Department of Electronic Systems

This paper presents an analysis on the performance of decode-and-forward cooperation transmissions in Rician fading channels. The users send the signals using multiband DS-CDMA transmission; the particular complete complementary code sets are used as spreading sets and to provide orthogonality and signal separation. These codes sets have an impulsive autocorrelation sum among each set and orthogonality along the set size. Under the constraint of half-duplex transmission, the performance of the proposed model is compared to the closed-form theoretical performance under similar transmission conditions and applying the suboptimum maximal ratio combiner model at the receiver stage. Then the proposed model is evaluated against a non-cooperative transmission framework using the same spreading sets over Rician fading channels with different K factors. The performance of the cooperative model is shown to be on the verge of the theoretical performance curves in terms of the probability of error. It also manifests an enhanced performance compared to the non-cooperative algorithm under different channel assumptions.

This paper proposes an adaptive combiner model that maximizes the performance of multi-rate cooperative relaying transmission in resources aware systems. Users’ cooperation is applied to smartly profit from the network resources to reach for efficient multi-rate transmissions. The low rate terminal uses a part of his bandwidth and power to relay the high rate user data side by side with his own data. This results in small performance degradation to the low rate user performance. The combiner model at the high rate terminal is carefully investigated for an optimized system performance and an adaptive combiner is proposed. The model calculates the weights adaptively for each bit based on a dynamic-blind method. The adaptive combiner is evaluated against the maximal-ratio combiner MRC and the equal gain combiner ECG to reach for the best model for the multi-rate relaying framework. Based on half-duplex mode, the probability of error performance is studied and compared for each combiner model under different channel conditions. The results show that the proposed adaptive combiner, characterized by its reliability and simplicity, leads to the best performance compared to the other models

Abstract—This paper develops a practical receiver suggested for cooperative systems using decode-and-forward transmission and compares it to the theoretical sub-optimum λ-MRC receiver model. The proposed receiver model adopts a channel blind λ-combiner and employs a practical estimation of the combiner’s weight λ that changes adaptively for each received bit. The λ estimation process relies on a dynamic-blind calculation performed on the incoming bit stream using an approximate formula. The accuracy of the estimated values of λ against the numerical (optimum) values is verified by comparing their effects on the performance curves. Next, the performance of the proposed receiver is evaluated against the sub-optimum receiver using the closed-form performance equations then verified using an actual implementation of the decode-and-forward cooperative algorithm. The use of the proposed receiver is shown to have reliable performance under different channel assignments and provides adaptiv...

This paper adopts an efficient strategy that allocates the network resources to the users according to the transmission rate requirement. Based on cooperation and the use of variable length spreading codes, the terminals requiring high transmission rate are supported by larger bandwidth and power in order to enhance their quality of service. The proposed resource allocation strategy assigns the terminals with low transmission rate to cooperate with those with high transmission rate by dedicating a part of their bandwidth and power to either relay their data or send a part of it. This will result in a slight limitation on the performance of the cooperating terminals (low rate). Two proposed schemes are presented namely the relaying strategy and re‐allocation strategy. These strategies are evaluated against the commonly used rate‐matching algorithm which constrains the terminals to fairly share the bandwidth regardless of the transmission rate. The proposed receiver is simple and effi...

This paper develops a new dynamic-blind estimation of the combiner's weight λ suggested for decode-and-forward cooperative based receivers and provides an empirical form for it. The estimation is performed by applying a monotonically decreasing formula on the incoming bit stream. The performance evaluation is investigated using the analytical probability of error (P e) closed-form of the decode-and-forward cooperative transmissions. An adjacency is illustrated between the performances using the values of λ obtained by the proposed estimation against the use of optimum values of λ found by iterative search. A practical implementation of the decode-and-forward cooperative algorithm is also considered and the performance analysis is studied for different channel assignments. It is demonstrated that the proposed dynamic-blind estimation of λ provides in an overall performance on the verge of that of using optimum values of the combiner's weight λ.

In this study, we introduce a novel unsupervised countermeasure for smart grid power systems, based on generative adversarial networks (GANs). Given the pivotal role of smart grid systems (SGSs) in urban life, their security is of particular importance. In recent years, however, advances in the field of machine learning, have raised concerns about cyber attacks on these systems. Power systems, among the most important components of urban infrastructure, have, for example, been widely attacked by adversaries. Attackers disrupt power systems using false data injection attacks (FDIA), resulting in a breach of availability, integrity, or confidential principles of the system. Our model simulates possible attacks on power systems using multiple generators in a step-by-step interaction with a discriminator in the training phase. As a consequence, our system is robust to unseen attacks. Moreover, the proposed model considerably reduces the well-known mode collapse problem of GAN-based mode...

