Showing papers by "Neal Patwari published in 2012"

A Fade-Level Skew-Laplace Signal Strength Model for Device-Free Localization with Wireless Networks

Abstract: Device-free localization (DFL) is the estimation of the position of a person or object that does not carry any electronic device or tag. Existing model-based methods for DFL from RSS measurements are unable to locate stationary people in heavily obstructed environments. This paper introduces measurement-based statistical models that can be used to estimate the locations of both moving and stationary people using received signal strength (RSS) measurements in wireless networks. A key observation is that the statistics of RSS during human motion are strongly dependent on the RSS "fade level” during no motion. We define fade level and demonstrate, using extensive experimental data, that changes in signal strength measurements due to human motion can be modeled by the skew-Laplace distribution, with parameters dependent on the position of person and the fade level. Using the fade-level skew-Laplace model, we apply a particle filter to experimentally estimate the location of moving and stationary people in very different environments without changing the model parameters. We also show the ability to track more than one person with the model.

Enhancing the accuracy of radio tomographic imaging using channel diversity

Abstract: Radio tomographic imaging (RTI) is an emerging device-free localization (DFL) technology enabling the localization of people and other objects without requiring them to carry any electronic device. Instead, the RF attenuation field of the deployment area of a wireless network is estimated using the changes in received signal strength (RSS) measured on links of the network. This paper presents the use of channel diversity to improve the localization accuracy of RTI. Two channel selection methods, based on channel packet reception rates (PRRs) and fade levels, are proposed. Experimental evaluations are performed in two different types of environments, and the results show that channel diversity improves localization accuracy by an order of magnitude. People can be located with average error as low as 0.10 m, the lowest DFL location error reported to date. We find that channel fade level is a more important statistic than PRR for RTI channel selection. Using channel diversity, this paper, for the first time, demonstrates that attenuation-based through-wall RTI is possible.

Follow @grandma: Long-term device-free localization for residential monitoring

Abstract: Device-free localization (DFL) enables localizing people by monitoring the changes in the radio frequency (RF) attenuation field of an area where a wireless network is deployed. Notably, this technology does not require people to participate in the localization effort by carrying any electronic device. This paper presents a DFL system for long-term residential monitoring. Due to the daily activities carried out by the people being monitored, the radio signals' propagation patterns change continuously. This would make a system relying only on an initial calibration of the radio environment highly inaccurate in the long run. In this paper, we use an on-line recalibration method that allows the system to adapt to the changes in the radio environment, and then provide accurate position estimates in the long run. A finite-state machine (FSM) defines when the person is located at specific areas-of-interest (AoI) inside the house (e.g. kitchen, bathroom, bed, etc.). Moreover, each time a state transition is triggered, the system tweets the new AoI in a Twitter account. The FSM allows extracting higher level information about the daily routine of the person being monitored, enabling interested parties (e.g. caretakers, relatives) to check that everything is proceeding normally in his life. In the long-term experiment carried out in a real domestic environment, the system was able to accurately and reliably localize the person during his daily activities.

Monitoring breathing via signal strength in wireless networks

Abstract: Systems and methods are disclosed for the use of sensor links in a network to estimate the breathing rate of a breathing subject within a structure, estimate the location of the subject within the structure, and detect if the subject is breathing. The structure may be a bed, a building, or a room in the building. The received signal strength of the sensor links is obtained and is then used in various breathing models to determine the breathing rate estimation, the location estimation, and the breathing detection.

Radio Tomographic Imaging for Ambient Assisted Living

Abstract: Accurate localization of people in indoor and domestic environments is one of the key requirements for ambient assisted living (AAL) systems. This chapter describes how the received signal strength (RSS) measurements collected by a network of static radio transceivers can be used to localize people without requiring them to wear or carry any radio device. We describe a technique named radio tomographic imaging (RTI), which produces real-time images of the change in the radio propagation field of the monitored area caused by the presence of people. People’s locations are inferred from the estimated RTI images. We show results from a long-term deployment in a typical single floor, one bedroom apartment. In order to deal with the dynamic nature of the domestic environment, we introduce methods to make the RTI system self-calibrating. Experimental results show that the average localization error of the system is 0.23 m. Moreover, the system is capable of adapting to the changes in the indoor environment, achieving high localization accuracy over an extended period of time.

Passive RFID tomographic imaging for device-free user localization

Abstract: Localization of users is an important part of location aware systems and smart environments. It forms a major data source for superimposed intention recognition systems. In RF device-free localization (DFL), the person being tracked does not need to wear a RF transmitter or receiver in order to be located. Instead, they are tracked using the changes in signal strength measured on static links in a wireless network. This work presents a new algorithm for RF DFL using passive RFID networks. We formulate and show how a tomographic imaging algorithm provides both low computational complexity and highly accurate position estimates. Using measurements conducted in an indoor environment with various human positions, we find the algorithm can locate the human with as low as 30 cm mean location error.

