This PDF file contains the front matter associated with SPIE Proceedings Volume 8899, including the Title Page, Copyright Information, Table of Contents, and the Conference Committee listing.

Computational ghost imaging is a technique that enables lensless single-pixel detectors to produce images. By illuminating a scene with a series of patterns from a digital light projector (DLP) and measuring the reflected or transmitted intensity, it is possible to retrieve a two-dimensional (2D) image when using a suitable computer algorithm. An important feature of this approach is that although the light travels from the DLP and is measured by the detector, the images produced reveal that the detector behaves like a source of light and the DLP behaves like a camera. By placing multiple single-pixel detectors in different locations it is possible to obtain multiple ghost images with different shading profiles, which together can be used to accurately calculate the three-dimensional (3D) surface geometry through a photometric stereo techniques. In this work we show that using four photodiodes and a 850nm source of illumination, high quality 3D images of a large toy soldier can be retrieved. The use of simplified lensless detectors in 3D imaging allows different detector materials and architectures to be used whose sensitivity may extend beyond the visible spectrum, at wavelengths where existing camera based technology can become expensive or unsuitable.

In this keynote address paper, we present an overview of our previously published work on using compressive sensing in multi-dimensional imaging. We shall examine a variety of multi dimensional imaging approaches and applications, including 3D multi modal imaging integrated with polarimetric and multi spectral imaging, integral imaging and digital holography. This Keynote Address paper is an overview of our previously reported work on 3D imaging with compressive sensing.

InAs/GaSb type-II superlattices (T2SLs) are attractive due to their potentially low dark currents and high responsivity. These low dark currents arise due to reduced Auger recombination caused by the spatial separation between the electrons and holes. Coupling these two aspects together leads to the potential of high operating temperature and high D*. An additional attraction of T2SLs is their wavelength tunability; the wavelength can be tuned between 3 to 12 μm, making them attractive for the militarily important MWIR and long-wave infrared (LWIR) bands. InAs/GaSb T2SLs are traditionally grown upon GaSb substrates due to lattice matching of the type-II material on GaSb. However, GaSb substrates are relatively small and expensive compared with GaAs, leading to increased cost. Additionally, the high absorption coefficient of GaSb requires the substrate to be removed prior to use in FPAs. We present an InAs/GaSb T2SL grown upon a GaAs substrate which operates at room temperature. A room temperature spectral response could be measured for the layer, with responsivity and shot and thermal noise limited specific detectivity (D*) of 0.45 A/W and 8.0x10⁸ cmHz^1/2/W, respectively, at a bias voltage of -0.3 V. This uncooled operation D* is the best to date compared with the literature for a p-i-n or n-i-p MWIR structure grown upon a GaAs substrate.

Previously real-time false-colored multispectral imaging was not available in a true snapshot single compact imager. Recent technology improvements now allow for this technique to be used in practical applications. This paper will cover those advancements as well as a case study for its use in UAV’s where the technology is enabling new remote sensing methodologies.

The objective of this paper is to describe the progress of a project designed to build on recent photonic capabilities in order to develop an ultra-wide band, true Arbitrary Waveform Generator (AWG) capable of providing radar quality signals in the 500MHz to 20GHz spectrum using photonic integration. Within this scope, it is planned to create a single channel radar environment simulator based on a photonic waveform generator, which will demonstrate the dynamic range, stability, and high signal fidelity required to simulate the modern complex radar environment. The paper will present recent measurements of critical parameters that are vital for the practical realisation of this system on a chip.

