{"title":"What to play next? A RNN-based music recommendation system","authors":"Miao Jiang, Ziyi Yang, Chen Zhao","doi":"10.1109/ACSSC.2017.8335200","DOIUrl":"https://doi.org/10.1109/ACSSC.2017.8335200","url":null,"abstract":"In the very recent years, development of music recommendation system has been a more heated problem due to a higher level of digital songs consumption and the advancement of machine learning techniques. Some traditional approaches such as collaborator filtering, has been widely used in recommendation systems, have helped music recommendation system to give music listeners a quick access to the music. However, collaborative filtering or model based algorithm have limitations in giving a better result with the ignorance of combination factor of lyrics and genre. In our paper, we will propose an improved algorithm based on deep neural network on measure similarity between different songs. The proposed method will make it possible that it could make recommendations in a large system to make comparisons by \"understand\" the content of songs. In this paper, we propose an end-end model, which is based on recurrent neural network to predict user's next most possible song by similarity. We will make experiments and evaluations based on Million Song Dataset and demonstrate how it outperformed the traditional methods.","PeriodicalId":296208,"journal":{"name":"2017 51st Asilomar Conference on Signals, Systems, and Computers","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133989741","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":"Multipulse subspace detectors","authors":"L. Scharf, Pooria Pakrooh","doi":"10.1109/ACSSC.2017.8335479","DOIUrl":"https://doi.org/10.1109/ACSSC.2017.8335479","url":null,"abstract":"In this paper we frame a fairly comprehensive set of spacetime detection problems, where a subspace signal modulates the mean-value vector of a multivariate normal measurement and nonstationary additive noise determines the covariance matrix. The measured spacetime data matrix consists of multiple measurements in time. As time advances, the signal component moves around in a subspace and the noise covariance matrix changes in scale.","PeriodicalId":296208,"journal":{"name":"2017 51st Asilomar Conference on Signals, Systems, and Computers","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134103398","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":"Reconstructing high-resolution cardiac MR movies from under-sampled frames","authors":"L. Cattell, C. Meyer, F. Epstein, G. Rohde","doi":"10.1109/ACSSC.2017.8335553","DOIUrl":"https://doi.org/10.1109/ACSSC.2017.8335553","url":null,"abstract":"In medicine, high-resolution magnetic resonance imaging can aid accurate diagnosis. However, high-resolution magnetic resonance imaging usually necessitates a longer acquisition time than low-resolution imaging, since the resolution of magnetic resonance images is determined by the extent of k-space that is sampled. Long scan times can induce motion artifacts in the images and lead to patient discomfort, and therefore, scan times should be kept as low as possible. Although a short acquisition time comes at the expense of spatial resolution, the resolution of magnetic resonance images can be increased using post-processing methods. In this work, we present one such method designed for cardiac magnetic resonance movies. Our method uses deformable image registration to capture the motion of the heart, and an additional term to account for changes in pixel intensity. We demonstrate that our method has the potential to reconstruct high-resolution cardiac magnetic resonance movies from highly under-sampled data, using only a single high-resolution reference frame.","PeriodicalId":296208,"journal":{"name":"2017 51st Asilomar Conference on Signals, Systems, and Computers","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130859550","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":"Energy efficient beam-alignment in millimeter wave networks","authors":"Muddassar Hussain, Nicolò Michelusi","doi":"10.1109/ACSSC.2017.8335545","DOIUrl":"https://doi.org/10.1109/ACSSC.2017.8335545","url":null,"abstract":"Millimeter wave communications rely on narrow-beam transmissions to cope with the strong signal attenuation at these frequencies, thus demanding precise beam-alignment between transmitter and receiver. The resulting signaling overhead may become excessive, especially in mobile environments. This paper addresses the energy efficient design of the beam-alignment protocol, with the goal of minimizing power consumption under a transmission rate constraint. The optimality of fractional search is proved, which allocates a given fraction of the interval of uncertainty on the mobile user's angular coordinate during beam-alignment. The fractional value is shown to be a function of the communication-sensing energy ratio: when large, a wider beam is selected and the fractional value approaches 1/2 (bisection); when small, a narrower