Channel Sounding for the Masses: Low Complexity GNU 802.11b Channel Impulse Response Estimation

Abstract: New techniques in cross-layer wireless networks are building demand for ubiquitous channel sounding, that is, the capability to measure channel impulse response (CIR) with any standard wireless network and node. Towards that goal, we present a software-defined IEEE 802.11b receiver and CIR measurement system with little additional computational complexity compared to 802.11b reception alone. The system implementation, using the universal software radio peripheral (USRP) and GNU Radio, is described and compared to previous work. We validate the CIR measurement system and present the results of a measurement campaign which measures millions of CIRs between WiFi access points and a mobile receiver in urban and suburban areas.

Directed by Directionality: Benefiting from the Gain Pattern of Active RFID Badges

Abstract: Tracking of people via active badges is important for location-aware computing and for security applications. However, the human body has a major effect on the antenna gain pattern of the device that the person is wearing. In this paper, the gain pattern due to the effect of the human body is experimentally measured and represented by a first-order directional gain pattern model. A method is presented to estimate the model parameters from multiple received signal strength (RSS) measurements. An alternating gain and position estimation (AGAPE) algorithm is proposed to jointly estimate the orientation and the position of the badge using RSS measurements at known-position anchor nodes. Lower bounds on mean squared error (MSE) and experimental results are presented that both show that the accuracy of position estimates can be greatly improved by including orientation estimates in the localization system. Next, we propose a new tracking filter that accepts orientation estimates as input, which we call the orientation-enhanced extended Kalman filter (OE-EKF), which improves tracking accuracy in active RFID tracking systems.

Histogram distance-based radio tomographic localization

Abstract: We present an interactive demonstration of histogram distance-based radio tomographic imaging (HD-RTI), a device-free localization (DFL) system that uses measurements of received signal strength (RSS) on static links in a wireless network to estimate the locations of people who do not participate in the system by wearing any radio device in the deployment area. Compared to prior methods of RSS-based DFL, using a histogram difference metric is a very accurate method to quantify the change in RSS on the link compared to historical metrics. The new method is remarkably accurate, and works with lower node densities than prior methods.

Demo abstract: Histogram distance-based radio tomographic localization

Abstract: We present an interactive demonstration of histogram distance-based radio tomographic imaging (HD-RTI), a device-free localization (DFL) system that uses measurements of received signal strength (RSS) on static links in a wireless network to estimate the locations of people who do not participate in the system by wearing any radio device in the deployment area. Compared to prior methods of RSS-based DFL, using a histogram difference metric is a very accurate method to quantify the change in RSS on the link compared to historical metrics. The new method is remarkably accurate, and works with lower node densities than prior methods.

Hidden Markov Estimation of Bistatic Range From Cluttered Ultra-wideband Impulse Responses

Abstract: Ultra-wideband (UWB) multistatic radar can be used for target detection and tracking in buildings and rooms. Target detection and tracking relies on accurate knowledge of the bistatic delay. Noise, measurement error, and the problem of dense, overlapping multipath signals in the measured UWB channel impulse response (CIR) all contribute to make bistatic delay estimation challenging. It is often assumed that a calibration CIR, that is, a measurement from when no person is present, is easily subtracted from a newly captured CIR. We show this is often not the case. We propose modeling the difference between a current set of CIRs and a set of calibration CIRs as a hidden Markov model (HMM). Multiple experimental deployments are performed to collect CIR data and test the performance of this model and compare its performance to existing methods. Our experimental results show an RMSE of 2.85 ns and 2.76 ns for our HMM-based approach, compared to a thresholding method which, if the ideal threshold is known a priori, achieves 3.28 ns and 4.58 ns. By using the Baum-Welch algorithm, the HMM-based estimator is shown to be very robust to initial parameter settings. Localization performance is also improved using the HMM-based bistatic delay estimates.

