Results 401-420 of 728.
Bookmark and Share    
Full Text
Peer Reviewed
See detailPatch-based Statistical Performance Analysis of Upsampling for Precise Super–Resolution
Aouada, Djamila UL; Al Ismaeil, Kassem UL; Ottersten, Björn UL

in 11th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISAPP'15) (2015, March)

All existent methods for the statistical analysis of super–resolution approaches have stopped at the variance term, not accounting for the bias in the mean square error. In this paper we give an original ... [more ▼]

All existent methods for the statistical analysis of super–resolution approaches have stopped at the variance term, not accounting for the bias in the mean square error. In this paper we give an original derivation of the bias term. We propose to use a patch-based method inspired by the work of (Chatterjee and Milanfar, 2009). Our approach, however, is completely new as we derive a new affine bias model dedicated for the multi-frame super resolution framework. We apply the proposed statistical performance analysis to the Upsampling for Precise Super–Resolution (UP-SR) algorithm. This algorithm was shown experimentally to be a good solution for enhancing the resolution of depth sequences in both cases of global and local motions. Its performance is herein analyzed theoretically in terms of its approximated mean square error, using the proposed derivation of the bias. This analysis is validated experimentally on simulated static and dynamic depth sequences with a known ground truth. This provides an insightful understanding of the effects of noise variance, number of observed low resolution frames, and super–resolution factor on the final and intermediate performance of UP–SR. Our conclusion is that increasing the number of frames should improve the performance while the error is increased due to local motions, and to the upsampling which is part of UP-SR. [less ▲]

Detailed reference viewed: 238 (13 UL)
Peer Reviewed
See detailExploiting Polarization for Spectrum Awareness in Cognitive Satellite Communications
Sharma, Shree Krishna UL; Chatzinotas, Symeon UL; Ottersten, Björn UL

in Software-Defined and Cognitive Radio Technologies for Dynamic Spectrum Access and Management (2015)

The continuously increasing demand of spectrum and current static spectrum allocation policies are rendering the available radio spectrum scarce. To address the problem of spectrum scarcity in the ... [more ▼]

The continuously increasing demand of spectrum and current static spectrum allocation policies are rendering the available radio spectrum scarce. To address the problem of spectrum scarcity in the satellite paradigm, cognitive Satellite Communications (SatComs) has been considered as a promising technique. In addition to the existing spectrum sharing dimensions such as frequency, time and space, polarization domain can be exploited as an additional degree of freedom in order to explore the spectral gaps in the under-utilized licensed spectrum. In this context, this chapter firstly provides an overview of the existing works in polarization-based spectrum sharing. Secondly, it presents the theoretical analysis of Energy Detection technique for dual polarized Additive White Gaussian Noise (AWGN) and Rayleigh fading channels considering the spectral coexistence scenarios of dual and hybrid satellite systems. Thirdly, it provides the comparison of different combining techniques in terms of the sensing performance in the considered dual polarized channels with the help of theoretical analysis and numerical results. Finally, it provides interesting future research directions in this domain. [less ▲]

Detailed reference viewed: 217 (10 UL)
Full Text
Peer Reviewed
See detailSecrecy Analysis on Network Coding in Bidirectional Multibeam Satellite Communications
Kalantari, Ashkan UL; Ottersten, Björn UL

in IEEE Transactions on Information Forensics and Security (2015)

Network coding is an efficient means to improve the spectrum efficiency of satellite communications. However, its resilience to eavesdropping attacks is not well understood. This paper studies the ... [more ▼]

Network coding is an efficient means to improve the spectrum efficiency of satellite communications. However, its resilience to eavesdropping attacks is not well understood. This paper studies the confidentiality issue in a bidirectional satellite network consisting of two mobile users who want to exchange message via a multibeam satellite using the XOR network coding protocol. We aim to maximize the sum secrecy rate by designing the optimal beamforming vector along with optimizing the return and forward link time allocation. The problem is non-convex, and we find its optimal solution using semidefinite programming together with a 1-D search. For comparison, we also solve the sum secrecy rate maximization problem for a conventional reference scheme without using network coding. Simulation results using realistic system parameters demonstrate that the bidirectional scheme using network coding provides considerably higher secrecy rate compared to that of the conventional scheme. [less ▲]

