توجه: محتویات این صفحه به صورت خودکار پردازش شده و مقاله‌های نویسندگانی با تشابه اسمی، همگی در بخش یکسان نمایش داده می‌شوند.
۱Filtering and Compression of Mel Sub–Band Energies for Speech Recognition in Noise
اطلاعات انتشار: پانزدهیمن کنفرانس مهندسی برق ایران، سال
تعداد صفحات: ۶
The Mel–frequency cepstral cofficients (MFCC) are commonly used in speech recognition systems. Bu they are high sensitive to presence of
external noise. In this paper, we propose a noise compensation method for Mel filter bank energies and so MFCC features. This compensation method is performed in two stages: Mel sub–band filtering and then compression of Mel–sub–band energies. In the compression step, we propose a sub–bqnd SNRdependent compression function. We use this function in place of logarithm function in conventional MFCC feature extraction in presence of additive noise. Results show that the proposed nethod significantly improves MFCC features performance in noisy conditions where it decreases average word error rate up to 3094 for isolated word recognition on three test sets of Aurora 2 database.<\div>

۲A New Similarity Difference Measure in Multi Agent Systems
نویسنده(ها): ، ،
اطلاعات انتشار: چهاردهمین کنفرانس بین المللی سالانه انجمن کامپیوتر ایران، سال
تعداد صفحات: ۶
In this paper, we present a new measure for evaluating similarity changes in a multi agent system. The similarity measure of the agents changes during the learning process. The similarity differences are because of any composition or decomposition of some agent sets. The presented measure, defines the changes of homogeneity of agents by composition and decomposition. The utility of the metrics is demonstrated in the experimental evaluation of multi agent foraging. The results show that while the similarity difference gets a positive value, the performance grow rapidly.<\div>

۳A New Wavelet Thresholding Method for Speech Enhancement Based on Symmetric Kullback–Leibler Divergence
نویسنده(ها): ، ،
اطلاعات انتشار: چهاردهمین کنفرانس بین المللی سالانه انجمن کامپیوتر ایران، سال
تعداد صفحات: ۶
Performance of wavelet thresholding methods for speech enhancement is dependent on estimating an exact threshold value in the wavelet sub–bands. In thispaper, we propose a new method for more exact estimating the threshold value. We proposed to determine the threshold value based on the symmetric Kullback–Leibler divergence between the probability distributions of noisy speech and noise wavelet coefficients. In the next step, we improved this value using segmental SNR. We used some of TIMIT utterances to assess the performance of the proposed threshold. The algorithm is evaluated using the PESQ score and the SNR improvement. In average, we obtain 2db SNR improvement and a PESQ score increase up to 0.7 in comparison to the conventional wavelet thresholding approaches.<\div>

۴Application of HMM Adaptation and Robust Features to Sub–Band Speech Recognition in Noise
اطلاعات انتشار: یازدهمین کنفرانس سالانه انجمن کامپیوتر ایران، سال
تعداد صفحات: ۵
In recent years, sub–band speech recognition has been found useful in robust speech recognition, especially for speech signals contaminated by band–limited noise. In sub–band speech recognition, full band speech is divided into several frequency sub–bands. Sub–band feature vectors or their generated likelihoods by corresponding sub–band recognizers are combined to give the result of recognition task. In this paper, we concatenate sub–band feature vectors, where we extract phase autocorrelation (PAC) MFCC and one type of group delay based MFCC, called MFPSCC, as noise robust features from each subband. Furthermore, we used a model adaptation method, named weighted projection measure (WPM), to adapt HMM Gaussian mean vectors to concatenated sub–band feature vectors in noisy conditions. The experimental results indicate that the proposed methods significantly improve the sub–band speech recognition system performance in presence of additive noise<\div>

۵Background noise control for speech enhancement applications
اطلاعات انتشار: یازدهمین کنفرانس سالانه انجمن کامپیوتر ایران، سال
تعداد صفحات: ۴
Speech enhancement systems need to have acceptable residual noise and acceptable speech distortion. Some of frequency based noise reduction systems use a spectral modification filter to modify the noisy speech spectrum and remove the background noise. In these methods distortion and musical noise are two well–known annoying effects. In this paper we propose to shape the residual noise in order to decrease annoying musical residual noise. In addition we propose to control the level of residual noise and to shape it. We show that by controlling the residual noise level we can achieve the enhanced speech signal with less distortion and less musical noise.<\div>