The latest developments in AlGaInN laser diode technology are reviewed for defence applications such as underwater telecommunications, sensor systems etc. The AlGaInN material system allows for laser diodes to be fabricated over a very wide range of wavelengths from u.v., i.e, 380nm, to the visible, i.e., 530nm, by tuning the indium content of the laser GaInN quantum well. Ridge waveguide laser diode structures are fabricated to achieve single mode operation with optical powers of >100mW in the 400-420nm wavelength range with high reliability. Visible light communications at high frequency (up to 2.5 Gbit/s) using a directly modulated 422nm Gallium-nitride (GaN) blue laser diode is reported. High power operation of AlGaInN laser diodes is also reviewed. We demonstrate the operation of a single chip, high power AlGaInN laser diode ‘mini-array’ consisting of a 3 stripe common p-contact configuration at powers up to 2.5W cw in the 408-412 nm wavelength range. Packaging of nitride laser diodes is substantially different compared to GaAs laser technology and new processes and techniques are required to optimize the optical power from a nitride laser bar. Laser bars of up to 4mm with 16 emitters have shown optical powers up to 4W cw at ~410nm with a common contact configuration. An alternative package configuration for AlGaInN laser arrays allows for each individual laser to be individually addressable allowing complex free-space and/or fibre optic system integration within a very small form-factor.

We review a wide range of absorbers based on patterned resistive sheets for use in mid-wave and long-wave infrared microbolometers. These structures range from wavelength selective dielectric coated Salisbury screens to patterned resistive sheets to stacked multi-spectral devices. For basic three color devices in the LWIR band we have designed and fabricated wavelength selective dielectric coated Salisbury screen (DSS) absorbers suitable for use in microbolometers. In order to produce wavelength selective narrowband absorption, the general design rules for DSS microbolometers show that the thickness of the air gap should be a half wavelength and the optical thickness of the dielectric support layer should be a quarter wavelength. This structure is also air gap tunable; i.e., by varying only air gap thickness, the center wavelength of the absorption curve is shifted. FTIR microscope measurements have been made on a number of the different devices demonstrating three color capability in the LWIR while maintain very high efficiency absorption. We have also shown that the use of a patterned resistive sheet consisting of a properly sized array of cross-shaped holes acts as a polarization independent frequency-selective absorber allowing a three-color system spanning the 7-14 micron band. For realistic metal layers the skin effect produces complex surface impedance that can be quite large in the LWIR band. We have shown that metal layers of thickness between one and three skin depths can act as the absorber layer, and have shown that thick metal layers can still produce excellent absorption in the LWIR. Holes in the dielectric support layer also reduce the thermal mass in the system without compromising spectral selectivity. Broadband designs using rectangular holes that produce substantially reduced thermal mass (over 50%) while maintaining efficient spectral absorption have also been found. Finally, we have considered multispectral stacked structures, including Jaumann absorbers and stacked dipole/slot patterned resistive sheets. These structures promise either two band (MWIR/LWIR) or two to three color LWIR in a multi-layer stacked pixel.

This paper reports the achievements so far attained in the development of high-performance CMOS SPADs for single-photon sensitive 2D imagers, based on photon-counting and 3D ranging cameras. The latters are based on both the direct in-pixel measurement of the Time-of-Flight of each photon bouncing bounce from the scene back to the camera and the “indirect” phase-resolved method to count reflected photons in well-defined time slots, synchronous to the active illumination of the scene. MiSPiA SPADs are the new state-of-the-art among SPADs in CMOS technologies.

Optical communications in the blue-green waveband offers the attractive prospect of secure, high-bandwidth communications for a variety of civil and military applications. Recent developments in blue-green lasers, spectral filters, and detection technologies are reviewed and their potential for compact, affordable systems is discussed.

Al_0.52In_0.48P is the largest bandgap material in III-V non-nitride semiconductors that is lattice matched to a readily available substrate (GaAs). Having a bandgap narrower than that of GaN enables it to detect wavelengths around 480 nm. Such wavelengths have the best transmittance underwater and may be used as a carrier in underwater communication systems. We present an Al_0.52In_0.48P homo-junction Separate-Absorption-Multiplication-Avalanche-Photodiode (SAMAPD) as a high sensitivity detector for such an application. By increasing the neutral and space-charge region thicknesses, the peak response wavelength can be tuned to longer wavelengths with a narrower full-width-half-maximum (FWHM). The quantum efficiency of the detector reduces with FWHM and this is compensated by having an avalanche gain. At room temperature, the SAM-APD has a dark current of <20 pA for a 210 μm radius device up to 99.9% of breakdown voltage. The structure gives a narrow spectral FWHM of 22 nm with centre wavelength of 482 nm. An external quantum efficiency of 33% and 6410% at 482 nm is obtained at bias voltage of -19 V and -92.6 V respectively.

