This paper proposes a dynamic orthogonal matching pursuit (OMP)-based localization for automotive radar systems. At each time instant, three dictionaries are designed based on prior information on targets’ approximate locations. We use mutual coherence as the criterion for designing each dictionary. The mutual coherence minimization problem over each dictionary is developed as a selection problem in the binary domain. Afterward, the OMP algorithm is used to perform the direction of arrival (DoA) estimation over each dictionary, and the estimated DoAs are fused together through averaging to obtain the final DoA estimation. We show that the proposed method significantly outperforms the uniform grid dictionary and improves localization accuracy.

This paper proposes to integrate reconfigurable intelligent surfaces (RISs) with unmanned aerial vehicles (UAVs) as a resilience mechanism to mitigate outages in UAV networks due to UAV and link failures. The inherent addition of RIS-aided links (UE-RIS-UAV links), combined with their reconfigurability, creates alternative paths for user equipment (UEs) to transmit signals to UAVs. The paper studies the problem of maximizing connectivity of UAV networks by jointly considering UE positioning, RIS-aided link selection, and phase shift design of RISs. To tackle it, we propose an efficient two-step solution. In the first step, we propose a supergradient method that locates the UEs in positions that improve their communication links until a certain connectivity threshold is satisfied. Given the optimized UE positioning, the second step jointly optimizes the RIS-aided link selection and RIS phase shift design using semidefinite programming (SDP). Through simulations, we illustrate the superiority of the proposed solution compared to the solutions available in the literature.

The widespread use of smartphones and Wi-Fi-enabled devices offers a promising alternative to vision-based crowd estimation methods by passively capturing Wi-Fi probe requests. However, recent techniques in MAC address randomization effectively prevent device tracking, ensuring better privacy protection while complicating traditional counting methods. This paper presents a machine learning-based approach for estimating crowd sizes using only the number of unique randomized MAC addresses observed over time, without any form of individual tracking. We treat the count of observed addresses as a time-series signal and apply MiniRocket to extract robust temporal features. These features, combined with statistical attributes of the signal, are used to train a machine learning model to classify crowd sizes. We validate our method across static, stable, and dynamic environments, achieving a Mean Absolute Percentage Error (MAPE) below 9% for dynamic crowds, 1% for static crowds, and 0.8% for stable real-world crowds, demonstrating accurate, scalable, and privacy-preserving crowd size estimation in near real-time.

As smartphones and other Wi-Fi-Enabled devices become increasingly common, they offer a practical way to estimate crowd size by collecting the probe request frames these devices periodically transmit. While this passive approach avoids relying on cameras or other intrusive sensors, the adoption of MAC address randomization—designed to protect user privacy—makes it difficult to reliably count how many unique devices are present. This paper presents a machine learning–based approach for device counting that leverages Channel State Information (CSI) features extracted from probe frames. Our method enables accurate device population estimation while preserving MAC-level privacy and avoiding persistent tracking or behavioral profiling. A Siamese neural network is trained to learn a discriminative similarity function between packet pairs, allowing the grouping of temporally co-occurring packets likely originating from the same device. To address the challenge of the limited number of packets available from each randomized MAC address, we use a lightweight augmentation strategy that interpolates between CSI samples to increase training density. We evaluate our approach in indoor and outdoor settings with varying device probing behaviors under static conditions. By leveraging packet-level voting and device-counting strategies, our model leverages packet-level voting and device-counting strategies to consistently achieve high counting accuracy—averaging over 98% in indoor environments and reaching 100% outdoors—while preserving device privacy by avoiding behavioral tracking and re-identification.

This paper proposes a model-based deep learning framework for Reconfigurable Intelligent Surface (RIS)-aided Orthogonal Frequency-Division Multiplexing (OFDM) radar interference mitigation without any assumption on the transmitted waveform. First, a modified Multiple Signal Classification (MUSIC) algorithm generates coarse estimates of target and interferer angles. These estimates feed a lightweight MLP that jointly optimizes RIS amplitude–phase settings via a custom loss balancing SINR enhancement with angle-estimation fidelity. A convolutional notch filtering step then deepens the null at the interference direction, and a subcarrier-pooling strategy fuses per-frequency spectra to ensure uniform suppression across the band. Simulation results under automotive radar conditions demonstrate robust interference mitigation, high angular resolution, and precise angle recovery in dense environments.

This paper addresses the limitations of GPS accuracy in vehicular networks, which is often insufficient for safety-critical applications. Cooperative positioning (CP) methods can mitigate GPS errors but encounter challenges in large networks due to high computational and communication complexity. To overcome these issues, we propose a star model CP method that reduces the observation vector size and an adaptive scheduling method that dynamically selects participating vehicles based on a safety-driven cost function, lowering computational and communication demands. This scheduler optimizes positioning frequency while reducing data exchange. Numerical evaluations show that our framework achieves comparable or improved positioning accuracy compared to methods that do not use scheduling, especially under high GPS error conditions.

Cooperative localization and tracking are expected to play a crucial role in supporting location-based services in 6G networks. This work shows that integrating reconfigurable intelligent surfaces (RISs) with sidelink communications between user equipments (UEs) can enhance tracking and localization accuracy in the absence of access points (APs). To achieve this, we consider a localization and tracking problem of RIS-assisted sidelink communications, where moving UEs are localized and tracked without relying on APs. Specifically, we first design orthogonal RIS phase shift vectors to separate RIS-aided (reflected) paths from direct sidelink communication paths at the receiving UE(s). The initial locations of the UEs are then derived from the estimated channel parameters, enabling the tracking of the UEs using an extended Kalman filter (EKF). We benchmark the performance of the localization with multiple RISs using the Cramér-Rao lower bound (CRLB), and we assess the EKF's performance using the root mean squared error (RMSE) metric. Simulation results indicate that the initial localization accuracy reaches the CRLB, and the EKF achieves an RMSE below 10 cm for 90% of the time.

The integration of reconfigurable intelligent surfaces (RISs) and unmanned aerial vehicle (UAV) communications has emerged as a promising solution for improving link quality and massive connectivity for beyond 5G wireless networks. This paper presents an innovative approach to maximizing connectivity of uplink multi-RIS-assisted UAV networks enabled by RIS placement and virtual partitioning, wherein RISs are deployed to assist in the communications between user-equipment (UEs) and UAVs. In the considered model, the UEs intend to transmit data to the UAVs, and RISs can assist in improving network connectivity by connecting the UEs to the blocked UAVs. First, exact and approximated closed-form (CF) expressions for signal-to-noise ratio (SNR) are derived based on aligned and non-aligned portions of the RISs. Then, we formulate the problem of maximizing the network connectivity that jointly considers 1) UE-RIS-UAV link selection and 2) RIS placement and virtual partitioning. This problem is a computationally expensive combinatorial optimization. Using the block coordinate descent (BCD) approach, we propose novel UE-RIS-UAV selection and RIS placement and partitioning methods. Specifically, we develop clustering and perturbation methods for UE-RIS-UAV selection, and derive a closed-form solution for the partitioning of the RISs. Moreover, for optimizing the RISs placement, Adam optimizer is used. Simulation results demonstrate that the proposed approaches yield a gain in the range of 12% to 45% compared to benchmark schemes. The finding emphasizes the potential of integrating RIS with UAV communications as a robust and reliable connectivity solution for future wireless communication systems.

Reconfigurable intelligent surfaces (RISs) are expected to be a main component of future 6G networks due to their capability to create a controllable wireless environment, achieve extended coverage, and improve localization accuracy. In this paper, we present a novel cooperative positioning use case of the RIS in mmWave frequencies and show that in the presence of RIS, together with sidelink communications, localization with zero access points (APs) is possible. We show that multiple (at least three) half-duplex single-antenna user equipments (UEs) can cooperatively estimate their positions through device-to-device communications with a single RIS as an anchor without the need for any APs. We start by formulating a three-dimensional positioning problem with Cramér-Rao lower bound (CRLB) derived for performance analysis. After that, we discuss the RIS profile design and the power allocation strategy between the UEs. Then, we propose low-complexity estimators for estimating the channel parameters and UEs’ positions. Finally, we evaluate the performance of the proposed estimators and RIS profiles in the considered scenario via extensive simulations and show that sub-meter level positioning accuracy can be achieved under multi-path propagation.

This paper studiesdirect localizationwhich is a technique used for positioning a source by directly searching within a planar grid. In this paper, we have developed an effective method involving sparse signal recovery using the$\ell _{0}$-pseudo-norm. Our novel contribution focuses on a cooperative direct localization technique that takes into account the ambiguity caused by reflections in the multipath environment. Specifically, we have factored in the distances between the source and the anchors using a mathematical relationship called theCayley-Mengerdeterminant. This determinant acts as a crucial component in deriving accurate range estimates. This relationship is added as an additional constraint to the sparse recovery problem which is NP-hard. To solve this NP-hard problem, we have employed a binary programming relaxation technique. Experiments reveal that our proposed approach significantly reduces localization errors when compared to traditional direct localization methods. This suggests that our cooperative direct localization method holds great potential for enhancing the accuracy and reliability of location estimation in challenging, multi-path scenarios.

This paper introduces an indoor localization method that utilizes fixed reflector objects within the environment, leveraging a base station (BS) or users equipped with Angle of Arrival (AoA) and Time of Arrival (ToA) measurement capabilities. The localization process consists of two phases. In the offline phase, using specific strategy effective reflector points within a specific region are identified. In the online phase, a maximization problem is solved to locate users based on BS measurements and information gathered during the offline phase. Through analysis and simulation results, we demonstrate that with the same number of training points, the performance of the proposed localization technique surpasses that of fingerprint-based techniques. Additionally, we show that localizing an unknown user does not require a large number of training points throughout the entire environment; it is sufficient to place these training points on the boundary of the environment. We introduce the reflectivity parameter ($n_{r}$), which quantifies the average number of first-order reflection paths from the transmitter to the receiver, and demonstrate its impact on localization accuracy. The log-scale accuracy ratio ($R_{a}$) is defined as the logarithmic function of the localization area divided by the localization ambiguity area, serving as an indicator of accuracy. We show that in scenarios where the Signal-to-Noise Ratio (SNR) approaches infinity, and without a line of sight (LoS) link,$R_{a}$is upper-bounded by$n_{r} \log _{2}\left ({{1 + \frac {\mathrm {Vol}({\mathcal {S}}_{A})}{\mathrm {Vol}({\mathcal {S}}_{\epsilon }({\mathcal {M}}_{s}))}}}\right)$, where$\mathrm {Vol}({\mathcal {S}}_{A})$and$\mathrm {Vol}({\mathcal {S}}_{\epsilon }({\mathcal {M}}_{s}))$represent the areas of the localization region and the area containing all reflector points with a probability of at least$1 - \epsilon $, respectively.

Advancements in machine learning, coupled with Wi-Fi sensing using channel state information (CSI), have emerged as a powerful approach for human activity and gesture recognition. Contrary to camera-based systems that rely on capturing images of individuals, passive Wi-Fi sensing provides a non-intrusive method for detecting and interpreting human movements without the need for capturing detailed images. In this paper, subcarrier masking contrastive transfer learning (SMCTL) is proposed for human gesture recognition from CSI measurements. The unsupervised feature extractor is pretrained on a large-scale dataset of CSI measurements. Then, the trained feature extractor and a classifier are fine-tuned end-to-end on a smaller dataset collected from a different environment for cross-domain adaptation. Performance of this method is evaluated for various numbers of fine-tuning samples, for which it outperforms baseline supervised and self-supervised deep learning models.

Time series classification is a challenging and essential task in many domains such as finance, healthcare, and industrial systems. Traditional methods often struggle with the complexity of multivariate time-series data and the need for effective feature selection. To address these issues, we propose a novel approach incorporating the Metropolis-Hastings algorithm within a Markov Chain Monte Carlo (MCMC) framework for optimized feature selection. Within the inner workings of the Metropolis-Hastings algorithm, we utilize convolutional kernels and randomized threshold exceedance rates for robust feature extraction. To further enrich the diversity and efficacy of the feature set, we seamlessly integrate Locality-Sensitive Hashing (LSH) techniques into the Metropolis-Hastings framework. As a result, our method not only sets a new standard for efficiency in time-series classification but also surpasses existing state-of-the-art methods across a diverse range of datasets.

This paper proposes an indoor localization strategy that utilizes fixed reflector points. The strategy assumes the presence of a base station (BS) with a known location, equipped with Angle of Arrival (AoA) and Time of Arrival (ToA) measurement capabilities. The localization process comprises two stages. In the initial stage, a systematic environmental scan identifies reflector points, facilitating the creation of a detailed map of their locations. Next, to achieve precise localization, a maximization problem is formulated. For each candidate location, the method assesses the likelihood of a set of measurements to the reflectors map. This approach accelerates user localization compared to fingerprint search algorithms. Simulation results provide compelling evidence for the effectiveness of this strategy in speeding up data gathering in the localization procedure.

In this paper, we study the problem of placing a reconfigurable intelligent surface (RIS) and tuning its reflected links to improve the resiliency and connectivity of uncrewed aerial vehicle (UAV) networks. We formulate an optimization problem of maximizing network connectivity that jointly optimizes RIS position, its phase shift, and UE-RIS-UAV link scheduling. Such problem is computationally expensive combinatorial optimization. To tackle this problem, we first design the phase control strategy at the RIS for the UE-RIS-UAV link scheduling. With such design, we propose an optimal linear search method, which has high computational complexity for large networks. Then, leveraging the convex relaxation method and the designed phase strategy, we propose another solution using semi-definite programming (SDP) optimization, which solves the problem in polynomial time. Simulation results show that our proposed solutions outperform other benchmark schemes.

Reconfigurable intelligent surface (RIS) is pivotal for beyond 5G networks in regards to the surge demand for reliable communication in unmanned aerial vehicle (UAV) networks. This paper presents an innovative approach to maximize connectivity of UAV networks using RIS deployment and virtual partitioning, wherein an RIS is deployed to assist in the communications between an user-equipment (UE) and blocked UAVs. Closed-form (CF) expressions for signal-to-noise ratio (SNR) of the two-UAV setup are derived and validated. Then, an optimization problem is formulated to maximize network connectivity by optimizing the 3D deployment of the RIS and its partitioning subject to predefined quality-of-service (QoS) constraints. To tackle this problem, we propose a method of virtually partitioning the RIS given a fixed 3D location, such that the partition phase shifts are configured to create cascaded channels between the UE and the blocked two UAVs. Then, simulated-annealing (SA) method is used to find the 3D location of the RIS. Simulation results demonstrate that the proposed joint RIS deployment and partitioning framework can significantly improve network connectivity compared to benchmarks, including RIS-free and RIS with a single narrow-beam link.

Reconfigurable intelligent surfaces (RISs) and sidelink communications have enabled radio positioning with zero access points (APs). In this paper, we study a positioning system with$K$single-antenna user equipments (UEs), a passive RIS, and zero APs. We show that$K$half-duplex UEs, under some blockage, can estimate their positions via two-way sidelink communications. We start by introducing the signal model. Then, we determine which paths are blocked using a deep neural network (DNN) and estimate the channel parameters. Next, we estimate the UEs' positions via one-dimensional search and trilateration. Finally, we evaluate the performance of the proposed algorithm through numerical studies.