Automatic Speaker Verification (ASV) is the process of identifying a person based on the voice presented to a system. Different synthetic approaches allow spoofing to deceive ASV systems (ASVs), whether using techniques to imitate a voice or recunstruct the features. Attackers try to beat up the ASVs using four general techniques; impersonation, speech synthesis, voice conversion, and replay. The last technique is considered as a common and high potential tool for spoofing purposes since replay attacks are more accessible and require no technical knowledge from adversaries. In this study, we introduce a novel replay spoofing countermeasure for ASVs. Accordingly, we used the Constant Q Cepstral Coefficient (CQCC) features fed into an autoencoder to attain more informative features and to consider the noise information of spoofed utterances for discrimination purpose. Finally, different configurations of the Siamese network were used for the first time in this context for classification. The experiments performed on ASVspoof challenge 2019 dataset using Equal Error Rate (EER) and Tandem Detection Cost Function (t-DCF) as evaluation metrics show that the proposed system improved the results over the baseline by 10.73% and 0.2344 in terms of EER and t-DCF, respectively.

Cardiovascular Disease (CVD) is considered as one of the principal causes of death in the world. Over recent years, this field of study has attracted researchers’ attention to investigate heart sounds’ patterns for disease diagnostics. In this study, an approach is proposed for normal/abnormal heart sound classification on the Physionet challenge 2016 dataset. For the first time, a fixed length feature vector; called i-vector; is extracted from each heart sound using Mel Frequency Cepstral Coefficient (MFCC) features. Afterwards, Principal Component Analysis (PCA) transform and Variational Autoencoder (VAE) are applied on the i-vector to achieve dimension reduction. Eventually, the reduced size vector is fed to Gaussian Mixture Models (GMMs) and Support Vector Machine (SVM) for classification purpose. Experimental results demonstrate the proposed method could achieve a performance improvement of 16% based on Modified Accuracy (MAcc) compared with the baseline system on the Physionet...

Столетие назад Алексей Николаевич Толстой (1883-1945) несколько лет провёл в эмиграции, вначале не приняв установившегося в России нового миропорядка, но затем постепенно склоняясь к возвращению на Родину. Всё его творчество так или иначе оказалось связанным с ситуацией коренного перелома в жизни страны. Смысл предлагаемой статьи-вкратце проследить путь выдающегося писателя. Первое десятилетие его творчества-это условно 1910-е годы. Начав печататься стихами с 1905-го и прозой с 1908-го, он вошёл в литературу серией повестей и рассказов так называемого заволжского цикла (1909-1911), что нашло продолжение в двух небольших романах «Чудаки» (1911) и «Хромой барин» (1912), а также в повести «Приключения Растёгина» (1913) и в ряде последующих вещей типа рассказа «Четыре века» (1914). Сквозь пелену чудачеств и экстравагантных происшествий их персонажей высвечивалась мысль об окончательно отмирающей усадебной России, существование которой превращалось в смехотворный анекдот, дурашливый фарс. О запустении угасающих дворянских гнёзд в повести «Петушок» (1909) констатируется: «Все вымерли, унеся с собой в сырую землю веселье, богатство и несбывшиеся мечты». М. Горький уже в 1910 году отметил в нём писателя «крупного, сильного, с жёсткой правдивостью изображающего разложение современного дворянства». Хиреет, выветривается помещичья порода, и параллельно тому Толстой показывает, как простонародье начинает поднимать голову. Оно теперь нередко и хитрее, и сильнее барина. Прежняя узда уже не держит его, и законов оно не чтит, выходя из повиновения. Впервые со всей отчётливостью крушение прежних порядков писатель обрисовал в рассказе «Архип» (1909), где, увидев беспомощность своего молодого хозяина, мужик может позволить в его адрес грубый окрик «Щенок!». А уж миром встать, чтобы спалить господскую усадьбу за грешки какие-это сам Бог велел. И в ряде последующих рассказов («Сватовство» 1910, «Прогулка» 1911) Толстой повествует, как осмелел простой люд. Грабит и бьёт, доходя порой до смертоубийства тех, кто формально ещё числится власть имущими. Хуже того, отмечается, как в глубинах низовой среды прорастает нечто замутнённое, тёмное, устрашающее. В том же рассказе «Архип» говорится о тех, что «на всё способны, в них бес сидит». Переполняет их злоба, слепая ненависть. Кто-то чинит ужасающий самосуд, кто-то со смехом убивает барина своего, в сущности, ни в чём не повинного. Матёрый казак толкует: «У нас в степи законы не писаны, колодцы глубокие-бросил туда человека, землицей засыпал, и пропал человек». Угрюмая натура с тяжёлой рукой, это ещё что. А то ведь глянет из человека и зверь настоящий-«мужик косматый, косолапый, глаза волчьи, налитые кровью». И повадки звериные, на что одна помещица замечает: «Мужик обращается в первобытное состояние». Таковы нелицеприятные грани иного уклада, зреющего в недрах народной России начала ХХ века. Но наряду с тем и всё очевиднее с середины 1910-х годов писатель с воодушевлением начинает говорить о поколении новых людей, которые стремились сбросить с себя оболочку прежнего существования, вырваться из оков обыденности, открыть для себя горизонты большой, незаурядной жизни-и если не сейчас же, то хотя бы в грядущем. Эти люди в своих исканиях несли заряд явственно современного мироощущения, и столь же явственным становился у Толстого новый стиль литературного письма (рассказы «Буря», «Для чего идёт снег», «Под водой»-все 1915, «Миссис Бризли» 1916). К той же середине 1910-х годов в творчестве писателя вызревает один из самых сокровенных мотивов: для обновляющейся человеческой натуры высшим притяжением жизни становится любовь (рассказ «Овражки» 1913, повесть «Большие неприятности» 1914, незаконченный роман «Егор Абозов» 1915).