Detailed reference viewed: 246 (40 UL)
Full Text
Peer Reviewed
See detailSpatial DCT-Based Channel Estimation in Multi-Antenna Multi-Cell Interference Channels
Alodeh, Maha UL; Chatzinotas, Symeon UL; Ottersten, Björn UL

in IEEE Transactions on Signal Processing (2015)

This work addresses channel estimation in multiple antenna multicell interference-limited networks. Channel state information (CSI) acquisition is vital for interference mitigation. Wireless networks ... [more ▼]

This work addresses channel estimation in multiple antenna multicell interference-limited networks. Channel state information (CSI) acquisition is vital for interference mitigation. Wireless networks often suffer from multicell interference, which can be mitigated by deploying beamforming to spatially direct the transmissions. The accuracy of the estimated CSI plays an important role in designing accurate beamformers that can control the amount of interference created from simultaneous spatial transmissions to mobile users. Therefore, a new technique based on the structure of the spatial covariance matrix and the discrete cosine transform (DCT) is proposed to enhance channel estimation in the presence of interference. Bayesian estimation and Least Squares estimation frameworks are introduced by utilizing the DCT to separate the overlapping spatial paths that create the interference. The spatial domain is thus exploited to mitigate the contamination which is able to discriminate across interfering users. Gains over conventional channel estimation techniques are presented in our simulations which are also valid for a small number of antennas. [less ▲]

Detailed reference viewed: 277 (23 UL)
See detailCooperative and Cognitive Satellite Systems
Chatzinotas, Symeon UL; Ottersten, Björn UL; DeGaudenzi, Riccardo

Book published by Elsevier (2015)

Detailed reference viewed: 400 (24 UL)
Full Text
Peer Reviewed
See detailConstructive Multiuser Interference in Symbol Level Precoding for the MISO Downlink Channel
Alodeh, Maha UL; Chatzinotas, Symeon UL; Ottersten, Björn UL

in IEEE Transactions on Signal Processing (2015)

This paper investigates the problem of interference among the simultaneous multiuser transmissions in the downlink of multiple-antenna systems. Using symbol-level precoding, a new approach to exploit the ... [more ▼]

This paper investigates the problem of interference among the simultaneous multiuser transmissions in the downlink of multiple-antenna systems. Using symbol-level precoding, a new approach to exploit the multiuser interference is discussed. The concept of exploiting the interference between spatial multiuser transmissions by jointly utilizing data information (DI) and channel state information (CSI), in order to design symbol-level precoders, is proposed. To this end, the interference between data streams is transformed under certain conditions into useful signal that can improve the signal to interference noise ratio (SINR) of the downlink transmissions. We propose a maximum ratio transmission (MRT) based algorithm that jointly exploits DI and CSI to glean the benefits from constructive multiuser interference. Subsequently, a relation between the constructive interference downlink transmission and physical layer multicasting is established. In this context, novel constructive interference precoding techniques that tackle the transmit power minimization (min-power) with individual SINR constraints at each user's receivers is proposed. Furthermore, fairness through maximizing the weighted minimum SINR (max-min SINR) of the users is addressed by finding the link between the min power and max min SINR problems. Moreover, heuristic precoding techniques are proposed to tackle the weighted sum rate problem. Finally, extensive numerical results show that the proposed schemes outperform other state of the art techniques. [less ▲]

Detailed reference viewed: 273 (40 UL)
Full Text
Peer Reviewed
See detailReal-time Curve-skeleton Extraction of Human-scanned Point Clouds
Garcia, Frederic; Ottersten, Björn UL

in International Conference on Computer Vision Theory and Applications (VISAPP 2015) (2015)