All the currently available unconditional security proofs on quantum key distribution, in particular for the BB84 protocol and its variants including continuous-variable ones, are invalid or incomplete at many points. In this paper we discuss some of the main known problems, particularly those on operational security guarantee and error correction. Most basic are the points that there is no security parameter in such protocols and it is not the case the generated key is perfect with probability ≥ 1 - ϵ under the trace distance criterion d ≤ ϵ, which is widely claimed in the technical and popular literature. The many serious security consequences of this error about the QKD generated key would be explained, including practical ramification on achievable security levels. It will be shown how the error correction problem alone may already defy rigorous quantitative analysis. Various other problems would be touched upon. It is pointed out that rigorous security guarantee of much more efficient quantum cryptosystems may be obtained by abandoning the disturbance-information tradeoff principle and utilizing instead the known KCQ (keyed communication in quantum noise) principle in conjunction with a new DBM (decoy bits method) principle that will be detailed elsewhere.

We report several vulnerabilities found in Clavis2, the flagship quantum key distribution (QKD) system from ID Quantique. We show the hacking of a calibration sequence run by Clavis2 to synchronize the Alice and Bob devices before performing the secret key exchange. This hack induces a temporal detection efficiency mismatch in Bob that can allow Eve to break the security of the cryptosystem using faked states. We also experimentally investigate the superlinear behaviour in the single-photon detectors (SPDs) used by Bob. Due to this superlinearity, the SPDs feature an actual multi-photon detection probability which is generally higher than the theoretically-modelled value. We show how this increases the risk of detector control attacks on QKD systems (including Clavis2) employing such SPDs. Finally, we review the experimental feasibility of Trojan-horse attacks. In the case of Clavis2, the objective is to read Bob's phase modulator to acquire knowledge of his basis choice as this information suffices for constructing the raw key in the Scarani-Acin-Ribordy-Gisin 2004 (SARG04) protocol. We work in close collaboration with ID Quantique and for all these loopholes, we notified them in advance. Wherever possible, we or ID Quantique proposed countermeasures and they implemented suitable patches and upgrade their systems.

This paper presents the physical prevention probability of an Intensity-Shift-Keying (ISK) Y00 quantum stream cipher against a polarity inversion attack, where the attacker in the middle of the communication line intercepts legitimate sender’s messages and resends false messages to the legitimate receiver by inverting some of signal polarities. Message falsification is recognized as a major issue in the field of mathematical encryption. Therefore, the attack should also be studied in the field of physical encryption. Y00 protocol was proposed by H. P. Yuen in 2000 to hide even ciphertexts from eavesdroppers under quantum noise of coherent light. Theoretical and experimental analyses of encryption strength of Y00 systems have also been studied against eavesdropping. However, there were not many studies about active attacks like message falsifications. Recent studies showed that the present ISK Y00 systems, whose communication bases are paired signals, may prevent the attack under the Known-Plaintext-Attack. To enhance the probability against the attack, a quadruple-signal-based ISK Y00 system was proposed, whose signal bases are sets of 4 signals. This study shows the proposed system has a prevention probability of 0.66665 per signal, while One-Time Pad used in BB84 cannot prevent message falsification under Known-Plaintext-Attack since polarity inversion directly falsifies the message.