۶Robust speech recognition using spectral subtraction and temporal structure normalization
نویسنده(ها): ، ،
اطلاعات انتشار: نوزدهمین کنفرانس مهندسی برق ایران، سال
تعداد صفحات: ۵
Filtering approaches in spectral domain and features domain have been shown their effectiveness for robust speech recognition. In this paper, we propose a two step filtering method. In the first step, spectral subtraction filter is applied to speech spectrum. In the second step, we design a temporal structure normalization filter in order to apply to features extracted from the filtered spectrum. Our results on Aurora2 show that the proposed method has higher recognition rate than both of spectral subtraction and temporal filtering methods<\div>

۷Class Dependent Feature Transformation for Intrusion Detection Systems
اطلاعات انتشار: نوزدهمین کنفرانس مهندسی برق ایران، سال
تعداد صفحات: ۶
Most of intrusion detection systems use primary and raw input features which are extracted from network connection without any preprocessing on the extracted features. In this paper, we propose a new feature transformation method based on class dependent approach for improving the accuracy of intrusion detection systems. In usual class dependent feature transformation methods the mapping process is accomplished using different mapping matrices for different classes of thedataset. In these methods, there is a difference between the train and test phases. In the training phase of class–dependent methods, samples of each class is mapped only using the corresponding matrix, however, in the test phase, each sampleis mapped using all of the transformed matrices. This may lead to some mistakes in classification. In this paper we modify the train and test phases on class dependent methods and propose a new linear feature transformation method. Unlike the usual class–dependent methods, the training and test phases of the proposed method are very similar. This similarity aids the classifier to learn more about dataset samples and transformation process. The performance of our proposed method is evaluated using three different indices, namely mutual information, maximum relevancy minimum redundancy criteria, and classification accuracy. The proposed method was evaluated on a benchmark intrusion detection dataset (NSLKDD dataset). The experimental results demonstrate that applying the proposed feature transformation method leads to higher classification accuracy and makes the IDS more capable of distinguishing intruders from normal users<\div>

۸Discriminative Transformations of Speech Features Based on Minimum Classification Error
اطلاعات انتشار: نوزدهمین کنفرانس مهندسی برق ایران، سال
تعداد صفحات: ۵
Feature extraction is an important step in pattern classification and speech recognition. Extracted features should discriminate classes from each other while being robust to the environmental conditions such as noise. For this purpose, some transformations are applied to features. In this paper, we propose a framework to improve independent feature transformations such as PCA (Principal Component Analysis), and HLDA (Heteroscedastic LDA) using the minimum classification error criterion. In this method, we modify full transformation matrices such that classification error is minimized for mapped features. We do not reduce feature vector dimension in this mapping. The proposed methods are evaluated for continuous phoneme recognition on clean and noisy TIMIT. Experimental results show that our proposed methods improve performance of PCA, and HLDA transformation for MFCC in both clean and noisy conditions.<\div>

۹Two–Microphone Speech Enhancement Using a Learned Binary Mask
اطلاعات انتشار: بیستمین کنفرانس مهندسی برق ایران، سال
تعداد صفحات: ۴
Ideal binary mask speech enhancement is shown to increase the speech quality as well as speech intelligibility. But, this property depends highly on the accurate separation ofspeech and masker time–frequency units of the input spectrum, which is a difficult task in real situations. Ordinary binary maskmethods are single–microphone methods and so, can obtain little information from the environment. In this paper, we devise a two–microphone method that uses a classifier to distinguishspeech–dominated and masker–dominated time–frequency units. The classifier uses simply computable two–microphone featureswhich enable it to be used in real–time scenarios. These proposed features empower the classifier to reach toclassification accuracies near 80%. This high accuracy in turn, empowers the Ideal binary mask mthod to obtain higher SNRI and NPLR values in comparison to state–of–the–art noisereduction methods. These results indicate that the proposed two–microphone features have high information content for speech\masker separation.<\div>

۱۰A Classifier Combination Approach for Farsi Accents Recognition
نویسنده(ها): ، ،
اطلاعات انتشار: بیستمین کنفرانس مهندسی برق ایران، سال
تعداد صفحات: ۵
Accent classification technologies directly influence the performance of automatic speech recognition (ASR) systems. In this paper, we evaluate three accent classificationapproaches: Phone Recognition followed by Language Modeling (PRLM) as a phonotactic approach; accent modeling using Gaussian Mixture Models (GMM) then selecting the mostsimilar model using Maximum Likelihood algorithm that is categorized in acoustic approaches a novel classifiercombination method which is proposed to improve the performance of accent classification for several regional accents. In the proposed approach, we use an ensemble methodin which each base classifier is a binary classifier that separates an accent from another one. We use the majority votealgorithm to combine the base classifiers. Results for five accents selected from FARSDAT speech database show that the proposed ensemble method outperforms PRLM and GMMbased approaches in the case of Farsi regional accent classifications.<\div>

