Multi-frequency multi-mode Terahertz screening for border checks

Final Report Summary - TERASCREEN (Multi-frequency multi-mode Terahertz screening for border checks)

Executive Summary:
The objective of TeraSCREEN was to develop and demonstrate in a live border control environment a safe, privacy respecting, high throughput security screening system which automatically detects and classifies potential threat objects concealed on a person. This would significantly improve both efficiency and security at border checks.
The project was designed to meet very demanding call requirements, with the consortium needing to undertake a series of developments, in many cases developing components from zero due to their unavailability or prohibitive cost. This led to a very ambitious project, both technically and considering the schedule and budget.
Unfortunately the main project objectives have not been achieved, but this is mainly due to lack of time and funding, as the intermediate results that have been achieved are very promising. All developed components have met the specifications set out at the beginning of the project, in many cases bringing the technology level with, or surpassing, the state-of-the-art. The components and processes developed can be exploited in their own right, also in multiple areas outside security.
Development issues with the components greatly reduced the time available for the rest of the project, leading to a project extension and a change in scope being required. Further issues at the integration stage, and with one of the main project partners entering administration, meant more changes in scope, with the end result being testing at subsystem level. Conclusions from this testing have defined clear paths of improvement for both subsystems, which will be done by the corresponding partners as and when funding becomes available. Given the level of completion of the project, it was not possible to come to a conclusion on the detection and classification potential of the integrated system.
Project Context and Objectives:
The challenge for any security screening system is to identify potentially harmful objects such as weapons and explosives concealed under clothing. Classical border and security checkpoints are no longer capable of fulfilling the demands of today’s ever growing security requirements, especially with respect to the high throughput generally required that entails a high detection rate of threat material and a low false alarm rate. Therefore, TeraSCREEN proposes to develop an innovative concept of multi-frequency multi-mode Terahertz (THz) and millimetre-wave (mm-wave) detection with new automatic detection and classification functionalities, which increases throughput while respecting privacy.
The overall objective of the project is to develop and demonstrate a safe, privacy respecting, high throughput security screening system which automatically detects and classifies potential threat objects concealed on a person at a live control point at Bristol International Airport. This will advance the state of the art, providing an innovative security screening solution for border and security checks that is effective both at the security level and at the concept of operation level.
The more specific scientific and technological objectives of the TeraSCREEN project are:
i) To identify the system-level architecture and requirements including the test and evaluation procedures;
ii) To identify the passive and active subsystem-level requirements based on the system-level ones, including the test and evaluation procedures;
iii) To develop innovative (advancing state of the art) components necessary for the subsystems, although using commercial-off-the-shelf (COTS) components when available and sharing the components that are common to both subsystems. Includes tests and characterisation;
iv) To integrate the components into modules for the subsystems. Includes tests and characterisation;
v) To integrate the modules into the passive and active subsystems. Includes tests and characterisation
- Passive subsystem: 360GHz and 94GHz real-time stand-off imaging;
- Active subsystem: 180GHz real-time stand-off imaging.
vi) To integrate the passive and active subsystems into the TeraSCREEN Prototype System
- The TeraSCREEN multi-frequency multi-mode Prototype System will include: 360GHz and 94GHz passive and 180GHz active real-time stand-off imaging;
- The different passive frequencies provide different information for different materials in the scene;
- The active feature provides high resolution images of the concealed objects and depth/thickness information.
vii) To develop the sensor data fusion and image processing, including the Automatic Object Detection and Classification, and the Privacy Enhancing algorithms complying with EC Regulation No 1147/2011, respecting fundamental rights and observing the principles recognised by the Charter of Fundamental Rights of the European Union;
viii) To develop the TeraSCREEN Prototype System user interface;
ix) To test and characterise the integrated TeraSCREEN Prototype System
- The system tests are to be specified in i) and conducted by the Industrial End-User’s VeriSys service.
x) To demonstrate the TeraSCREEN Prototype System in a Pilot Test at a live control point in Bristol International Airport. Includes system performance evaluation.
Project Results:
The main S&T results/foregrounds are:
• Development of the 40 nm GaAs mHEMT process
• Development of the InP DBHT process
• 140 GHz LNA including calibration and test circuits
• 360 GHz LNA including calibration and test circuits
• 360 GHz passive subsystem antenna array
• 360 GHz manifold and polarisation twist
• 360 GHz MIMO antenna for the active subsystem
• 180 GHz MIMO antenna for the active subsystem
• Frequency multiplier diodes
• 360GHz Schottky diode subharmonic mixer
• 90/180 GHz frequency doubler
• 30/90 GHz frequency tripler
• Integrated 10 GHz DDS for signal generation
• W-Band High Power Amplifier Development
• W-Band Medium Power Amplifier Development
• High Power W-band Frequency Multiplier
• Low-cost back end electronics for passive subsystem
• 360 GHz passive subsystem
• Optical modelling and redesign of existing passive imager
• 180 GHz active subsystem
• Combination of Mechanical Movement and MIMO Imaging
• Broadband Imaging at Millimetre Wave Frequencies
• Millimetre and Submillimetre Wave Component Testbeds
• Sensor data fusion and image processing
• Ethical compliance