beam is used to reduce the energy cost of sensing; finally, when smaller than 1/2, sensing is suboptimal. It is proved that the energy consumption under fractional search is smaller than that under bisection by at least a quantity proportional to the product of the minimum energy per radian used during beam-alignment, and the initial uncertainty on the mobile user's angular coordinate. Numerical results demonstrate a 2dB reduction in the average power consumption compared to bisection, for a wide range of rates.","PeriodicalId":296208,"journal":{"name":"2017 51st Asilomar Conference on Signals, Systems, and Computers","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132852443","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":"Distributed center and coverage region estimation in wireless sensor networks using diffusion adaptation","authors":"Sai Zhang, C. Tepedelenlioğlu, A. Spanias","doi":"10.1109/ACSSC.2017.8335575","DOIUrl":"https://doi.org/10.1109/ACSSC.2017.8335575","url":null,"abstract":"A fully distributed algorithm for estimating the center and coverage region of a wireless sensor network (WSN) is proposed. The proposed algorithm is useful in many applications, such as finding the required power for a certain level of connectivity in WSNs and localizing a service center in a network. The network coverage region is defined to be the smallest sphere that covers all the sensor nodes. The center and radius of the smallest covering sphere are estimated. The center estimation is formulated as a convex optimization problem using soft-max approximation. Then, diffusion adaptation is used for distributed optimization to estimate the center. After all the sensors obtain the center estimates, max consensus is used to calculate the radius distributively. The performance analysis of the proposed algorithm is provided, as a function of a design parameter controls the trade-off between the center estimation error and the convergence speed of the algorithm. Simulation results are provided.","PeriodicalId":296208,"journal":{"name":"2017 51st Asilomar Conference on Signals, Systems, and Computers","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133462192","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":"Array calibration in the presence of linear manifold distortion","authors":"B. Friedlander","doi":"10.1109/ACSSC.2017.8335541","DOIUrl":"https://doi.org/10.1109/ACSSC.2017.8335541","url":null,"abstract":"We consider the problem of calibrating an antenna array using received signals from known or unknown directions. A linear distortion model is used to characterize the mismatch between the true and assumed array manifolds. An iterative algorithm is presented for the joint estimation of the directions-of-arrival and the distortion matrix based on data collected during multiple time intervals.","PeriodicalId":296208,"journal":{"name":"2017 51st Asilomar Conference on Signals, Systems, and Computers","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132173171","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":"Impact of channel state information on wireless computing network control","authors":"Hao Feng, J. Llorca, A. Tulino, A. Molisch","doi":"10.1109/ACSSC.2017.8335394","DOIUrl":"https://doi.org/10.1109/ACSSC.2017.8335394","url":null,"abstract":"We consider the problem of efficient delivery of real-time computation services over wireless computing networks in which nodes are equipped with both communication and computation resources. We investigate the impact of channel state information (CSI) knowledge on the capacity of wireless computing networks, and design a distributed control policy that jointly schedules network flows for both routing and service processing, along with the corresponding allocation of communication and computation resources, using precise CSI knowledge at the transmitters. Compared to previous results, where only statistical CSI knowledge is considered, our results show that the availability of precise CSI at the transmitters increases the wireless computing network stability region, and that properly designed schemes that exploit such knowledge can support higher input rates and improve the overall cost-delay tradeoff.","PeriodicalId":296208,"journal":{"name":"2017 51st Asilomar Conference on Signals, Systems, and Computers","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132179373","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":"Brain language: Uncovering functional connectivity codes","authors":"V. Vergara, V. Calhoun","doi":"10.1109/ACSSC.2017.8335565","DOIUrl":"https://doi.org/10.1109/ACSSC.2017.8335565","url":null,"abstract":"The functional connectivity within a specific set of brain networks (or domain) can assume different configurations known as domain states that change with time. Recently, we proposed an information theoretical framework that models the finite set of domain states as elements of an alphabet. Significant bits of information have been found to be