Wireless local-area network (WLAN) sensing offers advantages over other approaches to human activity recognition (HAR) for Internet of Things (IoT) applications, including privacy as well as adaptability to non-line-of-sight scenarios. This is why HAR plays an important role in the upcoming IEEE 802.11bf Wi-Fi standard, which aims to bring the adoption of WLAN sensing to a much larger scale. In this paper, we propose CapsHAR, a model based on capsule networks, which uses channel state information (CSI) from Wi-Fi signals to accurately perform human activity recognition. We evaluate the capability of the model on a variety of datasets, including large and small-scale gestures, as well as compare its performance to a variety of models and approaches. We then extend the CapsHAR model into a distributed architecture in order to eliminate the communication overhead of sending CSI data from multiple access points (AP) to a single server. We propose the use of edge computing to run CapsHAR at each AP separately, then combine the outputs of the models through a Fresnel zone-based voting scheme which makes more efficient use of spatial diversity. Overall, the CapsHAR architecture consistently achieves classification accuracy surpassing that of the state-of-the-art models, demonstrating the viability of capsule networks for reliable HAR in Wi-Fi-based IoT applications.

Recently, fingerprint-based localization has become a promising solution for indoor positioning because of its great performance in complex multipath environments. However, the extensive time and labor effort of constructing the radio map has become the bottleneck that hinders the adaptation of fingerprint-based localization in practice. In this article, we propose a novel cost-efficient radio map construction scheme, which relies on the fingerprint measurements from only a small number of reference points (RPs) via graph signal sampling and recovery techniques. First, using the topological characteristics of RPs, we model the radio map as a graph and design the angle fingerprint for the band-limited graph signal. Subsequently, the radio map is built based on graph clustering, sampling set selection and signal recovery. Extensive simulations are performed in a geometry-based ray tracing signal propagation model, which demonstrates that the proposed method can recover the radio map with low-collection cost and outperform existing solutions in terms of fingerprint accuracy and localization performance.

Structured sparsity improves on traditional sparse modeling by suggesting that only a limited number of input variables is needed to describe the output variable. These methods go a step further by incorporating structured patterns in variable selection, such as groups or networks of input variables. This paper introduces a novel binary approximation method leveraging structured sparsity to enhance multi-anchor direct localization performance. By reformulating the sparse recovery as a quadratic unconstrained binary optimization (QUBO) problem, we address the significant computational complexity inherent in NP-hard problems. Employing compressed sensing and binary programming, our method reduces approximation errors and ensures that the line-of-sight component is consistently identified from a coherent grid point among anchors, thus enhancing localization accuracy. Our findings demonstrate a significant improvement in the accuracy of direct localization methods, underscoring the potential of structured sparsity and QUBO formulation to advance localization technologies in multipath environments.

With the ubiquitous availability of WiFi signals, WiFi-based localization methods have gained a lot of attention, especially in indoor environments. Generally, accurate indoor localization is challenging due to the multipath effect. Numerous methods have been proposed to increase the accuracy of localization for multipath environments. One of the recent methods is direct localization. This method includes a two-dimensional search in a planar geometry to directly localize the source without estimating any intermediate variable such as angle-of-arrival or time-of-flight. In this paper, we use a compressed sensing framework in the direct localization technique to estimate the location of a user in an indoor multipath environment. We form a penalized$\ell _{0}$-norm structure for this problem and then convert this structure to an Ising energy problem to take advantage of the efficient existing binary programming solvers. In this paper, the Ising energy problem is solved using Markov Chain Monte Carlo (MCMC). The evaluations show that our approach significantly improves the localization accuracy compared to other approaches in the literature.

Reconfigurable intelligent surfaces (RISs) have drawn considerable attention due to their ability to introduce controllable phase-shifts onto impinging electromagnetic waves and impose link redundancy. Meanwhile, unmanned aerial vehicles (UAVs) are expected to make future 6G networks more connected, but they are prone to several failures, which cause network disintegration. To harness the benefits of both, we study their integration to improve connectivity of multi-RIS-assisted UAV networks. We first propose to define the criticality of nodes, which reflects the importance of some nodes over other nodes. We then employ the algebraic connectivity metric, which is adjusted by the reflected links of the RISs and their criticality weights, to formulate the problem of maximizing the network connectivity. Such problem is a computationally expensive combinatorial optimization. Using a relaxation method where the discrete scheduling constraint of the problem is relaxed to be continuous, we propose two efficient solutions, namely semi-definite programming (SDP) optimization and Laplacian matrix perturbation, which both solve the problem in polynomial time. We rigorously derive the lower and upper bounds of the algebraic connectivity obtained from the perturbation solution. Simulation results compare the performance of the proposed solutions with different schemes, including without RISs, unoptimized link scheduling and phase shifts, greedy search, and optimal. The results show that the proposed schemes achieve considerably improved performance with low computational complexity compared to other schemes.

It is anticipated that integrating unmanned aerial vehicles (UAVs) with reconfigurable intelligent surfaces (RISs), resulting in RIS-assisted UAV networks, will offer improved network connectivity against node failures for the beyond 5G networks. In this context, we utilize a RIS to provide path diversity and alternative connectivity options for information flow from user equipment (UE) to UAVs by adding more links to the network, thereby maximizing its connectivity. This paper employs the algebraic connectivity metric, which is adjusted by the reflected links of the RIS, to formulate the problem of maximizing the network connectivity in two cases. First, we consider formulating the problem for one UE, which is solved optimally using a linear search. Then, we consider the problem of a more general case of multiple UEs, which has high computational complexity. To tackle this problem, we formulate the problem of maximizing the network connectivity as a semi-definite programming (SDP) optimization problem that can be solved efficiently in polynomial time. In both cases, our proposed solutions find the best combination between UE(s) and UAVs through the RIS. As a result, it tunes the phase shifts of the RIS to direct the signals of the UEs to the appropriate UAVs, thus maximizing the network connectivity. Simulation results are conducted to assess the performance of the proposed solutions compared to the existing solutions.

This paper proposes a solution for energy-efficient communication in reconfigurable intelligent surface (RIS)-assisted unmanned aerial vehicle (UAV) networks. The limited battery life of UAVs is a major concern for their sustainable operation, and RIS has emerged as a promising solution to reducing the energy consumption of communication systems. The paper formulates the problem of maximizing the energy efficiency of the network as a mixed integer nonlinear program, in which UAV placement, UAV beamforming, On-Off strategy of RIS elements, and phase shift of RIS elements are optimized. The proposed solution utilizes the block coordinate descent approach and a combination of continuous and binary genetic algorithms. Moreover, for optimizing the UAV placement, Adam optimizer is used. The simulation results show that the proposed solution outperforms the existing literature. Specifically, we compared the proposed method with the successive convex approximation (SCA) approach for optimizing the phase shift of RIS elements.

Photoplethysmogram (PPG) is a major indicator of a patient’s physiological status. PPG is generally studied using manually designed algorithms to detect its critical morphological points. However, existing algorithms for analyzing these signals do not serve the purpose effectively and accurately, particularly for abnormal signals. This paper proposes a multi-observation hidden semi-Markov model (HSMM) for PPG signal semantic segmentation, which leverages the information available in raw signal and its first and second derivatives simultaneously. The results indicate the feasibility of PPG signal segmentation with high accuracy using an HSMM and a small training dataset. Moreover, employing a multi-observation approach improves the accuracy significantly.

Machine learning and deep learning models for time series classification generally require a large volume of data to achieve superior performance. However, due to the lack of a sufficient amount of time series in many real-world applications, particularly health care, training these models is more challenging than expected. This paper introduces the Random Frequency Butchering (RFB) method to enhance the generalization performance of classification tasks on limited time series in health care. This approach generates a number of filters with random cutoff frequencies in the frequency domain. The concatenation of time series representations from these filters stacked with the original time series is then used to train an arbitrary time series classifier. The experimental results on the standard medical time series datasets show that the RFB time series representation can significantly enhance the classification performance of the MiniRocket, Inception-Net, and ResNet classifiers.

In the last decade, there has been a surge of research interest in feature extraction using random sampling. These techniques are fast and scalable and, at the same time, have practical favorability in low-sample size and high-dimensional training data. Convolutional Kitchen Sinks-based methods are promising random feature extractors for time series data. Since these methods are data-independent, many of the extracted features are redundant. To address this problem, we propose a simple and efficient feature selection method based on knee/elbow detection in the curve of ordered coefficients in linear regression. Our empirical studies show that without significant loss in accuracy, the proposed feature selector, on average, prunes more than 84 percent of randomly generated features.

WiFi sensing is an important part of the new WiFi 802.11bf standard, which can detect motion and measure distances. In recent years, some machine learning methods have been proposed for human activity recognition from WiFi signals. However, to the best of our knowledge, none of these methods have explored orientation prediction of the user using WiFi signals. Orientation prediction is particularly critical for human-machine interaction in an environment with multiple smart devices. In this paper, we propose a data collection setup and machine learning models for joint human orientation and activity recognition using WiFi signals from a single access point (AP) or multiple APs. The results show feasibility of joint orientation-activity recognition in an indoor environment with a high accuracy.

Direct localization is a technique used to estimate the location of a signal source by searching for it directly in a planar grid. However, due to the sparsity of the signal, a sparse recovery approach using ℓ0-norm can be effective. In this paper, we propose a cooperative direct localization method that takes into account the multi-path environment. To account for the effect of multi-path propagation, we consider the distances between the user and the anchors and describe this relationship using the Cayley-Menger determinant. This relationship is then introduced as an accordance function for the range estimates. We also employ a binary programming relaxation technique to solve this NP-hard optimization problem. Our experimental results show that this approach achieves a smaller localization error compared to existing direct localization methods.

Modulation that exploits orthogonality in time, frequency, and space (OTFS) is tolerant of delay and Doppler channel spreads. We exploit this fact to estimate, and hence compensate, hardware impairments that inhibit accurate localization using a single receiver. Our focus is on oscillator imperfections, which manifest as carrier frequency offset (CFO) and sampling frequency offset (SFO). By representing the channel state information (CSI) in the delay-Doppler domain, we show that these impairments are the equivalent of a Doppler shift under OTFS modulation and a Doppler spread under WiFi orthogonal frequency division multiplexing (OFDM). This observation informs a sparse modelling formulation for joint localization and clock error estimation that is readily solved using compressed sensing techniques. Through simulations, we verify that our proposed approach reaches sub-meter accuracy and contrast it to the OFDM-based localization approach.

A Reconfigurable Intelligent Surface (RIS) is a new technology containing elements that can be adjusted to modify wireless channel behavior. One of the areas in which RIS has proven to be useful is localization and object detection. In this paper, we utilize RIS to localize and recognize shapes of objects by changing its configurations and without collecting additional samples for calibration. We define a new problem based on the Cramer-Rao lower bound and propose a novel way of changing the configurations of the RIS using an Evolutionary Algorithm based on the random graph modeling and the Swendsen-Wang algorithm as a sampling method. Our analysis shows the convergence of the algorithm to the optimal value and our simulation results show the effectiveness of the method compared to other techniques.

For fingerprint-based localization, the time-consuming and labor-intensive construction of offline radio map is the bottleneck which hinders its large-scale implementation. In this paper, we propose a radio map extrapolation and localization algorithm by exploiting the angles and delays of specular multipath components. First, using the concept of virtual anchor nodes (VANs), we calculate the virtual transmitter (VT) of each multipath component according to angle and delay information measured at a known point. By estimating the positions of reflector with these VTs, the angles and delays of the uplink multipath components transmitted at other locations in the indoor environment are extrapolated. Then, a channel fingerprint composed of the extrapolated angles and delays is proposed to represent the channel response in the angle-delay domain, which is spatially unique and discriminative. Thus, the radio map constructed such can significantly reduce the labor and time costs. Lastly, a convolutional neural network (CNN) is applied for indoor localization. The performance of the proposed fingerprint extrapolation and localization method is validated through extensive simulations with a ray-tracing channel model, which exhibits promising localization performance for our proposed scheme with reduced construction costs.

Networks are frequently changing due to new technologies. To increase the network performance, companies migrate their existing network to a network with a new technology. Finding an efficient optimization algorithm is an important challenge in the network migration. In this paper, the network migration problem is considered as a set of circuit migration problems in which multiple technicians simultaneously migrate the endpoints of circuits in order to minimize the average latency and average technician travel cost. While average latency indicates how fast the sites can be upgraded, average travel cost estimates the required cost for modernizing the network. First, We derive binary linear program and binary quadratic program formulations for average latency and average technician travel cost, respectively. Then we use the linear scalarization method to obtain a multi-objective optimization problem for simultaneously minimizing both costs. Our approach for solving the derived multi-objective optimization problem is based on converting it to a quadratic unconstrained binary optimization problem (QUBO) using the penalty method. Subsequently, we exploit the third generation of Fujitsu Digital Annealer which is a hybrid system of hardware and software to minimize the derived QUBO. To investigate the performance of our proposed method, we study extensive network migration instances on the 75-node CONUS network topology. Simulation results indicate that both costs can efficiently be optimized using our proposed method. We also directly solve the obtained multi-objective optimization problem with Gurobi solver. The comparison results show that our proposed method outperforms the Gurobi solver.

Accurate indoor localization is a challenging problem in a multipath environment. In order to tackle this problem, several methods have been proposed. Direct localization is one of these methods that makes use of a two-dimensional search in a planar geometry. In this paper, we use a compressed sensing framework in the direct localization technique to estimate the location of a user in an indoor multipath environment. We form a penalized ℓ0-norm structure for this problem, and then convert this structure to an Ising energy problem.The Ising energy problem is solved using Markov Chain Monte Carlo (MCMC). Our simulation results show that our approach improves the estimation accuracy compared to the existing methods in the literature.

Anatomical movements of the human body can change the channel state information (CSI) of wireless signals in an indoor environment. These changes in the CSI signals can be used for human activity recognition (HAR), which is a pre-dominant and unique approach due to preserving privacy and flexibility of capturing motions in non-line-of-sight environments. Existing models for HAR generally have a high computational complexity, contain very large number of trainable parameters, and require extensive computational resources. This issue is particularly important for implementation of these solutions on devices with limited resources, such as edge devices. In this paper, we propose a lightweight human activity recognition (LiteHAR) approach which, unlike the state-of-the-art deep learning models, does not require extensive training of a large number of parameters. This approach uses randomly initialized convolution kernels for feature extraction from CSI signals without training the kernels. The extracted features are then classified using Ridge regression classifier, which has a linear computational complexity and is very fast. LiteHAR is evaluated on a public benchmark dataset and the results show its high classification performance with a much lower computational complexity in comparison with the complex deep learning models.

Channel State Information (CSI) based fingerprinting is surfacing as an accurate and robust method of indoor localization. However, the high-dimensional nature of CSI data impedes its adoption in multi access point (AP) systems. To reap the rewards of cooperative localization with privacy and limited system complexity in mind, we propose a federated learning (FL) architecture. Each AP has an individual model and a shared model, where the individual model parameters are unique to each AP and the shared model parameters are communicated to a central server for aggregation. The server averages the models and sends them back to each AP, which use this joint model as a regularization term. To capture the spatio-temporal characteristics of CSI, we propose a convolutional neural network (CNN) as each AP’s individual model and a multi layer perceptron (MLP) as the shared model. Extensive experimental studies verify the superiority of the proposed edge computing approach compared to the exiting methods in the literature. We use commercial off-the-shelf APs collecting CSI data in multiple indoor environments and compare the proposed system to a state-of-the-art deep learning model. Our approach shows significant improvement in the localization accuracy for both individual APs and aggregate predictions.

Reconfigurable intelligent Surfaces (RISs) have recently gained a lot of attention due to their capability in creating a smarter controllable radio environment. In this paper, we present a novel cooperative positioning (CP) technique to improve positioning accuracy in RIS-assisted systems with device-to-device (D2D) communications. We consider a scenario with one base station (BS), one RIS, and$K$mobile stations (MSs). We assume that the line-of-sight (LoS) between the BS and all the MSs is blocked and the communication has to be established through the RIS. The proposed method uses beam sweeping in the downlink communication to determine the direction of each MS relative to the RIS location. In the uplink communication, the MSs communicate with each other and the power received from each MS is used to determine the distances between every pair of MSs. We then formulate an optimization problem to minimize the mismatch between the distance measurements subject to the relative direction information, and find the location estimates using a neural network. Finally, we test the performance of the proposed method through simulations.