1.3.1. 140 GHz LNA
Development and consolidation of the new 40 nm GaAs mHEMT commercial technology for low noise submillimetre applications. In order to accomplish the active device characterization a TRL calibration kit was designed (frequency band DC÷110 GHz) and a small-signal noise model was extracted up to 110 GHz and extrapolated up to 200 GHz.
A 140GHz LNA test vehicle (central frequency of 140 GHz, 30 GHz bandwidth, a small signal gain of 20 dB together with a noise figure of 4.5 dB) was successfully designed and measured. This result is extremely important since it shows that the effort of the OMMIC foundry for the technology development and the UROM MMIC designers oriented to the mHEMT commercial process moved in the right direction.
The 140 GHz LNA was characterised, with state of the art noise performance of NF = 4 dB with an associated gain of 20 dB at 140 GHz using on-wafer measurements. A micro-photograph of the 140 GHz LNA MMIC, developed in the TeraSCREEN project, can be seen in Figure 1.

1.3.2. 360 GHz LNA
Development and consolidation of the new 40 nm GaAs mHEMT commercial technology for low noise submillimetre applications. In order to accomplish the active device characterization a TRL calibration kit was designed (frequency band DC÷400 GHz) and a scalable small-signal noise model was extracted up to 110 GHz and extrapolated up to 400 GHz.
A 360GHz LNA test vehicle (central frequency of 360 GHz, 30 GHz bandwidth, with predicted small signal gain of 15 dB together with a noise figure of 7.5 dB) was successfully designed and developed. A micro-photograph of the 360 GHz LNA MMIC, developed in the TeraSCREEN project, can be seen in Figure 2.

1.3.3. 360 GHz passive subsystem antenna array
The TeraSCREEN passive antenna system was designed in order to have minimum dimension, and in this way obtained a better resolution thanks to the reduced pixel distance. This was achieved by means of a manifold section that assembled the mixer blocks and the rectangular antenna array. By using this section, the separation between the antennas was not dependent on the mixer block design, which benefited the achievable distance resolution. Moreover, another fact that reduced the dimension of the antenna system, and therefore mitigated the problems related to blockage inside the ALFA3 prototype imager, was the use of extra compact twists. These twists allowed the assembly between the mixer blocks and the manifold with almost no influence on the receiver footprint. The frontend antenna has been manufactured using two different techniques (electroerode erosion and Metal Additive Layer Manufacturing). This has given us the chance to compare them. Both techniques gave similar results, confirming that 3D manufacture is a promising technique not only for mechanical structures but also for High Frequency applications. A photograph of two 8-channel 360 GHz antenna array blocks, developed in the TeraSCREEN project, can be seen in Figure 3.

1.3.4. Frequency multiplier diodes
TeraSCREEN has supported the development of high power frequency multipliers required for LO sources and as sources in FMCW radars at high frequencies. New diode technology developed at TCL has now demonstrated a 180 GHz frequency doubler operating at 35% efficiency. These devices have also provided 83 mW of output power at 180 GHz. The second stage multipliers have demonstrated frequency doublers at 360 GHz with an efficiency of 22% and an output power of 12 mW. A photograph of eight 90/180 GHz frequency multipliers, containing the frequency multiplier diodes, developed in the TeraSCREEN project, (connected to eight 360 GHz subharmonic mixers), can be seen in Figure 4.

