Search

This paper considers the capacity limit of an uplink wideband CDMA system. In order to make the studied results useful for the performance assessment of real cellular networks, various realistic assumptions are included in the problem. An equivalent discrete-time channel model is derived based on sufcient statistic for optimal decoding of transmit messages. The capacity regions are then characterized considering nite constellation and Gaussian input assumptions. For further insight, an analysis on the asymptotic capacity is considered, in which the conditions to simultaneously achieve the individual capacities are derived., QC 20150608

PHY layer authentication of a wireless sender has gained much interest recently. In this paper, we consider the famous Alice, Bob and Eve model and investigate (for the first time) the feasibility of using time-varying clock offsets for sender-node-authentication at Bob. Specifically, we exploit the fact (and de-facto problem) that clock offset between every node pair is unique; moreover, the two clock offsets between any two node pairs drift independently and randomly over time. Therefore, an explicit mechanism is needed to track the time-varying clock offsets. To this end, we model oscillator drift as brownian motion frequency and phase drift, and present a novel framework which is based on interplay between a hypothesis testing device and a bank of two Kaiman filters; one KF (KFh0) tracks Alice's clock while other KF (KFh1) tracks Eve's clock. Building on aforementioned framework, we then propose a novel sender-node-authentication method (so-called MHF method) by means of which Bob can automatically accept (reject) a received packet if it is sent by Alice (Eve). Finally, simulation results are presented which corroborate the efficiency of the proposed method., QC 20150223

We investigate the issue of distributed receiver cooperation in a multiple-relay network with memoryless independent fading channels, where the channel state information can't be obtained. The received signals at distributed receiving nodes are first compressed or quantized before being sent to the decoder via rate-limited cooperation channels for joint processing. We focus on the low SNR regime and analyze the capacity bounds using network equivalence theory and a multiple-layer binning peaky frequency shift keying (FSK). When the received signals at the relaying nodes are in low SNR regime and the cooperation rates are not sufficiently high, compressed/quantized observations at relaying nodes become useless and only decoding can help., This paper has been accepted for publication in IEEE GlobalSIP, Atlanta, GA, Dec. 2014.QC20141209

The framework of network equivalence theory developed by Koetter et al. introduces a notion of channel emulation to construct noiseless networks as upper/lower bounding models for the original noisy network. This paper presents scalable upper bounding models for wireless networks, by firstly extending the ``one-shot'' bounding models developed by Calmon et al. and then integrating them with network equivalence tools. A channel decoupling method is proposed to decompose wireless networks into decoupled multiple-access channels (MACs) and broadcast channels (BCs). The main advantages of the proposed method is its simplicity and the fact that it can be extended easily to large networks with a complexity that grows linearly with the number of nodes. It is demonstrated that the resulting upper bounds can approach the capacity in some setups., QC 20140619, VR International Postdoc

In this paper we generalize analytical performance models for proportional fair scheduling in OFDMA/LTE networks. We address the issue of modelling multiple fading interferers present in practical deployments. Specifically, we elaborate on the stochastic modelling of SINR-distribution for which we derive the rate expectation of instantaneously scheduled resources. The resulting analytical performance model is validated by means of simulations considering realistic network deployments. Compared with related work, our model demonstrates a significantly higher accuracy for long-term rate estimation. We illustrate the utility of such high-precision models by studying the impact on terminal assignment in fractional frequency reuse. Simply by using a suitable estimation model, cell-edge throughput can be improved up to 50%., QC 20150123

Speech is the most important communication modality for human interaction. Automatic speech recognition and speech synthesis have extended further the relevance of speech to man-machine interaction. Environment noise and various distortions, such as reverberation and speech processing artifacts, reduce the mutual information between the message modulated inthe clean speech and the message decoded from the observed signal. This degrades intelligibility and perceived quality, which are the two attributes associated with quality of service. An estimate of the state of these attributes provides important diagnostic information about the communication equipment and the environment. When the adverse effects occur at the presentation side, an objective measure of intelligibility facilitates speech signal modification for improved communication. The contributions of this thesis come from non-intrusive quality assessment and intelligibility-enhancing modification of speech. On the part of quality, the focus is on predictor design for limited training data. Paper A proposes a quality assessment model for bounded-support ratings that learns efficiently from a limited amount of training data, scales easily with the sampling frequency, and provides a platform for modeling variations in the individual subjective ratings. The predictive performance of the model for the mean of the subjective quality ratings compares favorably to the state-of-art in the field. Patterns in the spread of the individual ratings are captured in the feature space of the training data. Paper B focuses on enhancing predictive performance for the mean of the quality variable when the signal feature space is sparsely sampled by the training data. Using a Gaussian Processes framework, the deterministic signal-based feature set is augmented with a stochastic feature that is hypothesized to be jointly distributed with the target quality rating. An uncertainty propagation mechanism ensures that the variance of, QC 20140523