This paper presents a practical and robust approach for upright human curve-skeleton extraction. Curve skeletons are object descriptors that represent a simplified version of the geometry and topology of ... [more ▼]

This paper presents a practical and robust approach for upright human curve-skeleton extraction. Curve skeletons are object descriptors that represent a simplified version of the geometry and topology of a 3-D object. The curve-skeleton of a human-scanned point set enables the approximation of the underlying skeletal structure and thus, to estimate the body configuration (human pose). In contrast to most curve-skeleton extraction methodologies from the literature, we herein propose a real-time curve-skeleton extraction approach that applies to scanned point clouds, independently of the object’s complexity and/or the amount of noise within the depth measurements. The experimental results show the ability of the algorithm to extract a centered curve-skeleton within the 3-D object, with the same topology, and with unit thickness. The proposed approach is intended for real world applications and hence, it handles large portions of data missing due to occlusions, acquisition hindrances or registration inaccuracies. [less ▲]

Detailed reference viewed: 192 (2 UL)
Peer Reviewed
See detailUser Scheduling in Cooperative Satellite Systems
Christopoulos, Dimitrios UL; Chatzinotas, Symeon UL; Ottersten, Björn UL

in Cooperative and Cognitive Satellite Systems (2015)

Detailed reference viewed: 171 (13 UL)
Full Text
Peer Reviewed
See detailPower Control in Cognitive Radio Networks Using Cooperative Modulation and Coding Classification
Tsakmalis, Anestis UL; Chatzinotas, Symeon UL; Ottersten, Björn UL

in Proceedings of 10th International Conference on Cognitive Radio Oriented Wireless Networks (CROWNCOM) 2015 (2015)

In this paper, a centralized Power Control (PC) scheme aided by interference channel gain learning is proposed to allow a Cognitive Radio (CR) network to access the frequency band of a Primary User (PU ... [more ▼]

In this paper, a centralized Power Control (PC) scheme aided by interference channel gain learning is proposed to allow a Cognitive Radio (CR) network to access the frequency band of a Primary User (PU) operating based on an Adaptive Coding and Modulation (ACM) protocol. The main idea is the CR network to constantly probe the band of the PU with intelligently designed aggregated interference and sense whether the Modulation and Coding scheme (MCS) of the PU changes in order to learn the interference channel gains. The coordinated probing is engineered by the Cognitive Base Station (CBS), which assigns appropriate CR power levels in a binary search way. Subsequently, each CR applies a Modulation and Coding Classification (MCC) technique and sends the sensing information through a control channel to the CBS, where all the MCC information is combined using a fusion rule to acquire an MCS estimate of higher accuracy and monitor the probing impact to the PU MCS. After learning the normalized interference channel gains towards the PU, the CBS selects the CR power levels to maximize total CR network throughput while preserving the PU MCS and thus its QoS. The effectiveness of the proposed technique is demonstrated through numerical simulations. [less ▲]

Detailed reference viewed: 170 (11 UL)
Full Text
Peer Reviewed
See detailPower control in cognitive radio networks using cooperative modulation and coding classification
Tsakmalis, Anestis UL; Chatzinotas, Symeon UL; Ottersten, Björn UL

in Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST (2015), 156

In this paper, a centralized Power Control (PC) scheme aided by interference channel gain learning is proposed to allow a Cognitive Radio (CR) network to access the frequency band of a Primary User (PU ... [more ▼]