Beyond OFDM: Best-effort dynamic spectrum access using filterbank multicarrier

Abstract: Orthogonal frequency division multiplexing (OFDM), widely recommended for sharing the spectrum among different nodes in a dynamic spectrum access network, imposes tight timing and frequency synchronization requirements. We examine the use of filterbank multicarrier (FBMC), a somewhat lesser known and understood alternative, for dynamic spectrum access. FBMC promises very low out-of-band energy of each subcarrier signal when compared to OFDM. In order to fully understand and evaluate the promise of FBMC, we first examine the use of special pulse shaping filters of the FBMC PHY layer in reliably transmitting data packets at a very high rate. Next, to understand the impact of FBMC beyond the PHY layer, we devise a distributed and adaptive medium access control (MAC) protocol that coordinates data packet traffic among the different nodes in the network in a best effort manner. Using extensive simulations, we show that FBMC consistently achieves at least an order of magnitude performance improvement over OFDM in several aspects including packet transmission delays, channel access delays, and effective data transmission rate available to each node. Using measurements of power spectral density and high data rate transmissions from a transceiver that we build using our National Instruments hardware platform, we show that while FBMC can decode/distinguish all the received symbols without any errors, OFDM cannot. In summary, FBMC offers a much higher performing alternative to OFDM for networks that dynamically share the spectrum among multiple nodes.

Experimental performance evaluation of location distinction for MIMO links

Abstract: A radio channel-based location distinction system monitors physical layer measurements of received signals to detect if a transmitter has changed position since its previous transmission. This paper explores the design space for MIMO-based location distinction systems. Using extensive channel measurements collected with two different MIMO testbeds, we make several observations about the tradeoffs inherent in MIMO location distinction, and the scaling of performance with respect to bandwidth, history size and insertion delay, and number of antenna elements. We show that MIMO location distinction is very reliable. For example, a 2×2 MIMO channel with a bandwidth of 80 MHz allows a 64-fold reduction in miss rate over the single-input single-output (SISO) channel for a fixed false alarm rate, achieving false alarm rates as low as 4 × 10−4 for a 2.4 × 10−4 probability of missed detection.

Kernel Methods for RSS-Based Indoor Localization

Abstract: Indoor localization algorithms have been proposed using various methods, such as angle of arrival, time of flight, and received signal strength (RSS) This chapter explores the features and advantages of kernel‐based localization Kernel methods simplify received signal strength (RSS)‐based localization by providing a means to learn the complicated relationship between RSS measurement vector and position The chapter discusses their use in self‐calibrating indoor localization systems It reviews four kernel‐based localization algorithms and presents a common framework for their comparison The chapter shows results from two simulations and from an extensive measurement data set, which provide a quantitative comparison and intuition into their differences It compares the performance of the different kernel‐based localization algorithms and describes the environment along with the processing of the experimental data and the evaluation procedure for each data set

Hidden Markov estimation of bistatic range from cluttered ultra-wideband impulse responses

Abstract: Multistatic radar is used for target detection and tracking in buildings and rooms. Target detection and tracking relies on accurate knowledge of the excess delay of the multipath component which travels from the transmitter, to the target, and then to the receiver. If the environment creates many multipath components, individual ultra-wideband (UWB) impulses overlap. We seek to accurately estimate a target's excess delay by considering the difference between the channel impulse response (CIR) and a known CIR of the static environment as a hidden Markov model (HMM). Experimental CIRs are used to test the performance of this model, as well as energy detection. The RMSE from applying a HMM was 2.7 ns and 2.8 ns and from applying energy detection was 5.3ns and 5.2 ns for two experimental setups.

Detecting receiver attacks in VRTI-based device free localization

Abstract: Variance-based Radio Tomographic Imaging (VRTI) is an emerging technology that locates moving objects in areas surrounded by simple and inexpensive wireless sensor nodes. VRTI uses human motion induced variation in RSS and spatial correlation between link variations to locate and track people. An artificially induced power variations in the deployed network by an adversary can introduce unprecedented errors in localization process of VRTI and, given the critical applications of VRTI, can potentially lead to serious consequences including loss of human lives. In this paper, we tackle the problem of detecting malicious receivers that report false RSS values to induce artificial power variations in a VRTI system. We use the term “Receiver Attack” to refer to such malicious power changes. We use a combination of statistical hypothesis testing and heuristics to develop real-time methods to detect receiver attack in a VRTI system. Our results show that we can detect receiver attacks of reasonable intensity and identify the source(s) of malicious activity with very high accuracy.