PHY layer authentication of a wireless sender has gained much interest recently. In this paper, we consider the famous Alice, Bob and Eve model and investigate (for the first time) the feasibility of using time-varying clock offsets for sender-node-authentication at Bob. Specifically, we exploit the fact (and de-facto problem) that clock offset between every node pair is unique; moreover, the two clock offsets between any two node pairs drift independently and randomly over time. Therefore, an explicit mechanism is needed to track the time-varying clock offsets. To this end, we model oscillator drift as brownian motion frequency and phase drift, and present a novel framework which is based on interplay between a hypothesis testing device and a bank of two Kaiman filters; one KF (KFh0) tracks Alice's clock while other KF (KFh1) tracks Eve's clock. Building on aforementioned framework, we then propose a novel sender-node-authentication method (so-called MHF method) by means of which Bob can automatically accept (reject) a received packet if it is sent by Alice (Eve). Finally, simulation results are presented which corroborate the efficiency of the proposed method., QC 20150223

We investigate the issue of distributed receiver cooperation in a multiple-relay network with memoryless independent fading channels, where the channel state information can't be obtained. The received signals at distributed receiving nodes are first compressed or quantized before being sent to the decoder via rate-limited cooperation channels for joint processing. We focus on the low SNR regime and analyze the capacity bounds using network equivalence theory and a multiple-layer binning peaky frequency shift keying (FSK). When the received signals at the relaying nodes are in low SNR regime and the cooperation rates are not sufficiently high, compressed/quantized observations at relaying nodes become useless and only decoding can help., This paper has been accepted for publication in IEEE GlobalSIP, Atlanta, GA, Dec. 2014.QC20141209

In this paper, we propose a new algorithm for recovery of low-rank matrices from compressed linear measurements. The underlying idea of this algorithm is to closely approximate the rank function with a smooth function of singular values, and then minimize the resulting approximation subject to the linear constraints. The accuracy of the approximation is controlled via a scaling parameter δ, where a smaller δ corresponds to a more accurate fitting. The consequent optimization problem for any finite δ is nonconvex. Therefore, to decrease the risk of ending up in local minima, a series of optimizations is performed, starting with optimizing a rough approximation (a large δ) and followed by successively optimizing finer approximations of the rank with smaller δ's. To solve the optimization problem for any δ > 0, it is converted to a new program in which the cost is a function of two auxiliary positive semidefinite variables. The paper shows that this new program is concave and applies a majorize-minimize technique to solve it which, in turn, leads to a few convex optimization iterations. This optimization scheme is also equivalent to a reweighted Nuclear Norm Minimization (NNM). For any δ > 0, we derive a necessary and sufficient condition for the exact recovery which are weaker than those corresponding to NNM. On the numerical side, the proposed algorithm is compared to NNM and a reweighted NNM in solving affine rank minimization and matrix completion problems showing its considerable and consistent superiority in terms of success rate., QC 20150417

In this paper we generalize analytical performance models for proportional fair scheduling in OFDMA/LTE networks. We address the issue of modelling multiple fading interferers present in practical deployments. Specifically, we elaborate on the stochastic modelling of SINR-distribution for which we derive the rate expectation of instantaneously scheduled resources. The resulting analytical performance model is validated by means of simulations considering realistic network deployments. Compared with related work, our model demonstrates a significantly higher accuracy for long-term rate estimation. We illustrate the utility of such high-precision models by studying the impact on terminal assignment in fractional frequency reuse. Simply by using a suitable estimation model, cell-edge throughput can be improved up to 50%., QC 20150123

The framework of network equivalence theory developed by Koetter et al. introduces a notion of channel emulation to construct noiseless networks as upper/lower bounding models for the original noisy network. This paper presents scalable upper bounding models for wireless networks, by firstly extending the ``one-shot'' bounding models developed by Calmon et al. and then integrating them with network equivalence tools. A channel decoupling method is proposed to decompose wireless networks into decoupled multiple-access channels (MACs) and broadcast channels (BCs). The main advantages of the proposed method is its simplicity and the fact that it can be extended easily to large networks with a complexity that grows linearly with the number of nodes. It is demonstrated that the resulting upper bounds can approach the capacity in some setups., QC 20140619, VR International Postdoc

PHY layer authentication of a wireless sender has gained much interest recently. In this paper, we consider the famous Alice, Bob and Eve model and investigate (for the first time) the feasibility of using time-varying clock offsets for sender-node-authentication at Bob. Specifically, we exploit the fact (and de-facto problem) that clock offset between every node pair is unique; moreover, the two clock offsets between any two node pairs drift independently and randomly over time. Therefore, an explicit mechanism is needed to track the time-varying clock offsets. To this end, we model oscillator drift as brownian motion frequency and phase drift, and present a novel framework which is based on interplay between a hypothesis testing device and a bank of two Kaiman filters; one KF (KFh0) tracks Alice's clock while other KF (KFh1) tracks Eve's clock. Building on aforementioned framework, we then propose a novel sender-node-authentication method (so-called MHF method) by means of which Bob can automatically accept (reject) a received packet if it is sent by Alice (Eve). Finally, simulation results are presented which corroborate the efficiency of the proposed method., QC 20150223