Continuous-variable quantum key distribution is proven in theory secure against general attacks, but side channel is still a crucial problem for practical setup, since security proofs do not take into account all possible experimental imperfections. In this paper, we consider a loophole that links to electronics limitation of homodyne detection. By using this loophole, we propose a saturation attack combined with intercept-resend attack on the practical continuous-variable quantum key distribution using Gaussian-modulated coherent state protocol. Under this attack, Eve can launch a full intercept-resend attack and further influence the excess noise estimated by Alice and Bob. We analyse this saturation attack with operating protocol and show that our attack could render secret key without being discovered. We also propose a countermeasure against such saturation attack.

We present the results of a Swiss project dedicated to the development of high speed quantum key distribution and data encryption. The QKD engine features fully automated key exchange, hardware key distillation based on finite key security analysis, efficient authentication and wavelength division multiplexing of the quantum and the classical channel and one-time pas encryption. The encryption device allows authenticated symmetric key encryption (e.g AES) at rates of up to 100 Gb/s. A new quantum key can uploaded up to 1000 times second from the QKD engine.

We investigate an extended version of the Bragg reflection waveguide (BRW) with air gaps as one of the layers. This design has the potential of drastically simplifying the epitaxial structure for integrated nonlinear optical elements at the expense of more complicated structuring. This approach would afford much more flexibility for designing and varying BRW structures. Here, we discuss an extension of the established theory for BRW slabs and report our results of applying Marcatili's method for rectangular waveguides to the BRW case. With this analytic approach we can estimate the effective index of the modes orders of magnitudes faster than with full numerical techniques, such as finite-difference time-domain (FDTD) or finite elements. Initial results are mixed; while phase-matched designs have been found, they currently have no significant advantage over other schemes.

Randomness is of fundamental importance in various fields, such as cryptography, numerical simulations, or the gaming industry. Quantum physics, which is fundamentally probabilistic, is the best option for a physical random number generator. In this article, we will present the work carried out in various projects in the context of the development of a commercial and certified high speed random number generator.

Randomness is crucial for a variety of applications, ranging from gambling to computer simulations, and from cryptography to statistics. However, many of the currently used methods for generating randomness do not meet the criteria that are necessary for these applications to work properly and safely. A common problem is that a sequence of numbers may look random but nevertheless not be truly random. In fact, the sequence may pass all standard statistical tests and yet be perfectly predictable. This renders it useless for many applications. For example, in cryptography, the predictability of a randomly" chosen password is obviously undesirable. Here, we review a recently developed approach to generating true | and hence unpredictable | randomness.

An overview will be given of various approaches to implementing a quantum repeater for quantum communication over large distances. This includes a discussion of systems and protocols that are experimentally feasible and thus realizable in the midterm in order to go beyond the current limit of a few hundred km given by direct quantum-state transmissions. At the same time, these schemes should be, in principle, scalable to arbitrary distances. In this context, the influence of various elements and strategies in a quantum repeater protocol on the final fidelities and rates shall be addressed: initial entanglement distribution, Bell measurements, multiplexing, postselection, quantum memories, and quantum error detection/correction. Solely on the hardware side, the differences in using just single quanta or instead employing many quanta for the flying (photons) and the stationary (atoms) qubits will be pointed out.

As society becomes more reliant on electronic communication and transactions, ensuring the security of these interactions becomes more important. Digital signatures are a widely used form of cryptography which allows parties to certify the origins of their communications, meaning that one party, a sender, can send information to other parties in such a way that messages cannot be forged. In addition, messages are transferrable, meaning that a recipient who accepts a message as genuine can be sure that if it is forwarded to another recipient, it will again be accepted as genuine. The classical digital signature schemes currently employed typically rely on computational complexity for security. Quantum digital signatures offer the potential for increased security. In our system, quantum signature states are passed through a network of polarization maintaining fiber interferometers (a multiport) to ensure that recipients will not disagree on the validity of a message. These signatures are encoded in the phase of photonic coherent states and the choice of photon number, signature length and number of possible phase states affects the level of security possible by this approach. We will give a brief introduction into quantum digital signatures and present results from our experimental demonstration system.

