Pub Date : 2002-12-10DOI: 10.1109/NEUREL.2002.1057978
I. Reljin, B. Reljin
Assuming the shared buffer in ATM node with bursty input traffic, we have derived a new competitive neural network algorithm for buffer management. The algorithm is an extended version of the previous one, based on input prebuffers solution. The cell loss rate obtained is considerably lower comparing to the round-robin control in such a case. The interarrival times of output streams are considered by fractal/multifractal analysis, as well.
{"title":"Neural network control of shared ATM buffer","authors":"I. Reljin, B. Reljin","doi":"10.1109/NEUREL.2002.1057978","DOIUrl":"https://doi.org/10.1109/NEUREL.2002.1057978","url":null,"abstract":"Assuming the shared buffer in ATM node with bursty input traffic, we have derived a new competitive neural network algorithm for buffer management. The algorithm is an extended version of the previous one, based on input prebuffers solution. The cell loss rate obtained is considerably lower comparing to the round-robin control in such a case. The interarrival times of output streams are considered by fractal/multifractal analysis, as well.","PeriodicalId":347066,"journal":{"name":"6th Seminar on Neural Network Applications in Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128637705","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 : 2002-12-10DOI: 10.1109/NEUREL.2002.1057996
J. Antonijevic, J. Jovkovic
The ITU-R Recommendations are the international technical standards developed by the Radiocommunication Sector of the International Telecommunication Union (ITU). ITU-R is based on reading the diagrams defined in Recommendation ITU-R P.370-7. The main problem with ITU-R method is the extrapolation of the curves with the calculated h/sub e/ that differs from the values shown on the diagrams in Rec. ITU-R P.370-7. We examined a MLP neural model of the propagation curves and proved its accuracy on a few examples providing the best structure of the neural network.
{"title":"Neural model of the propagation curves from ITU-R P.370-7","authors":"J. Antonijevic, J. Jovkovic","doi":"10.1109/NEUREL.2002.1057996","DOIUrl":"https://doi.org/10.1109/NEUREL.2002.1057996","url":null,"abstract":"The ITU-R Recommendations are the international technical standards developed by the Radiocommunication Sector of the International Telecommunication Union (ITU). ITU-R is based on reading the diagrams defined in Recommendation ITU-R P.370-7. The main problem with ITU-R method is the extrapolation of the curves with the calculated h/sub e/ that differs from the values shown on the diagrams in Rec. ITU-R P.370-7. We examined a MLP neural model of the propagation curves and proved its accuracy on a few examples providing the best structure of the neural network.","PeriodicalId":347066,"journal":{"name":"6th Seminar on Neural Network Applications in Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129921357","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 : 2002-12-10DOI: 10.1109/NEUREL.2002.1057993
D. Raković
It is shown that any quantum system has a formal dynamical structure of Hopfield-like associative neural network. Besides, it is pointed out that investigations in the field of microwave resonance therapy of the acupuncture system, as well as research of the interactions of consciousness with microscopic and macroscopic environment, imply the existence of local and nonlocal macroscopic biophysical affects, with tremendous potential implications in the fields of medicine, psychology, biology, physics, engineering, and philosophy/religion.
{"title":"Hopfield-like quantum associative neural networks and (quantum) holistic psychosomatic implications","authors":"D. Raković","doi":"10.1109/NEUREL.2002.1057993","DOIUrl":"https://doi.org/10.1109/NEUREL.2002.1057993","url":null,"abstract":"It is shown that any quantum system has a formal dynamical structure of Hopfield-like associative neural network. Besides, it is pointed out that investigations in the field of microwave resonance therapy of the acupuncture system, as well as research of the interactions of consciousness with microscopic and macroscopic environment, imply the existence of local and nonlocal macroscopic biophysical affects, with tremendous potential implications in the fields of medicine, psychology, biology, physics, engineering, and philosophy/religion.","PeriodicalId":347066,"journal":{"name":"6th Seminar on Neural Network Applications in Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123376947","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 : 2002-12-10DOI: 10.1109/NEUREL.2002.1057959
D. Kopčanski, S. Odri, D. Petrovacki
This paper involves the evolution of neural networks based on parameters gained from a backpropagation training algorithm. The bridging of layers of neural networks is suggested as a new process in this evolutionary process, in addition to cell division and degeneration of synapses and cells. The process of bridging the layers was performed in three different ways in order to find an appropriate algorithm.