In this paper, a centralized Power Control (PC) scheme aided by interference channel gain learning is proposed to allow a Cognitive Radio (CR) network to access the frequency band of a Primary User (PU) operating based on an Adaptive Coding and Modulation (ACM) protocol. The main idea is the CR network to constantly probe the band of the PU with intelligently designed aggregated interference and sense whether the Modulation and Coding scheme (MCS) of the PU changes in order to learn the interference channels. The coordinated probing is engineered by the Cognitive Base Station (CBS), which assigns appropriate CR power levels in a binary search way. Subsequently, each CR applies a Modulation and Coding Classification (MCC) technique and sends the sensing information through a control channel to the CBS, where all the MCC information is combined using a fusion rule to acquire an MCS estimate of higher accuracy and monitor the probing impact to the PU MCS. After learning the normalized interference channel gains towards the PU, the CBS selects the CR power levels to maximize total CR network throughput while preserving the PU MCS and thus its QoS. The effectiveness of the proposed technique is demonstrated through numerical simulations. © Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2015. [less ▲]

Detailed reference viewed: 217 (5 UL)
Full Text
Peer Reviewed
See detailMulticast multigroup beamforming for per-antenna power constrained large-scale arrays
Christopoulos, Dimitrios UL; Chatzinotas, Symeon UL; Ottersten, Björn UL

in IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC (2015), 2015-August

Large in the number of transmit elements, multi-antenna arrays with per-element limitations are in the focus of the present work. In this context, physical layer multigroup multicasting under per-antenna ... [more ▼]

Large in the number of transmit elements, multi-antenna arrays with per-element limitations are in the focus of the present work. In this context, physical layer multigroup multicasting under per-antenna power constrains, is investigated herein. To address this complex optimization problem low-complexity alternatives to semi-definite relaxation are proposed. The goal is to optimize the per-antenna power constrained transmitter in a maximum fairness sense, which is formulated as a non-convex quadratically constrained quadratic problem. Therefore, the recently developed tool of feasible point pursuit and successive convex approximation is extended to account for practical per-antenna power constraints. Interestingly, the novel iterative method exhibits not only superior performance in terms of approaching the relaxed upper bound but also a significant complexity reduction, as the dimensions of the optimization variables increase. Consequently, multicast multigroup beamforming for large-scale array transmitters with per-antenna dedicated amplifiers is rendered computationally efficient and accurate. A preliminary performance evaluation in large-scale systems for which the semi-definite relaxation constantly yields non rank-1 solutions is presented. © 2015 IEEE. [less ▲]

Detailed reference viewed: 190 (6 UL)
Full Text
Peer Reviewed
See detailCognitive interference alignment for spectral coexistence
Sharma, Shree Krishna UL; Chatzinotas, Symeon UL; Ottersten, Björn UL

in Cognitive Radio and Networking for Heterogeneous Wireless Networks (2015)

Interference Alignment (IA) has been widely recognized as a promising interference mitigation technique since it can achieve the optimal degrees of freedom in certain interference limited channels. In the ... [more ▼]

Interference Alignment (IA) has been widely recognized as a promising interference mitigation technique since it can achieve the optimal degrees of freedom in certain interference limited channels. In the context of Cognitive Radio (CR) networks, this technique allows the coexistence of two heterogeneous wireless systems in an underlay cognitive mode. The main concept behind this technique is the alignment of the interference on a signal subspace in such a way that it can be filtered out at the non-intended receiver by sacrificing some signal dimensions. This chapter starts with an overview of IA principle, Degree of Freedom (DoF) concept, and the classification of existing IA techniques. Furthermore, this chapter includes a discussion about IA applications in CR networks.Moreover, a generic system model is presented for allowing the coexistence of two heterogeneous networks using IA approach while relevant precoding and filtering processes are described. In addition, two important practical applications of the IA technique are presented along with the numerical results for underlay spectral coexistence of (i) femtocell-macrocell systems, and (ii) monobeam-multibeam satellite systems. More specifically, an uplink IA scheme is investigated in order to mitigate the interference of femtocell. User Terminals (UTs) towards the macrocell Base Station (BS) in the spatial domain and the interference of multibeam satellite terminals towards themonobeam satellite in the frequency domain. [less ▲]

Detailed reference viewed: 139 (2 UL)
Full Text
Peer Reviewed
See detailCellular-broadcast service convergence through caching for CoMP cloud RANs
Chatzinotas, Symeon UL; Christopoulos, Dimitrios; Ottersten, Björn UL

in arXiv preprint arXiv:1504.08274 (2015)

