An Impulsive Noise-Resistant Target Localization Approach With Unknown Model Parameter Learning

We present a novel hybrid localization algorithm for wireless sensor networks in the absence of knowledge regarding the transmit power and path-loss exponent. Transmit power and the path-loss exponent are critical parameters for target localization algorithms in wireless sensor networks, which help extract target position information from the received signal strength. In the absence of information on transmit power and path-loss exponent, it is critical to estimate them for reliable deployment of conventional target localization algorithms. In this paper, we propose a simultaneous estimation of transmit power and path-loss exponent based on Kalman filter. The unknown transmit power and path-loss exponent are estimated using a Kalman filter with the tentatively estimated target position based solely on angle information. Subsequently, the target position is refined using a hybrid method incorporating received signal strength measurements based on the estimated transmit power and path-loss exponent. Our proposed algorithm accurately estimates transmit power and path-loss exponent and yields almost the same target position accuracy as the simulation results confirm, as the hybrid target localization algorithms with known transmit power and path-loss exponent. Simulation results confirm the proposed algorithm achieves 99.7% accuracy of the target localization performance with known transmit power and path-loss exponent, even in the presence of severe received signal strength measurement noise.

This paper investigates the problem of received signal strength (RSS) based target localization with sensor position uncertainties and unknown model parameters. Transmit power and path-loss exponent (PLE) are two key parameters of the path-loss model that is critical for RSS schemes to work well. In practice, however, these parameters change dynamically and are difficult to estimate in real time. Four combinations of the two parameters are considered: both parameters are known; only one of the two is known; and both are unknown. We develop new semidefinite programming (SDP) methods to relax the nonconvex maximum likelihood estimator (MLE) problem for each case, and these methods are further optimized when sensor position uncertainties are taken into consideration. We present a comprehensive set of simulation results to compare the performance of the proposed scheme with existing methods.

In this correspondence, we investigate unmanned aerial vehicle (UAV)-base stations (BSs) assisted and received signal strength (RSS) based mobile station (MS) localization. A practical air-to-ground path loss model is utilized, where the path loss exponent (PLE) varies with the elevation angle and altitude of UAV, and the accurate PLE estimate is often difficult to obtain. With unknown and unequal PLEs for different UAVs, the UAVs assisted localization problem becomes nonlinear and non-convex, which cannot be solved by the existing methods. We propose a piecewise convex approximation aided localization (PCAL) approach to convert the localization problem into linear and convex, without requiring the knowledge of PLE. The proposed PCAL approach with unknown and unequal PLEs achieves much higher accuracy than the existing methods which require perfectly known and equal PLE, due to its higher robustness against shadowing. In addition, a grid search aided ambiguity elimination (GSAE) method, which is more effective than the state-of-the-art methods, is proposed to determine the final MS localization estimate based on multiple tentative estimates derived from PCAL. The effectiveness of PCAL is also verified by the Cramer-Rao lower bound (CRLB) derived.

In this paper, we present an estimation scheme for the transmit power(TP) and path loss exponent(PLE) for hybrid target localization in wireless sensor networks. In the absence of knowledge on the TP and PLE, hybrid target localization schemes that utilize both received signal strength (RSS) and angle of arrival(AOA) measurements have significantly reduced performances. To estimate the TP and PLE in the presence of a severe measurement noise, moving-averaged RSS measurements are applied to a least-squares algorithm utilizing a tentatively estimated target position using an AOA-only algorithm.

A weighted hybrid indoor positioning method based on path loss exponent estimation

Localization of wireless devices via the received signal strength (RSS) has become very popular in recent years owing to its economical and uncomplicated implementation. The performance of RSS based location systems is highly dependent on the knowledge of the propagation channel. An unrealistic assumption in many current studies is the possession of the exact knowledge of the path-loss exponent (PLE). This letter deals with the estimation of location coordinates in unknown path-loss model. First, the non-linear path-loss equations are linearized and the solution is obtained using linear least square (LLS) solution. Second, in order to underline the effects of inaccurate PLE assumption in noisy channels on location coordinates, error analysis is done when incorrect PLE is assumed, hence a closed form expression is derived for mean square error (MSE) and bias. Next, we transform the LLS problem into a single variable minimization problem to estimate jointly the location and PLE. Simulation results show that incorrect PLE assumption has superlative impact on MSE and bias of location estimates. Further simulation shows that the proposed technique exhibits acceptable performance while estimating the PLE however the estimates are biased. In order to achieve higher accuracy, the obtained estimates are used as an initial estimate to our previously proposed iterative algorithm [1]. The resultant MSE is then compared with the Cramer-Rao bound and the improved convergence is demonstrated.