1.3.5. 360 GHz Schottky diode subharmonic mixers
TeraSCREEN has supported the development of a high sensitivity and broadband mixer required as the first stage component in the front-end receiver at 360 GHz. A bespoke mixer design developed at STFC has now demonstrated a 360 GHz subharmonic mixer with outstanding radiometer sensitivity (NETD) below 0.5K at the required bandwidth, which is beyond the system requirement, and is also superior to European LNA performance at the same frequency range up to date. Another achievement through TeraSCREEN is the reliability and repeatability of these devices, in total 32 components have been made and tested with highly repeatable performance. There are number of factors that define the actual mixer performance at high frequencies, from diode quality to micro assembly and high precision machining etc. We have demonstrated this through the whole process. A photograph of eight 360 GHz subharmonic mixers, containing the Schottky diodes, developed in the TeraSCREEN project (connected to eight 90/180 GHz frequency multipliers) can be seen in Figure 4.

1.3.6. Integrated DDS for signal generation
Development of a DDS with 20 GHz clock frequency and 12 bit input digital frequency word as well as 6 bit digital phase word with < 3W dissipated power and a 6 bit DAC. The design has a transistor count of 4500. With the parameters indicated above one can achieve a frequency resolution of 5 MHz and a SFDR > 30 dBc for most of the frequencies. The settling time is < 300 ns. The DDS is useful for the active subsystem, because it generates an ideal linear sweep in frequency with high speed. The expected phase-noise is expected to be comparable to SoA synthesizers. A micro-photo of the DDS design, developed in the TeraSCREEN project, can be seen in Figure 5.

1.3.7. W-Band Power Amplifier
Development of a power amplifier (PA) operating at 85 GHz – 95 GHz with output power levels of Pout > 130 mW (> 21 dBm) on the chip and an associated gain of 5 dB and more than 15 % power-added efficiency. Although the bandwidth of the PA is less than written in the DoW the output power is close to double that written in the DoW. The targeted efficiency exceeds the planned 15% over a bandwidth of several GHz. Also the gain specification is fulfilled across the desired frequency range. A micro-photo of the W-band power amplifier design, developed in the TeraSCREEN project, can be seen in Figure 6.

1.3.8. High Power W-Band Frequency Multiplier
Development of a InP DHBT frequency tripler for the W-Band. The output power is Pout > 10 dBm which is below the required 40 mW. Therefore, a 2nd MMIC has been developed to boost the power up to 16 dBm, which fulfils the DoW specifications. The tripler exhibits a conversion gain of 0 dB with 12 GHz bandwidth, which is narrower as compared to the DoW. The medium power amplifier developed within the project provides 16 dBm output power with a bandwidth of at least 10 GHz. A micro-photo of the high-power W-band frequency multiplier design, developed in the TeraSCREEN project, can be seen in Figure 7.

1.3.9. Passive subsystem 360 GHz module integration
The final run of the passive subsystem Rx module packaging and assembly was successfully performed using the final components developed in WP2: the 360 GHz passive antenna array; the 360 GHz manifold and polarisation twist; the 360 GHz subharmonic mixer; the 90/180 GHz frequency doubler; and the back-end electronics. These were integrated together with an OTS Gunn diode oscillator.
A measured NETD of 375 mK was obtained in a 400 Hz video bandwidth when measured without the manifold and polarisation twist, and a NETD of 773 mK was measured when the manifold and polarisation twist was added. The specification for the single receiver channel performance is a NETD < 1 K and the target was < 0.5 K. Therefore, the measured results demonstrate that the passive subsystem architecture meets the required performance level. Critically, the results demonstrate that the 360 GHz subharmonic mixer performance is excellent within the complete subsystem and confirms that the packaged and assembled mixer performance meets the system requirements. In future, a higher level of integration would eliminate the need for the manifold and polarisation twist, thus confirming that the target specification can be met.
A photo of the passive subsystem 360 GHz module integration, developed in the TeraSCREEN project, can be seen in Figure 8.