Orthogonal time frequency space (OTFS) modulation is a new technique that transmits symbols in delay-Doppler plane and can be used for both communication and sensing. A Reconfigurable Intelligent Surface (RIS) is an array of elements that can be controlled to produce desirable behavior over the signals and enhance communication. In this paper, we utilize RIS to increase the resolution of the OTFS-based radar systems, while decreasing the transmission time of base station. We solve the problem of detection by sparse recovery and treating it as an inverse problem. Our simulation results show the effectiveness of the RIS in increasing the resolution of detection over the range and velocity of the targets.

The downlink time-difference-of-arrival (DL-TDOA), which is used for positioning in 3GPP NR, is the time interval that is measured by a user equipment (UE) between the reception of the downlink signals from two different cells. The measurement of the DL-TDOA might be challenging, especially at a cell center, where signals from remote base stations (BSs) are usually very weak. Reconfigurable intelligent Surfaces (RISs) are expected to be part of future communication networks because of their capability to create a smarter controllable radio environment. In this paper, we study whether RIS can replace the function of a remote cell in the DL-TDOA measurement, hence maintaining the localization procedure fully within a single cell. We consider a scenario with one BS and one RIS, and show that the TDOA between the line-of-sight path and the reflected path through the RIS can replace the DL-TDOA measurement in the 3GPP NR recommendations. The DL-TDOA and the time-of-flight measurements between the BS and the UE suffice to accurately localize the UE. The proposed algorithm uses one round trip time (RTT) observation and one TDOA observation in millimeter wave (mmWave) frequencies. We present an extended Kalman filter positioning and tracking algorithm to localize users. Simulation results show that the positioning accuracy of RIS-enabled localization matches that of the two-cell structure while being a cost-effective solution.

Due to their capability in creating a controllable wireless environment, extending coverage and improving localization accuracy, reconfigurable intelligent surfaces (RISs) are expected to be a main component of future 6G networks. In this paper, we present a novel cooperative positioning (CP) use-case of the RIS in mmWave frequencies. We show that two mobile stations (MSs) are able to estimate their positions through device-to-device (D2D) communications, and processing the signals reflected from the RIS. We start by building the system model based on the uniform linear array (ULA) architecture of the RIS elements. Then, we derive the Fisher information matrix (FIM) and the Cramér-Rao lower bound (CRLB) for calculating the MSs positioning error. After that, we optimize the RIS configuration to minimize the CRLB. Finally, simulation results compare the localization performance of random phases at the RIS with the optimal configuration.

Dropout is a well-known regularization method by sampling a sub-network from a larger deep neural network and training different sub-networks on different subsets of the data. Inspired by the dropout concept, we propose EDropout as an energy-based framework for pruning neural networks in classification tasks. In this approach, a set of binary pruning state vectors (population) represents a set of corresponding sub-networks from an arbitrary original neural network. An energy loss function assigns a scalar energy loss value to each pruning state. The energy-based model (EBM) stochastically evolves the population to find states with lower energy loss. The best pruning state is then selected and applied to the original network. Similar to dropout, the kept weights are updated using backpropagation in a probabilistic model. The EBM again searches for better pruning states and the cycle continuous. This procedure is a switching between the energy model, which manages the pruning states, and the probabilistic model, which updates the kept weights, in each iteration. The population can dynamically converge to a pruning state. This can be interpreted as dropout leading to pruning the network. From an implementation perspective, unlike most of the pruning methods, EDropout can prune neural networks without manually modifying the network architecture code. We have evaluated the proposed method on different flavors of ResNets, AlexNet,$l_{1}$pruning, ThinNet, ChannelNet, and SqueezeNet on the Kuzushiji, Fashion, CIFAR-10, CIFAR-100, Flowers, and ImageNet data sets, and compared the pruning rate and classification performance of the models. The networks trained with EDropout on average achieved a pruning rate of more than 50% of the trainable parameters with approximately < 5% and < 1% drop of Top-1 and Top-5 classification accuracy, respectively.

Many indoor location estimation algorithms compare the received signal strengths from WiFi access points to a prerecorded dataset, which may be composed of carefully mea-sured and curated data through a massive calibration campaign, and a large volume of crowd-sourced data captured by casual users. The crowd-sourced data usually contains valuable, but at the same time noisy, measurements where the noise might be in the features and/or the labels of data points. The treatment of such a dataset is challenging, as improper application of the dataset might result in inaccurate location estimation. In this paper, we propose a learning algorithm that makes the K-Nearest Neighbour (KNN) regression robust to noises both in features and labels of the training data. An intuition on why the learning algorithm should work is provided and the effectiveness of the algorithm is shown by experiments in a real environment.

Networks are frequently changing due to new technologies. The growing demand for bandwidth is forcing many carriers to migrate their existing network to a network with a new technology in order to increase network performance. Telecommunication companies are looking for optimization algorithms to efficiently manage their network migration. In this paper, the network migration problem is considered as a set of circuit migration problems in which two technicians simultaneously migrate the two ends of a circuit in order to minimize the total accumulated sites in-service and total technician travels. While total accumulated sites in-service indicates how fast the sites can be upgraded during the migration process, total technician travels estimates the required cost. We first formulate our target problem as a constrained binary quadratic program which is NP-hard in general. Our approach for solving the derived optimization problem is based on converting it to a quadratic unconstrained binary optimization problem (QUBO) using the penalty method. Subsequently, we exploit Digital Annealer which is a massively parallel hardware architecture to minimize the derived QUBO. To evaluate our proposed method, we study extensive network migration instances on the 75-node CONUS network topology.

Resampling increasing the variance of the tracking algorithm in Particle Filtering (PF). Instead of utilizing resampling procedures that rely on asymptotic convergence properties, we show that intelligently selecting and replicating a set of samples can better represent the posterior approximation and improve the overall performance of the PF. To this end, we formulate the resampling procedure as an integer program that minimizes an upper bound on the Kullback-Leibler divergence (KLD) between the resampled distribution and the posterior approximation. We then transform the problem into an Ising energy minimization problem, which we are able to efficiently solve. Applying our novel paradigm to a challenging sequential importance resampling (SIR) simulation shows faster convergence over the number of resampled particles and a 35% improvement in the median KLD for a fixed number of particles.

A typical deep neural network (DNN) has a large number of trainable parameters. Choosing a network with proper capacity is challenging and generally a larger network with excessive capacity is trained. Pruning is an established approach to reducing the number of parameters in a DNN. In this paper, we propose a framework for pruning DNNs based on a population-based global optimization method. This framework can use any pruning objective function. As a case study, we propose a simple but efficient objective function based on the concept of energy-based models. Our experiments on ResNets, AlexNet, and SqueezeNet for the CIFAR-10 and CIFAR-100 datasets show a pruning rate of more than 50% of the trainable parameters with approximately < 5% and < 1% drop of Top-1 and Top-5 classification accuracy, respectively.

Pruning is one of the major methods to compress deep neural networks. In this paper, we propose an Ising energy model within an optimization framework for pruning convolutional kernels and hidden units. This model is designed to reduce redundancy between weight kernels and detect inactive kernels/hidden units. Our experiments using ResNets, AlexNet, and SqueezeNet on CIFAR-10 and CIFAR-100 datasets show that the proposed method on average can achieve a pruning rate of more than 50% of the trainable parameters with approximately < 10% and < 5% drop of Top-1 and Top-5 classification accuracy, respectively.

Localization in wireless networks is possible by measuring some characteristics of the propagating signal related to the position of the user, which is always corrupted by noise components. In this paper, a correntropy based algorithm is proposed for localization and tracking of a mobile station in wireless networks. The performance of the proposed algorithm is compared with the Least Mean Square (LMS) and Least Mean P-norm (LMP) algorithms, and its preference aspects are discussed. The results show that, using correntropy can bring robustness to localization and improve performance in many realistic scenarios such as fat-tail noise distributions, one of the serious bottlenecks of the next-generation 5G wireless communications systems. In addition, it is shown that, the Gaussian kernel of the correntropy function reduces the sensitivity of the algorithm to the learning rate.

The access point (AP) deployment is a fundamental task for constructing an accurate localization system. Existing literature mainly deals with the AP placement problem using optimal geometry analysis since the target-AP geometry will affect the localization performance. However, some non-ideal phenomena in practical scenario, e.g., the existence of obstacles, array orientation and path loss, will degrade the accuracy of angle-of-arrival (AoA) estimation as well as the localization accuracy. In this article, we reformulate the AP planning incorporating these factors. We decompose the problem into two subproblems, namely AP selection problem and error minimization problem. The AP selection problem selects the minimum number of APs to satisfy a desired localization accuracy, aided by a refined orientation updating procedure. We design a centralized and a distributed error minimization algorithm to further decrease the localization error. The centralized algorithm shows superiority in time efficiency. Nevertheless, the case with large number of APs may lead to excessive computational cost. Accordingly, we further devise the distributed algorithm which is adaptive to large-scale deployment. Numerical studies in indoor environments with barriers are conducted to verify our proposed approach.

Vehicular networks are critical pieces in support of advanced intelligent transportation systems (ITS). These networks are formed by vehicles that can be connected to one another as well as to the infrastructure, and are subject to constant topology changes, disconnections, and data congestion. Each ITS application could have a different set of communication requirements, such as delay, bandwidth, and packet delivery ratio. Meeting these heterogeneous requirements in the complex dynamic environment of vehicular networks is a challenge. This paper develops a new framework for application-driven vehicular networks using 5G network slicing. We present the architecture of the proposed solution and design algorithms for heterogeneous traffic in a dynamic vehicular environment. Our simulations on realistic vehicular scenarios show significant improvements in network performance compared to the state-of-the-art approaches.

We present a resource management framework for load-coupled downlink cellular OFDMA networks considering the load factor of an individual base station (BS) per resource block (RB), i.e., the number of adjacent sub-carriers (SCs), as the variable of interest in the resource management problem. The load factor of a BS per RB, which corresponds to the fraction of active SCs in the BS per RB, is an indicator of the level of resource consumption, and it affects the interference caused to that RB reused in other BSs, and thereby, results in a load-coupled OFDMA system. We first propose two distributed schemes to minimize: (i) the total load factor of the BSs (which would in turn increase the number of supportable users in the system), and (ii) the total downlink transmit power level of the BSs. Then, we derive the necessary and sufficient conditions for checking the feasibility of given target-rate requirements (also referred to as demand vector) for users. Accordingly, an iterative and distributed scheme is proposed to check the feasibility of a given demand vector. Next, for a priority-based load-coupled network, we propose a priority-based gradual removal algorithm to support the maximal number of low-priority users while satisfying the demands of the high-priority users. To evaluate the performance of our proposed schemes for resource management and admission control in load-coupled OFDMA networks, the theoretical investigations are complemented with Monte Carlo simulations.

Cellular V2X, the “Vehicle to Everything” standard, defines a framework for information exchange among vehicles and other network entities. In one of the main modes, LTE V2X relies on a central scheduler to minimize the consumed resources in a conflict-free manner. This NP-hard problem leads to a scheduling strategy that assigns separate resources to the conflicting links and can enjoy from opportunistic resource reuse. In this paper, a novel polynomial-time heuristic, MUCS, is introduced, which models this scheduling as a Vehicle Routing Problem. The existing conflict-free schedulers face major incompetence in satisfying the LTE and 5G V2X requirements mainly due to their reliance on simplifications that are unnatural to the V2X environment. On the contrary, MUCS can flexibly accommodate general Device-to-Device topologies as the basis of V2X networks without imposing any packet segmentation. This way, MUCS minimizes the control information overhead in the cellular V2X standard. Due to scalability and a high quality resource utilization (near-optimal in certain conditions), compared to the existing literature, MUCS is desirable for LTE V2X and the 5G networks.

In this paper, we present a novel cooperative positioning (CP) technique to improve localization accuracy in vehicular environments. The proposed method uses position information obtained from the Global Positioning System (GPS) and Angle-of-Arrivals (AoAs) between the nodes estimated using Millimeter wave (mmWave) V2X communication and MUSIC algorithm. We propose two methods to improve the positioning accuracy: pairwise cooperation and angular Multi-dimensional scaling (AMDS). Simulation results show that the proposed methods outperform other CP methods in the literature.

Macro base stations are densely overlaid by small cells to satisfy the demands of user equipment in heterogeneous networks. Due to their dense deployment, some small cells are not directly connected to macro base stations and thus backhaul connections are required to connect small cells to macro base stations. Millimeter wave backhauls which have high bandwidths are preferred for small cell backhaul communication, since they can increase the capacity of network considerably. In this context, association of user equipment to base stations becomes challenging due to the backhaul architecture. Considering environmental concerns, energy efficiency is a vital criterion in designing user association algorithms. In this paper, we study the user association problem aiming at the maximization of energy efficiency. We develop centralized and distributed user association algorithms based on sequentially minimizing the power consumption. We evaluate the performance of the proposed algorithms under two scenarios and show that they achieve higher energy efficiency compared to the existing algorithms in the literature, while maintaining high spectral efficiency and backhaul load balancing.

Neural network pruning is an important technique for creating efficient machine learning models that can run on edge devices. We propose a new, highly flexible approach to neural network pruning based on Gibbs distributions. We apply it with Hamiltonians that are based on weight magnitude, using the annealing capabilities of Gibbs distributions to smoothly move from regularization to adaptive pruning during an ordinary neural network training schedule. This method can be used for either unstructured or structured pruning, and we provide explicit formulations for both. We compare our proposed method to several established pruning methods on ResNet variants and find that it outperforms them for unstructured, kernel-wise, and filter-wise pruning.

Data representation can facilitate training of deep neural network when limited data is available. We have previously proposed the radial transform sampling method as a data representation technique for training neural networks. In this paper, a probabilistic framework to analyze radial transform is presented. To further elaborate it, performance of training deep neural networks on radial transform generated images for semantic segmentation of kidneys in abdominal computed tomography is evaluated. Our results show that the proposed representation method can achieve higher performance than other similar methods by translating a semantic segmentation problem to a classification problem when limited annotated images are available.

While Wi-Fi is of great potential for indoor localization, the access points (APs) deployment in realistic indoor environments is particularly challenging due to the impact of various obstacles, e.g., walls, pillars or bookcases. The diverse obstacles create the troublesome non-line-of-sight and the multipath effect, which deteriorate the localization accuracy. In this paper, we study the effect of obstacles on the localization error and formulate the AP planning problem as a AP selection problem. This problem is decomposed into two subproblems, i.e., AP selection problem and error minimization problem. The AP selection problem aims to choose the minimum number of APs to satisfy the preset accuracy requirement. Furthermore, the error minimization problem improves the localization performance through optimizing the AP positions and array orientations. Extensive simulations show that our proposed method is adaptive to the obstacles and it achieves higher localization accuracy compared with the existing deployment method.

This paper presents a ℓ0-norm minimization based angle-of-arrival (AoA) estimation approach using an Ising model (AIM). The proposed method takes advantage of the fact that incoming signals are spatially sparse and that the angular domain can be discretized. To solve the NP-hard problem, we formulate the ℓ0norm minimization as an Ising energy minimization problem. The minimization of the Ising energy results in the maximization of the Gibbs distribution. The performance of the proposed method in identifying and resolving the angle of arrival of incoming signals is studied for different scenarios, in all of which the proposed method outperforms the state-of-the-art methods.