۱۱Adaptive fuzzy controller design to improve the performance of ABS in the presence of suspension system effects
اطلاعات انتشار: اولین همایش ملی مهندسی برق ایران، سال
تعداد صفحات: ۸
In this paper, an adaptive fuzzy controller is synthesized to improve the anti–lock braking system (ABS) performance in the presence of suspension systemeffects. The ABS performance in generating the maximum braking force is mainly affected by the tire normal force applied the suspension system. In thisstudy, an optimization is carried out to obtain the optimal value of longitudinal slip reference value corresponding to the suspension normal force andvehicle speed. The slip optimal value is then imported as the reference input of the fuzzy controller. In this manner, the fuzzy controller is adapted to the tirenormal force, suspension deflection and vehicle speed. Therefore, this controller is aimed at tracking the slip optimal value which is updated by on–line data from suspension system and braking system states. In this study, two quarter car models are used to simulate the ABS and suspension systems behavior.The first being a quarter car suspension model with two degrees of freedom which describes the sprung and unsprung mass vertical dynamic and the latter a quarter car braking model with two degrees of freedom which describe the braking dynamics (vehicle longitudinal dynamic). Simulation of this study is derived during hard braking on a poor quality white noise road. The stopping distance is compared in the cases of fuzzy and adaptive fuzzy ABS control situations while a passive suspension is employed. The results show that the adaptive fuzzy control strategy can be helpful to increase the ABS controller performance so that it decreases the stopping distance in the semi–real situation.<\div>

۱۲Beri and Herfindahl indices of dietary diversity; can they will be used for measuring food security in sanction situation
اطلاعات انتشار: دومین کنفرانس بین المللی اقتصاد در شرایط تحریم، سال
تعداد صفحات: ۶
Iranian economy is under sanction since 35 years back. In a sanction situation monitoring food security condition is one of the most important issues. One of the methods to measure food security in a community is application diet diversity indices. However, one of the most important questions in this regard is that which index is suitable one for the purpose of measurement and whether these measures will produce different results? To achieve research objectives in this research work an endeavor is made to measure food security situation in rural areas of Gorgan, Golestan province of Iran with the help of Beri and Herfindahl Indices and compare results obtained with application of these two indices. Results of both indices showed that 152 households consumed diversified diet, 114 households consumed relatively diversified diet and 1 household consumed a non– diversified diet. Therefore we can conclude that while application of these two indices produced different mathematical results (values of indices) the end results and interpretation is similar thus both of them can be used as correct and reliable indices of dietary diversity.<\div>

۱۳Diet Diversity in sanction situation; Case study of rural areas of Gorgan
اطلاعات انتشار: دومین کنفرانس بین المللی اقتصاد در شرایط تحریم، سال
تعداد صفحات: ۶
Since 35 years back Iranian economy is experiencing different type of sanctions, in recent years the severity of sanctions increased, as for this fact, it is important to study food security situation under sanction condition in local and micro communities. Food security is defined as situation in which every member of a community has physical as well as economic access to adequate and healthy food for an active life. Diet diversity is commonly used as an index of food security. In this research work an attempt is made to gauge food security situation in rural areas of Gorgan. For the purpose of this research Beri index is used as an index of diversity. Results of cluster analysis showed that 152 households are classified in first cluster with average Beri index of 0.74 which means these 152 household consumed a diversified diet. Cluster analysis further classified 114 households with average Beri index of 0.64 in cluster 2 which means a relatively diversified diet and only one household with Beri index of 0.36 is classified in cluster three which is non–diversified diet.<\div>

۱۴Effect of PEG molecular weight and coagulation temperature on amoxicillin separation by polysulfone nanofiltration membrane
اطلاعات انتشار: هفتمین کنگره ملی مهندسی شیمی، سال
تعداد صفحات: ۸
Today, hundreds of antobiotics are produced in huge quantities for healthcare and diseases therapy. Besides to their economic value, entrance of these substances to the environment can change the organisms proceeds, leading to environmental pollution and ecosystem damages. In this study, flat sheet Polysulfone asymmetric membranes were prepared using immersion precipitation method. Raw ultrafiltration membranes were synthesized using wide range of Polyethylene glycol (PEG) molecular weights at different coagulation bath temperatures (CBTs). The membranes were modified using UV–assisted polymerization technique and their performance in the separation of amoxicillin (AMX), as the model pharmaceutical, were studied.The results showed that the membranes made with higher molecular weight of PEG were more susceptible for modification, so while they had higher flux, also showed relatively high amoxicillin retention. Moreover membranes made in higher coagulation temperatures had better separation performance. The best performance were obtained from the membrane made of high molecular weight of PEG (PEG–6000), high bath temperature (60 ° C). In this paper, a fairly comprehensive discussion of these effects on membrane structure and function was represented.<\div>