1.3.10. Back-end electronics design for the 360 GHz passive subsystem
The back-end electronics design for the 360 GHz passive subsystem was made based on the simulations made in ADS. A PCB was manufactured and tested together with a mechanical housing to shield the electronics from emi and for good thermal design. The application of Eccosorb onto the mechanical housing walls and lid eliminated any problems with cavity resonances and oscillations.
The test results of the back-end electronics design were in accordance with the design specifications, a reasonably flat IF gain was achieved as well as reasonable 1/f noise. The IF amplification was approximately 58 dB; the bandwidth was 0.1 MHz to 15 GHz; the input return loss < -13 dB; and detector diode matching < -8 dB. A Y factor test showed an effective NETD contribution < 50 mK. A photo of the passive subsystem’s back-end electronics, developed in the TeraSCREEN project, can be seen in Figure 9.

1.3.11. 360 GHz passive subsystem integration
Many problems were experienced during the 360 GHz passive subsystem integration and a lot of delays were incurred. However, good progress started to be made towards the end of the project as the problem issues were addressed by the project. In the end, not all 16 channels were operational. However, 11 channels were giving reasonable results at the module level after the passive subsystem integration was performed.
Passive subsystem 360 GHz channels could achieve a NETD < 1K with the manifold and polarisation twist and an NETD < 0.5K without the manifold and polarization twist. These results confirmed that the components and electronics developed and integrated into the passive subsystem meet the required performance. As mentioned previously, in future, by having a higher level of integration to eliminate the need for the manifold and polarisation twist, to comfortably meet the required target specification for good passive imaging.
It is worth noting that the components and electronics required for the passive subsystem integration at 360 GHz were delivered in larger prototype series of 16 and 32 off. This in itself is an achievement for a research project. Admittedly, there were problems with a spread of performance, with channel performance ranging from good to poor, and a channel yield that was only 11 out of 16 (68.75%). Nevertheless, the potential for a larger passive subsystem integration was demonstrated.
Other achievements in the passive subsystem integration were the development of a data acquisition board, a very low-noise power supply board and a National Instruments DAQ and Labview environment to acquire, process and display the pixel data and images obtained from the array of channels.
A photo of the passive subsystem integration front-end, developed in the TeraSCREEN project, can be seen in Figure 10.

1.3.12. Optical modelling and redesign of existing passive imager
The passive subsystem designed and realized in TS combines an existing 94 GHz imager with a 360 GHz receiver. Even an additional 220 GHz receiver has been considered. Their position in the overall system had to be optimized to check the imager footprint over all scan positions, and is a key point in the design. Therefore, the footprint expected for the receivers has been obtained based on the optimization of the 360 GHz array placement. It has been taken into account that the 94 GHz receiver is at the system focus while the other receivers are out of focus. Therefore, the system based on the three receiver frequencies has been optimized, obtaining the simulated footprint and frequency response available over the entire image plane, confirming the best array-optical system integration. The simulation model and simulation results for the 16 channel 360 GHz receiver are shown in Figure 11.