The modelling of the Automatic Target Detection, Recognition and Identification performance in systems of multiple sensors and/or platforms is important in many respects. For example, in the selection of sensors or sensor combinations of sufficient effectiveness to achieve operational requirements, or for understanding how the system might be best exploited. It is possible that a sensing system may be comprised of sensors of several different types, including active and passive approaches in the radio frequency and optical portions of the spectrum. Some may have well-understood performance, whereas others may be only poorly characterised. A simulation framework has been developed examining sensor options across different sensor types, parameterisations, search strategies, and applications. The framework is based around Bayesian Decision Theoretic principles along with simple sensor models and search environment. It uses Monte-Carlo simulation to derive statistical measures of performance for systems. The framework has been designed to encompass detection, recognition and identification problems and also to treat sensor characterisation. The modelling framework has been applied to a number of illustrative problems. These range from simple target detection scenarios using sensors of differing performance or of different regional search schemes, through to examinations of: the number of measurements required to reach threshold performance; the effects of sensor measurement cost; issues relating to the poor characterisation of sensors within the system, and; the performance of combined detection and recognition sensor systems. Results are presented illustrating these effects. These generally show that the method is able to quantify qualitative expectations of performance, and is sufficiently powerful to highlight some unexpected aspects of operation.

Generalised wide are search and surveillance is a common-place tasking for multi-sensory equipped autonomous systems. Here we present on a key supporting topic to this task - the automatic interpretation, fusion and detected target reporting from multi-modal sensor information received from multiple autonomous platforms deployed for wide-area environment search. We detail the realization of a real-time methodology for the automated detection of people and vehicles using combined visible-band (EO), thermal-band (IR) and radar sensing from a deployed network of multiple autonomous platforms (ground and aerial). This facilities real-time target detection, reported with varying levels of confidence, using information from both multiple sensors and multiple sensor platforms to provide environment-wide situational awareness. A range of automatic classification approaches are proposed, driven by underlying machine learning techniques, that facilitate the automatic detection of either target type with cross-modal target confirmation. Extended results are presented that show both the detection of people and vehicles under varying conditions in both isolated rural and cluttered urban environments with minimal false positive detection. Performance evaluation is presented at an episodic level with individual classifiers optimized for maximal each object of interest (vehicle/person) detection over a given search path/pattern of the environment, across all sensors and modalities, rather than on a per sensor sample basis. Episodic target detection, evaluated over a number of wide-area environment search and reporting tasks, generally exceeds 90%+ for the targets considered here.

Rockwell Collins France (RCF) radar department is currently developing, in close collaboration with TNO in The Hague, The Netherlands, a Frequency Modulated Continuous Wave (FMCW) radar sensor dedicated to Obstacle Warning function and potentially to air traffic detection. The sensor combines flood light illumination and digital beam forming to accommodate demanding detection and coverage requirements. Performances have been evaluated in flight tests and results prove that such a radar sensor is a good candidate for the Sense Function of Sense and Avoid Systems onboard UAV.

Defence Research and Development Canada – Atlantic (DRDC Atlantic) is currently involved in research on the topic of northern Maritime Domain Awareness (MDA). One project, entitled Situational Information for Enabling Development of Northern Awareness (SEDNA), includes research on the exploitation of MDA data in northern areas. One aspect of this research is to utilize wide area MDA data to provide awareness to an unattended, land-based system. Wide area MDA is attained through the use of space-based AIS (SAIS) data, a data feed used by the Canadian Department of National Defence and supplied by the commercial provider exactEarth Ltd. The land-based surveillance system used is the remote northern system constructed within the DRDC Northern Watch Technology Demonstration Project. Northern Watch is a multi-year project intended to show state-of-the-art, unattended, surveillance capabilities in the Canadian north. The link between the SAIS and Northern Watch is provided by a research infrastructure that consists of an assembly of data sources, users, applications, and product management techniques that collectively support research in areas such as information management and MDA data exploitation. High-level descriptions of the systems are provided along with elaboration on the alerting algorithm, the notifications that would be sent to the Northern Watch southern command site, and the resulting actions that could be taken by the Northern Watch surveillance system.