۱۵A Holonic Multi Agent System For Operating Smart Grid Market
نویسنده(ها): ،
اطلاعات انتشار: چهارمین کنفرانس بین المللی رویکردهای نوین در نگهداشت انرژی، سال
تعداد صفحات: ۹
Smart Grid is known as the next generation of power systems. This technology is the main solution to challenges such as increasing electric demand, aging utility infrastructure and workforce, and the environmental impact of greenhouse gases produced during electric generation. The agent based functions are highly autonomous and operate independently while managing energy consumption and production by human is autonomous and independently in power systems. The smart grid consists of several operation layers and in each layer there are several operating segments that must be coordinated with each other. These operating segments can map to an organization to reduce the complexity of coordination between them. A holonic multi agent system presents hierarchical organization structure with decentralized control. This paper proposes a model of holonic multi agent system (HMAS) for smart grid market to manage electricity agents in Smart grid operation. This approach controls the smart grid complexity by intelligent agents in a holonic organization mounted on smart grid infrastructure. The presented approach utilizes the advantages of using the HMAS technology for managing smart grid energy market and trading strategies to obtain satisfying energy exchange and energy price between the production units and loads in the smart grid.<\div>

۱۶Estimation of QoE by QoS parameters in Multimedia services over IP Networks
اطلاعات انتشار: کنفرانس بین المللی پژوهش های کاربردی در فناوری اطلاعات، کامپیوتر و مخابرات، سال
تعداد صفحات: ۸
Satisfying the end users is the important goal of service providers recently due to the growing use of video and television–based services provided through next generation IP networks, such as IPTV, OTT–TV, Video on Demand, Web TV, Time Shifted TV, Catch up TV and Video Conference; and competition in this area, therefore Quality of experience (QoE) can be considered The most suitable criterion for this purpose.In this article we investigate the relationship between QoE and QoS parameters in IPTV systems thus we can estimate QoE by using Quality of service parameters. Proposed method for estimating QoE is a no reference model and has low processing time so it can estimate the quality of experience in real–time. We discuss in details the proposed model and implementation of our testbed and finally we evaluate this model. Our evaluation shows that the accuracy of this model is acceptable for our context.<\div>

۱۷Nano silica chromic acid\wet SiO2 and NaNO2 as an efficient reagent system for synthesis of azo dyes based on 1–naphthol at room temperature and solvent–free conditions
اطلاعات انتشار: Iranian Journal of Catalysis، اول،شماره۲، ۲۰۱۱، سال
تعداد صفحات: ۴
A convenient, rapid, and one–pot method for the synthesis of azo dyes has been developed. In this protocol, diazotization reagent (ArN2+ –CrO3–SiO2) was prepared via grinding of aromatic amines, NaNO2, wet SiO2 and nano silica chromic acid (Nano–SCA) without solvent at room temperature. The obtained diazotization reagent was sufficiently stable to be kept at room temperature in the dry state for long time. Azo dyes were prepared by coupling of ArN2+ –CrO3–SiO2 with 1–naphthoxide in good to excellent yields. Mild and heterogeneous reaction conditions, high stability of diazonium salt, easy procedure, short reaction time and high yields are some important advantages of this protocol.

۱۸Preparation of Novel Thin–Film Composite Nanofiltration Membranes for Separation of Amoxicillin
اطلاعات انتشار: Journal of Nano Structures، چهارم،شماره۲، ۲۰۱۴، سال
تعداد صفحات: ۱۲
Several novel composite membranes were prepared to separate and recycle amoxicillin from pharmaceutical wastewater via nanofiltration process. The synthesis of these membranes included three stages: 1– preparation of polysulfone ultrafiltration membranes as a support via phase separation process, 2– modification of its surface by interfacial polymerization as a selective layer (polyamide), and 3– self–assembly of TiO2 nanoparticles on the selective layer as an anti–fouling agent. The rejection of all nanofiltration membranes was more than 99% and only its flux was changed proportional to different conditions. In the presence and absence of TiO2 nanoparticles, the pure water flux of polyamide thin–film membrane also obtained 44.4 and 38.4 L\h.m2 at 4 bar pressure, respectively. These were equal to 34 L\h.m2 for amoxicillin solutions. The results showed that TiO2 nanoparticles increased hydrophilicity of polyamide selective layer and therefore,nanoparticlesdecreasedthefoulinglevel.SEMimages illustrated the excellent establishment of polyamide layer and distribution of TiO2nanoparticles on the selective layer. The properties of membrane surface were taken into consideration by using AFM, indicating the increment of surface roughnesswith interfacial polymerization and TiO2 nanoparticles self–assembly. The pore sizeofmembranes was in the nanoscale (2.653 and 2.604 nm without and with TiO2nanoparticles self–assembly, respectively).