Localization of people using received signal strength measurements in wireless sensor networks

Abstract: Location information of people is valuable for many applications including logistics, healthcare, security and smart facilities. This dissertation focuses on localization of people in wireless sensor networks using radio frequency (RF) signals, specifically received signal strength (RSS) measurements. A static sensor network can make RSS measurements of the signal from a transmitting badge that a person wears in order to locate the badge. We call this kind of localization method radio device localization. Since the human body causes RSS changes between pairwise sensor nodes of a static network, we can also use RSS measurements from pairwise nodes of a network to locate people, even if they are not carrying any radio device. We call this device-free localization (DFL). The first contribution of this dissertation is to radio device localization. The human body has a major effect on the antenna gain pattern of the transmitting badge that the person is wearing, however, existing research on device localization ignores this effect. In this work, the gain pattern due to the effect of the human body is experimentally measured and represented by a first-order gain pattern model. A method is presented to estimate the model parameters from multiple received signal strength (RSS) measurements. An alternating gain and position estimation (AGAPE) algorithm is proposed to jointly estimate the orientation and the position of the badge using RSS measurements at known-position anchor nodes. Lower bounds on mean squared error (MSE) and experimental results are presented that both show that the accuracy of position estimates can be greatly improved by including orientation estimates in the localization system. Finally, a new tracking filter that accepts orientation estimates as input is developed, which is called the orientation-enhanced extended Kalman filter (OE-EKF). Experimental results show that this new method using the localization estimates from AGAPE algorithm improves tracking accuracy in radio device localization systems. In the field of DFL, this dissertation has two major contributions: (1) improving the robustness of variance-based DFL methods that can locate human motion; (2) developing a new DFL system that is capable of locating both moving and stationary people without using "empty-room" offline calibration. For the first contribution, two robust estimators for variance-based radio tomographic imaging (VRTI) – subspace variance-based radio tomography (SubVRT), and least squares variance-based radio tomography (LSVRT) are proposed. Human motion in the vicinity of a wireless link causes variations in the link received signal strength (RSS). DFL systems, such as VRTI, use these RSS variations in a static wireless network to locate and track people in the area of the network. However, intrinsic motion, such as branches moving in the wind and rotating or vibrating machinery, also causes RSS variations which degrade the performance of a localization system. The first robust estimator SubVRT uses subspace decomposition, and the second estimator LSVRT uses a least squares formulation on the "empty-room" calibration measurements. Experimental results show that both estimators reduce localization root mean squared error by about 40% compared to VRTI. In addition, the Kalman filter tracking results from both estimators are more robust to large errors compared to tracking results from VRTI. The second contribution in DFL is a new localization system, which we call kernel distance-based radio tomographic imaging (KRTI). Since many DFL systems including VRTI cannot locate stationary people, we present and evaluate a system that can locate stationary and moving people, with or without calibration, by quantifying the difference between two histograms of signal strength measurements. From five experiments, we show that our KRTI localization system performs better than the state-of-the-art device-free localization systems in different non-line-of-sight environments.

Incorporating link correlations in models and algorithms for localization in wireless sensor networks

Abstract: In wireless sensor networks, knowing the location of the wireless sensors is critical in many remote sensing and location-based applications, from asset tracking, and structural monitoring to geographical routing. For a majority of these applications, received signal strength (RSS)-based localization algorithms are a cost effective and viable solution. However, RSS measurements vary unpredictably because of fading, the shadowing caused by presence of walls and obstacles in the path, and non-isotropic antenna gain patterns, which affect the performance of the RSS-based localization algorithms. This dissertation aims to provide efficient models for the measured RSS and use the lessons learned from these models to develop and evaluate efficient localization algorithms. The first contribution of this dissertation is to model the correlation in shadowing across link pairs. We propose a non-site specific statistical joint path loss model between a set of static nodes. Radio links that are geographically proximate often experience similar environmental shadowing effects and thus have correlated shadowing. Using a large number of multi-hop network measurements in an ensemble of indoor and outdoor environments, we show statistically significant correlations among shadowing experienced on different links in the network. Finally, we analyze multi-hop paths in three and four node networks using both correlated and independent shadowing models and show that independent shadowing models can underestimate the probability of route failure by a factor of two or greater. Second, we study a special class of algorithms, called kernel-based localization algorithms, that use kernel methods as a tool for learning correlation between the RSS measurements. Kernel methods simplify RSS-based localization algorithms by providing a means to learn the complicated relationship between RSS measurements and position. We present a common mathematical framework for kernel-based localization algorithms to study and compare the performance of four different kernel-based localization algorithms from the literature. We show via simulations and an extensive measurement data set that kernel-based localization algorithms can perform better than model-based algorithms. Results show that kernel methods can achieve an RMSE up to 55 % lower than a model-based algorithm. Finally, we propose a novel distance estimator for estimating the distance between two nodes a and b using indirect link measurements, which are the measurements made between a and k, for k ≠ b and b and k, for k ≠ a. Traditionally, distance estimators use only direct link measurement, which is the pairwise measurement between the nodes a and b. The results show that the estimator that uses indirect link measurements enables better distance estimation than the estimator that uses direct link measurements.