1.3.13. Passive subsystem
The passive subsystem imager developed in the project was tested and characterised after the passive subsystem integration. A Labview environment was used to monitor and view the images obtained from the 360 GHz receiver array, and custom software, with image pre-processing, to monitor and view the images obtained from the 94 GHz receiver array. Images of reasonable quality are obtained at 94 GHz whilst images at 360 GHz are currently poor.
Test plans were followed, and a series of tests were performed by ICTS at the ACR lab facility in Norrköping, Sweden. Some problems were identified for the 360 GHz passive imager pertaining to the antenna configuration, and the material properties of some of the components inside the equipment. These problems, now identified, will be addressed and it is hoped that good images at 360 GHz can be achieved in future.
Interesting results were obtained for the 94 GHz passive subsystem, for example:
• The best imaging position was the chest and the worst was the extremities
- as expected in a passive imager
• Metal, Plasticine and Playdoh were all detected
• Beeswax, a common simulant for plastic explosive at 94 GHz, was not detected
• A bag of water was easily detected when cold but not at all when warm
- illustrating a common problem with passive imagers
• Testpieces were not detected on the extremities
• The limit of detection was metal 10 cm x 5 cm on the chest
• Generally, the shape of the testpiece was not transmitted, but the shape of a long thin bread knife was accurately imaged
• Penetration was excellent with a metal testpiece being successfully imaged through 1 layer of leather + 2 layers of fleece
Figure 12 shows some image results from the 94 GHz passive subsystem. From left to right the images show:
1. Raw data image with no object
2. Raw data image with a metal plate placed on the chest
3. Pre-processed image with a simulant placed on the stomach at 3m range
4. Pre-processed image with steel wool placed on the chest at 4m range
- this is hidden under one leather and two fleece layers
5. Pre-processed image with steel wool reduced in size to the limit of detection
6. Pre-processed image with a bread knife

1.3.14. Millimetre and submillimetre wave component testbed (passive)
The passive imaging subsystem integration platform was based on the ALFA3 Prototype. The research performed in WP4 established that this equipment was not particularly suitable for the test of components and modules at frequencies above perhaps 150 GHz. Furthermore, the platform was not suitable for multi-frequency operation. These aspects were addressed so that the equipment has become more flexible and will in time be suitable for testing components and modules up to much higher frequencies, e.g. 360 GHz and higher.
After the completion of the TS project, the ALFA3 Prototype itself can be further exploited as an integration platform test bed for testing mm-wave components, radiometers, etc., in a passive imaging system to confirm functionality and performance.

1.3.15. Active subsystem
The active subsystem development has driven numerous activities to cope with the amount of data which is generated by the 3D imaging system at high resolution. Especially the combination of mirror based beam focus and digital MIMO beam forming is an interesting concept which can be used for many other scenarios. The back-end had to be designed for high repetition frame-rates and high no. of channels. Beside the system developed within the CONSORTIS project, the TS system is the only available person screening system which uses active FMCW radar at such high frequencies. The multiplier technology allows the use of high bandwidths based on microwave signal generation. Part of the active subsystem module integration and MIMO antenna are shown in Figure 14.

1.3.16. Combination of mechanical movement and MIMO imaging
The active subsystem designed and realized in TeraSCREEN combines the mirror based focus as used for passive imaging with digital beamforming used for active radar arrays. This allows a high dynamic range and a reasonable update rate. At the same time the system costs are kept low. The operating principles of this imaging method are shown in Figure 15.

1.3.17. Millimetre and submillimetre wave component testbed (active)
One of the key points within the TS project was to develop advanced millimetre wave components, whose performance can be demonstrated in the imaging system. The active subsystem now provides a subsystem suited for demonstrating components at 180 and 360 GHz. The mixers and multipliers from STFC and TCL provided at 180 GHz already allow an impressive dynamic range at such a high frequency. As soon as the LNA designed in TS is available and the 90 GHz power amplifiers are finally working the active subsystem can be used for system experiments to support dissemination of those modules.