We investigate the issue of distributed receiver cooperation in a multiple-relay network with memoryless independent fading channels, where the channel state information can't be obtained. The received signals at distributed receiving nodes are first compressed or quantized before being sent to the decoder via rate-limited cooperation channels for joint processing. We focus on the low SNR regime and analyze the capacity bounds using network equivalence theory and a multiple-layer binning peaky frequency shift keying (FSK). When the received signals at the relaying nodes are in low SNR regime and the cooperation rates are not sufficiently high, compressed/quantized observations at relaying nodes become useless and only decoding can help., This paper has been accepted for publication in IEEE GlobalSIP, Atlanta, GA, Dec. 2014.QC20141209

In this paper we generalize analytical performance models for proportional fair scheduling in OFDMA/LTE networks. We address the issue of modelling multiple fading interferers present in practical deployments. Specifically, we elaborate on the stochastic modelling of SINR-distribution for which we derive the rate expectation of instantaneously scheduled resources. The resulting analytical performance model is validated by means of simulations considering realistic network deployments. Compared with related work, our model demonstrates a significantly higher accuracy for long-term rate estimation. We illustrate the utility of such high-precision models by studying the impact on terminal assignment in fractional frequency reuse. Simply by using a suitable estimation model, cell-edge throughput can be improved up to 50%., QC 20150123

Since wireless communication has become a standard feature in the daily life, smartphones and tablets among other things are integrated with the Bluetooth technology. While in some parts of the day wireless communication can be used for searching the internet and share information on social networks without the need of having a secure connection, there are some other parts where the security might become of high importance. When the technology gets integrated in companies the security problem becomes more evident. This is because when the radio signals spread in the medium they can be accessed by anyone that is in reach in the network and the information that was sent may not be intended for everyone. To secure the network from unintended users becomes important when handling fragile information, which companies may deal with daily. This paper gives an introduction on which security features and techniques that already exist in some personal area networks. From this it has been clear that a security feature could be implemented on the baseband layer of Bluetooth to increase the secrecy during the transmission since at the moment security is only implemented on higher layers using encryption algorithms. This paper proposes a conceptual idea of improving the secrecy in the network by using a wiretap code that is implemented before the error-correction coding in the Bluetooth's baseband. By disabling the ARQ scheme in Bluetooth one can modulate the channel as a Packet Erasure Channel that will lose packet with a certain probability. By using a nested code structure, the message can then be securely sent by using a higher rate than what the eavesdropper can recover due to the amount of errors the received signal will have. The performance of the concept is evaluated with the secrecy throughput, secrecy outage and the leakage.

In this paper, we focus on a Cognitive Relay Network (CRN) where only the information of the average channel gains is available at the secondary source. A distributed multiple relay selection scheme is proposed for the secondary transmission. Since the secondary source does not fully control the selection but lets each secondary relay decide by itself, it doesn't know which and how many relays will participate in the forwarding. This brings a serious uncertainty in the number of forwarding relays. We analyze this uncertainty and derive the closed-form expression of the outage probability of the secondary transmission under the uncertainty. Finally, we evaluate the proposed relay selection scheme by means of simulation. The simulation shows the appropriateness of our analytical model. In addition, although having lower interference constraint at each relay, the performances of the proposed multi-relay CRN scheme are shown to be strictly superior to the single relay CRN. Moreover, the simulation results suggest that it is necessary to spend the general overhead for secondary networks on knowing the exact QoS requirement of the primary user (i.e. the interference violation probability) to apply our proposed relay selection scheme with the statistical total interference constraint., QC 20150123

Jamming is a well-known reliability threat for mass-market wireless networks. With the rise of safety-critical applications this is likely to become a constraining issue in the future. Thus, the design of accurate jamming detection algorithms becomes important to react to ongoing jamming attacks. With respect to experimental work, jamming detection has been mainly studied for sensor networks. However, many safety-critical applications are also likely to run over 802.11-based networks where the proposed approaches do not carry over. In this paper we present a jamming detection approach for 802.11 networks. It uses metrics that are accessible through standard device drivers and performs detection via machine learning. While it allows for stand-alone operation, it also enables cooperative detection. We experimentally show that our approach achieves remarkably high detection rates in indoor and mobile outdoor scenarios even under challenging link conditions., QC 20150123

Parallel and distributed simulation have been extensively researched for a long time. Nevertheless, many simulation models are still executed sequentially. We attribute this to the fact that many of those models are simply not capable of being executed in parallel since they violate particular constraints. In this paper, we analyze the INET model suite, which enables network simulation in OMNeT++ with regard to parallelizability. We uncovered several issues preventing parallel execution of INET models. We analyzed those issues and developed solutions allowing INET models to be run in parallel. A case study shows the feasibility of our approach. Though there are parts of the model suite that we didn’t investigate yet and the performance can still be improved, the results show parallelization speedup for most configurations. The source code of our implementation is available through our web site at code.comsys.rwth-aachen.de, QC 20150123