Using the classical received signal strength (RSS)-distance log-normal model in wireless sensor network (WSN) applications poses a series of characteristic challenges derived from (a) the model’s structural limitations when it comes to explaining real observations, (b) the inherent hardware (HW) variability typically encountered in the low-cost nodes of WSNs, and (c) the inhomogeneity of the deployment environment. The main goal of this article is to better characterize how these factors impact the model parameters, an issue that has received little attention in the literature. For that matter, I qualitatively elaborate on their effects and interplay, and present the results of two quantitative empirical studies showing how much the parameters can vary depending on (a) the nodes used in the model identification and their position in the environment, and (b) the antenna directionality. I further show that the path loss exponent and the reference power can be highly correlated. In view of all this, I argue that real WSN deployments are better represented by random model parameters jointly accounting for HW and local environmental characteristics, rather than by deterministic independent ones. I further argue that taking this variability into account results in more realistic models and plausible results derived from their usage. The article contains example values of the mean and standard deviation of the model parameters, and of the correlation between the path loss exponent and the reference power. These can be used as a guideline in other studies. Given the sensitivity of localization algorithms to the proper model selection and identification demonstrated in the literature, the structural limitations of the log-normal model, the variability of its parameters and their interrelation are all relevant aspects that practitioners need to be aware of when devising optimal localization algorithms for real WSNs that rely on this popular model.

In recent years, the knowledge of location is of prime importance for success of many location-based services (LBS) applications such as indoor navigation, emergency management, elderly tracking, etc. There are many popular approaches in the literature which utilize received signal strengths (RSSs) in wireless sensor network (WSN) to track the moving target. Many times for the applications wherein localization accuracy to the scale of few centimetres is required, the application of one localization approach is not sufficient. The important reason is the highly fluctuating nature of RSS’s, and dynamicity in target motion and surrounding environment. Although generalized regression neural network (GRNN) has very nice estimation ability, we believe that the addition of the concept centroid to the GRNN architecture and fusing it with kalman filter (KF) can further improve its target localization performance. This paper proposes two robust range-free RSS based target localization algorithms namely, CGRNN+KF, and CGRNN+UKF, obtained by the fusion of fundamentals of centroid, GRNN, and KF. From the results obtained, it is observed that the proposed CGRNN+KF, and CGRNN+UKF algorithms yield positioning accuracy of few centimetres and effectively deal with the highly fluctuating nature of RSS’s, and dynamicity in target motion and surrounding environment.

Compressive sensing is a sub-Nyquist sampling technique for efficient signal acquisition and reconstruction of sparse or compressible signals. In order to account for the sparsity of the underlying signal of interest, it is common to use sparsifying priors such as Bernoulli–Gaussian-inverse Gamma (BGiG) and Gaussian-inverse Gamma (GiG) priors on the components of the signal. With the introduction of variational Bayesian inference, the sparse Bayesian learning (SBL) methods for solving the inverse problem of compressive sensing have received significant interest as the SBL methods become more efficient in terms of execution time. In this paper, we consider the sparse signal recovery problem using compressive sensing and the variational Bayesian (VB) inference framework. More specifically, we consider two widely used Bayesian models of BGiG and GiG for modeling the underlying sparse signal for this problem. Although these two models have been widely used for sparse recovery problems under various signal structures, the question of which model can outperform the other for sparse signal recovery under no specific structure has yet to be fully addressed under the VB inference setting. Here, we study these two models specifically under VB inference in detail, provide some motivating examples regarding the issues in signal reconstruction that may occur under each model, perform comparisons and provide suggestions on how to improve the performance of each model.

In the field of marine robotics, the problem of rangebased underwater target localization can be defined as that of localizing an unknown — fixed or moving — target from a surface vehicle called the tracker, using range information available about the target. In this set-up, the motion of the tracker (relative to the target) has a significant effect on its ability to localize the target. In general, without any prior knowledge about the target motion, this is a very challenging problem and one that requires, at the outset, a thorough observability analysis of the model adopted for localization system design. For a restricted class of target motions generated using constant linear speed and piecewise constant course rates throughout a given observation window, the target's initial position, constant speed, initial course angle, and course rate, completely characterize its motion. The conditions under which one can estimate any combinations of these four parameters using range information has not been fully studied, except for a simple isolated case where the initial target position is the only unknown parameter. In this paper, using range information, we address the observability properties of the target localization problem for a target moving along a straight line or an arc of a circumference with constant linear and angular speeds on a given observation window. In the first case there are three target parameters (initial position, linear speed, and course angle), while in the second case there are four (initial position, linear speed, angular speed, and course angle). To make the problem mathematically tractable, we consider simple kinematics models for the tracker (we assume a constant linear speed) and target with planar position and course angle as state variables. Under these conditions, for the aforementioned two cases, we investigate the observability properties of range-based target localization problem for various target parameter combinations. In particular, when the target is moving along a straight line, for most of the unknown target parameter combinations, we show that observability can be achieved by a nonzero constant tracker's course rate, which is a simple condition. In the case where the target moves along a circular path, with the knowledge of the angular speed, we derive sufficient conditions on the trackers input to achieve observability, which are more demanding and less straightforward, and do not lend themselves to a simple geometrical explanation.