Cardiovascular disease (CVD) is considered as one of the main causes of death in the world. Accordingly, scientists look for methods to recognize normal/abnormal heart patterns. Over recent years, researchers have been interested in to investigate CVDs based on heart sounds. The physionet 2016 corpus is presented to provide a standard database for researchers in this field. In this study we proposed an approach for normal/abnormal heart sound detection, based on i-vector features on phiysionet 2016 corpus. In this method, a fixed length vector, namely i-vector, is extracted from each record, and then Principal Component Analysis (PCA) is applied. Then Variational AuotoEncoders (VAE) is used to reduce dimensions of the obtained i-vector. After that, this i-vector and its transmitted version by PCA and VAE are used for training two Gaussian Mixture Models (GMMs). Finally, test set is scored using these trained GMMs. In the next step we applied a simple global threshold to classify the...

We propose a multi-layer variational autoencoder method, we call HR-VQVAE, that learns hierarchical discrete representations of the data. By utilizing a novel objective function, each layer in HR-VQVAE learns a discrete representation of the residual from previous layers through a vector quantized encoder. Furthermore, the representations at each layer are hierarchically linked to those at previous layers. We evaluate our method on the tasks of image reconstruction and generation. Experimental results demonstrate that the discrete representations learned by HR-VQVAE enable the decoder to reconstruct high-quality images with less distortion than the baseline methods, namely VQVAE and VQVAE-2. HR-VQVAE can also generate high-quality and diverse images that outperform state-of-the-art generative models, providing further verification of the efficiency of the learned representations. The hierarchical nature of HR-VQVAE i) reduces the decoding search time, making the method particularly suitable for high-load tasks and ii) allows to increase the codebook size without incurring the codebook collapse problem.

We address the video prediction task by putting forth a novel model that combines (i) our recently proposed hierarchical residual vector quantized variational autoencoder (HR-VQVAE), and (ii) a novel spatiotemporal PixelCNN (ST-PixelCNN). We refer to this approach as a sequential hierarchical residual learning vector quantized variational autoencoder (S-HR-VQVAE). By leveraging the intrinsic capabilities of HR-VQVAE at modeling still images with a parsimonious representation, combined with the ST-PixelCNN's ability at handling spatiotemporal information, S-HR-VQVAE can better deal with chief challenges in video prediction. These include learning spatiotemporal information, handling high dimensional data, combating blurry prediction, and implicit modeling of physical characteristics. Extensive experimental results on the KTH Human Action and Moving-MNIST tasks demonstrate that our model compares favorably against top video prediction techniques both in quantitative and qualitative evaluations despite a much smaller model size. Finally, we boost S-HR-VQVAE by proposing a novel training method to jointly estimate the HR-VQVAE and ST-PixelCNN parameters.

— The purpose of this paper is to introduce and initially evaluate an experimental sensory substitution device (SSD), which converts color to audible sound. Our system, called Colorophone provides continuous information about the color, light intensity and distance. The color sonification method is inspired by the human visual system. While designing sonification method we took into consideration mismatch in sensory bandwidth, possible sensory overload, nature of sensory spatiotemporal continuity and cross-modal correspondences between colors and sounds. We conducted a preliminary experimental evaluation of the color and object recognition abilities as well as orientation ability. We show that blindfolded participants can easily acquire the color information coded in the auditory stimuli generated by the Colorophone system after a short learning session. Our preliminary experiments indicate that sonified color information helps in object identification and enhances orientation ability. The Colorophone proved to be an easy to understand and intuitive visual-to-auditory coding system, which shows promising results in future visual rehabilitation of the blind.

Designing of color sonification systems provides a possibility of contribution to various fields ranging from rehabilitation of visually impaired through color perception, multisensory art experience to consciousness studies. The design process itself requires understanding and integrating knowledge from many difficult and inherently different branches of science and the resulting sonification method will be highly dependent on the purpose of the system. We present work in progress on designing and experimental verification of color sonification method that will be implemented in Colorophone-a wearable assistive device for the visually impaired, which enables perception of the information about color through sound. Although our system shows promising results in color and object recognition, we would like to enhance the existing color sonification method by designing a framework for experimental verification of our color sonification algorithm. The goal of this paper is therefore to briefly describe our way of thinking in order to provide the basis for the discussion.