Jamming is a well-known reliability threat for mass-market wireless networks. With the rise of safety-critical applications this is likely to become a constraining issue in the future. Thus, the design of accurate jamming detection algorithms becomes important to react to ongoing jamming attacks. With respect to experimental work, jamming detection has been mainly studied for sensor networks. However, many safety-critical applications are also likely to run over 802.11-based networks where the proposed approaches do not carry over. In this paper we present a jamming detection approach for 802.11 networks. It uses metrics that are accessible through standard device drivers and performs detection via machine learning. While it allows for stand-alone operation, it also enables cooperative detection. We experimentally show that our approach achieves remarkably high detection rates in indoor and mobile outdoor scenarios even under challenging link conditions., QC 20150123

Parallel and distributed simulation have been extensively researched for a long time. Nevertheless, many simulation models are still executed sequentially. We attribute this to the fact that many of those models are simply not capable of being executed in parallel since they violate particular constraints. In this paper, we analyze the INET model suite, which enables network simulation in OMNeT++ with regard to parallelizability. We uncovered several issues preventing parallel execution of INET models. We analyzed those issues and developed solutions allowing INET models to be run in parallel. A case study shows the feasibility of our approach. Though there are parts of the model suite that we didn’t investigate yet and the performance can still be improved, the results show parallelization speedup for most configurations. The source code of our implementation is available through our web site at code.comsys.rwth-aachen.de, QC 20150123

In this paper, we focus on a Cognitive Relay Network (CRN) where only the information of the average channel gains is available at the secondary source. A distributed multiple relay selection scheme is proposed for the secondary transmission. Since the secondary source does not fully control the selection but lets each secondary relay decide by itself, it doesn't know which and how many relays will participate in the forwarding. This brings a serious uncertainty in the number of forwarding relays. We analyze this uncertainty and derive the closed-form expression of the outage probability of the secondary transmission under the uncertainty. Finally, we evaluate the proposed relay selection scheme by means of simulation. The simulation shows the appropriateness of our analytical model. In addition, although having lower interference constraint at each relay, the performances of the proposed multi-relay CRN scheme are shown to be strictly superior to the single relay CRN. Moreover, the simulation results suggest that it is necessary to spend the general overhead for secondary networks on knowing the exact QoS requirement of the primary user (i.e. the interference violation probability) to apply our proposed relay selection scheme with the statistical total interference constraint., QC 20150123

Parallelization is crucial for ecient execution of large-scale network simulation. Today's computing clusters commonly used for that purpose are built from a large amount of multiprocessor machines. The traditional approach to utilize all CPU cores in such a system is to partition the network and distribute the partitions to the cores. This, however, does not incorporate the presence of shared memory into the design, such that messages between partitions on the same computing node have to be serialized and synchronization becomes more complex. In this paper, we present an approach that combines the shared-memory parallelization scheme Horizon [9] with the standard approach to distributed simulation to leverage the strengths of today's computing clusters. To further reduce the synchronization overhead, we introduce a novel synchronization algorithm that takes domain knowledge into account to reduce the number of synchronization points. In a case study with a UMTS LTE model, we show that both contributions combined enable much higher scalability achieving almost linear speedup when simulating 1,536 LTE cells on 1,536 CPU cores., QC 20150123

The present paper studies the asymptotic performance of dual-hop amplify-and-forward multiple-input multiple-output relay communication systems. In the corresponding setup, a relay amplifies the signal received from a source, retransmitting it towards a destination, while the direct source-destination link is absent. Ergodic achievable rates under separate decoding, along with the average bit error rate under various detection schemes are derived in the regime where the number of antennas at each terminal grows without bound. To overcome the mathematical difficulty of averaging over both channel realizations and input signals we apply large-system analysis based on the replica method from statistical physics. The validity of the large-system analysis is further verified through Monte Carlo simulations, providing particularly good accuracy at low SNR., QC 20140901

The present paper studies the asymptotic performance of dual-hop amplify-and-forward multiple-input multiple-output relay communication systems. In the corresponding setup, a relay amplifies the signal received from a source, retransmitting it towards a destination, while the direct source-destination link is absent. Ergodic achievable rates under separate decoding, along with the average bit error rate under various detection schemes are derived in the regime where the number of antennas at each terminal grows without bound. To overcome the mathematical difficulty of averaging over both channel realizations and input signals we apply large-system analysis based on the replica method from statistical physics. The validity of the large-system analysis is further verified through Monte Carlo simulations, providing particularly good accuracy at low SNR., QC 20140901