۱۹Detecting Flood–based Attacks against SIP Proxy Servers and Clients using Engineered Feature Sets
نویسنده(ها): ، ،
اطلاعات انتشار: International Journal Information and Communication Technology Research، هشتم،شماره۱، Winter ۲۰۱۶، سال
تعداد صفحات: ۹
Session Initiation Protocol (SIP) is the main signaling protocol of the next generation networks. The security issues of SIP–based entities (i.e. proxy servers and clients) have a direct impact on the perceived quality of experience of end users in multimedia sessions. In this paper, our focus is on the SIP flooding attacks including denial of service and distributed denial of service attacks. After classifying various types of SIP attacks based on their sources, we extract four feature sets based on the specification of its attack group, as well as the normal behavior of the SIP state machine specified in RFC 3261. We then minimize the number of derived features in each set to reduce the computational complexity of our proposed approach. This facilitates employing the engineered feature sets in embedded SIP–based devices such as cell phones and smart TVs. We evaluate the performance of the proposed feature sets in detecting SIP attack sequence. For this, we design and implement a real test–bed for SIP–based services to generate normal and attack traffics. The experimental results confirm that the engineered feature sets perform well in terms of detection accuracy and false alarm rates in classifying benign and anomaly traffic in various attack scenarios.

۲۰Language Model Adaptation Using Dirichlet Class Language Model Based on Part–of–Speech
نویسنده(ها): ، ،
اطلاعات انتشار: Journal of Information Systems and Telecommunication، دوم،شماره۱(پياپي ۵)، Jan-Mar ۲۰۱۴، سال
تعداد صفحات: ۶
Language modeling has many applications in a large variety of domains. Performance of this model depends on its adaptation to a particular style of data. Accordingly, adaptation methods endeavour to apply syntactic and semantic characteristics of the language for language modeling. The previous adaptation methods such as family of Dirichlet class language model (DCLM) extract class of history words. These methods due to lake of syntactic information are not suitable for high morphology languages such as Farsi. In this paper, we present an idea for using syntactic information such as part–of–speech (POS) in DCLM for combining with one of the language models of n–gram family. In our work, word clustering is based on POS of previous words and history words in DCLM. The performance of language models are evaluated on BijanKhan corpus using a hidden Markov model based ASR system. The results show that use of POS information along with history words and class of history words improves performance of language model, and decreases the perplexity on our corpus. Exploiting POS information along with DCLM, the word error rate of the ASR system decreases by 1.2% compared to DCLM.

۲۱SIP Vulnerability Scan Framework
نویسنده(ها): ، ،
اطلاعات انتشار: Journal of Information Systems and Telecommunication، دوم،شماره۴(پياپي ۸)، Oct-Dec ۲۰۱۴، سال
تعداد صفحات: ۹
The purpose of this paper is to provide a framework for detecting vulnerabilities in SIP (Session Initiation Protocol) networks. We focused our studies on the detection of SIP DoS related vulnerabilities in VoIP infrastructures because of their generalization. We try to find weaknesses in SIP enabled entities that an attacker by exploiting them is able to attack the system and affect it. This framework is provided by the concept of penetration testing and is designed to be flexible and extensible, and has the capability to customize for other similar session based protocols. To satisfy the above objectives, the framework is designed with five main modules for discovery, information modeling, operation, evaluation and report. After setting up a test–bed as a typical VoIP system to show the validity of the proposed framework, this system has been implemented as a SIP vulnerability scanner. We also defined appropriate metrics for gathering the performance statistics of SIP components. Our test–bed is deployed by open–source applications and used for validation and also evaluation of the proposed framework. The main contributions of this paper are its non–destructive manner in identifying vulnerabilities and incorporating the penetration testing ideas and steps in the overall architecture of our framework. We also defined appropriate metrics that help us to identify vulnerabilities in a black box penetration testing.
نمایش نتایج ۱ تا ۲۱ از میان ۲۱ نتیجه