Macro base stations are densely overlaid by small cells to satisfy the demands of user equipments in heterogeneous networks. Due to their dense deployment, some small cells are not directly connected to macro base stations and thus backhaul connections are required to connect small cells to macro base stations. Millimeter wave backhauls which have high bandwidths are preferred for small cell backhaul communication, since they can increase the capacity of network considerably. In this context, association of user equipments to base stations becomes challenging due to the backhaul architecture. Considering environmental concerns, energy efficiency is a vital criterion in designing the user association algorithms. In this paper, we study the user association problem aiming at the maximization of energy efficiency given a specific spectral efficiency target. Firstly, we derive a quadratic upper bound on the total power consumption in heterogeneous network (access network and backhaul). Then we develop an energy and spectrum efficient user association method based on minimizing the derived quadratic upper bound, which indeed turns our target problem into a generalized quadratic assignment problem. Since this problem is NP-hard in general, we propose a heuristic algorithm for solving it. Extensive simulations show that our approach provides an enhanced level of energy efficiency compared to other well-known alternatives in the literature.

Indoor localization is important for a variety of applications, such as location-based services, mobile social networks, and emergency response. Fusing spatial information is an effective way to achieve accurate indoor localization with little or with no need for extra hardware. However, the existing indoor localization methods that make use of spatial information are either computationally expensive or sensitive to the completeness of landmarks. In this article, we propose a novel, low-cost, high-accuracy indoor localization method based on a landmark graph. The experimental results show that the proposed method outperforms the state-of-the-art methods.

In this paper, amplify-and-forward beamforming (AFB) is considered to establish a communication, through wireless sensor networks (WSNs) of K sensor nodes, from a source to a receiver in the presence of both scattering and interference. All sources send their data to the WSN during the first time slot, while the nodes forward a properly weighted version of their received signals during the second slot. These weights are properly selected to maximize the desired power while completely canceling the interference signals. We show, however, that they depend on information locally unavailable at each node, making the zero-forcing beamformer (ZFB) unsuitable for WSNs, due to the prohibitive data exchange overhead and the power depletion it would require. To address this issue, we exploit the asymptotic expression at large K of the ZFB weights that is locally computable at every node and, further, well-approximates their original counterparts. The performance of the resulting new distributed ZFB (DZFB) version is analyzed and compared with the conventional ZFB and two other distributed AFB benchmarks: the monochromatic (i.e., single-ray) AFB whose design neglects the presence of scattering and the bichromatic AFB which relies on an efficient two-ray channel approximation valid only for low angular spread (AS). We show that the proposed DZFB outperforms its monochromatic and bichromate counterparts while incurring much less overhead and power depletion than ZFB. We show also that it is able to provide optimal performance even in highly scattered environments as in the latter.

Indoor localization has been an active research area in the last decade. Various localization systems have been proposed based on different types of signals, including but not limited to WiFi, ultrawideband, inertial measurements, and visual signals. Fingerprinting-based methods are among the most popular methods due to their accuracy and ease of deployment. However, a disadvantage to fingerprinting-based methods is the training phase in which fingerprints have to be collected at known locations and stored for future localization inquiries. Recently a few methods have been proposed to alleviate this problem. In this paper, we focus on the possibility of reducing the burden of the training phase for visual (image-based) indoor localization systems by proposing a system that automatically generates the image-location database. The proposed system will be referred to as automatic visual fingerprinting that can be paired with any indoor image-based localization method. We will also validate the proposed system through extensive experiments.

The knowledge of the floor-level location of a user in a multi-storey building is important for many applications, especially for emergency response. Existing floor identification systems suffer from a variety of limitations such as low accuracy, the need for a time-consuming site survey, assumption of user encounters, knowledge of the initial floor, and/or poor applicability. In this paper, we propose a novel, zero-effort, deep learning-based floor identification system, called ZeeFi. The proposed system uses the widely-available smartphone sensing to identify on which floor a user is located. By recognizing the ground floor automatically, the proposed system does not require site survey, initial floor knowledge, and other assumptions. To achieve accurate floor identification performance, we have developed a deep learning-based method. Experimental results show that the proposed system outperforms the state-of-the-art systems, and is very promising for large-scale deployment.

Intelligent Transportation System's main objective is to provide sustainable and optimized means of transportation for citizens of urban centers. Identifying the travel modes used by citizens, in real- time, allows these systems to better adapt transportation infrastructure and services according to user needs and possibilities of interaction. Many works have explored the use of machine learning algorithms and ensemble methods, combined with general and domain specific feature extraction techniques, for this task. More recently, several works evaluated the use of deep learning algorithms while none of them has explored the use of Recurrent Neural Networks (RNNs) in conjunction with domain feature extraction to enable the development of flexible and lightweight travel mode detection solutions based on multiple smartphone sensor readings. In this paper, we propose a deep RNN architecture for building online travel mode detection models using Long-Short Term Memory (LSTM) cells and evaluate its performance using real mobility data collected with a wide variety of sensors. The experiments show that the proposed architecture allows the generation of models with high accuracy and lower memory consumption and computation cost than state-of-the-art supervised machine learning (ML) approaches.

Label imbalance is an unavoidable issue for deep learning in the field of medical imaging. In terms of semantic segmentation, diseased tissues occur infrequently with very limited volumes, which exerts a strong bias to the learning of convolutional neural networks (CNNs). Dice loss functions are often applied for training to alleviate the problem, however they become very unstable when the dataset is extremely imbalanced. In this work, we propose a novel method that first locates the abnormal tissues, and performs segmentation in local regions. We also propose the Ising conditional random fields (CRFs) for post- processing, and apply the Digital Annealer (DA) for optimization. Our experiments demonstrate that the proposed approach is effective in learning with very imbalanced data.

Cyber attacks on the power systems can mislead the control center to produce incorrect state and topology estimate. State and topology attacks can have harmful impacts on the operation of a power system. The problem of placing secure phasor measurement units (PMUs) to detect these attacks has been studied in the literature. Specifically, it has been shown that placing secure PMUs to disable undetectable state and topology attacks can enhance the security of the system against cyber attacks. Placing secure PMUs is indeed the minimum vertex cover problem. Since the cost of deploying PMUs is high, it is important to place the secure PMUs efficiently in order to maximize the ability of detecting cyber attacks while reducing the costs. In this paper, we use Digital Annealer to solve the vertex cover problem. Digital Annealer is a hardware architecture for solving combinatorial optimization problems. We have performed numerous numerical experiments and noticed that our approach has an enhanced level of optimality compared to other well-known alternatives in the literature.

Multipath interference causes the antenna array of an anchor to estimate several angles of arrival (AoA) for a single user. The resulting ambiguity regarding the line of sight (LoS) component can lead to severe errors in location estimation. This work formulates the problem within an outlier rejection framework: a set of candidate locations are computed by considering all AoAs at all anchors; with the observation that LoS AoA vary less than non-LoS AoA, over several time instances an inlier set can be found that clusters around the true user location. This work then derives an Ising model representation for finding the inlier set and solves the NP-hard problem using the Digital Annealer (DA). Simulations show that this approach improves median localization accuracy by 48.5% when compared to the state-of-the-art in localization methods.

Overfitting is a major problem in training machine learning models, specifically deep neural networks. This problem may be caused by imbalanced datasets and initialization of the model parameters, which conforms the model too closely to the training data and negatively affects the generalization performance of the model for unseen data. The original dropout is a regularization technique to drop hidden units randomly during training. In this paper, we propose an adaptive technique to wisely drop the visible and hidden units in a deep neural network using Ising energy of the network. The preliminary results show that the proposed approach can keep the classification performance competitive to the original network while eliminating optimization of unnecessary network parameters in each training cycle. The dropout state of units can also be applied to the trained (inference) model. This technique could compress the number of parameters up to 41.18% and 55.86% for the classification task on the MNIST and Fashion-MNIST datasets, respectively.

Cellular V2X, the “Vehicle to Everything” standard, defines a framework for practically feasible information exchange among vehicles and other network entities. This interaction is proved to bring in substantial economic and ecological benefits. LTE V2X uses a portion of uplink frame as a resource pool and in the main mode, relies on a central scheduler for allocating these resources to the users. As an important resource management problem, besides the optimal resource scheduling, finding a proper lower bound for required resources in this mode is NP-hard. Network management entities and service providers require this lower bound to determine the minimum size of the uplink frame slice to be allocated to the V2X resource pool. In this paper, we take advantage of Digital Annealer potentials to target this problem at a new size range with considerably enhanced level of optimality that has not been achievable so far due to computational complexities. The Digital Annealer enables us to find the minimum required size of the V2X slice in the uplink frame with high speed through our maximum clique and minimum cover graph theoretic formulations of this problem. Our approach considerably enhances performance compared to its closest alternatives in the literature.

Semi-flocking algorithms have been demonstrated to be efficient in maneuvering MSNs in multi-target tracking tasks. In many real-world applications, targets can be assigned with different priorities according to their importance of being tracked. However, existing semi-flocking algorithms normally assume the importance of all targets to be identical, which may not allocate resources in an efficient manner. In this paper, we propose a target evaluation method that incorporates priorities of the targets in the assessment process. Based on the evaluation results, mobile agents decide to track a target or continue to scan the terrain via a probabilistic task switching mechanism. Simulation results indicate a higher effectiveness of the proposed method in target tracking and area coverage when compared with two existing semi-flocking algorithms.

Medical datasets are often highly imbalanced with over-representation of prevalent conditions and poor representation of rare medical conditions. Due to privacy concerns, it is challenging to aggregate large datasets between health care institutions. We propose synthesizing pathology in medical images as a means to overcome these challenges. We implement a deep convolutional generative adversarial network (DCGAN) to create synthesized chest X-rays based upon a modest sized labeled dataset. We used a combination of real and synthesized images to train deep convolutional neural networks (DCNNs) to detect pathology across five classes of chest X-rays. The comparative study of DCNNs trained with the combination of real and synthesized images showed that these networks can outperform similar networks trained solely with real images in pathology classification. This improved performance is largely attributable to the balancing of the dataset using DCGAN synthesized images, where classes that are lacking in example images are preferentially augmented.

This paper aims at the problem of time-of-flight (ToF) estimation using channel state information (CSI) obtainable from commercialized multiple-input-multiple-output orthogonal frequency-division multiplexing (MIMO-OFDM) wireless local area network (WLAN) receivers. It was often claimed that the CSI phase is contaminated with errors of known and unknown natures rendering the ToF-based positioning difficulty. To search for an answer, we take a bottom-up approach by first understanding CSI, its constituent building blocks, and the sources of error that contaminate it. We then model these effects mathematically. The correctness of these models is corroborated based on the CSI collected in extensive measurement campaign, including radiated, conducted, and chamber tests. Knowing the nature of contaminations in the CSI phase and amplitude, we proceed with introducing pre-processing methods to clean CSI from those errors and make it usable for range estimation. To check the validity of the proposed algorithms, the MUSIC super-resolution algorithm is applied to post-processed CSI to perform range estimates. The results substantiate that a median accuracy of 0.7, 0.8, and 0.9 m is achievable in a highly multipath line-of-sight environment where the transmitter and the receiver are 5, 10, and 15 m apart.

The next generation of cellular networks promises to be the platform for ubiquitous, precise, and accurate location-awareness. Although positioning using radio-frequency waves has many advantages in terms of cost and performance, propagation phenomena such as multipath propagation and shadowing can immensely deteriorate this performance leading to incorrect estimation of a user's location. This paper introduces LocHunt, an algorithm that addresses such a problem. Concretely, at each time-instance, angle-of-arrivals (AoAs) of multipath waves from several anchors are estimated and all the candidate locations of the user are derived. Repeating this operation for a span of time, the algorithm exploits the heuristics that those candidate locations obtained from non-line-of-sight (NLoS) paths exhibit large spatial variations, whereas the true location of the user is manifested as a denser cluster. Subsequently, the peak of the underlying empirical probability density function corresponds to an estimate of the user's location, which can be found by the mean-shift algorithm followed by the connected-components algorithm. The proposed data-driven approach is compared to the state-of-the-art, showing improved localization accuracy with up to one order of magnitude reduction in computational complexity. Channel state information (CSI), similar to that exposed by 802.11n MIMO-OFDM systems, is used for testing the algorithms and is generated by a modified Winner 2 (WIM2) channel simulator.

We propose a new collaborative beamforming (CB) solution robust against major implementation impairments over dual-hop transmissions from a source to a destination communicating through a wireless sensor network (WSN) of K nodes. In the first time slot, the source sends its signal to the WSN while, in the second, each node forwards its received signal after multiplying it by a properly selected beamforming weight. The latter aims to minimize the received noise power while maintaining the desired power equal to unity. These weights depend a priori on some channel state information (CSI) parameters. Hence, the latter have to be estimated locally at each node thereby resulting in implementation errors that could severely hinder CB performance. Exploiting an efficient asymptotic approximation at large K, we develop alternative CB solutions that not only account for estimation errors, but also adapt to different implementation scenarios and wireless propagation environments ranging from monochromatic (i.e., scattering-free) to polychromatic (i.e., P-DCB) ones. Besides, in contrast to existing techniques, our new CB solutions are distributed (i.e., DCB) in that they do not require any information exchange among nodes, thereby dramatically improving both WSN spectral and power efficiencies. Simulation results confirm that the proposed DCB techniques are much more robust against implementations errors than their benchmarks at much lower complexity.

Deep learning models have a large number of free parameters that must be estimated by efficient training of the models on a large number of training data samples to increase their generalization performance. In real-world applications, the data available to train these networks is often limited or imbalanced. We propose a sampling method based on the radial transform in a polar coordinate system for image augmentation to facilitate the training of deep learning models from limited source data. This pixel-wise transform provides representations of the original image in the polar coordinate system by generating a new image from each pixel. This technique can generate radial transformed images up to the number of pixels in the original image to increase the diversity of poorly represented image classes. Our experiments show improved generalization performance in training deep convolutional neural networks with radial transformed images.

Medical datasets are often highly imbalanced with over-representation of common medical problems and a paucity of data from rare conditions. We propose simulation of pathology in images to overcome the above limitations. Using chest X-rays as a model medical image, we implement a generative adversarial network (GAN) to create artificial images based upon a modest sized labeled dataset. We employ a combination of real and artificial images to train a deep convolutional neural network (DCNN) to detect pathology across five classes of chest X-rays. Furthermore, we demonstrate that augmenting the original imbalanced dataset with GAN generated images improves performance of chest pathology classification using the proposed DCNN in comparison to the same DCNN trained with the original dataset alone. This improved performance is largely attributed to balancing of the dataset using GAN generated images, where image classes that are lacking in example images are preferentially augmented.

V2X, the "Vehicle to Everything" communication, facilitates communication among vehicles and other network entities, which is verified to bring in substantial economic and ecological benefits. In the main mode, LTE V2X relies on a central scheduler. As an important problem from the resource management viewpoint, minimizing the consumed resources in a conflict-free manner is NP-hard. In this paper, a novel point of departure is proposed for this NP-hard problem and a polynomial-time heuristic, MUCS, is introduced that models scheduling as a Vehicle Routing Problem. The existing multichannel methods seem to be incapable of satisfying LTE and 5G V2X communication demands and requirements due to their reliance on inapplicable simplifications. Our technique enables contiguous link scheduling, as opposed to most of the existing multichannel methods, which also remarkably performs better. The proposed heuristic MUCS eliminates the need for such simplifying constraints and can be employed in general V2X network topologies. Due to its scalability and a close-to-optimal resource utilization compared to the existing literature, MUCS is desirable for the cellular V2X, i.e., LTE V2X and the upcoming 5G systems.