Parallelization is crucial for ecient execution of large-scale network simulation. Today's computing clusters commonly used for that purpose are built from a large amount of multiprocessor machines. The traditional approach to utilize all CPU cores in such a system is to partition the network and distribute the partitions to the cores. This, however, does not incorporate the presence of shared memory into the design, such that messages between partitions on the same computing node have to be serialized and synchronization becomes more complex. In this paper, we present an approach that combines the shared-memory parallelization scheme Horizon [9] with the standard approach to distributed simulation to leverage the strengths of today's computing clusters. To further reduce the synchronization overhead, we introduce a novel synchronization algorithm that takes domain knowledge into account to reduce the number of synchronization points. In a case study with a UMTS LTE model, we show that both contributions combined enable much higher scalability achieving almost linear speedup when simulating 1,536 LTE cells on 1,536 CPU cores., QC 20150123

Speech is the most important communication modality for human interaction. Automatic speech recognition and speech synthesis have extended further the relevance of speech to man-machine interaction. Environment noise and various distortions, such as reverberation and speech processing artifacts, reduce the mutual information between the message modulated inthe clean speech and the message decoded from the observed signal. This degrades intelligibility and perceived quality, which are the two attributes associated with quality of service. An estimate of the state of these attributes provides important diagnostic information about the communication equipment and the environment. When the adverse effects occur at the presentation side, an objective measure of intelligibility facilitates speech signal modification for improved communication. The contributions of this thesis come from non-intrusive quality assessment and intelligibility-enhancing modification of speech. On the part of quality, the focus is on predictor design for limited training data. Paper A proposes a quality assessment model for bounded-support ratings that learns efficiently from a limited amount of training data, scales easily with the sampling frequency, and provides a platform for modeling variations in the individual subjective ratings. The predictive performance of the model for the mean of the subjective quality ratings compares favorably to the state-of-art in the field. Patterns in the spread of the individual ratings are captured in the feature space of the training data. Paper B focuses on enhancing predictive performance for the mean of the quality variable when the signal feature space is sparsely sampled by the training data. Using a Gaussian Processes framework, the deterministic signal-based feature set is augmented with a stochastic feature that is hypothesized to be jointly distributed with the target quality rating. An uncertainty propagation mechanism ensures that the variance of, QC 20140523

The present paper studies the asymptotic performance of dual-hop amplify-and-forward multiple-input multiple-output relay communication systems. In the corresponding setup, a relay amplifies the signal received from a source, retransmitting it towards a destination, while the direct source-destination link is absent. Ergodic achievable rates under separate decoding, along with the average bit error rate under various detection schemes are derived in the regime where the number of antennas at each terminal grows without bound. To overcome the mathematical difficulty of averaging over both channel realizations and input signals we apply large-system analysis based on the replica method from statistical physics. The validity of the large-system analysis is further verified through Monte Carlo simulations, providing particularly good accuracy at low SNR., QC 20140901

In this paper, we focus on a Cognitive Relay Network (CRN) where only the information of the average channel gains is available at the secondary source. A distributed multiple relay selection scheme is proposed for the secondary transmission. Since the secondary source does not fully control the selection but lets each secondary relay decide by itself, it doesn't know which and how many relays will participate in the forwarding. This brings a serious uncertainty in the number of forwarding relays. We analyze this uncertainty and derive the closed-form expression of the outage probability of the secondary transmission under the uncertainty. Finally, we evaluate the proposed relay selection scheme by means of simulation. The simulation shows the appropriateness of our analytical model. In addition, although having lower interference constraint at each relay, the performances of the proposed multi-relay CRN scheme are shown to be strictly superior to the single relay CRN. Moreover, the simulation results suggest that it is necessary to spend the general overhead for secondary networks on knowing the exact QoS requirement of the primary user (i.e. the interference violation probability) to apply our proposed relay selection scheme with the statistical total interference constraint., QC 20150123

Parallel and distributed simulation have been extensively researched for a long time. Nevertheless, many simulation models are still executed sequentially. We attribute this to the fact that many of those models are simply not capable of being executed in parallel since they violate particular constraints. In this paper, we analyze the INET model suite, which enables network simulation in OMNeT++ with regard to parallelizability. We uncovered several issues preventing parallel execution of INET models. We analyzed those issues and developed solutions allowing INET models to be run in parallel. A case study shows the feasibility of our approach. Though there are parts of the model suite that we didn’t investigate yet and the performance can still be improved, the results show parallelization speedup for most configurations. The source code of our implementation is available through our web site at code.comsys.rwth-aachen.de, QC 20150123

Parallelization is crucial for ecient execution of large-scale network simulation. Today's computing clusters commonly used for that purpose are built from a large amount of multiprocessor machines. The traditional approach to utilize all CPU cores in such a system is to partition the network and distribute the partitions to the cores. This, however, does not incorporate the presence of shared memory into the design, such that messages between partitions on the same computing node have to be serialized and synchronization becomes more complex. In this paper, we present an approach that combines the shared-memory parallelization scheme Horizon [9] with the standard approach to distributed simulation to leverage the strengths of today's computing clusters. To further reduce the synchronization overhead, we introduce a novel synchronization algorithm that takes domain knowledge into account to reduce the number of synchronization points. In a case study with a UMTS LTE model, we show that both contributions combined enable much higher scalability achieving almost linear speedup when simulating 1,536 LTE cells on 1,536 CPU cores., QC 20150123