A robust Cubature Kalman filter for nonlinear systems subject to randomly occurring measurement anomalies without a priori statistic

Robust Bayesian learning approach for massive MIMO channel estimation

Orthogonal frequency division multiplexing (OFDM) is adopted in most wireless communication systems, and the performance of OFDM is affected by impulse noise (IN). Better channel estimation (CE) performance is required to detect the channel information on the receiving side. The OFDM system is considered with the number of subcarriers carrying the data and pilot symbols. In the OFDM modulator, the symbols of the frequency domain are transformed into the signals of the time domain using inverse discrete Fourier transformation (IDFT). Effective impulsive noise mitigation is crucial for improving the effectiveness of OFDM communication systems, and it improves the signal-to-noise ratio (SNR) at the receiver. A cyclic prefix is then appended before transmission through the channel. In the receiver noise, the effect of IN is mitigated jointly with CE using the Bayesian matching pursuit (BMP) and Moth-flame algorithm (MFA). In this approach, the IN and channel information is considered as a sparse vector. Here, the data symbols used are considered an unknown parameter. The approach incorporating all subcarriers has a lower mean square error (MSE) of IN estimate for impulsive noise reduction. The MFA algorithm is used to optimize the sensor matrix in BMP. The sparsity of the channel and the impulse response are observed in the time domain to represent the channel and IN together. It exactly recovers the sparse signal for finding the convex objectives of the sparse minimizer, and the sparse solution is obtained with fixed point updates. The proposed BMP algorithm improves the CE by explicitly considering the existence of IN. The BMP is a greedy algorithm that selects the most correlated residuals at each column. Under mutual incoherence, it recovers the signal with a higher probability. The simulated outcomes proved that the proposed CE and noise mitigation model achieved better performance based on the bit error rate (BER), throughput and MSE.

With growing technology, Location Based Services (LBSs) are becoming a vital part of life. The rapid development of mobile internet has offered an opportunity for Wi-Fi indoor positioning to come under the spotlight due to its low cost. Here, indoor positioning is performed using Wi-Fi. The earlier solutions like GPS does not provide efficient positioning in indoor environment due to weak signal reception which makes Wi-Fi more reliable. However, nowadays the accuracy of the traditional indoor positioning algorithms like trilateration cannot meet the practical requirement. This paper proposes the performance analysis of indoor positioning algorithms using Wi-Fi technology based on Received Signal Strength (RSS) from the access points in the network. The positioning algorithms are based on location fingerprinting consisting of two stages: The offline database creation and the online positioning. The RSS values obtained by the mobile device at reference points are sent to the server where all the location fingerprints of all reference points are stored in a database. Coordinates of user position are calculated using positioning algorithms and the position is displayed in terms of 2Dcoordinates of the room and indoor map. Basic Fingerprint algorithm, RSS Fingerprint with minimum Euclidean Distance algorithm and RSS Fingerprint with Euclidean Distance and k-nearest neighbors' algorithm were implemented. Moreover comparison of their performance was carried out in terms of accuracy of position. From the result, it is found that an average accuracy of 0.4962 meters was achieved by RSS Fingerprint with minimum Euclidean distance algorithm.

Localization of mobile phones is important to location-based mobile services, but achieving good location estimation of mobile phones is difficult especially in environment whose path loss exponent is unknown. In this paper, we present a Wi-Fi localization solution specifically designed for dense WLANs with unknown path loss exponent. In order to leverage between the computational cost and localization accuracy, our solution establishes a neighbor selection scheme based on the Voronoi diagram to identify a subset of Access Points (APs) to participate in localization. It considers the identified subset of APs and a mobile phone to be located as a mass-spring system. Provided with information of known coordinates of APs, the solution estimates the path loss exponent of the physical environment, infers inter-distances between APs and the mobile phone from Wi-Fi signals received, and implements spring relaxation algorithm to approximate the geographical location of the mobile phone, where this location estimation is fed back to refine the estimated exponent iteratively. Extensive simulation results confirm that our solution is able to provide location estimation with an attractive average accuracy of below 2 m in a typical Wi-Fi setup.