Methods for automated person detection and person tracking are essential core components in modern security and surveillance systems. Most state-of-the-art person detectors follow a statistical approach, where prototypical appearances of persons are learned from training samples with known class labels. Selecting appropriate learning samples has a significant impact on the quality of the generated person detectors. For example, training a classifier on a rigid body model using training samples with strong pose variations is in general not effective, irrespective of the classifiers capabilities. Generation of high-quality training data is, apart from performance issues, a very time consuming process, comprising a significant amount of manual work. Furthermore, due to inevitable limitations of freely available training data, corresponding classifiers are not always transferable to a given sensor and are only applicable in a well-defined narrow variety of scenes and camera setups. Semi-supervised learning methods are a commonly used alternative to supervised training, in general requiring only few labeled samples. However, as a drawback semi-supervised methods always include a generative component, which is known to be difficult to learn. Therefore, automated processes for generating training data sets for supervised methods are needed. Such approaches could either help to better adjust classifiers to respective hardware, or serve as a complement to existing data sets. Towards this end, this paper provides some insights into the quality requirements of automatically generated training data for supervised learning methods. Assuming a static camera, labels are generated based on motion detection by background subtraction with respect to weak constraints on the enclosing bounding box of the motion blobs. Since this labeling method consists of standard components, we illustrate the effectiveness by adapting a person detector to cameras of a sensor network. While varying the training data and keeping the detection framework identical, we derive statements about the sample quality.

In modern tracking systems the ability to obtain high quality, high resolution appearance of the tracked target is often highly desirable. However, the reality of operational deployment often means that imaging systems deployed for this task suffer from limitations reducing effective image quality. These limitations can be attributed to a range of causes such as low quality video sensors, system noise, high target dynamics and other environmental noise factors. Despite the advantages of the super-resolution techniques the problem of handling complex motion still remains a challenging task for the effective super-resolution implementation. The computational complexity and large memory requirements required for the implementation of super-resolution imaging largely restrict the usage of these techniques in real-time hardware implementations. In order to improve visual quality of the tracked target and overcome these limitations, we propose a simple yet effective solution that integrates a super-resolution imaging approach based on combination of the Sum of the Absolut Differences (SAD) and gradient-descent motion estimation techniques into a novel tracking approach. In addition, the proposed approach demonstrates robustness in improved target appearance modeling that assists the overall tracking system. The presented results demonstrate this significant improvement in visual target representation whilst tracking over high dynamic scenes. The implementation simplicity of the proposed approach makes it an attractive solution for realization on low power hardware. Such a system can be deployed on small unmanned aerial vehicles (UAV) or other hardware where size, weight and power (SWaP) is of a particular concern.

The wide availability of previously acquired, geo-referenced imagery enables automatic video based solutions for high precision object geo-localization and cooperative visualization. We present a system which geo-references objects seen in UAV video streams, distributes this information in a sensor network and visualizes them on modern smartphones using augmented reality techniques. The feasibility of the approach was experimentally validated using Mini-UAV ("MD-400") and high altitude UAV video footage in combination with modern off-the-shelve smartphones. Applications are widespread and include for instance crisis and disaster management or military applications.