Jamming is a well-known reliability threat for mass-market wireless networks. With the rise of safety-critical applications this is likely to become a constraining issue in the future. Thus, the design of accurate jamming detection algorithms becomes important to react to ongoing jamming attacks. With respect to experimental work, jamming detection has been mainly studied for sensor networks. However, many safety-critical applications are also likely to run over 802.11-based networks where the proposed approaches do not carry over. In this paper we present a jamming detection approach for 802.11 networks. It uses metrics that are accessible through standard device drivers and performs detection via machine learning. While it allows for stand-alone operation, it also enables cooperative detection. We experimentally show that our approach achieves remarkably high detection rates in indoor and mobile outdoor scenarios even under challenging link conditions., QC 20150123

Parallel and distributed simulation have been extensively researched for a long time. Nevertheless, many simulation models are still executed sequentially. We attribute this to the fact that many of those models are simply not capable of being executed in parallel since they violate particular constraints. In this paper, we analyze the INET model suite, which enables network simulation in OMNeT++ with regard to parallelizability. We uncovered several issues preventing parallel execution of INET models. We analyzed those issues and developed solutions allowing INET models to be run in parallel. A case study shows the feasibility of our approach. Though there are parts of the model suite that we didn’t investigate yet and the performance can still be improved, the results show parallelization speedup for most configurations. The source code of our implementation is available through our web site at code.comsys.rwth-aachen.de, QC 20150123

In this paper, we focus on a Cognitive Relay Network (CRN) where only the information of the average channel gains is available at the secondary source. A distributed multiple relay selection scheme is proposed for the secondary transmission. Since the secondary source does not fully control the selection but lets each secondary relay decide by itself, it doesn't know which and how many relays will participate in the forwarding. This brings a serious uncertainty in the number of forwarding relays. We analyze this uncertainty and derive the closed-form expression of the outage probability of the secondary transmission under the uncertainty. Finally, we evaluate the proposed relay selection scheme by means of simulation. The simulation shows the appropriateness of our analytical model. In addition, although having lower interference constraint at each relay, the performances of the proposed multi-relay CRN scheme are shown to be strictly superior to the single relay CRN. Moreover, the simulation results suggest that it is necessary to spend the general overhead for secondary networks on knowing the exact QoS requirement of the primary user (i.e. the interference violation probability) to apply our proposed relay selection scheme with the statistical total interference constraint., QC 20150123

Jamming is a well-known reliability threat for mass-market wireless networks. With the rise of safety-critical applications this is likely to become a constraining issue in the future. Thus, the design of accurate jamming detection algorithms becomes important to react to ongoing jamming attacks. With respect to experimental work, jamming detection has been mainly studied for sensor networks. However, many safety-critical applications are also likely to run over 802.11-based networks where the proposed approaches do not carry over. In this paper we present a jamming detection approach for 802.11 networks. It uses metrics that are accessible through standard device drivers and performs detection via machine learning. While it allows for stand-alone operation, it also enables cooperative detection. We experimentally show that our approach achieves remarkably high detection rates in indoor and mobile outdoor scenarios even under challenging link conditions., QC 20150123

Parallelization is crucial for ecient execution of large-scale network simulation. Today's computing clusters commonly used for that purpose are built from a large amount of multiprocessor machines. The traditional approach to utilize all CPU cores in such a system is to partition the network and distribute the partitions to the cores. This, however, does not incorporate the presence of shared memory into the design, such that messages between partitions on the same computing node have to be serialized and synchronization becomes more complex. In this paper, we present an approach that combines the shared-memory parallelization scheme Horizon [9] with the standard approach to distributed simulation to leverage the strengths of today's computing clusters. To further reduce the synchronization overhead, we introduce a novel synchronization algorithm that takes domain knowledge into account to reduce the number of synchronization points. In a case study with a UMTS LTE model, we show that both contributions combined enable much higher scalability achieving almost linear speedup when simulating 1,536 LTE cells on 1,536 CPU cores., QC 20150123

The present paper studies the asymptotic performance of dual-hop amplify-and-forward multiple-input multiple-output relay communication systems. In the corresponding setup, a relay amplifies the signal received from a source, retransmitting it towards a destination, while the direct source-destination link is absent. Ergodic achievable rates under separate decoding, along with the average bit error rate under various detection schemes are derived in the regime where the number of antennas at each terminal grows without bound. To overcome the mathematical difficulty of averaging over both channel realizations and input signals we apply large-system analysis based on the replica method from statistical physics. The validity of the large-system analysis is further verified through Monte Carlo simulations, providing particularly good accuracy at low SNR., QC 20140901