Many popular fingerprinting-based indoor localization methods, such as WiFi-based localization systems, rely on a dataset of fingerprints labelled by known locations (fingerprint-location dataset) to be able to localize a user by comparing the queried fingerprint with the fingerprints in the dataset. Generating and updating such a dataset is a burden and requires significant amount of time, human effort and expertise. In this work, we propose a system to build such a dataset from scratch using crowdsourced data. Consequently we reduce the required time and effort, by leveraging the information shared by the crowd. The proposed system is based on localization of user motion, hence called LocaMotion. LocaMotion takes a supervised learning perspective towards localization of user-sent motion data. In other words, LocaMotion is formalised as a classifier that extracts and assigns features from a user motion data to a sequence of points in the building. Training and testing procedures for LocaMotion will be discussed in details followed by extensive experiments to demonstrate its validity and success to localize motion patterns and generate useful fingerprint-location datasets.

Locomotion activity recognition (LAR) is important for a number of applications, such as indoor localization, fitness tracking, and aged care. Existing methods usually use handcrafted features, which requires expert knowledge and is laborious, and the achieved result might still be suboptimal. To relieve the burden of designing and selecting features, we propose a deep learning method for LAR by using data from multiple sensors available on most smart devices. Experimental results show that the proposed method, which learns useful features automatically, outperforms conventional classifiers that require the hand-engineering of features. We also show that the combination of sensor data from four sensors (accelerometer, gyroscope, magnetometer, and barometer) achieves a higher accuracy than other combinations or individual sensors.

Human activity recognition is important for a large number of applications including indoor localization. Existing methods usually involve manually-designed features, which require expert knowledge and are laborious. Also, previous works use only the accelerometer for activity recognition, which may fail to recognize some complex activities. In this paper, we propose a deep learning-based method for locomotion activity recognition by using the combination of data from multiple smartphone built-in sensors. Eight types of locomotion activities are identified including the new `False Motion' activity introduced for the first time in this work. Experimental results show that the proposed method, which learns useful features automatically, outperforms conventional classifiers that require hand-engineering of features. Also, using data from multiple sensors helps to improve recognition accuracy by about 10% compared to that using accelerometer data only.

The problem of joint load management and admission control (JLAC) in underlay OFDMA-based cognitive radio networks (CRNs) is studied. The fraction of the active sub-carriers in a base station (BS) in each resource block (RB) is defined as the load factor of the BS per RB. In the JLAC problem, we simultaneously minimize the secondary users' outage ratio and the total load factors of the BSs via the RBs, subject to the constraint that the primary users are protected. This problem is a NP-hard problem. We first relax it into a convex optimization problem. Then, by employing the optimal solution to the relaxed JLAC problem, we derive a heuristic algorithm to gradually remove the most adversative secondary users imposing the most interference to the primary users. The performance of the proposed algorithm is studied in terms of the outage ratio of secondary users and the total load factors of the BSs via the RBs through extensive simulations.

Wi-Fi fingerprinting is the most widely used technique for indoor localization of Android smartphones. This technique has been under intensive research recently and several methods have been investigated, ranging from traditional k-Nearest-Neighbors (kNN) to sophisticated methods based on neural networks and deep learning. In this paper, we investigate the problem of indoor localization using Wi-Fi fingerprinting in a large-scale schema, where instead of localizing the user on one floor or inside one building, the fingerprinting database includes thousands of buildings and/or Wi-Fi access points (AP). We will investigate the complexity of the simple weighted kNN algorithm and will show that the current algorithms are not efficient enough to run on such a huge database, on the smartphone or on a cloud computer. Then, we propose efficient algorithms to select a portion of the database to work on (a.k.a reduced radiomap) and find the complexity of these algorithms. We have implemented our proposed algorithms for reducing the radiomap size on a cloud server using structured query language (SQL).

Pathfinding in hospitals is challenging for patients, visitors, and even employees. Many people have experienced getting lost due to lack of clear guidance, large footprint of hospitals, and confusing array of hospital wings. In this paper, we propose Halo; An indoor navigation application based on voice-user interaction to help provide directions for users without assistance of a localization system. The main challenge is accurate detection of origin and destination search terms. A custom convolutional neural network (CNN) is proposed to detect origin and destination search terms from transcription of a submitted speech query. The CNN is trained based on a set of queries tailored specifically for hospital and clinic environments. Performance of the proposed model is studied and compared with Levenshtein distance-based word matching.

A fast and robust algorithm for motion estimation based on stereo vision is proposed. The algorithm consists of two consecutive stages. First, interest points are found and tracked between four images from two consecutive time steps. A novel robust method for tracking features is proposed for this step to address shortcomings of the traditional methods for indoor hand-held/wearable scenarios. Next, motion of the camera in time is estimated via optimizing the total reprojection error associated with the robust features found using the first stage. A key assumption is that motion between the two time steps is sufficiently small or equivalently, the frames are captured close enough in time. The proposed Visual Ego-motion Estimation (VEE), is robust due to robust feature tracking in the first step, and fast due to efficient reprojection error minimization used to estimate the six-degree-of-freedom (6DoF) motion of the camera. Experiments prove comparable/superior results compared to the state-of-the-art methods in the literature especially for indoor handheld scenarios.

In this paper, amplify-and-forward beamforming (AFB) is considered to establish a reliable communication between devices in networked smart cities. All sources send their data during the first time slot while the cooperative terminals forward a properly weighted version of their received signals during the second. These zero-forcing AFB (ZFB) weights are properly selected to maximize the desired power while completely canceling the interference signals. We show, however, that their implementation requires a huge terminals' information exchange, making ZFB unsuitable for smart cities where the overhead and power restrictions are very stringent. To address this issue, we exploit the asymptotic expression at large K of the ZFB weights whose computation requires much less information exchange and, further, well-approximate their original counterparts. The performance of the proposed beamforming is analyzed and compared to ZFB and monochromatic (i.e., single-ray) AFB (MB) whose design neglects the presence of scattering.

Summary Mobile social networks are presenting new opportunities for content dissemination. User location, mobility, and social communities can be used to deliver delay‐tolerant content. Most existing dissemination methods either fail to consider the user's interest in their protocol design or just allow exchange of contents between nodes that have a similar interest. If nodes with similar interest never encounter each other, then contents might not be exchanged with them. In this paper, we propose a method in which user location, mobility, and interest as well as certain limiting factors such as relay buffer size and communication overhead are used to select a set of relays for targeted advertisement distribution. Besides, our method does not depend on nodes with similar interest encountering one another. In the proposed method, a distribution agent exploits user location, mobility, and social networks as well as the interest of destinations to select a group of relays to carry content to the targeted destinations. Users move among social communities and carry advertisements to their peers in other communities. We have also developed an optimization problem for advertisement selection and scheduling. Since the problem is NP‐hard, we propose a heuristic solution. Evaluation of the proposed approach shows that the algorithm is not sensitive to network resource variation and readily outperforms popular methods reported in the literature in terms of delay, delivery ratio, and interest compatibility.

The Received Signal Strength (RSS) based fingerprinting approaches for indoor localization pose a need for updating the fingerprint databases due to dynamic nature of the indoor environment. This process is hectic and time-consuming when the size of the indoor area is large. The semi-supervised approaches reduce this workload and achieve good accuracy around 15 percent of the fingerprinting load but the performance is severely degraded if it is reduced below this level. We propose an indoor localization framework that uses unsupervised manifold alignment. It requires only 1 percent of the fingerprinting load, some crowd sourced readings, and plan coordinates of the indoor area. The 1 percent fingerprinting load is used only in perturbing the local geometries of the plan coordinates. The proposed framework achieves less than 5 m mean localization error, which is considerably better than semi-supervised approaches at very small amount of fingerprinting load. In addition, the few location estimations together with few fingerprints help to estimate the complete radio map of the indoor environment. The estimation of radio map does not demand extra workload rather it employs the already available information from the proposed indoor localization framework. The testing results for radio map estimation show almost 50 percent performance improvement by using this information as compared to using only fingerprints.

Vehicular safety applications require reliable and up-to-date knowledge of the local neighborhood. Under IEEE 802.11p, this is attained through single-hop broadcasts of safety beacons in the control channel. However, high transmission power and node mobility can cause regions of node density to form rapidly. In such situations, excessive load on the control channel must be avoided to prevent performance degradation for safety applications. Existing congestion control schemes aim to reach a fair distribution of available channel resources, but fail to account for the differing quality of service (QoS) requirements of vehicles in different driving contexts. This context depends on many factors, including the relative position and velocity of its neighbors. The problem of adapting each vehicle's transmission probability under a slotted p-persistent vehicular broadcast medium access control (MAC) protocol is formulated as a network utility maximization (NUM) problem which takes the driving context into account. A distributed algorithm is proposed to solve this problem in a decentralized manner, its convergence is analyzed, and its performance is evaluated through simulations.

Signal strength variation across diverse devices is a major problem with RSS-based indoor localization systems. This paper aims to solve this problem by considering two factors: the instability of collected RSS samples and linear shift of RSS patterns collected by different devices. We propose different techniques to handle the uncertainty of samples. Furthermore, we propose an automatic linear transformation algorithm that relies on the linear relationship across diverse devices. The algorithm finds the set of nearest neighbor fingerprints for an online point through a series of linear transformations. A localizer engine is then used to detect user's location. The proposed system is automatic, has low computational complexity and does not require any training period. Experimental results indicate the proposed system is reliable with very high positional accuracy.

In this paper, we propose an image-based localization system, applicable for a number of indoor scenarios including office buildings, airports, chain stores, etc. In such applications, text/numbers are suitable distinctive landmarks for localization. The proposed system takes advantage of OCR to read the text/numbers and provide a rough estimate using the floor plan. Next, it performs OCR-aided stereo feature matching to refine the estimate by solving a PnP problem. Experiments show that this system achieves a median localization error of less than 50 cm for test positions located as far as 7 meters from a 20cm by 30cm number plate using different test devices in a university building scenario.

We investigate the problem of graph matching to translate topological indoor localization to geographical localization, by modeling the building map and the semantic maps as graphs. A matching algorithm based on hidden Markov models is proposed. The matching algorithm is tested on both simulations and real data and accuracies as high as 94% on real data is achieved, while the matching is shown to be robust to noise, scale variance and partial matching via simulations.

A low complexity and infrastructure free indoor positioning system is proposed, by tracking the users on a set of user profiles via Viterbi algorithm. A modification on the implementation is also proposed to further optimize the complexity of tracking, by using the sparsity of the state transition probability matrix. Our algorithm only requires users to hold a smartphone, in arbitrary orientation/gesture and walk in the building. Since we only use the sensors embedded on the smartphones, this approach does not require any installed infrastructure in the building. The algorithm is tested on real-case scenarios with successfully obtaining less than 2.5 meters of median error.

Crowdsourcing allows for rapid deployment of indoor localization systems. However, compared to the conventional methods, crowdsourcing might collect fewer received signal strength (RSS) values, hence result in greater influence to outliers in RSS values. In this paper, we propose an algorithm to detect such outliers and to substitute them with more suitable RSS values. In particular, we investigate the relationship of RSS values between adjacent locations using a signal propagation model and show that the outliers can be corrected using a signal propagation model. We propose the Signal Propagation-based Outlier Reduction Technic (SPORT) for identifying and adjusting outlier values in both the offline training phase and the online localization phase. Experimental results show that SPORT greatly smoothens the radio map and improves the location accuracy.

One major bottleneck in the practical implementation of received signal strength (RSS) based indoor localization systems is the extensive deployment efforts required to construct the radio maps through fingerprinting. In this paper, we aim to design an indoor localization scheme that can be directly employed without building a full fingerprinted radio map of the indoor environment. By accumulating the information of localized RSSs, this scheme can also simultaneously construct the radio map with limited calibration. To design this scheme, we employ a source data set that possesses the same spatial correlation of the RSSs in the indoor environment of interest. The knowledge of this data set is then transferred to a limited number of calibration fingerprints and one or several RSS observations with unknown locations, in order to perform direct localization of these observations using manifold alignment. We test two different source data sets, namely a simulated radio propagation map and the environment's plan coordinates. For moving users, we exploit the correlation of their observations to improve their localization accuracy. The online testing in two indoor environments shows that the plan coordinates achieve better results than the simulated radio maps, and a negligible degradation with 70-85 percent reduction in the calibration load.

In this paper, we consider the problem of minimizing the completion delay for instantly decodable network coding (IDNC) in wireless multicast and broadcast scenarios. We are interested in this class of network coding due to its numerous benefits, such as low decoding delay, low coding and decoding complexities, and simple receiver requirements. We first extend the IDNC graph, which represents all feasible IDNC coding opportunities, to efficiently operate in both multicast and broadcast scenarios. We then formulate the minimum completion delay problem for IDNC as a stochastic shortest path (SSP) problem. Although finding the optimal policy using SSP is intractable, we use this formulation to draw the theoretical guidelines for the policies that can minimize the completion delay in IDNC. Based on these guidelines, we design a maximum weight clique selection algorithm, which can efficiently reduce the IDNC completion delay in polynomial time. We also design a quadratic-time heuristic clique selection algorithm, which can operate in real-time applications. Simulation results show that our proposed algorithms significantly reduce the IDNC completion delay compared to the random and maximum-rate algorithms, and almost achieve the global optimal completion delay performance over all network codes in broadcast scenarios.

This paper studies the effect of mobility on the sensing performance of a cognitive radio network with mobile nodes. The secondary nodes sense the spectrum using a distributed compressive sensing approach to detect the available channels. Distributed compressive sensing is suggested to reduce the number of samples by exploiting correlation between the samples. Channel occupancy at the two nodes will be jointly estimated and a channel available at the location of both nodes is chosen for communication. We show that mobility can be exploited to further decrease the number of samples by increasing the average level of correlation among the sensed samples over time.

This paper presents a novel cooperative positioning (CP) method to increase localization accuracy in vehicular ad-hoc networks (VANET). The proposed method uses a semi-extended Kalman filter to fuse position data and distance information of nodes in the network. This paper also introduces a new distance measurement method using the time-difference-of-arrival in positive orthogonal codes. Simulation results show that the proposed method outperforms other cooperative positioning systems, both for low and high traffic densities.

In this paper, we propose a homography-aware semi-supervised formulation for the logo-based indoor localization problem using smartphone cameras. Our method labels unmatched feature points detected inside the logo parts of query images with their estimated 3D coordinates. The 3D coordinates are computed using the homography estimated from the matched features. We demonstrate the accuracy improvement and lower localization error variance resulted from our semi-supervised approach via experiments in an indoor scenario.

In this paper, we study the performance of sensing in mobile sensor networks with imperfect knowledge of neighborhood mobility. We examine the impact of exchanging incorrect mobility information on the cost of sensing and the required target coverage. The study is performed for two target coverage models: an independent coverage model and a Markovian one. We demonstrate via extensive simulations that a small amount of the mobility information is required to be successfully exchanged between the mobile sensors to provide the required coverage of targets. Finally, we conduct an experiment at the University of Toronto campus to show the impact of the mobility knowledge in sensing. The experiment results match our expectation in terms of the target coverage, sensing cost, and trade-off between the two metrics.

The main limitation of deploying/updating Received Signal Strength (RSS) based indoor localization is the construction of fingerprinted radio map, which is quite a hectic and time-consuming process especially when the indoor area is enormous and/or dynamic. Different approaches have been undertaken to reduce such deployment/update efforts, but the performance degrades when the fingerprinting load is reduced below a certain level. In this paper, we propose an indoor localization scheme that requires as low as 1% fingerprinting load. This scheme employs unsupervised manifold alignment that takes crowd sourced RSS readings and localization requests as source data set and the environment's plan coordinates as destination data set. The 1% fingerprinting load is only used to perturb the local geometries in the destination data set. Our proposed algorithm was shown to achieve less than 5 m mean localization error with 1% fingerprinting load and a limited number of crowd sourced readings, when other learning based localization schemes pass the 10 m mean error with the same information.