{"title":"Bridging of layers of neural networks","authors":"D. Kopčanski, S. Odri, D. Petrovacki","doi":"10.1109/NEUREL.2002.1057959","DOIUrl":"https://doi.org/10.1109/NEUREL.2002.1057959","url":null,"abstract":"This paper involves the evolution of neural networks based on parameters gained from a backpropagation training algorithm. The bridging of layers of neural networks is suggested as a new process in this evolutionary process, in addition to cell division and degeneration of synapses and cells. The process of bridging the layers was performed in three different ways in order to find an appropriate algorithm.","PeriodicalId":347066,"journal":{"name":"6th Seminar on Neural Network Applications in Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123112106","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 : 2002-12-10DOI: 10.1109/NEUREL.2002.1057973
A. Tsakoumis, S. Vladov, V. Mladenov
In this paper we consider daily load forecast problem and explore the idea that similar conditions to those at the forecasting moment have normally existed. before. If the load conditions change relatively slowly, then the yesterday's load curve can be used as an indicator of the load conditions of the present day; so it is assumed the robustness of the model. To test the idea of the robustness two models are considered. The first model uses the self-organizing map (SOM) to form network weights. The map is trained on the load data of ten months. The forecast is received by connecting load data of the previous day to a weight vector that contains a forecast for the target day. The second model that we suggest here is a considerable simplification of the first one and is based on the idea of the nearest neighbor.
{"title":"Daily load forecasting based on previous day load","authors":"A. Tsakoumis, S. Vladov, V. Mladenov","doi":"10.1109/NEUREL.2002.1057973","DOIUrl":"https://doi.org/10.1109/NEUREL.2002.1057973","url":null,"abstract":"In this paper we consider daily load forecast problem and explore the idea that similar conditions to those at the forecasting moment have normally existed. before. If the load conditions change relatively slowly, then the yesterday's load curve can be used as an indicator of the load conditions of the present day; so it is assumed the robustness of the model. To test the idea of the robustness two models are considered. The first model uses the self-organizing map (SOM) to form network weights. The map is trained on the load data of ten months. The forecast is received by connecting load data of the previous day to a weight vector that contains a forecast for the target day. The second model that we suggest here is a considerable simplification of the first one and is based on the idea of the nearest neighbor.","PeriodicalId":347066,"journal":{"name":"6th Seminar on Neural Network Applications in Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126436685","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 : 2002-12-10DOI: 10.1109/NEUREL.2002.1057980
V. Litovski, M. Andrejević
Artificial neural networks (ANNs) are applied for modeling of the nonlinear negative resistance being an element in the Chua's circuit. The properties of double hook attractor are presented. ANNs are used for application of the black-box modeling concept in the time domain. The ANNs topology, the testing signal used for excitation, together with the complexity of the ANN are considered. The generalization property of the neural networks is exploited to implement the model as an element of the double hook attractor.
{"title":"ANN application in modeling of Chua's circuit","authors":"V. Litovski, M. Andrejević","doi":"10.1109/NEUREL.2002.1057980","DOIUrl":"https://doi.org/10.1109/NEUREL.2002.1057980","url":null,"abstract":"Artificial neural networks (ANNs) are applied for modeling of the nonlinear negative resistance being an element in the Chua's circuit. The properties of double hook attractor are presented. ANNs are used for application of the black-box modeling concept in the time domain. The ANNs topology, the testing signal used for excitation, together with the complexity of the ANN are considered. The generalization property of the neural networks is exploited to implement the model as an element of the double hook attractor.","PeriodicalId":347066,"journal":{"name":"6th Seminar on Neural Network Applications in Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129128550","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 : 2002-12-10DOI: 10.1109/NEUREL.2002.1057988
P.B. Dejan, P. Mirjana
The neural prostheses (NPs) for grasping were developed to assist some daily living activities of hemiplegic subjects after a stroke. A NP that also controls the elbow joint movements could benefit even more to some hemiplegic subjects. NP users require an effective automatic control and practical command interface. The control that we developed is based on the following hypotheses: once the task and preferred strategy for movement are selected, then by using the voluntary (natural) control that drives the proximal segment (shoulder joint), the synergistic (artificial) control drives the distal segment (elbow joint). We confirmed in experiments that reproducible synergies between the shoulder and elbow joint movement exist. Here, we describe a method for determining synergies between joint movements while reaching by applying an inductive learning (IL) technique. This method relies on the hierarchical mutual information classifier algorithm. The synergy is a map obtained by IL between the flex ion/extension (F/E) angular velocities at the shoulder and elbow joints. As the two other shoulder joint rotations are independent from the F/E synergy; thus the results of this study are applicable to a general 3D movement.
