THE PROBLEM

Covert evidence gathering has not seen major changes in decades. Law enforcement Agencies (LEAs) are even today using conventional, man-power based techniques such as interviews and searches to gather forensic evidence. As sophistication of both criminals and their crimes are increasing, investigators must improve the means available for gathering a body of compelling evidence of a suspect’s involvement in a crime. Concealed surveillance devices have been instrumental in this direction, providing irrefutable evidence that can play an important part in bringing criminals to justice. However, current video surveillance systems are usually bulky and complicated, and rely on complex, expensive infrastructure to supply power, bandwidth and illumination.

A current digital IP surveillance system

A current digital IP surveillance system.

EXISTING TECHNOLOGICAL ADVANCES

ADVANCES IN IMAGING SENSORS

Recent years have seen significant advances in the surveillance industry, in both hardware and software, but these were often targeted to other markets and applications. Sensor technology has evolved, resulting in smaller modules with significantly higher resolution and image quality. At the same time platforms that host the sensors have also improved, embedding more powerful processing units, enabling more complex but also more energy-demanding operations. However, the imaging community is, understandably, focusing on cameras for mobile phones, where the figures of merit are resolution, image quality, and extremely low profile. Power consumption, while an important parameter, is often a secondary aspect. A mobile phone with its camera on would consume its entire power supply in less than two hours. Industrial surveillance cameras are even more power hungry, typically requiring 10W for their operation, even without night illumination.

Important advances have been also achieved in the signal processing domain. New vision algorithms have been produced to facilitate a rich semantic understanding of events occurring in a parking lot, a city corner, a highway, an office room, or a house where an elderly person is living alone (face recognition, object recognition to name a few). However, many recent breakthroughs are oriented towards the advantages that the “Cloud” and “Big Data” offer and require extremely high processing power, such as backend server farms, and are not available in conventional surveillance systems. Low power consumption for surveilance has yet to emerge as a central topic for R&D.

CURRENT STATE

The gradual integration of new technologies in the standard law enforcement procedure for fighting organized crime has led to a dramatic increase in the availability of equipment within the reach of law enforcement agencies for covert surveillance. Historically seen as the preserve of specialist units and associated with investigations into serious and organised crime, covert deployments are now widely made. New technologies have enabled better quality of video surveillance, miniaturization of devices, multiple communication channels and increased autonomy with the use of energy harvesters. However, video surveillance is currently still expensive, power-hungry, complicated and used mainly for recording with only limited real-time scene interpretation analytics. Despite the abundance of commercial devices, what is missing is an autonomous, intelligent sensor; a small, smart device, easily concealable, intelligent enough to provide the required evidence effectively and cost effectively. Such a device must be able to respond to operational requirements, recording only when predefined events occur, minimizing cost, complexity, installing time, and reliance on often unavailable infrastructure. To multiply the capabilities of such sensors, secure and intelligent communications are needed; secure to respect the unfavourable operation conditions and intelligent to safeguard the energy preserves against unnecessary communications attempts.

CONCERNS

ETHICAL AND LEGAL CONCERNS

It is well known that covert surveillance techniques imply interference with fundamental rights of the persons subject to surveillance, and in particular to the potential impact on their right to privacy and to the protection of their personal data. Such rights are guaranteed by several international and European human right conventions and declarations, as well as by the European Union and national data protection legislations. Given the likely negative impact of the video surveillance on the privacy and data protection, it is essential that the human rights impact of the use of a covert technique is considered in advance of each deployment. Not all socially accepted values are encoded into the law, but remain an important reference point in the form of the ethical principles. Surveillance technology might interfere with such social values, endangering the acceptance of the surveillance technologies in the democratic societies. It is, thus, crucial to identify the ethical questions and principles applicable to surveillance, and assess the impact of the surveillance on such principles. It is also important to consider the application of the general principles of the criminal justice, such as the principles of the due process of law or the principle of presumption of innocence.

The integrity of the acquired digital evidence plays a predominant role in the digital process of forensic investigation. Proper chain of custody must include information on how the evidence was collected, transported, analysed, preserved, and handled. It must document where, when and how the digital evidence was discovered, collected, handled with, when and who came in contact with the evidence and whether it is altered in any way. If a link is missing in this chain, it could be deemed compromised and may be rejected by the court.