This paper presents an analytical study of the stable throughput for multiple broadcast sessions in a multi-hop wireless tandem network with random access. Intermediate nodes leverage on the broadcast nature of wireless medium access to perform inter-session network coding among different flows. This problem is challenging due to the interaction among nodes, and has been addressed so far only in the saturated mode where all nodes always have packet to send, which results in infinite packet delay. In this paper, we provide a novel model based on multi-class queueing networks to investigate the problem in unsaturated mode. We devise a theoretical framework for computing maximum stable throughput of network coding for a slotted ALOHA-based random access system. Using our formulation, we compare the performance of network coding and traditional routing. Our results show that network coding leads to high throughput gain over traditional routing. We also define a new metric, network unbalance ratio (NUR), that indicates the unbalance status of the utilization factors at different nodes. We show that although the throughput gain of the network coding compared to the traditional routing decreases when the number of nodes tends to infinity, NUR of the former outperforms the latter. We carry out simulations to confirm our theoretical analysis.

Reliable local information dissemination is the primary concern for periodic safety broadcasting in VANETs. We propose a sublayer in the application layer of the WAVE stack to increase the reliability of safety applications. Our design uses rebroadcasting of network coded safety messages, which considerably improves the overall reliability. It also tackles the synchronized collision problem stated in the IEEE 1609.4 standard as well as congestion problem and vehicle-to-vehicle channel loss. We propose a discrete phase type distribution to model the time transitions of a node state. Based on this model, a tight loss probability upper bound for the network coding algorithm is derived. Numerical results based on our analysis as well as ns-2 simulations show that our method outperforms the previous repetition-based algorithms.

In this paper, we study the multicast completion and decoding delay minimization problems for instantly decodable network coding (IDNC) in the case of lossy feedback. When feedback loss events occur, the sender falls into uncertainties about packet reception at the different receivers, which forces it to perform partially blind selections of packet combinations in subsequent transmissions. To determine efficient selection policies that reduce the completion and decoding delays of IDNC in such an environment, we first extend the perfect feedback formulation in our previous works to the lossy feedback environment, by incorporating the uncertainties resulting from unheard feedback events in these formulations. For the completion delay problem, we use this formulation to identify the maximum likelihood state of the network in events of unheard feedback and employ it to design a partially blind graph update extension to the multicast IDNC algorithm in our earlier work. For the decoding delay problem, we derive an expression for the expected decoding delay increment for any arbitrary transmission. This expression is then used to find the optimal policy that reduces the decoding delay in such lossy feedback environment. Results show that our proposed solutions both outperform previously proposed approaches and achieve tolerable degradation even at relatively high feedback loss rates.

Reliable multi-hop forwarding in vehicular networks is required by many critical intelligent transportation system (ITS) safety applications. Cooperative vehicular multi-hop schemes achieve reliability using broadcast transmissions and multiple forwarding relays at each hop. However, packet duplication must be controlled to circumvent the broadcast storm problem. This paper presents the cooperative POC-based forwarding (CPF) protocol for highway vehicular networks, which extends the repetition-based POC-MAC protocol for multi-hop transmissions. At each forwarding hop, multiple cooperating relays form a virtual relay and schedule their transmissions to correspond to a single POC codeword, thereby adhering to the POC-MAC. CPF exploits spatial diversity while mitigating the effect of hidden terminals. By allocating separate POC-based schedules for multi-hop packets and the periodic broadcast of safety heartbeat packets, the CPF protocol reduces the interference between the two. The performance of the CPF protocol is studied through analysis using a Markov model and through ns-2 simulations.

In this paper, we show that knowledge of mobility helps in sensing and coverage in vehicular sensor networks. First, we propose a mathematical formulation for the stationary sensing-coverage problem in terms of the maximization of a utility function. Then, we propose a method to solve the sensing problem for mobile sensors. We solve the two problems via the branch and bound approximation algorithm. Simulation results show that mobility improves vehicular sensing by selecting an optimal number of sensors to provide the same quality of coverage over an interval of time. In addition, we study the model by using probabilistic node availability.

Vehicular ad hoc networks are expected to provide a reliable networking platform for cooperative safety communication systems. Those systems are of a broadcast nature and require to deliver both safety messages and vehicle tracking information while being interfered by other types of lower priority messages on the same channel. Vehicle tracking information are necessary to enable safety communication systems and intelligent transportation systems. Due to the large number of communicating vehicles and the amount of traffic exchanged in the broadcast mode, network congestion often occurs in vehicular communication systems. In this paper, a new methodology for avoiding such a network congestion is proposed. We identify the sparsity of the vehicle tracking information and propose a novel information recovery scheme. The proposed scheme reduces the amount of data exchanged due to vehicle tracking packets while providing a robust information reception at the receivers. Essentially, it utilizes compressive sensing to transmit a few measurements of the vehicles velocity vector that allows perfect recovery of the original vector at the receiver with a minimal error. Extensive simulation results demonstrate the effectiveness of applying compressive sensing in recovering vehicles tracking information, and relieving the network from unnecessary congestion due to the large amount of data exchanged.

Reliable multi-hop forwarding in vehicular networks is required by many critical Intelligent Transportation System (ITS) safety applications. Cooperative vehicular multi-hop schemes achieve reliability using broadcast transmissions and multiple forwarding relays at each hop. However, packet duplication must be controlled to circumvent the broadcast storm problem. This paper proposes the Cooperative POC-based Forwarding (CPF) protocol, which extends the repetition-based POC-MAC protocol to handle multi-hop transmissions. Multiple cooperating relays at each forwarding hop schedule their transmissions to correspond to a single POC codeword. The transmission adhere to the transmission patterns of the POC-MAC. The proposed scheme exploits spatial diversity while mitigating the effect of hidden terminals on broadcast transmissions. By allocating separate POC-based schedules for multi-hop packets and the periodic broadcast of safety heartbeat packets, the CPF protocol reduces the interference between the two. A Markov model is presented for the end-to-end probability of reception of multi-hop packets. The performance of the CPF protocol is compared with other multi-hop schemes through ns-2 simulations.

In this paper, we aim to identify the strategies that maximize and monotonically increase the density of coding opportunities in instantly decodable network coding (IDNC). Using the graph representation of IDNC, we first derive an expression for the exact evolution of the edge set size after the transmission of any arbitrary coded packet. From the derived expression, we show that sending commonly wanted packets for all the receivers can maximize the number of coding opportunities. Since guaranteeing such property in IDNC is usually impossible, this strategy does not guarantee the achievement of our target. Consequently, we further investigate the problem by deriving an expression for the expected edge set size evolution after ignoring the identities of the packets requested by the different receivers and considering only their numbers. This expression was then employed to show that serving the maximum number of receivers with largest numbers of missing packets and erasure probabilities tends to maximize and monotonically increase the expected density of coding opportunities. Simulation results justify our theoretical findings. Finally, we validate the importance of our work through two case studies showing that our\ignore{ identified} strategy outperforms several well-known IDNC solutions in optimizing the IDNC completion delay and receiver goodput.

An indoor tracking and navigation system based on measurements of received signal strength (RSS) in wireless local area network (WLAN) is proposed. In the system, the location determination problem is solved by first applying a proximity constraint to limit the distance between a coarse estimate of the current position and a previous estimate. Then, a Compressive Sensing-based (CS--based) positioning scheme, proposed in our previous work , , is applied to obtain a refined position estimate. The refined estimate is used with a map-adaptive Kalman filter, which assumes a linear motion between intersections on a map that describes the user's path, to obtain a more robust position estimate. Experimental results with the system that is implemented on a PDA with limited resources (HP iPAQ hx2750 PDA) show that the proposed tracking system outperforms the widely used traditional positioning and tracking systems. Meanwhile, the tracking system leads to 12.6 percent reduction in the mean position error compared to the CS-based stationary positioning system when three APs are used. A navigation module that is integrated with the tracking system provides users with instructions to guide them to predefined destinations. Thirty visually impaired subjects from the Canadian National Institute for the Blind (CNIB) were invited to further evaluate the performance of the navigation system. Testing results suggest that the proposed system can be used to guide visually impaired subjects to their desired destinations.

The growing interest for location-based services motivates many researchers to study different localization techniques for indoor environments. The main objective of these studies is to find a balance point between the accuracy of the scheme and its deployment/training cost. RSS-based schemes and in particular Graph-based Semi-Supervised Learning (G-SSL) constitute a group of techniques which has low setup cost and good localization accuracy. In this paper, we analyze the G-SLL scheme and show that, despite its high performance, the G-SSL method (in its original format) is not a very accurate model for a localization problem. Based on this observation and to improve the accuracy of localization, we propose an alternative approach which incorporates our knowledge of wireless signal propagation into the label propagation mechanism. Experimental results are then used to evaluate the performance of the proposed scheme compared to the original G-SSL.

One major bottleneck in the practical implementation of received signal strength (RSS) based indoor localization systems is the extensive deployment load required to construct radio maps through fingerprinting. Several works aimed to employ radio propagation models as alternative to fingerprinting but the different sources of inaccuracies in the generation of these models result in high localization errors. In this paper, we propose an indoor localization scheme that can be directly deployed and employed without building a full radio map of the indoor environment. The proposed scheme employs the information from a radio propagation simulator and limited number of calibration measurements to perform direct localization using manifold alignment. For moving users, we exploit the correlation of their reported observations to improve the localization accuracy. The online performance evaluation shows that our algorithm achieves localization errors in the order of 2.5 to 3 m with as low as 15% - 30 % of the complete fingerprinting load.

There is a recent interest in developing algorithms for the reconstruction of jointly sparse signals, which arises in a large number of applications such as sensor networks. In many of these applications, we encounter extremely large problem sizes for which algorithms with low computational complexity are required. Recently, an algorithm called iterative hard thresholding has been proposed, which is faster than the ℓ1-minimization and greedy algorithms for compressed sensing. In this work, we extend the iterative hard thresholding algorithm to jointly sparse signals and will investigate the performance of our proposed algorithm analytically by giving conditions under which the exact reconstruction could happen. We will show that our algorithm is faster than the state of the art algorithms for jointly sparse signals while showing similar performance.

In this paper, we propose a coding strategy that maximizes the density of the coding opportunities in instantly decodable network coding (IDNC). Using the graph representation of IDNC, we derive the expression for the expected evolutions of coding opportunities after the transmission of any arbitrary coded packet and show that serving the maximum number of receivers, with the largest numbers of missing packets and erasure probabilities, tends to both maximize the expected number of coding opportunities and increase the expected coding density almost monotonically. Simulation results justify our theoretical findings and demonstrate the importance of maintaining high coding density when optimizing long-term parameters.

The recent growing interest for indoor Location-Based Services (LBSs) has created a need for more accurate and real-time indoor positioning solutions. The sparse nature of location finding makes the theory of Compressive Sensing (CS) desirable for accurate indoor positioning using Received Signal Strength (RSS) from Wireless Local Area Network (WLAN) Access Points (APs). We propose an accurate RSS-based indoor positioning system using the theory of compressive sensing, which is a method to recover sparse signals from a small number of noisy measurements by solving an `1-minimization problem. Our location estimator consists of a coarse localizer, where the RSS is compared to a number of clusters to detect in which cluster the node is located, followed by a fine localization step, using the theory of compressive sensing, to further refine the location estimation. We have investigated different coarse localization schemes and AP selection approaches to increase the accuracy. We also show that the CS theory can be used to reconstruct the RSS radio map from measurements at only a small number of fingerprints, reducing the number of measurements significantly. We have implemented the proposed system on a WiFi-integrated mobile device and have evaluated the performance. Experimental results indicate that the proposed system leads to substantial improvement on localization accuracy and complexity over the widely used traditional fingerprinting methods.

M-LWWF/M-LWDF scheduling schemes have attracted much interest due to their ability to both stabilize queues whenever possible and control delay through parameter selection. However, a good implementation of these schedulers would require a mechanism to minimize the required fraction of the bandwidth while satisfying its stability and delay requirements. To the best of our knowledge, previous works on these scheduling policies did not address the problem of minimizing the bandwidth utilization while satisfying delay constraints. In this paper, we explore the solution of this problem using a joint bandwidth and weight adaptation approach. We characterize the problem solution space for M-LWWF and M-LWDF scheduling, assuming time-varying traffic. We also show that, starting from any point in the solution space, simple dynamic bandwidth and weight updates can surely lead to the convergence to the optimal operation point in this space. Based on these characteristics, we propose a dynamic parameter adaptation algorithm that is able to track the time-varying optimal operation points for dynamic traffic and channel conditions. Simulation results show the efficiency of our proposed algorithm in tracking the optimal operation points in dynamic traffic and channel settings.

In this paper, we consider the problem of minimizing the broadcast completion delay for instantly decodable network coding with limited feedback. We first extend the stochastic shortest path formulation of the full feedback scenario in to the limited feedback scenario. We then show that the resulting formulation is more complicated to solve than the original one but has its same properties and structure. Based on this result, we design four variants of the algorithm employed in with four different approaches to deal with un-acknowledged transmissions. We finally compare these four algorithms through extensive simulations and show that the algorithm that temporarily avoids all un-acknowledged transmissions in subsequent coding decisions can result in tolerable degradation against the full feedback performance while using much lower feedback.

In this paper, we study the effect of probabilistic and prolonged packet feedback loss events on the broadcast completion time of instantly decodable network coding (IDNC). These feedback loss events result in a lack of knowledge about the reception status at different subsets of receivers, which creates a challenge in selecting efficient IDNC packet combinations in subsequent transmissions. To solve this problem for both probabilistic and prolonged feedback loss, we first identify the different possibilities of feedback loss events at the sender and determine their probabilities in both cases. Given these probabilities and the nature of the IDNC completion time problem, we design three blind instantly decodable network coding approaches that perform coding decisions similar to the algorithms proposed in, but on blindly updated graphs to account for feedback events. These three approaches are then compared through extensive simulations. Results show that the full consideration and the full negligence of all the attempted packet requests with probabilistic and prolonged feedback loss events, respectively, in subsequent coding decisions can achieve a tolerable degradation against the perfect feedback performance for relatively high feedback loss probabilities and periods.

In this paper, we introduce the novel concept of mobility diversity as the diversity gained by transmitting the information of the nodes of a mobile network over different network topologies. Due to the mobility of the nodes, different network topologies emerge which can benefit the information transmission throughout the network. In traditional diversity schemes, such as frequency, time, or spatial diversity, a signal is transmitted over different diversity dimensions (e.g., different frequency bands, different time intervals, or different spatial paths) to combat the destructive effects of fading in each individual channel. In a mobile wireless network, the nodes can exploit the topology diversity to communicated with their corresponding destinations more reliably as compared to the case when the topology of the network is fixed. In fact, in a fixed topology, the probability of a source having a poor connectivity to its destination is higher than the case when there are multiple topologies over which the communication can occur.

This paper designs a distributed Transmission Opportunity (TXOP) adaptation algorithm for IEEE802.11e Enhanced Distributed Channel Access (EDCA). Each node measures its throughput in a window and compares it with a target value. If the measured throughput is higher than the target value, the node reduces its TXOP, otherwise if the measured value is less than the target throughput, the node increases its TXOP. We show that the target throughput can be achieved in a globally stable manner.

In this paper, we study the effect of packet feedback loss events on the broadcast completion delay performance of instantly decodable network coding. These feedback loss events result in a continuous lack of knowledge about the reception status at different subsets of receivers. This lack of knowledge creates a challenge in selecting efficient packet combinations in subsequent transmissions. To solve this problem, we first identify the different possibilities of unheard feedback events at the sender and determine their probabilities. Given these probabilities and the nature of the problem, we design three partially blind instantly decodable network coding approaches that perform coding decisions similar to the algorithms proposed in [1], [2], but on blindly updated graphs to account for unheard feedback events. These three approaches are then compared through extensive simulations. Results show that re-considering all the attempted packet requests, with unheard feedback, in subsequent coding decisions can achieve a tolerable degradation against the perfect feedback performance for relatively high feedback loss probabilities.

