title":"An EEG and Machine Learning based Method for the Detection of Major Depressive Disorder","authors":"Elif Izci, Mehmet Akif Özdemir, A. Akan, Mehmet Akif Özçoban, M. K. Arıkan","doi":"10.1109/SIU53274.2021.9477800","DOIUrl":"https://doi.org/10.1109/SIU53274.2021.9477800","url":null,"abstract":"Major depressive disorder (MDD) is a common mood disorder encountered worldwide. Early diagnosis has great importance to prevent the negative effects on the person. The aim of this study is to develop an objective method to differentiate MDD patients from healthy controls. Electroencephalography (EEG) signals taken from 16 MDD patients and 16 healthy subjects are analyzed according to the regions of the brain, and time-domain, frequency-domain, and nonlinear features were extracted. The feature sets are classified using five different classification algorithms. As a result of the study, a classification accuracy of 89.5% was yielded using the Bagging classifier with 7 features calculated from the central EEG channels.","PeriodicalId":160693,"journal":{"name":"2021 29th Signal Processing and Communications Applications Conference (SIU)","volume":"138 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121711490","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}

{"title":"Semi-Supervised Segmentation of Multi-vendor and Multi-center Cardiac MRI","authors":"Mahyar Bolhassani, Ilkay Öksüz","doi":"10.1109/SIU53274.2021.9477818","DOIUrl":"https://doi.org/10.1109/SIU53274.2021.9477818","url":null,"abstract":"Automatic segmentation of the heart cavity is an essential task for the diagnosis of cardiac diseases. In this paper, we propose a semi-supervised segmentation setup for leveraging unlabeled data to segment Left-ventricle, Right-ventricle, and Myocardium. We utilize an enhanced version of residual U-Net architecture on a large-scale cardiac MRI dataset. Handling the class imbalanced data issue using dice loss, the enhanced supervised model is able to achieve better dice scores in comparison with a vanilla U-Net model. We applied several augmentation techniques including histogram matching to increase the performance of our model in other domains. Also, we introduce a simple but efficient semi-supervised segmentation method to improve segmentation results without the need for large labeled data. Finally, we applied our method on two benchmark datasets, STACOM2018, and M&Ms 2020 challenges, to show the potency of the proposed model. The effectiveness of our proposed model is demonstrated by the quantitative results. The model achieves average dice scores of 0.921, 0.926, and 0.891 for Left-ventricle, Right-ventricle, and Myocardium respectively.","PeriodicalId":160693,"journal":{"name":"2021 29th Signal Processing and Communications Applications Conference (SIU)","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123879715","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}