Cellular and Broadcast services have been traditionally treated independently due to the different market requirements, thus resulting in different business models and orthogonal frequency allocations ... [more ▼]

Cellular and Broadcast services have been traditionally treated independently due to the different market requirements, thus resulting in different business models and orthogonal frequency allocations. However, with the advent of cheap memory and smart caching, this traditional paradigm can converge into a single system which can provide both services in an efficient manner. This paper focuses on multimedia delivery through an integrated network, including both a cellular (also known as unicast or broadband) and a broadcast last mile operating over shared spectrum. The subscribers of the network are equipped with a cache which can effectively create zero perceived latency for multimedia delivery, assuming that the content has been proactively and intelligently cached. The main objective of this work is to establish analytically the optimal content popularity threshold, based on a intuitive cost function. In other words, the aim is to derive which content should be broadcasted and which content should be unicasted. To facilitate this, Cooperative Multi- Point (CoMP) joint processing algorithms are employed for the uni and broad-cast PHY transmissions. To practically implement this, the integrated network controller is assumed to have access to traffic statistics in terms of content popularity. Simulation results are provided to assess the gain in terms of total spectral efficiency. A conventional system, where the two networks operate independently, is used as benchmark. [less ▲]

Detailed reference viewed: 136 (6 UL)
Full Text
Peer Reviewed
See detailTemplate-Based Statistical Shape Modelling on Deformation Space
Demisse, Girum UL; Aouada, Djamila UL; Ottersten, Björn UL

in 22nd IEEE International Conference on Image Processing (2015)

A statistical model for shapes in $\mathbb{R}^2$ or $\mathbb{R}^3$ is proposed. Shape modelling is a difficult problem mainly due to the non-linear nature of its space. Our approach considers curves as ... [more ▼]

A statistical model for shapes in $\mathbb{R}^2$ or $\mathbb{R}^3$ is proposed. Shape modelling is a difficult problem mainly due to the non-linear nature of its space. Our approach considers curves as shape contours, and models their deformations with respect to a deformable template shape. Contours are uniformly sampled into a discrete sequence of points. Hence, the deformation of a shape is formulated as an action of transformation matrices on each of these points. A parametrized stochastic model based on Markov process is proposed to model shape variability in the deformation space. The model's parameters are estimated from a labeled training dataset. Moreover, a similarity metric based on the Mahalanobis distance is proposed. Subsequently, the model has been successfully tested for shape recognition, synthesis, and retrieval. [less ▲]

Detailed reference viewed: 269 (52 UL)
Full Text
Peer Reviewed
See detailCellular-Broadcast Service Convergence through Caching for CoMP Cloud RANs
Christopoulos, Dimitrios UL; Chatzinotas, Symeon UL; Ottersten, Björn UL

in 22nd IEEE Symposium on Communications and Vehicular Technology in the Benelux (2015)

Cellular and Broadcast services have been traditionally treated independently due to the different market requirements, thus resulting in different business models and orthogonal frequency allocations ... [more ▼]

Cellular and Broadcast services have been traditionally treated independently due to the different market requirements, thus resulting in different business models and orthogonal frequency allocations. However, with the advent of cheap memory and smart caching, this traditional paradigm can converge into a single system which can provide both services in an efficient manner. This paper focuses on multimedia delivery through an integrated network, including both a cellular (also known as unicast or broadband) and a broadcast last mile operating over shared spectrum. The subscribers of the network are equipped with a cache which can effectively create zero perceived latency for multimedia delivery, assuming that the content has been proactively and intelligently cached. The main objective of this work is to establish analytically the optimal content popularity threshold, based on a intuitive cost function. In other words, the aim is to derive which content should be broadcasted and which content should be unicasted. To facilitate this, Cooperative Multi- Point (CoMP) joint processing algorithms are employed for the uni and broad-cast PHY transmissions. To practically implement this, the integrated network controller is assumed to have access to traffic statistics in terms of content popularity. Simulation results are provided to assess the gain in terms of total spectral efficiency. A conventional system, where the two networks operate independently, is used as benchmark [less ▲]