shared among domains, but specific domain codification was not explored. This work describes a method to identify code words used to transmit and receive information between the cerebrum and the cerebellum based on dynamic domain connectivity estimated from functional magnetic resonance imaging (fMRI). Following the theory of jointly typical sets, the developed method identifies the codeword length and the specific combination of domain states on each codeword. Resting state fMRI data was taken from 121 subjects with no significant age difference between males and females. Group independent component analysis was utilized to identify important brain networks and group them in a cerebellum and six other domains representing the cerebrum. The amount of information between the cerebellum, the executive control and sensorimotor domains showed a statistically significant number of bits. The proposed method quantified specific temporal sequences of domain states encoded within bits shared between cerebellum and cerebrum.","PeriodicalId":296208,"journal":{"name":"2017 51st Asilomar Conference on Signals, Systems, and Computers","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128882215","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":"Glaucoma detection using texture features extraction","authors":"N. Kavya, K. Padmaja","doi":"10.1109/ACSSC.2017.8335600","DOIUrl":"https://doi.org/10.1109/ACSSC.2017.8335600","url":null,"abstract":"Glaucoma is a second leading cause of the disease in the world. The World Health Organization has estimated that by 2020, about 80 million people would suffer from glaucoma. As the disease progresses, it leads to structural changes in the Optic Nerve Head (ONH). Optic Nerve Head is the region which consists of Optic Cup and Optic Disc. The region of interest is extracted from the fundus image by using Hough Transformation. It is an automated way of segmentation used to obtain the accurate results and it replaces the manual segmentation. The k-mean clustering also used for segmentation which is another approach. From the segmented ONH, the different features like Gray Level Cooccurrence Matrix (GLCM) and Markov Random Field (MRF) are extracted. As the structural changes taken place in ONH, the texture and the intensity values also changes. The features are used to classify the images as normal and glaucoma. The algorithm speed increases by applying the technique on region of interest instead of using complete image directly. Hence the algorithm results about 94% of accuracy in segmentation using Hough Transform, 84% for segmentation using k-means clustering and about 86% for classification using support vector machine.","PeriodicalId":296208,"journal":{"name":"2017 51st Asilomar Conference on Signals, Systems, and Computers","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115189321","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 robust adaptive binaural beamformer for hearing devices","authors":"Jinjun Xiao, Z. Luo, I. Merks, Zhang Tao","doi":"10.1109/ACSSC.2017.8335691","DOIUrl":"https://doi.org/10.1109/ACSSC.2017.8335691","url":null,"abstract":"In a hearing aid application, the performance of an adaptive beamformer is sensitive to errors in the acoustic transfer function or noise estimation. The estimation errors can be caused by factors including the head movement of the hearing aid users, non-stationarity of the environment, or imperfect array calibration. With such inaccurate information, the beamformer not only provides less noise reduction, but also causes undesirable speech distortions. To improve the beamformers performance in such conditions, robust beam-forming algorithms have been proposed in the literature. In particular, to minimize the speech distortions caused by imperfect look direction estimation, a robust binaural beam-forming algorithm has been proposed in [1]. This algorithm utilizes constrained optimization to improve the protection of the target speech. The beamforming adaption is based on a low-complexity iteration method called Alternating Direction Method of Multipliers (ADMM). This paper aims for evaluating the proposed binaural beamforming algorithm by comparing its performance with the Minimum Variance Distortionless Response (MVDR) beamformer in various noisy environments. The benefit of the proposed algorithm is demonstrated through both objective and subjective evaluations. In particular, multiple-microphone recordings on a pair of binaural hearing aids on a mannequin are used. Objective perception scores are calculated and compared. In addition, a subjective evaluation of speech intelligibility using normal-hearing listeners is conducted. Both the objective and subjective evaluation results show the robustness of the proposed algorithm.","PeriodicalId":296208,"journal":{"name":"2017 51st Asilomar Conference on Signals, Systems, and Computers","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124224351","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}