Network coding has recently attracted attention as a substantial improvement to packet retransmission schemes in wireless multicast broadcast services (MBS). Since the problem of finding the optimal network code maximizing the bandwidth efficiency is hard to solve and hard to approximate, two main network coding heuristic schemes, namely opportunistic and full network coding, were suggested in the literature to improve the MBS bandwidth efficiency. However, each of these two schemes usually outperforms the other in different receiver, demand, and feedback settings. The continuous and rapid change of these settings in wireless networks limits the bandwidth efficiency gains if only one scheme is always employed. In this paper, we propose an adaptive scheme that maintains the highest bandwidth efficiency obtainable by both opportunistic and full network coding schemes in wireless MBS. The proposed scheme adaptively selects, between these two schemes, the one that is expected to achieve the better bandwidth efficiency performance. The core contribution in this adaptive selection scheme lies in our derivation of performance metrics for opportunistic network coding, using random graph theory, which achieves efficient selection when compared to appropriate full network coding parameters. To compare between different complexity levels, we present three approaches to compute the performance metric for opportunistic coding using different levels of knowledge about the opportunistic coding graph. For the three considered approaches, simulation results show that our proposed scheme almost achieves the bandwidth efficiency performance that could be obtained by the optimal selection between the opportunistic and full coding schemes.

The low rank feature of location estimation in Wireless Sensor Networks (WSNs) makes it feasible to use nuclear norm minimization as an accurate and fast solution for low-dimensional embedding problems. In this paper, a novel localization algorithm for WSNs is proposed by using nuclear norm for rank minimization. We formulate the location finding problem from only a small fraction of random entries of Euclidean Distance Matrix (EDM) as a low-rank matrix recovery problem, subject to a set of linear equality constraints. We show that a measurement matrix using orthogonal projection obeys the RIP and thus, supports a sufficient condition for the recovery of the low-rank matrix with overwhelming probability. For simplicity, Singular Value Thresholding (SVT) algorithm, a standard convex optimization approach, is used for the nuclear norm minimization. Simulation results demonstrate that in a 100 m × 100 m area, for a small scale network with 100 nodes, only 20% of measurements is needed to achieve a 0.5 m localization error, while 3% needed to achieve a 0.05 m error for a comparatively large scale network with 1000 nodes.

In this paper, we introduce the concept of generalized instantly decodable network coding (G-IDNC) to further minimize decoding delay in wireless broadcast, compared to strict instantly decodable network coding (S-IDNC), studied in. G-IDNC loosens the strict instant decodability constraint in order to target more receivers while preserving the attractive properties of S-IDNC. We show that the minimum decoding delay problem for G-IDNC can be formulated as a maximum weight clique problem over a well structured graph. Since finding the maximum weight clique of a graph is NP-hard, we design a simple heuristic G-IDNC algorithm with sub-optimal performance. However, simulation results show that both proposed optimal and heuristic G-IDNC algorithms considerably outperform several other S-IDNC and G-IDNC optimal and heuristic approaches.

We propose a gesture recognition system based primarily on a single 3-axis accelerometer. The system employs dynamic time warping and affinity propagation algorithms for training and utilizes the sparse nature of the gesture sequence by implementing compressive sensing for gesture recognition. A dictionary of 18 gestures is defined and a database of over 3,700 repetitions is created from 7 users. Our dictionary of gestures is the largest in published studies related to acceleration-based gesture recognition, to the best of our knowledge. The proposed system achieves almost perfect user-dependent recognition and a user-independent recognition accuracy that is competitive with the statistical methods that require significantly a large number of training samples and with the other accelerometer-based gesture recognition systems available in literature.

In this paper we propose a novel wide-band spectrum sensing scheme using compressive sensing. The wide-band signal is fed into a number of wide-band filters and the outputs of the filters are used to reconstruct the vector of channel energies through the ℓ1 norm minimization. An energy detection is then performed by comparing the obtained vector to a vector of energy thresholds to decide about the occupancy of each channel. Performance of the proposed approach is compared to the current wide-band spectrum sensing algorithms as well as the conventional channel-by-channel scanning method.

In this paper, we consider the problem of minimizing the mean completion delay in wireless broadcast for instantly decodable network coding. We first formulate the problem as a stochastic shortest path (SSP) problem. Although finding the packet selection policy using SSP is intractable, we use this formulation to draw the theoretical properties of efficient selection algorithms. Based on these properties, we propose a simple online selection algorithm that efficiently minimizes the mean completion delay of a frame of broadcast packets, compared to the random and greedy selection algorithms with a similar computational complexity. Simulation results show that our proposed algorithm indeed outperforms these random and greedy selection algorithms.

The sparse nature of location finding problem makes the theory of compressive sensing desirable for indoor positioning in Wireless Local Area Networks (WLANs). In this paper, we address the received signal strength (RSS)-based localization problem in WLANs using the theory of compressive sensing (CS), which offers accurate recovery of sparse signals from a small number of measurements by solving an ¿1-minimization problem. A pre-processing procedure of orthogonalization is used to induce incoherence needed in the CS theory. In order to mitigate the effects of RSS variations due to channel impediments, the proposed positioning system consists of two steps: coarse localization by exploiting affinity propagation, and fine localization by the CS theory. In the fine localization stage, access point selection problem is studied to further increase the accuracy. We implement the positioning system on a WiFi-integrated mobile device (HP iPAQ hx4700 with Windows Mobile 2003 Pocket PC) to evaluate the performance. Experimental results indicate that the proposed system leads to substantial improvements on localization accuracy and complexity over the widely used traditional fingerprinting methods.

In this paper, we propose a QoS tracking algorithm to support different QoS requirements to accommodate delay-sensitive and throughput-sensitive traffics in wireless mesh networks. The proposed scheme is carried out in a totally distributed manner by adjusting the minimum contention window size. Moreover, we investigate whether we can satisfy all QoS requirements in a stable manner through stability analysis. The simulation results show that the proposed algorithm can satisfy both average delay bound and minimum throughput requirements without sacrifice of total system efficiency.

We consider the problem of designing cooperative driver assistance and collision warning systems in vehicular networks. In this problem each car has a small size state information message that should be received by its neighborhood within a short lifetime of L timeslots. Because of the safety nature of the application, communication reliability (success probability) and delay are of critical importance. In a repetition-based MAC scheme, each car retransmits its safety message w times in a time frame of L timeslots. Based on the proposed opportunistic network coding in this paper, given the local feedback information and already heard messages, each node tries to find the best message combining strategy such that the number of nodes that can instantly decode an uncoded packet is maximized. Simulation results show that the proposed scheme outperforms the random linear network coding as well as the uncoded case in terms of message loss probability. Also it results in lower average message reception delay compared to random linear network coding.

In the context of vehicular safety and entertainment applications, we focus on the design of a reliable medium access control scheme. Each vehicle is willing to form a network and regularly communicate with the other vehicles in its vicinity, forming a Vehicular Ad hoc Network (VANET). The information that is being communicated is short lived, and requires a very low probability of failure. In this article we assume that there is no central entity managing the medium access for the cluster of vehicles. Several code-based repetition-based QoS provisioning MAC schemes have been proposed for this scenario. We present a mathematical description of the frame failure probability, our performance metric, and explain its application in finding optimal codes.

This paper presents a new foundational approach to reason about available bandwidth estimation as the analysis of a min-plus linear system. The available bandwidth of a link or complete path is expressed in terms of aservice curve, which is a function that appears in the network calculus to express the service available to a traffic flow. The service curve is estimated based on measurements of a sequence of probing packets or passive measurements of a sample path of arrivals. It is shown that existing bandwidth estimation methods can be derived in the min-plus algebra of the network calculus, thus providing further mathematical justification for these methods. Principal difficulties of estimating available bandwidth from measurements of network probes are related to potential nonlinearities of the underlying network. When networks are viewed as systems that operate either in a linear or in a nonlinear regime, it is argued that probing schemes extract the most information at a point when the network crosses from a linear to a nonlinear regime. Experiments on the Emulab testbed at the University of Utah, Salt Lake City, evaluate the robustness of the system-theoretic interpretation of networks in practice. Multinode experiments evaluate how well the convolution operation of the min-plus algebra provides estimates for the available bandwidth of a path from estimates of individual links.

In this paper we propose a new class of repetition- based MAC protocols for Vehicular Ad-hoc Networks. The design can be used for safety applications and position information dissemination. An erasure broadcast channel is considered in which several vehicles are in a cluster and each vehicle attempts to send its own safety message to all other vehicles. A distributed feedback mechanism has been introduced to propagate the network transmission and reception information throughout the network. Based on the feedback information, a packet coding algorithm, inspired by index coding, is proposed to efficiently reduce the number of transmissions and contentions. Simulations show that the proposed protocol yields a greater average probability of successful transmission.

In this paper, a novel multiple target localization approach is proposed by exploiting the compressive sensing theory, which indicates that sparse or compressible signals can be recovered from far fewer samples than that needed by the Nyquist sampling theorem. We formulate the multiple target locations as a sparse matrix in the discrete spatial domain. The proposed algorithm uses the received signal strengths (RSSs) to find the location of targets. Instead of recording all RSSs over the spatial grid to construct a radio map from targets, far fewer numbers of RSS measurements are collected, and a data pre-processing procedure is introduced. Then, the target locations can be recovered from these noisy measurements, only through an ¿1-minimization program. The proposed approach reduces the number of measurements in a logarithmic sense, while achieves a high level of localization accuracy. Analytical studies and simulations are provided to show the performance of the proposed approach on localization accuracy.

The recent research in cluster-based MAC and routing schemes for Vehicle Ad Hoc Networks (VANETs) motivates the necessity for a stable VANET clustering algorithm. Due to the highly mobile nature of VANETs, mobility must play an integral role in cluster formation. We present a novel, mobility-based clustering scheme for Vehicle Ad hoc Networks, which utilizes the Affinity Propagation algorithm in a distributed manner. The proposed algorithm considers typical vehicular mobility during cluster formation, which produces clusters with high stability. Simulation results confirm the superior performance of the proposed algorithm, when compared to other accepted mobility-based clustering techniques. Clustering performance is measured in terms of average cluster head duration, average cluster member duration, average rate of cluster head change, and average number of clusters.

In this paper, we study the possibility of throughput improvement through precoding in OFDMA based wireless systems with limited channel feedback. Precoding can increase the overall system throughput by selecting the use of higher modulation and coding schemes (if there are any) in each physical resource unit (PRU) under a maximum bit error rate constraint. For a specific form of precoding that we proposed in, we analytically derive a formula for the maximum PRU throughput. Sufficient conditions guaranteeing that our proposed technique outperforms the conventional technique are derived. Finally, numerical results show the amount of throughput gain achieved by our proposed technique.

Peak to average power ratio (PAPR) and symbol loss rate (SLR) are two challenges of multicarrier based communications that have recently drawn much attention. High SLR renders the system unreliable and high PAPR is associated with power inefficiency and nonlinearity of the system. There are rich literatures studying these two issues separately but, unfortunately, only a few works have studied simultaneous reductions of PAPR and SLR. This paper studies the problem of reducing the PAPR while keeping the SLR at minimum. In, we derived the conditions for the minimum SLR in On-Off channels. The algorithm proposed in this paper simultaneously satisfies the conditions derived in and reduces the PAPR substantially. This paper differs from previous techniques in the sense that none of the previously proposed techniques are capable of reducing PAPR substantially while achieving the minimum symbol loss rate. We compare our algorithm with the optimum selected mapping PAPR reduction method, which is well known in the literature for having a strong reduction capability. The comparison is done in terms of Complementary Cumulative Distribution Function (CCDF) of the PAPR of the multicarrier signal. The simulation results show that our algorithm can achieve a stronger PAPR reduction while maintaining the minimum SLR.

Broadcast communications is critically important in vehicular networks. Many safety applications need safety warning messages to be broadcast to all vehicles present in an area. Design of a medium access control (MAC) protocol for vehicular networks is an interesting problem because of challenges posed by broadcast traffic, high mobility, high reliability and low delay requirements of these networks. In this article, we propose a topology-transparent broadcast protocol and present a detailed mathematical analysis for obtaining the probability of success and the average delay. We show, by analysis and simulations, that the proposed protocol outperforms two existing protocols for vehicular networks with topology-transparent properties and provides reliable broadcast communications for delivering safety messages under load conditions deemed to be common in vehicular environments.

MAC layer random network coding (MRNC) was proposed as an alternative to HARQ for reliable data transmission in WiMAX unicast. It has been shown that MRNC achieves a higher transmission efficiency than HARQ as it avoids the problem of ACK/NAK packet overhead and the additional redundancy resulting from their loss. However, it did not address the problem of restricting the number of transmissions to an upper bound which is important for delay sensitive applications. In this paper, we investigate a more structured MAC layer coding scheme that achieves the optimum performance in the delay sensitive traffic context while achieving the same overhead level as MRNC. We first formulate the delay sensitive traffic satisfaction, in such an environment, as a minimax optimization problem over all possible coding schemes. We then show that the MAC layer systematic network coding (MSNC), which transmits the packets once uncoded and employs random network coding for retransmissions, achieves the optimum performance for delay sensitive applications while achieving the same overhead level as MRNC.

In wireless multicast, the receivers are interested in obtaining only a subset of the packets transmitted by the access node. Consequently, it is intuitively assumed that random network coded packet retransmissions will result in a lower bandwidth efficiency compared to opportunistic network coded retransmissions as the former involves the delivery of unwanted packets. In the first part of this paper, we show, through simulations, that the random network coded retransmission (RNCR) scheme outperforms the opportunistic network coded retransmission (ONCR) scheme in terms of bandwidth efficiency in a wide range of multicast settings. Motivated by this result, we propose an adaptive algorithm that can dynamically select, from the RNCR and ONCR schemes, the one that is expected to achieve a better performance for each multicast frame. Simulation results show that the proposed algorithm almost achieves the optimal performance that can be obtained by combining these two retransmission schemes.

In this paper, a novel compressive sensing for manifold learning protocol (CSML) is proposed for localization in wireless sensor networks (WSNs). Intersensor communication costs are reduced significantly by applying the theory of compressive sensing, which indicates that sparse signals can be recovered from far fewer samples than that needed by the Nyquist sampling theorem. We represent the pair-wise distance measurement as a sparse matrix. Instead of sending full pair-wise measurement data to a central node, each sensor transmits only a small number of compressive measurements. And the full pair-wise distance matrix can be well reconstructed from these noisy compressive measurements in the central node, only through an ¿1-minimization algorithm. The proposed method reduces the overall communication bandwidth requirement per sensor such that it increases logarithmically with the number of sensors and linearly with the number of neighbors, while achieves high localization accuracy. CSML is especially suitable for manifold learning based localization algorithms. Simulation results demonstrate the performance of the proposed protocol on both the localization accuracy and the communication cost reduction.

Due to the high round trip delay in satellite systems, the retransmission of lost packets using conventional ARQ schemes is performed in a very rigid manner and after a very long time of the initial packet transmission. This results in a high average packet delay and packet drop rate in broadcast streaming applications. Moreover, conventional ARQ schemes are generally inefficient in broadcast scenarios. In this paper, we propose a network coded ARQ protocol that performs both proactive and reactive packet retransmissions in hybrid satellite systems. The proposed protocol employs a network coding approach to generate efficient proactive retransmission packets without the knowledge of lost packets. This not only allows the transmission of these coded retransmissions before the arrival of the initial packets to their destinations but also achieves more efficient packet recovery compared to conventional ARQ. Reactive retransmissions in response to packet acknowledgments are then employed if one or more packets are still lost. Simulation results show considerable gains for our proposed protocol over the selective repeat ARQ protocol in terms of average packet delay, packet drop rate and goodput.