Detailed reference viewed: 152 (7 UL)
Full Text
Peer Reviewed
See detailTo AND or To OR: On Energy-Efficient Distributed Spectrum Sensing with Combined Censoring and Sleeping
Maleki, Sina UL; Leus, Geert; Chatzinotas, Symeon UL et al

in IEEE Transactions on Wireless Communications (2015)

Distributed spectrum sensing improves the detection reliability of a cognitive radio network but generally comes at the price of a large power consumption. Since cognitive radios are generally low-power ... [more ▼]

Distributed spectrum sensing improves the detection reliability of a cognitive radio network but generally comes at the price of a large power consumption. Since cognitive radios are generally low-power sensors with limited batteries, a combined censoring and sleeping scheme is considered as an energy-efficient algorithm for distributed spectrum sensing. Each sensor switches off its sensing module with a specific sleeping rate. When the sensor is on, a censoring policy is employed in order to send the sensing result to the fusion center. The result is only transmitted, if it is deemed to be informative. Hence, the energy consumption of each sensor, including the sensing and transmission energies, is reduced. The underlying sensing parameters are derived by minimizing the maximum average energy consumption per sensor subject to a lower-bound on the global probability of detection and an upper-bound on the global probability of false alarm. We analyze the problem for the OR and the AND rule and provide a performance analysis for a case study based on the IEEE 802.15.4/ZigBee standard. It is shown that the combined censoring and sleeping scheme achieves a significant energy saving compared to the case where no censoring or sleeping is taken into account. [less ▲]

Detailed reference viewed: 230 (8 UL)
Full Text
Peer Reviewed
See detailFeasibility Study of Full-duplex Relaying in Satellite Networks
Shankar, Bhavani UL; Zheng, Gan; Maleki, Sina UL et al

in The 16th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2015 (2015)

With the successful implementation of full-duplex radio prototypes, traditional orthogonal half-duplex communications is deemed to be inefficient in certain terrestrial applications. While full-duplex ... [more ▼]

With the successful implementation of full-duplex radio prototypes, traditional orthogonal half-duplex communications is deemed to be inefficient in certain terrestrial applications. While full-duplex techniques are gaining interest in terrestrial communications, thanks to the trend of short-distance and low-power transmissions, their application to satellite communications has drawn little attention. Motivated by this, the paper explores the use of the full-duplex relaying operation on-board the satellite in a DVB-S2 compliant network. Self-interference, whose management is the key component of a full-duplex communication, is the focus of study in this paper. Modelling the effects of self-interference and power amplifier nonlinearities on the quality of the received signal in undertaken. Subsequently, closed-form expressions for the various interference components are derived. The numerical evaluations of derived expressions rely on realistic link budgets and indicate substantial gains in spectral efficiency when self-interference can be well calibrated and measured. This confirms that the satellite full-duplex communications could be a promising solution for the efficient use of satellite spectrum, at least from a technical point of view. [less ▲]

Detailed reference viewed: 191 (4 UL)
Full Text
Peer Reviewed
See detailPerformance Analysis of Noncoherent Frame Synchronization in Satellite Communications with Frequency Uncertainty
Mazzali, Nicolò UL; Stante, Giulio; Shankar, Bhavani UL et al

in Proceedings on IEEE Symposium on Communications and Vehicular Technology in the Benelux (2015)

Emerging markets for satellite communications, such as maritime and aeronautical applications, require reliable communications in challenging conditions and low-cost, low-complexity user terminals. In ... [more ▼]

