Nicolas Amiot

PyLayers: An open source dynamic simulator for indoor propagation and localization

PyLayers is a new open source radio simulator. It has been designed to evaluate localization algorithm performances through the realistic simulation of location-dependent parameters (LDPs) in heterogeneous mobile radio networks. The radio channel is synthesized by using a novel graph-based ray tracing method which has been introduced in order to improve performances in mobile ray-tracing scenarios where geometrical information reuse from one mechanical time-step to another is advantageous. PyLayers can synthesized the narrow band, wide band or ultra wide band (UWB) channel impulse response and thus allows to produced various kind of location dependent parameters as the widely used LDPs received power an time of arrival. Realistic movement of pedestrian agents into the building layout is modeled with a virtual forces approach. The simulated data can be directly exploited with one of the original built-in localization algorithms or be exported to various standards file extensions for external post-processing. Examples of typical PyLayers outputs are provided.

Qualitative analysis of RSSI behavior in cooperative wireless body area networks for mobility detection and navigation applications

In this paper, we account for radio-location experiments aiming at both indoor navigation and mobility detection applications for Wireless Body Area Networks (WBAN). This measurement campaign involved IEEE 802.15.4-compliant integrated radio devices organized within a full mesh topology over on-body and off-body links. The latter devices produce peer-to-peer Received Signal Strength Indicators (RSSI) that could feed ranging, positioning or tracking algorithms. An in-depth behavioral analysis of the collected time-stamped radio-location metrics is thus proposed with respect to the captured human mobility (including body shadowing). Based on our observations and interpretations, practical insights are finally drawn in terms of system and algorithms design.

Cooperative and heterogeneous indoor localization experiments

In this paper we present the results of real-life localization experiments performed in an unprecedented cooperative and heterogeneous wireless context. These measurements are based on ZigBee and orthogonal frequency division multiplexing (OFDM) devices, respectively endowed with received signal strength indicator (RSSI) and round trip delay (RTD) estimation capabilities. More particularly we emulate a multi-standard terminal, moving in a typical indoor environment, while communicating with fixed OFDM-based femto-base stations (Femto-BSs) and with other mobiles or fixed anchor nodes (through peer-to-peer links) forming a wireless sensor network (WSN). We introduce the measurement functionalities and metrics, the scenario and set-up, providing realistic connectivity and obstruction conditions. Out of the experimental data, preliminary positioning results based on cooperative and geometric algorithms are finally discussed, showing benefits through mobile-to-mobile cooperation, selective hybrid data fusion and detection of unreliable nodes.

A Hybrid Positioning Method Based on Hypothesis Testing

We consider positioning in the scenario where only two reliable range estimates, and few less reliable power observations are available. Such situations are difficult to handle with numerical maximum likelihood methods which require a very accurate initialization to avoid being stuck into local maxima. We propose to first estimate the support region of the two peaks of the likelihood function using a set membership method, and then decide between the two regions using a rule based on the less reliable observations. Monte Carlo simulations show that the performance of the proposed method in terms of outlier rate and root mean squared error approaches that of maximum likelihood when only few additional power observations are available.

Improved mobility modeling for indoor localization applications

This paper presents a novel mobility model to perform realistic simulations of human movements and behaviors. The proposed model is based on discrete event simulation and graph theory. The proposed model is implemented in a wireless propagation simulator and used to evaluate various wireless network protocols including: propagation, localization and communication.

Assessment of cooperative and heterogeneous indoor localization algorithms with real radio devices

In this paper we present results of real-life localization experiments performed in an unprecedented cooperative and heterogeneous wireless context. The experiments covered measurements of different radio devices packed together on a trolley, emulating a multi-standard Mobile Terminal (MT) along representative trajectories in a crowded office environment. Among all the radio access technologies involved in this campaign (including LTE,WiFi...), the focus is herein put mostly on Impulse Radio - Ultra Wideband (IR-UWB) and ZigBee sub-systems, which are enabled with peer-to-peer ranging capabilities based on Time of Arrival (ToA) estimation and Received Signal Strength (RSS) measurements respectively. Single-link model parameters are preliminarily drawn and discussed. In comparison with existing similar campaigns, new algorithms are also applied to the measurement data, showing the interest of advanced decentralized message-passing techniques, heterogeneous geometric positioning with hypothesis testing, context-aware localization with e.g., mobility learning or channel-dependent Non Line of Sight (NLoS) mitigation.

A semidefinite programming approach to hybrid localization using RSSI and TOA

In this paper, we exploit the semidefinite programming and the Cramer Rao lower bound techniques to study the hybrid fusion of RSSI and TOA. A semidefinite program is developed to fuse these two location-dependent parameters. The Cramer Rao lower bound expressions are also developed in order to assess theoretical performances of the RSSI and TOA fusion. In order to evaluate this localization scheme, Monte Carlo simulations are carried out in a generic environment using realistic parameters extracted from an ultra wide band measurement campaign. The semidefinite approach is compared to the weighted least-squares, the maximum likelihood, and the Cramer Rao lower bound for the different schemes (i.e. sole RSSI, sole TOA, and hybrid RSSI+TOA). The importance of the fusion of RSSI and TOA is highlighted while assessing the different factors influencing the positioning accuracy.

Theoretical performances assessment of hybrid localization techniques

In this paper, expressions of Cramer Rao lower bounds (CRLB) are derived for non-hybrid and hybrid localization techniques based on RSSI, TOA, and TDOA under Gaussian uncorrelated measurements assumption. Simulations are then carried out using realistic radio parameters values in order to compare different non-hybrid and hybrid localization techniques. CRLBs expressions and simulations show that the number, the nature, the precision of measurements, and the anchors positions are the key parameters that influence positioning accuracy. Hence, the CRLB is proposed as a criterion of localization and a simulation based proof is given.

Localization and Fingerprint of Radio Signals Employing a Multichannel Photonic Analog-to-Digital Converter

The fingerprint and localization of radio signals employing a multichannel photonic analog-to-digital converter (ADC) is proposed, analyzed, and demonstrated in a laboratory experiment. The photonic ADC detects the radio signals with high sensitivity in a large bandwidth without down-conversion stages. This is of special interest when processing emerging low-power wireless standards like ultra-wideband (UWB) radio. The optical processing in the multichannel photonic ADC is tailored for the localization and fingerprint of generic radio transmitters when orthogonal-frequency division multiplexing (OFDM) modulation is employed in the transmission. The photonic ADC includes engineered optical and electrical amplification. The experimental work demonstrates that detection of radio signals with -65 dBm power with signal-to-noise ratio better than 20 dB is feasible, which is in good accordance with the theoretical analysis. The multichannel photonic ADC comprises five optical channels which are precisely time-aligned in optical domain achieving 0.23-m spatial resolution (median) in the localization of radio transmitters. The experimental work also demonstrates that photonic-ADC processing is adequate for OFDM-based UWB radio-signal fingerprint including estimation of the average power, frequency band of operation, and time-frequency hopping pattern if applicable. UWB transmitter localization has been experimentally demonstrated with 0.4-m error.

Localisation of ultra-wide band radio signals by time-multiplexed photonic analog-to-digital processing

A time-multiplexed multichannel photonic-ADC is proposed and demonstrated for the detection and localisation of ultra-wideband radio transmitters. The experimental results indicate successful operation with a mean positioning error of 0.9 m.