1.3.18. Broadband imaging at millimetre wave frequencies
The demonstrator of the active subsystem realized in TS covers 15 GHz bandwidth centred at 180 GHz. This allows full 3D imaging as well as the potential capability for material identification. With a successful delivery of the 90 GHz amplifiers developed by GU the subsystem could also be updated to 360 GHz. An example image is shown in Figure 17.
Measurements were conducted by attaching testpieces to the body of the model using methods that leave little or no signature in themselves. In this test, the testpieces were secured using wide mesh thin netting or rubber bands. All measurements were conducted at a distance of 4.5 m. The tests were performed by ICTS directly at Fraunhofer FHR.
Figure 16 shows what appears to be a side view - this is the ‘Range’ view. The Range view is produced by reconstructing an image using the ‘time of flight’ of reflections from surfaces back to the imager. Figure 18 shows the Range view collected from the side of the body without a test piece and with test piece at the upper arm.
The Range view cannot be used at a checkpoint for detecting concealed items because it only shows the profile of surfaces nearest the imager whereas a more traditional millimetre wave picture shows an image made by different areas of the body reflecting at different angles and intensities. However, these images show the good resolution in the time of flight (TOF) axis and this capability can be combined with improved standard reflection images to enable the determination of the shape and depth of concealed items. This, in turn, will assist the imager to bring down the nuisance alarm rate.
The imager was tested with the following materials:
• Metal wool (100% reflection)
• Plasticine (dense, clay, particle based)
• Beeswax (traditional testpiece for 94GHz passive imagers)
• Polyamide 66 (active imager simulant)
• Playdoh (flour based)
• Water (liquid)
All had approximately the same dimensions of 18x15x1 cm. Metal wool, plasticine and water were detected. Beeswax, Polyamide 66 and Playdoh were not observed.
Due to multiple reflections caused by partly transmissive material, it is possible to distinguish metallic and non-metallic objects. The key issue in this case is the refractive index contrast of the material compared to the surrounding medium. If this is small the dynamic range of the image needs to be sufficiently high, which is not yet the case for the current active subsystem.
Experiments were conducted to see whether the testpieces could be visualized under clothing. There were several indications that the penetration at 180GHz is poor. The radar image generated is compared for typical scenarios, a person wearing a shirt then when wearing a jacket. The imaging through the shirt clearly shows the outline of the person (Figure 19 left), while on the other image (Figure 19 right) only the outline of the jacket can be seen. The used frequency band appears to be too high for penetration through thick clothing. To avoid this, lower frequency bands should be used.
The active imaging system was developed to a stage where tests were possible. However, the planned performance could not be obtained since the component development within TeraSCREEN was heavily delayed. However, an alternative concept, based on commercial components, made it possible to obtain an operational system that can be upgraded as soon as resources are available for that purpose.

1.3.19. Sensor data fusion and image processing
An algorithm for automatic detection of hidden threats for passive data has been developed and tested with 94 GHz images. The algorithm works quite well for locating threats on chest, stomach, lower and upper back, etc. To enhance the robustness of the approach, the outputs of all frames in a single sequence are fused. The algorithm is designed in a way that its performance can be tuned with more data from 94 GHz images or even be adapted to other frequencies like the 360 GHz passive subsystem.
An algorithm for the active data was developed by using simulated data as well as last-minute real data. Data reading, data display and analysis of 2d projections have been done. Two approaches for the classification of threats are proposed. The intensity based filtering detects threats with a sliding window approach according to a material database. As an alternative solid object, like guns and knives may be detected based on their shape. The performance of the developed algorithm remains to be tuned with more images from the scanner.
A fusion algorithm has been designed and implemented based on the initial data from the 94 GHz passive subsystem and the 180 GHz active subsystem. Partial results from all frequencies, passive and active are weighed together and fused in order to obtain the final detection and classification result. The system is capable of reading and processing the data of the active and passive subsystems in parallel. The level of robustness of the algorithms will depend on the data being available. Lots of data is required for testing and finalising parameters of the algorithms.

1.3.20. Ethical compliance
The ethical aspects and privacy issues that arise in the context of the application of screening systems have been systematically identified. This was done in such a way so that the prototype development can address both passenger safety aspects and the fulfilment of their fundamental rights to privacy and freedom. The most prominent ethical concerns regarding screening systems have been identified as health risks and violations of bodily privacy. The TeraSCREEN prototype concept addresses both concerns through the use of non-ionising radiation for which no health concerns could be identified and through the use of automatic threat detection and item recognition algorithms that have a privacy preserving effect. Further ethical aspects that could be identified for similar screening systems were data protection concerns. Hence, for the TeraSCREEN prototype, data protection guidelines were established. In particular, the prototype must not store, retain, export or print images and it must not transmit images via internet or any Ethernet accessible for non-authorized people.