Speech is the most important communication modality for human interaction. Automatic speech recognition and speech synthesis have extended further the relevance of speech to man-machine interaction. Environment noise and various distortions, such as reverberation and speech processing artifacts, reduce the mutual information between the message modulated inthe clean speech and the message decoded from the observed signal. This degrades intelligibility and perceived quality, which are the two attributes associated with quality of service. An estimate of the state of these attributes provides important diagnostic information about the communication equipment and the environment. When the adverse effects occur at the presentation side, an objective measure of intelligibility facilitates speech signal modification for improved communication. The contributions of this thesis come from non-intrusive quality assessment and intelligibility-enhancing modification of speech. On the part of quality, the focus is on predictor design for limited training data. Paper A proposes a quality assessment model for bounded-support ratings that learns efficiently from a limited amount of training data, scales easily with the sampling frequency, and provides a platform for modeling variations in the individual subjective ratings. The predictive performance of the model for the mean of the subjective quality ratings compares favorably to the state-of-art in the field. Patterns in the spread of the individual ratings are captured in the feature space of the training data. Paper B focuses on enhancing predictive performance for the mean of the quality variable when the signal feature space is sparsely sampled by the training data. Using a Gaussian Processes framework, the deterministic signal-based feature set is augmented with a stochastic feature that is hypothesized to be jointly distributed with the target quality rating. An uncertainty propagation mechanism ensures that the variance of, QC 20140523

In this paper, the privacy problem of a parallel distributed detection system vulnerable to an eavesdropper is proposed and studied in the Neyman-Pearson formulation. The privacy leakage is evaluated by a metric related to the Neyman-Pearson criterion. We will show that it is sufficient to consider a deterministic likelihood-ratio test for the optimal detection strategy at the eavesdropped sensor. This fundamental insight helps to simplify the problem to find the optimal privacy-constrained distributed detection system design. The trade-off between the detection performance and privacy leakage is illustrated in a numerical example., QC 20150122. QC 20160314

In this paper, eavesdropping in parallel distributed sequential detections is considered. The privacy risk is evaluated by the minimal achievable Bayesian risk of a greedy and informed eavesdropper who is curious about the hypothesis realization. We propose a novel metric based on Bayesian risk to take the detection performance and privacy risk with different weights into account. We formulate and study the privacy-concerned parallel distributed Bayesian sequential detection problem under a finite time-horizon assumption. Solving this problem will lead to the optimal distributed sequential detection design which achieves the minimal privacy-concerned Bayesian risk. The study shows that it is not sufficient to consider a deterministic likelihood-ratio test for a remote decision maker at an active time index in the optimal privacy-concerned system design. However, properties of the optimal design indicate that the standard method can be extended to solve the proposed problem., QC 20160209

In this paper, the privacy problem of a tandem distributed detection system vulnerable to an eavesdropper is proposed and studied in the Bayesian formulation. The privacy risk is evaluated by the detection cost of the eavesdropper which is assumed to be informed and greedy. For the sensors whose operations are constrained to suppress the privacy risk, it is shown that the optimal detection strategies are likelihood-ratio tests. This fundamental insight allows for the optimization to reuse known algorithms extended to incorporate the privacy constraint. The trade-off between the detection performance and privacy risk is illustrated in an example., QC 20150123. QC 20160314

In this paper, the differential privacy problem in parallel distributed detections is studied in the Bayesian formulation. The privacy risk is evaluated by the minimum detection cost for the fusion node to infer the private random phenomenon. Different from the privacy-unconstrained distributed Bayesian detection problem, the optimal operation point of a remote decision maker can be on the boundary of the privacy-unconstrained operation region or in the intersection of privacy constraint hyperplanes. Therefore, for a remote decision maker in the optimal privacy-constrained distributed detection design, it is sufficient to consider a deterministic linear likelihood combination test or a randomized decision strategy of two linear likelihood combination tests which achieves the optimal operation point in each case. Such an insight indicates that the existing algorithm can be reused by incorporating the privacy constraint. The trade-off between detection and privacy metrics will be illustrated in a numerical example., QC 20151211. QC 20160209

In this paper, eavesdropping in parallel distributed sequential detections is considered. The privacy risk is evaluated by the minimal achievable Bayesian risk of a greedy and informed eavesdropper who is curious about the hypothesis realization. We propose a novel metric based on Bayesian risk to take the detection performance and privacy risk with different weights into account. We formulate and study the privacy-concerned parallel distributed Bayesian sequential detection problem under a finite time-horizon assumption. Solving this problem will lead to the optimal distributed sequential detection design which achieves the minimal privacy-concerned Bayesian risk. The study shows that it is not sufficient to consider a deterministic likelihood-ratio test for a remote decision maker at an active time index in the optimal privacy-concerned system design. However, properties of the optimal design indicate that the standard method can be extended to solve the proposed problem., QC 20160209