Any successful technical implementation does not necessarily warrant its admissibility in court. A strict and formal process needs to be followed, which can vary significantly among Member State, in order to make evidence, provided by a surveillance system, acceptable in court for the cause of prosecuting crime. Initially the device needs to be certified and controlled by the authorised crime prosecuting and investigating authorities, in order at first to be employed by them. Its findings need to be examined and cross-tested, in order to warrant their technical usability in court. Even more importantly, the relevant legal provisions need to be examined to ensure that such findings can be used as evidence and the recording procedure is completely transparent. In the court, all the technical data regarding the recorded data will need to be made available to the defendor’s legal side, in order to be able to be examined with their own technical advisors, and maybe contest their consistency, if applicable.

EUROPE 2020

EU’s growth strategy for the coming decade

INNOVATION UNION

A Europe 2020 flagship initiative aimed at securing Europe’s global competitiveness

HORIZON 2020

A financial instrument implementing the Innovation Union (the biggest EU R&I programme ever with nearly €80 billion of funding available over 2014 to 2020)

THANK YOU EU!

FORENSOR is co-funded by the Horizon 2020 Framework Programme of the European Union

THE PROPOSED SOLUTION

MAIN GOAL

We aim to develop and validate a novel, ultra-low-power, miniaturised, low-cost, wireless, autonomous sensor (“FORENSOR”) for evidence gathering, able to operate for up to two months without infrastructure. FORENSOR will be manageable remotely, will preserve the availability and the integrity of the evidence collected, and comply with all legal and ethical standards, in particular those related to privacy and personal data protection. Secure and intelligent communications let such sensors join their forces towards robust evidence management and real time monitoring and control operations. The combination of built-in intelligence with ultra-low power consumption will make this device a true breakthrough for combating crime.

SPECIFIC OBJECTIVES

ULTRA-LOW POWER OPERATION

Enabling operation for several weeks, even in infrastructure-less environments.

Allowing acquisition of suspected criminal activity only.

Allowing the sensor to adjust to different operational requirements and use cases. The FORENSOR will work in both remote, low-activity environments, working for months on a single battery w/wo power harvesting, and in crowded environments where the sensor will be able to operate autonomously for several weeks.

For receiving crime alerts, monitoring the sensor’s operational status, managing the operating profiles and the recorded data.

For smooth collaboration among sensors, reliable evidence management and real time operation between the sensors and the monitoring station.

For easy deployment and concealment in environments.

To ensure the availability and the integrity of the evidence recorded and, therefore, ensure that the chain of custody remains intact until the data is presented in a court of law.

To ensure abidance to all relevant legal and ethical standards, in particular those related to privacy and personal data protection.

To make viable the deployment of swarms of sensors.

DETAILS

FORENSOR will deliver a unique combination of an ultra-low-power visual sensor and built-in intelligence for understanding the events occurring in the observed scene. FORENSOR will filter irrelevant events such as moving trees, shadows and illumination changes at the sensor level, and recognize specific objects and activities such as a person standing, walking or running, a car speeding, someone climbing a fence or reaching for a specific object. For example, if real-time analysis identifies that a prescribed activity occurs, the sensor is activated and several video frames are recorded and potentially transmitted in an optimal way (through a multi-criterion routing algorithm), in terms of security and energy effectiveness towards the monitoring centre. With this strategy, the higher processing stages of the visual sensor stay idle for up to 99% of the time, depending on the operational situation, allowing the sensor to operate on a single battery for several weeks. Note that based on unique technology introduced, the lower level algorithms embedded in the CMOS sensor’s silicon pixel operate 24/7 continuously, consuming 1-5 millwatts only. In addition, the developed devices will be low-cost allowing the deployment of dense wireless sensor networks. To support data integrity, FORENSOR will investigate existing and will develop new, decentralized storage and validation schemes among the sensors that will ensure the availability and credibility of the collected evidence in Court.

A simple low-power vision system

Prototype of a simple low-power vision system with low-level event detection capabilities.

VALIDATION

FORENSOR will be tested and validated in real-life scenarios carefully selected by the LEA End Users. Initially, the visual sensor will be developed and training sequences will be captured to enable scene analysis and then the FORENSOR framework will be tested and validated for more than 18 months. Such scenarios are already analysed and present both considerable threat and impact to the involved users.