{"title":"Control for neural prostheses: neural networks for determining biological synergies","authors":"P.B. Dejan, P. Mirjana","doi":"10.1109/NEUREL.2002.1057988","DOIUrl":"https://doi.org/10.1109/NEUREL.2002.1057988","url":null,"abstract":"The neural prostheses (NPs) for grasping were developed to assist some daily living activities of hemiplegic subjects after a stroke. A NP that also controls the elbow joint movements could benefit even more to some hemiplegic subjects. NP users require an effective automatic control and practical command interface. The control that we developed is based on the following hypotheses: once the task and preferred strategy for movement are selected, then by using the voluntary (natural) control that drives the proximal segment (shoulder joint), the synergistic (artificial) control drives the distal segment (elbow joint). We confirmed in experiments that reproducible synergies between the shoulder and elbow joint movement exist. Here, we describe a method for determining synergies between joint movements while reaching by applying an inductive learning (IL) technique. This method relies on the hierarchical mutual information classifier algorithm. The synergy is a map obtained by IL between the flex ion/extension (F/E) angular velocities at the shoulder and elbow joints. As the two other shoulder joint rotations are independent from the F/E synergy; thus the results of this study are applicable to a general 3D movement.","PeriodicalId":347066,"journal":{"name":"6th Seminar on Neural Network Applications in Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131343534","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 : 2002-12-10DOI: 10.1109/NEUREL.2002.1057963
V. Ilić
This paper discusses evolutionary autonomous agents controlled by neural networks. An hierarchical model of neural networks is suggested, and we also present ENAA software that simulates evolutionary training of agents, which move inside the arena and perform a given task.
{"title":"Evolutionary neuro autonomous agents","authors":"V. Ilić","doi":"10.1109/NEUREL.2002.1057963","DOIUrl":"https://doi.org/10.1109/NEUREL.2002.1057963","url":null,"abstract":"This paper discusses evolutionary autonomous agents controlled by neural networks. An hierarchical model of neural networks is suggested, and we also present ENAA software that simulates evolutionary training of agents, which move inside the arena and perform a given task.","PeriodicalId":347066,"journal":{"name":"6th Seminar on Neural Network Applications in Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127446709","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 : 2002-12-10DOI: 10.1109/NEUREL.2002.1057994
B. Milovanovic, V. Markovic, Z. Marinković, Z. Stankovic
This paper is an overview of the results in neural networks application in the microwave circuits modeling achieved within the Laboratory for Microwave Technique and Satellite Television at the Faculty of Electronic Engineering in Nis, Yugoslavia. Neural networks are applied in modeling either of passive or active structures. Modeling is performed using not only simple multilayer perceptron network but also advanced knowledge based neural network structures.
{"title":"Microwave circuits modeling using neural networks: overview of the results achieved at the faculty of electronic engineering in Nis","authors":"B. Milovanovic, V. Markovic, Z. Marinković, Z. Stankovic","doi":"10.1109/NEUREL.2002.1057994","DOIUrl":"https://doi.org/10.1109/NEUREL.2002.1057994","url":null,"abstract":"This paper is an overview of the results in neural networks application in the microwave circuits modeling achieved within the Laboratory for Microwave Technique and Satellite Television at the Faculty of Electronic Engineering in Nis, Yugoslavia. Neural networks are applied in modeling either of passive or active structures. Modeling is performed using not only simple multilayer perceptron network but also advanced knowledge based neural network structures.","PeriodicalId":347066,"journal":{"name":"6th Seminar on Neural Network Applications in Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124080130","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 : 2002-12-10DOI: 10.1109/NEUREL.2002.1057989
S. Micera, J. Carpaneto, P. Dario, M. Popovic
The feasibility of predicting elbow position from shoulder angular trajectories during pointing movements was analyzed. Aiming to achieve this result a hybrid strategy (composed of statistical and soft computing algorithms) was developed. Using a statistical procedure we first clustered the different trajectories and then a neuro-fuzzy system was trained for each group. The results show the feasibility of this approach in terms of mean errors in the prediction of the elbow velocity and position.
{"title":"Statistical and soft-computing techniques for the prediction of upper arm articular synergies","authors":"S. Micera, J. Carpaneto, P. Dario, M. Popovic","doi":"10.1109/NEUREL.2002.1057989","DOIUrl":"https://doi.org/10.1109/NEUREL.2002.1057989","url":null,"abstract":"The feasibility of predicting elbow position from shoulder angular trajectories during pointing movements was analyzed. Aiming to achieve this result a hybrid strategy (composed of statistical and soft computing algorithms) was developed. Using a statistical procedure we first clustered the different trajectories and then a neuro-fuzzy system was trained for each group. The results show the feasibility of this approach in terms of mean errors in the prediction of the elbow velocity and position.","PeriodicalId":347066,"journal":{"name":"6th Seminar on Neural Network Applications in Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123434185","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：