Potential Impact:
The TeraSCREEN project potentially has an impact on several levels for millimetre and submillimetre wave technology for the EU. Namely at the semiconductor processing level, the component level, the module level and the system level.
At the semiconductor processing level, it is important for the EU to have a European foundry that can compete globally in the area of high frequency MMIC technology. OMMIC has key enabling technology in this area. The D004IH process, that has been aided and improved by the TeraSCREEN project, is such an example. The miniaturization required for processes suitable for designs working at the 360 GHz frequency required a lot of technical improvement to the process in order to solve key fabrication challenges so that it is possible to produce and handle the new products with good yield.
The developed 40 nm mHEMT commercial technology can be applied for designs up to 500 GHz. This will be a key process technology for the design of imaging and telecommunication systems in the EU with EU technology, and is expected to be a foundation for a number of key EU research projects in the future, as well as new products for companies within the EU.
Furthermore, the InP DHBT process technology at the Goethe-Universität Frankfurt am Main in cooperation with Ferdinand-Braun-Institut has been further developed and improved. The process has provided the basis for state-of-the-art component performance at W-band frequencies. This process technology and competence is also expected to have impact for EU researchers and product developers.
Finally, the Schottky diode technology developed by the Science and Technology Facilities Council and Teratech Components provides the EU with a global leader for submillimetre wave Schottky diode based component development and sourcing.
Regarding the component level, several components have been developed in this project beyond current state-of-the-art, and thus have an impact on the level of European technology available in this field and in other fields where the technology is also applicable, such as wireless telecommunication and space applications.
The availability of 140 GHz and 360 GHz LNAs, which will give rise to standard products, after some optimization runs, will be useful for many purposes, such as high-resolution imagery, radar and remote sensing in optically challenging environments, and high data rate communications. The expertise gained in this field, during the TeraSCREEN project, by the research group of Roma "Tor Vergata", together with the 40 nm OMMIC technology, will be the key starting point for the improvement of the technology in order to develop wireless telecommunication systems in the D and G band, and for participation in new EU research projects such as ULTRAWAVE.
The Schottky diode technology developments by Science and Technology Facilities Council and Teratech Components have led to frequency multiplied source and mixer components 360 GHz and 180 GHz. Such components are important for European business interests in Test and Measurement, Radar, Imaging, and NDT applications. The same applies for the high-power W-band frequency multipliers based on InP DHBT technology. Furthermore, in terms of European non-dependence the MMIC development at Goethe-Universität Frankfurt am Main, using the InP DHBT technology developed is of great interest to Europe. For example, there are currently no PA components available with similar performance to those achieved in TeraSCREEN from European institutions. Even the Integrated DDS is of strategic importance for radar, communications, and measurement systems applications. Given the fact that the performance of the integrated DDS is verified independently, it could serve European companies in the above fields and provide substantial advantage over competitors. The phase control word allows for the realization of digital modulation schemes, which would be a unique feature for high frequency DDS components.
The antenna designs by Anteral and Fraunhofer FHR in TeraSCREEN have aimed at designs suitable for improving fabrication techniques. As tolerance and fabrication aspects are improving, there becomes scope for new and more complicated antenna designs that can exploit conventional fabrication techniques, but also the 3D printing methods that may reduce cost, fabrication time and weight, whilst providing similar performance, or even better performance in some cases. It is important that the EU maintains its strong position in such an important field that is applicable in so many different technology areas.
Concerning the module and system levels, it is highly important for the EU to develop capabilities in the field of millimetre and submillimetre wave technology. Competence in this technology area is required at all levels, from process technology and MMIC technology, to packaging, integration, and systems. The module integration in TeraSCREEN proved to be a challenge. However, it enabled the cooperation of expertise and created new research contact networks within the consortium to improve the level of knowledge and competence for Europe. For the research institute RISE Acreo the work was a significant step in a new direction allowing the institute to extend its role as a non-profit research organisation to support a sustainable society. Participation also helped the participating research groups at Fraunhofer maintain their leading position in the area of high frequency active imaging technology.
The TeraSCREEN work has led to component test beds for both the active and passive subsystems that could be used by research organisations and companies, within the EU, that are interested in testing their developed mm-wave components and modules, in such systems to confirm functionality and performance. TeraSCREEN provided the partners developing components to test them as an integrated part of an operational subsystem, this providing an opportunity for them to evaluate their technology improvements in a millimetre wave test system.