Two-hop network architectures, with wireline and 802.16a backhauls, respectively, and 802.11 repeaters/relays, were proposed for high-speed rail. The resulting infrastructure cost in the former, and complexity of dual mode wireless relays in the latter, urge the need for more technically and economically efficient solutions. In this paper, we first propose a two-hop wireless network architecture for high-speed rail employing 802.16j. Due to its backward compatibility with 802.16e, the use of 802.16j not only mitigates the restrictions of the previous two-hop heterogeneous solutions but also allows a third direct communication link from the base-station to the trains, thus providing opportunities for throughput improvements. We then propose a network coded downlink transmission scheme over the proposed network architecture to both eliminate the undesirable ARQ overhead in high-speed rail communications and better exploit relay diversity. We refer to our proposed scheme as network coded information raining. Simulation results show the merits of our proposed solutions.

This paper studies the problem of minimizing symbol loss probability while keeping the system throughput above a certain threshold in downlink transmission of future OFDMA based wireless networks that rely on imperfect one-bit channel state feedback. To solve this problem, we study different preceding classes and propose a new class of preceding matrices that can gain a better result. This work is different from previous OFDM preceding literature in two main aspects. First, it addresses a more practical case where one-bit channel state feedback is available at the base station. Second, it compares precoding classes and proposes a new one. We prove analytically that our proposed precoding class has a lower symbol loss probability than the existing classes. Numerical evaluations show that a large gain in symbol loss probability is achieved by our class in comparison with the other classes.

In this paper, we introduce distributed traffic regulation for wireless local area networks (WLANs) operating in ad hoc mode. We use the concept of service curve to determine the minimum guaranteed service given to a backlogged connection. In a WLAN, the service curve depends on the number of active nodes, their traffic load in active periods, and the back-off procedure used for contention resolution. We will show that service curve can be used for distributed traffic regulation. We use some of the data packets, denoted as the probing packets, to estimate the service curve. The call is accepted if the service curve is higher than a preselected threshold, called the universal lower bound. The universal lower bound is independent of the number of nodes and traffic fluctuation and acts as a worst-case reference point for the network performance.

In this paper, a new information theoretic algorithm is proposed for signal enumeration in DS-CDMA networks. The approach is based on the predictive description length (PDL) of the observation vector. The PDL is the length of a predictive code of observations. For signal detection, the PDL criterion is computed for the candidate models and is minimized to determine the number of signals. The proposed technique uses the maximum likelihood (ML) estimate of the correlation matrix. The only information used in the ML estimation of the correlation matrix is the multiplicity of the smallest eigenvalue. The PDL algorithm has a signal-to-noise ratio resolution threshold that is smaller than that of the minimum description length (MDL). The proposed method can be used on-line and can be applied to time-varying and non-stationary systems.

In this paper, a new technique is proposed for estimating the total spatial power spectral density (PSD) caused by multiple incoherently distributed sources. Our approach is based on the fact that the array covariance matrix can be represented through the moments of the spatial PSD. Based on this representation, we develop a computationally efficient technique to estimate the moments from the array covariance matrix. The so-obtained moments are then used to estimate the total spatial PSD.

We present a distributed scheduling algorithm for provisioning of guaranteed link bandwidths in ad hoc mesh networks. The guaranteed link bandwidths are necessary to provide deterministic end-to-end bandwidth guarantees. Using Time Division Multiple Access (TDMA), links are assigned slots in each frame and during each slot a number of non-conflicting links can transmit simultaneously. The bandwidth of each link is given by the number of slots assigned to it the frame and the modulation used in the slots. Our scheduling algorithm has two parts. The first part of the algorithm is an iterative procedure that finds locally feasible schedules by exchanging link scheduling information between nodes. The iterative procedure is based on the distributed Bellman-Ford algorithm running on the conflict graph, whose partial view is available at every node. The second part of the algorithm is a wave based termination procedure used to detect when all nodes are locally scheduled and a new schedule should be activated. We use analysis to show the worst case convergence time of the algorithm and simulations to show performance of the algorithm in practice.

We present a novel cooperative vehicle position estimation algorithm, which can achieve higher levels of accuracy and reliability than existing GPS based positioning solutions by making use of inter-vehicle distance measurements taken by a radio ranging technology. Our algorithm uses signal strength based inter-vehicle distance measurements, road maps, vehicle kinematics, and Extended Kalman Filtering to estimate relative positions of vehicles in a cluster. We have preformed analysis of our algorithm examining its performance bounds, computational complexity and communication overhead requirements. Also, we have shown that the accuracy of our algorithm is superior to previous proposed localization algorithms.

Time division multiple access (TDMA) based medium access control (MAC) protocols provide QoS with guaranteed access to wireless channel. However, in multihop wireless networks, these protocols may introduce delay when packets are forwarded from an inbound link to an outbound link on a node. Delay occurs if the outbound link is scheduled to transmit before the inbound link. The total round trip delay can be quite large since it accumulates at every hop in the path. This paper presents a method that finds schedules with minimum round trip scheduling delay. We show that the scheduling delay can be interpreted as a cost collected over a cycle on the conflict graph. We use this observation to formulate a min-max program for the delay across a set of multiple paths. The min-max delay program is NP-complete since the transmission order of links is a vector of binary integer variables. We design heuristics to select appropriate transmission orders. Once the transmission orders are known, a modified Bellman-Ford algorithm is used to find the schedules. The simulation results confirm that the proposed algorithm can find effective min-max delay schedules.

Significant research has been dedicated to methods that estimate the available bandwidth in a network from traffic measurements. While estimation methods abound, less progress has been made on achieving a foundational understanding of the bandwidth estimation problem. In this paper, we develop a min-plus system theoretic formulation of bandwidth estimation. We show that the problem as well as previously proposed solutions can be concisely described and derived using min-plus system theory, thus establishing the existence of a strong link between network calculus and network probing methods. We relate difficulties in network probing to potential non-linearities of the underlying systems, and provide a justification for the distinctive treatment of FIFO scheduling in network probing.

Motivated by the results in (K. Navaie et al., 2006) on the self-similarity of the downlink interference in heterogenous service DS-CDMA networks, in this paper, we propose a model-based linear adaptive-predictive method to estimate the level of interference for optimizing the system throughput and minimizing the delay for non-real-time data transmission. We use a fractional Gaussian noise (fGn) model in an appropriate time-scale to represent the self-similarity in the downlink interference. The estimated interference is utilized to allocate the available power to non-real-time services. In doing so, we use a utility-based optimization scheme and dynamic programming for time-domain optimal scheduling of non-real-time traffic. Simulation results validate the fGn model and show a substantial improvement in the delay fairness and a significant increase in the average cell throughput using our proposed scheme; and confirm that the interference model is valid for a broad range of arrival rates of non-real time traffic.

We present diversity coded directed diffusion (DCDD), a reliable and energy efficient transport protocol for sensor networks. In DCDD, the sink uses a number of receivers- called ldquoprongsrdquo-that connect to it with reliable links. Sensors split observations into many fragments and generate parity fragments with an FEC algorithm. The fragments are then distributed over the paths and simultaneously sent to the sink. The sink can reconstruct the observations if it receives a portion of the fragments that is of the same size as their original observation. We use the ns-2 simulator to examine the ability of DCDD to increase end-to-end reliability, as well as the effect of DCDD on energy consumption in the network. Our simulations show that the network where DCDD is used outperforms the network in which the sensors use only MAC retransmissions to increase reliability. DCDD makes the energy use in the network more fair and at the same time it increases the end-to-end reliability in the network. DCDD also decreases the delay in the network.

We propose a distributed algorithm that uses inter-vehicle distance estimates, made using a radio-based ranging technology, to localize a vehicle among its neighbours. Given that the inter-vehicle distance estimates contain noise, our algorithm reduces the residuals of the Euclidean distance between the vehicles and their measured distances, allowing it to accurately estimate the position of a vehicle within a cluster. In this paper, we show that our proposed algorithm outperforms previously proposed algorithms and present its performance in a simulated vehicular environment.

We study the problem of using path diversification to provide low probability of packet loss (PPL) in wireless networks. Path diversification uses erasure codes and multiple paths in the network to transmit packets. The source uses Forward Error Correction (FEC) to encode each packet into multiple fragments and transmits the fragments to the destination using multiple disjoint paths. The source uses a load balancing algorithm to determine how many fragments should be transmitted on each path. The destination can reconstruct the packet if it receives a number of fragments equal to or higher than the number of fragments in the original packet. We study the load balancing algorithm in two general cases. In the first case, we assume that no knowledge of the performance along the paths is available at the source. In such a case, the source decomposes traffic uniformly among the paths; we call this case blind load balancing. We show that for low PPL, blind load balancing outperforms single-path transmission. In the second case, we assume that a feedback mechanism periodically provides the source with information about the performance along each path. With that information, the source can optimally distribute the fragments. We show how to distribute the fragments for minimized PPL, and maximized efficiency given a bound on PPL. We evaluate the performance of the scheme through numerical simulations.

In this paper, we show that the total downlink interference in heterogeneous wireless DS-CDMA networks follows an asymptotically self-similar (as-s) process. The as-s model is valid for the interference under certain conditions on channel variations and traffic characteristics that cover a range of practical situations. We derive these conditions and generalize earlier results, obtained for data-centric cellular networks, to heterogeneous cellular networks. Simulation results for actual cases confirm analytical results, and show that non-uniform spatial distribution of users and their soft-hand-off status do not affect the nature of this self-similar process. Furthermore, we discuss the impact of the analysis developed in this paper in designing appropriate mechanisms for controlling radio resources in such networks.

This paper introduces a novel technique to maximize the lifetime of fault tolerant sensor networks. The proposed architecture uses multipath diversity in the network layer and erasure codes. We use a distributed sink where information arrives at the sink via multiple proxy nodes, called "prongs" in this paper. The sender node uses erasure coding and splits each packet into multiple fragments and transmits the fragments over multiple parallel paths. The erasure coding allows the sink to reconstruct the original packet even if some of the fragments are lost. Occasionally, the sink broadcasts a query to awaken the sensors and to allow them to collect information about probability of packet loss and energy consumption in the network. The awakened sensors then use the collected information to distribute their data among different prongs so as to maximize the network lifetime, while keeping reliability in the network above a certain level

With the growing interest in broadband wireless access (BWA) and demand for mobile high-speed connection, there is a need to extend wireless connectivity to passengers travelling in highspeed vehicles. In this paper, we use the IEEE 802.16 standard as a backhaul communication technology for broadband wireless access to railway systems. The proposed architecture uses relay elements located in the vicinity of train track to repeat the signal between the base station and the mobile vehicle. The signal transmitted from the base station is received by repeaters and relayed to the train, and vice versa. We propose spatio-temporal scheduling as a means to increase downlink throughput. The proposed spatio-temporal scheduler distributes data traffic among the repeaters in the vicinity of the train and on the route of the train. Simulation results show that a substantial improvement can be obtained when data are scheduled in both temporal and spatial dimensions.

In this paper, we develop a cross-layer model for downlink interference in heterogeneous DS-CDMA wireless cellular networks. In this model, interference is described as a function of application layer parameters (traffic characteristics) and physical layer variations (channel characteristics). We show that for a heterogeneous service DS-CDMA network, downlink interference is a second-order self-similar process and thus has long-range dependence. We then use the predictive structure of total downlink interference to maximize non-realtime data throughput. We use fractional Gaussian noise (fGn) to model the self-similarity of downlink interference. In the proposed method, the base-station uses an optimal linear predictor, based on the fGn model, to estimate the level of interference. The estimated interference is then used to allocate power to users. To maximize data throughput, we use time domain scheduling. The simulation studies confirm the self-similarity of downlink interference and validate the fGn model. The simulation results also show a substantial performance improvement using the proposed predictive-adaptive scheme and confirm that the interference model is still valid after applying the proposed method.

In this letter, we propose a new robust MUD, called /spl alpha/ detector, for non-Gaussian noise. We consider the Gaussian-mixture model for non-Gaussian or impulsive noise. Our technique outperforms the decorrelator and the minimax detectors in highly impulsive noise. The proposed method uses a parametric cost function, where the parameter /spl alpha/ is selected using the difference between the asymptotic variance of estimation error of the /spl alpha/ detector and that of the minimax detector.

In this paper, we propose a link layer design for mobile hotspots. We design a novel system architecture that enables high-speed Internet access in railway systems. The proposed design uses a number of repeaters placed along the track and multiple antennas installed on the roof of a vehicle. Each packet is decomposed into smaller fragments and relayed to the vehicle via adjacent repeaters. We also use erasure coding to add parity fragments to original data. This approach is called information raining since fragments are rained upon the vehicle from adjacent repeaters. We investigate two instances of information raining. In blind information raining, all repeaters awaken when they sense the presence of the vehicle. The fragments are then blindly transmitted via awakened repeaters. A vehicle station installed inside the train is responsible for aggregating a large enough number of fragments. In the throughput-optimized information raining, the vehicle station selects a bipartite matching between repeaters and roof-top antennas and activates only a subset of the repeaters. It also dictates the amount of transmission power of each activated repeater. Both the bipartite matching and power allocations are individually shown to be NP-complete. Matching heuristics based on the Hungarian algorithm and Gale-Shapley algorithm are proposed. A simplex-type algorithm is proposed as the power allocation heuristics.

This paper develops a new network provisioning and resource allocation scheme. We introduce the concept of the effective burstiness curve (EBC), which is defined as a percentile of the maximum burstiness curve. For a fixed service rate, EBC represents the size of a buffer for which the probability of buffer overflow is arbitrarily small. We show that EBC is a convex nonincreasing function of the service rate. We also introduce the empirical effective burstiness curve (EEBC), an estimator of EBC, which can be obtained with a water-filling algorithm. For discrete queue size, EEBC can be evaluated with a recursive algorithm. The technique is applied to MPEG4 encoded video traces.

where he taught and conducted research in both Antenna Arrays for Cellular Networks and ATM/MPLS Networks.During this period, he was also a consultant to Iran Telecommunications Research Center and directed various projects funded by private sector.Dr Valaee'

In this paper, a new information theoretic algorithm is proposed for signal enumeration in DS-CDMA networks. The approach is based on the predictive description length (PDL), that is, the length of a predictive code of observations. The PDL cost is computed for the candidate models and is minimized to determine the number of signals. The proposed technique uses the maximum likelihood (ML) estimate of the correlation matrix. The only information used in the ML estimation of the correlation matrix is the multiplicity of the smallest eigenvalue, therefore the method is applicable to blind multiuser detection. The PDL algorithm has a signal-to-noise ratio resolution threshold that is smaller than that of the minimum description length (MDL). The proposed method can be used on-line and can be applied to time-varying and non-stationary systems.

The leaky-bucket regulator has several potential roles in the operation of future transport networks; among them, the bounding of possible source trajectories in implementations of worst-case approaches to network design. It seems plausible that there will be applications whose specific traffic characteristics are known a priori neither to the user nor to the network; in such cases, a recursive algorithm for setting the leaky-bucket parameters may prove useful. We devise such an algorithm here. The leaky-bucket parameters are computed recursively over a limited period of observation of the source behavior. We provide an explicit characterization of the dynamics of the estimator, and the results of a simulation study of performance in the case of real source trajectories.

A new algorithm for the processing of signals by an array of sensors is proposed. The objective is to find the number and the directions of arrival (DOA) of signals impinging on a linear array. The predictive stochastic complexity criterion of Rissanen (1986) is used to select the best model order. To reduce the computational load, the algorithm operates with a suboptimal estimator while maintaining the consistency of the estimator. The proposed method is on-line and can be utilized in time-varying systems for target tracking. The method can be used for both correlated and uncorrelated signals.>