In this paper, the privacy problem of a parallel distributed detection system vulnerable to an eavesdropper is proposed and studied in the Neyman-Pearson formulation. The privacy leakage is evaluated by a metric related to the Neyman-Pearson criterion. We will show that it is sufficient to consider a deterministic likelihood-ratio test for the optimal detection strategy at the eavesdropped sensor. This fundamental insight helps to simplify the problem to find the optimal privacy-constrained distributed detection system design. The trade-off between the detection performance and privacy leakage is illustrated in a numerical example., QC 20150122. QC 20160314

In this paper, the privacy problem of a tandem distributed detection system vulnerable to an eavesdropper is proposed and studied in the Bayesian formulation. The privacy risk is evaluated by the detection cost of the eavesdropper which is assumed to be informed and greedy. For the sensors whose operations are constrained to suppress the privacy risk, it is shown that the optimal detection strategies are likelihood-ratio tests. This fundamental insight allows for the optimization to reuse known algorithms extended to incorporate the privacy constraint. The trade-off between the detection performance and privacy risk is illustrated in an example., QC 20150123. QC 20160314

In this paper, the differential privacy problem in parallel distributed detections is studied in the Bayesian formulation. The privacy risk is evaluated by the minimum detection cost for the fusion node to infer the private random phenomenon. Different from the privacy-unconstrained distributed Bayesian detection problem, the optimal operation point of a remote decision maker can be on the boundary of the privacy-unconstrained operation region or in the intersection of privacy constraint hyperplanes. Therefore, for a remote decision maker in the optimal privacy-constrained distributed detection design, it is sufficient to consider a deterministic linear likelihood combination test or a randomized decision strategy of two linear likelihood combination tests which achieves the optimal operation point in each case. Such an insight indicates that the existing algorithm can be reused by incorporating the privacy constraint. The trade-off between detection and privacy metrics will be illustrated in a numerical example., QC 20151211. QC 20160209

In this paper, the privacy problem of a parallel distributed detection system vulnerable to an eavesdropper is proposed and studied in the Neyman-Pearson formulation. The privacy leakage is evaluated by a metric related to the Neyman-Pearson criterion. We will show that it is sufficient to consider a deterministic likelihood-ratio test for the optimal detection strategy at the eavesdropped sensor. This fundamental insight helps to simplify the problem to find the optimal privacy-constrained distributed detection system design. The trade-off between the detection performance and privacy leakage is illustrated in a numerical example., QC 20150122. QC 20160314

In this paper, the privacy problem of a tandem distributed detection system vulnerable to an eavesdropper is proposed and studied in the Bayesian formulation. The privacy risk is evaluated by the detection cost of the eavesdropper which is assumed to be informed and greedy. For the sensors whose operations are constrained to suppress the privacy risk, it is shown that the optimal detection strategies are likelihood-ratio tests. This fundamental insight allows for the optimization to reuse known algorithms extended to incorporate the privacy constraint. The trade-off between the detection performance and privacy risk is illustrated in an example., QC 20150123. QC 20160314

In this paper, eavesdropping in parallel distributed sequential detections is considered. The privacy risk is evaluated by the minimal achievable Bayesian risk of a greedy and informed eavesdropper who is curious about the hypothesis realization. We propose a novel metric based on Bayesian risk to take the detection performance and privacy risk with different weights into account. We formulate and study the privacy-concerned parallel distributed Bayesian sequential detection problem under a finite time-horizon assumption. Solving this problem will lead to the optimal distributed sequential detection design which achieves the minimal privacy-concerned Bayesian risk. The study shows that it is not sufficient to consider a deterministic likelihood-ratio test for a remote decision maker at an active time index in the optimal privacy-concerned system design. However, properties of the optimal design indicate that the standard method can be extended to solve the proposed problem., QC 20160209

In this paper, the differential privacy problem in parallel distributed detections is studied in the Bayesian formulation. The privacy risk is evaluated by the minimum detection cost for the fusion node to infer the private random phenomenon. Different from the privacy-unconstrained distributed Bayesian detection problem, the optimal operation point of a remote decision maker can be on the boundary of the privacy-unconstrained operation region or in the intersection of privacy constraint hyperplanes. Therefore, for a remote decision maker in the optimal privacy-constrained distributed detection design, it is sufficient to consider a deterministic linear likelihood combination test or a randomized decision strategy of two linear likelihood combination tests which achieves the optimal operation point in each case. Such an insight indicates that the existing algorithm can be reused by incorporating the privacy constraint. The trade-off between detection and privacy metrics will be illustrated in a numerical example., QC 20151211. QC 20160209

This paper addresses text-independent speaker identification (SI) based on line spectral frequencies (LSFs). The LSFs are transformed to differential LSFs (MLSF) in order to exploit their boundary and ordering properties. We show that the square root of MLSF has interesting directional characteristics implying that their distribution can be modeled by a mixture of von-Mises Fisher (vMF) distributions. We analytically estimate the mixture model parameters in a fully Bayesian treatment by using variational inference. In the Bayesian inference, we can potentially determine the model complexity and avoid overfitting problem associated with conventional approaches based on the expectation maximization. The experimental results confirm the effectiveness of the proposed SI system., QC 20140912