The project studied an approach of combining mechanical and electronic beam forming. This concept allows the development of imaging systems even at submillimetre wave frequencies, which would not be possible only relying on lower-cost CMOS or SiGe technology. While completely integrated systems are feasible today up to 80 GHz, massive scientific work has to be performed to reach higher frequency bands. Here the modular structure of the TS subsystem can be used as a platform for integrating components with a higher level of integration, and extend their frequency range by making use of the multiplier technologies developed within the project. In this way, first images can be generated, at reasonable frame-rates, before such frequency ranges can be reached by silicon MMIC technology.
Another impact to consider for the EU is that the TeraSCREEN project has provided a demonstrator for full 3D imaging at the highest frequency which has been used so far for passenger screening. This results in a test-bed for detection algorithms as well as a valuable source for scanning data for further research. Other research groups as well as commercial players in the field are curious to see the capabilities of passenger screening at such high frequencies. The correct choice of the frequency band is an on-going discussion. So far, commercial systems have always been restricted to frequency bands where cheap integrated components are available. As MMIC development progresses, the results from TeraSCREEN will be of high value for future system approaches for system integrators in the EU. Combining the TeraSCREEN project results with project results from other relevant EU projects, such as CONSORTIS, and building on these foundations, will provide a EU knowledge platform that can be exploited.
The TeraSCREEN project addressed ethical issues. The project did not achieve many results regarding item recognition, however, there is a great potential to address many remaining privacy issues for state-of-the-art body scanners, e.g. false alarms for prostheses or medical devices. This may result in a considerable improvement of privacy protection e.g. at European Airports. A point of contact was provided by TeraSCREEN for the provision of expertise to CEN/Cenelec JWG 8 for developing an EU standard for "privacy by design" for security technologies. Such a standard would be recommended and important for the EU.
Generally, the expertise gained from many of the developments in this project has allowed / will allow the project partners to become involved in and propose new EU research projects within the different H2020 calls.
Partners have also gained experience in the volume production of devices with extensive process development for improved performance, yield and to lower the cost of manufacturing these devices. This means that the partners are well placed to supply components, in large numbers if necessary, for future projects for security, communications, and space customers.
Another impact is that of job and expertise creation: there are currently too few specialists available in the technological areas targeted in TeraSCREEN. However, the number of scientists with a basic know-how has been increased in the duration of the project by the work undertaken by the university and institute partners, which is in turn influencing university curricula. In the future this will constitute an elite group of highly qualified people capable of generating and undertaking innovative research as well as implementing practical solutions in this field.
The main dissemination activities that have taken place during the project are presentations of the project and project results at international scientific conferences and industrial workshops, and the corresponding publications in the conference proceedings. The project has been included in the “EU Research for a Secure Society” yearly publication and a journal paper is pending acceptance. TeraSCREEN has also been present at several major trade exhibitions in the aerospace, defence and security fields. A public project website was set up at the beginning of the project and has been kept updated throughout, both with information related to the project and with the different dissemination activities that have taken place. Press releases were also published in the media and on the partner websites at the start of the project. Additionally, project research has been used by the academic and several of the research partners as material for their Masters level courses. The project objective, as well as its intermediate and final results, have also been used as a demonstration of up-to-date research activities carried out at one of the university partners to attract new students in Electronic Engineering.
Concerning exploitation, the project results at the component and module level are being integrated into the business plans of the partners, making the resulting products and services available to customers. In most cases the technology is applicable to other business areas besides security, such as space, communications, test and measurement, agri-food, medical, high-resolution imagery, and radar and remote sensing in optically challenging environments.
The passive subsystem aims to be exploited as an integration platform test bed for research organisations and companies that are interested in testing mm-wave components, radiometers, etc., in a passive imaging system to confirm functionality and performance.
The active subsystem will serve as a millimetre-wave demonstrator, data source and test-bench for many future activities at the developing partner.
The expertise gained during the project is also being exploited by the different project partners via participation in other research projects, addressing different areas such as wireless telecommunications.
List of Websites:
www.fp7-terascreen.com

Final Report Summary - TERASCREEN (Multi-frequency multi-mode Terahertz screening for border checks)

Share this page

Download