Emerging markets for satellite communications, such as maritime and aeronautical applications, require reliable communications in challenging conditions and low-cost, low-complexity user terminals. In this perspective, the synchronization chain is a fundamental part of the receiver and its design becomes critical. When a quadricorrelator is used as first stage of the synchronization chain so as to perform a coarse frequency estimation, then a noncoherent algorithm for frame synchronization can be employed as second stage. A popular algorithm for frame synchronization is the post-detection integration (PDI), whose noncoherent version (NCPDI) is quite robust to non-negligible residual frequency offsets. In this paper we will assess the performance of NCPDI in presence of a random uniformly-distributed residual frequency offset and will find a closed-form approximation for the probability of detection. Moreover, we will optimize the design parameters of NCPDI for a system operating at very low signal-to-noise ratio (VL-SNR) by minimizing the probability of missed detection for a given probability of false alarm. [less ▲]

Detailed reference viewed: 191 (8 UL)
Full Text
Peer Reviewed
See detailDistributed Coordinated Beamforming for Multi-cell Multigroup Multicast Systems
Pennanen, Harri UL; Christopoulos, Dimitrios; Chatzinotas, Symeon UL et al

in arXiv preprint arXiv:1511.03561 (2015)

This paper considers coordinated multicast beamforming in a multi-cell wireless network. Each multiantenna base station (BS) serves multiple groups of single antenna users by generating a single beam with ... [more ▼]

This paper considers coordinated multicast beamforming in a multi-cell wireless network. Each multiantenna base station (BS) serves multiple groups of single antenna users by generating a single beam with common data per group. The aim is to minimize the sum power of BSs while satisfying user-specific SINR targets. We propose centralized and distributed multicast beamforming algorithms for multi-cell multigroup systems. The NP-hard multicast problem is tackled by approximating it as a convex problem using the standard semidefinite relaxation method. The resulting semidefinite program (SDP) can be solved via centralized processing if global channel knowledge is available. To allow a distributed implementation, the primal decomposition method is used to turn the SDP into two optimization levels. The higher level is in charge of optimizing inter-cell interference while the lower level optimizes beamformers for given inter-cell interference constraints. The distributed algorithm requires local channel knowledge at each BS and scalar information exchange between BSs. If the solution has unit rank, it is optimal for the original problem. Otherwise, the Gaussian randomization method is used to find a feasible solution. The superiority of the proposed algorithms over conventional schemes is demonstrated via numerical evaluation. [less ▲]

Detailed reference viewed: 161 (9 UL)
Full Text
Peer Reviewed
See detailEnsemble of Example-Dependent Cost-Sensitive Decision Trees
Bahnsen, Alejandro Correa; Aouada, Djamila UL; Ottersten, Björn UL

in arXiv preprint arXiv:1505.04637 (2015)

Several real-world classification problems are example-dependent cost-sensitive in nature, where the costs due to misclassification vary between examples and not only within classes. However, standard ... [more ▼]

Several real-world classification problems are example-dependent cost-sensitive in nature, where the costs due to misclassification vary between examples and not only within classes. However, standard classification methods do not take these costs into account, and assume a constant cost of misclassification errors. In previous works, some methods that take into account the financial costs into the training of different algorithms have been proposed, with the example-dependent cost-sensitive decision tree algorithm being the one that gives the highest savings. In this paper we propose a new framework of ensembles of example-dependent cost-sensitive decision-trees. The framework consists in creating different example-dependent cost-sensitive decision trees on random subsamples of the training set, and then combining them using three different combination approaches. Moreover, we propose two new cost-sensitive combination approaches; cost-sensitive weighted voting and cost-sensitive stacking, the latter being based on the cost-sensitive logistic regression method. Finally, using five different databases, from four real-world applications: credit card fraud detection, churn modeling, credit scoring and direct marketing, we evaluate the proposed method against state-of-the-art example-dependent cost-sensitive techniques, namely, cost-proportionate sampling, Bayes minimum risk and cost-sensitive decision trees. The results show that the proposed algorithms have better results for all databases, in the sense of higher savings. [less ▲]

Detailed reference viewed: 109 (12 UL)