Interferometric fiber optic gyroscopes belong to the class of inertial sensors. Due to their high accuracy they are used for absolute position and rotation measurement in manned/unmanned vehicles, e.g. submarines, ground vehicles, aircraft or satellites. The important system components are the light source, the electro optical phase modulator, the optical fiber coil and the photodetector. This paper is focused on approaches to realize a stable light source and fiber coil. Superluminescent diode and erbium doped fiber laser were studied to realize an accurate and stable light source. Therefor the influence of the polarization grade of the source and the effects due to back reflections to the source were studied. During operation thermal working conditions severely affect accuracy and stability of the optical fiber coil, which is the sensor element. Thermal gradients that are applied to the fiber coil have large negative effects on the achievable system accuracy of the optic gyroscope. Therefore a way of calculating and compensating the rotation rate error of a fiber coil due to thermal change is introduced. A simplified 3 dimensional FEM of a quadrupole wound fiber coil is used to determine the build-up of thermal fields in the polarization maintaining fiber due to outside heating sources. The rotation rate error due to these sources is then calculated and compared to measurement data. A simple regression model is used to compensate the rotation rate error with temperature measurement at the outside of the fiber coil. To realize a compact and robust optical package for some of the relevant optical system components an approach based on ion exchanged waveguides in thin glass was developed. This waveguides are used to realize 1x2 and 1x4 splitter with fiber coupling interface or direct photodiode coupling.

Tracking ground targets using low cost ground-based sensors is a challenging field because of the limited capabilities of such sensors. Among the several candidates, including seismic and magnetic sensors, the acoustic sensors based on microphone arrays have a potential of being useful: They can provide a direction to the sound source, they can have a relatively better range, and the sound characteristics can provide a basis for target classification. However, there are still many problems. One of them is the difficulty to resolve multiple sound sources, another is that they do not provide distance, a third is the presence of background noise from wind, sea, rain, distant air and land traffic, people, etc., and a fourth is that the same target can sound very differently depending on factors like terrain type, topography, speed, gear, distance, etc. Use of sophisticated signal processing and data fusion algorithms is the key for compensating (to an extend) the limited capabilities and mentioned problems of these sensors. It is hard, if not impossible, to evaluate the performance of such complex algorithms analytically. For an effective evaluation, before performing expensive field trials, well-designed laboratory experiments and computer simulations are necessary. Along this line, in this paper, we present an object-oriented modeling and simulation framework which can be used to generate simulated data for the data fusion algorithms for tracking multiple on-road targets in an unattended acoustic sensor network. Each sensor node in the network is a circular microphone array which produces the direction of arrival (DOA) (or bearing) measurements of the targets and sends this information to a fusion center. We present the models for road networks, targets (motion and acoustic power) and acoustic sensors in an object-oriented fashion where different and possibly time-varying sampling periods for each sensor node is possible. Moreover, the sensor’s signal processing and detection blocks are modeled using a parametric approach by associating a receiver operating characteristics (ROC) curve to the whole process, which results in false alarms as well as missed detections. The proposed simulation environment can be used for ground-truth and synthetic data generation for road-constraint multiple target tracking in an unattended acoustic sensor network.

While there is a variety of approaches and algorithms for optimizing the mission of an unmanned moving sensor, there are much less works which deal with the implementation of several sensors within a human organization. In this case, the management of the sensors is done through at least one human decision layer, and the sensors management as a whole arises as a bi-level optimization process. In this work, the following hypotheses are considered as realistic: Sensor handlers of first level plans their sensors by means of elaborated algorithmic tools based on accurate modelling of the environment; Higher level plans the handled sensors according to a global observation mission and on the basis of an approximated model of the environment and of the first level sub-processes. This problem is formalized very generally as the maximization of an unknown function, defined a priori by sampling a known random function (law of model error). In such case, each actual evaluation of the function increases the knowledge about the function, and subsequently the efficiency of the maximization. The issue is to optimize the sequence of value to be evaluated, in regards to the evaluation costs. There is here a fundamental link with the domain of experiment design. Jones, Schonlau and Welch proposed a general method, the Efficient Global Optimization (EGO), for solving this problem in the case of additive functional Gaussian law. In our work, a generalization of the EGO is proposed, based on a rare event simulation approach. It is applied to the aforementioned bi-level sensor planning.