Solutions for the self-Adaptation of Communicating Systems in Operation – SACSO

The proposed method is based on four steps:
1/ Definition of the performance to optimize
2/ Definition of indicators linked to targeted performance
3/ Design of the electrical circuitry to measure indicators
3/ Design of the electrical circuitry to give auto-adaptive capabilities to the system

To demonstrate the validity of our developed concepts, we work on three different demonstrators:

• The first demonstrator will be an NFC front-end interface, on which we will build a self -adaptation of the system to magnetic field in the environment. A static adaptation of the system (IC & antenna) to its dedicated application will be a starting point, then a dynamic adaptation will be considered to improve performance in an unstable magnetic environment. Parasitic modeling such as TAG distance, TAG antenna size, user hand, battery load level, etc., will be part of the second step of the project objectives described above. During this project, we plan to develop a complete demonstrator (NFC reader) including an NFC IC with our embedded solution.
• The Second demonstrator is a sensor dedicated to the Intra-Ocular Pressure (IOP) measurement. This sensor allows us to record the IOP during 24 hours to offer an efficient and reliable diagnosis and help the treatment of glaucoma. This device is very sensitive to environment context and especially to the specificity of each patient. Our objective in the SACSO project is to develop solutions allowing the sensor to correct or to adapt the measure according to the considered use case.
• The Third demonstrator is based on RF transceiver to investigate the suitable solution to manage and optimize the power consumption of this critical blocs in portable medical devices.

• Development of a complete model of the NFC transmission chain. The characteristic of this type of systems is its high sensitivity to the application context. In literature, no solution exists. The proposed model was submitted to an international conference.
• Development of a Biomechanical Emulator of Eye (BEE) to emulate the biomechanical features and parasitic effects for the measurement of intraocular pressure.
• Development of a static self-calibration for RF circuits in order to find the best trade off between power consumption and performance. The key point of this method is that the required measurements to achieve the goal of self-calibration are obtained by non-intrusive embedded sensors. This approach allows us to preserve the functionality of the circuit and perform the self-calibration in a single iteration without complex test instruments.

The proposed approach would be validated thanks to demonstrator covering a large part of existing features of portable medical devices. Based on this validation we will generalize the approach to be able to transfer it onto any kinds of application

1. M.Dieng et al, « Accurate and Efficient Analytical Electrical Model of Antenna for NFC Applications », IEEE international Newcas Conference, June 2013

2. A.Deluthault, V. Kerzérho, S. Bernard, M. Renovell, F. Soulier and P. Cauvet “Self-Adaptive System for Medical Application”, poster Groupement de Recherche SoC-SiP, Juin 2013

3. M. Dieng, T. Kervaon, P.H. Pugliesi-Conti, M. Comte, S. Bernard, V. Kerzérho, F. Azaïs, M. Renovell “Study a self-adaptation strategy system for Near Field Communication (NFC) transceiver”, poster Groupement de Recherche SoC-SiP, Juin 2013

In the context of high performance systems and critical applications, the general objective of the project is to design Self-Adaptive Systems which are capable of monitoring the surrounding environment and adapt themselves to different scenarios and requirements. In other words, a Self-Adaptive System must be able to provide the required high performances regardless the application mode and despite the changing environmental conditions.
The fundamental concept of self-adaptation by itself is not completely new: self-calibration techniques applied after manufacturing have been developed to deal with the many sources of technological variations and self-compensation techniques applied during the lifetime of the system have been developed to deal with the ageing effects. In this project, we address two non-classical cases of self-adaptability: self-adaptation to the application and self-adaptation to the environment.
Concerning self-adaptation to the application, on one hand, the performances of the complete system within the application are limited by the performances of each individual component or sub-system. On the other hand, the performances of the individual components or sub-systems are optimized for a large range of future systems or potential applications and not for a specific one. The originality of the approach proposed in this project is to ‘anticipate’ the system deployment while designing the components, i.e. the components are designed with an integrated ‘Self-Adaptation Circuitry’ which enables them to autonomously modify their electrical characteristics when they are deployed into the application, in order to optimize the performances of the complete system.
Concerning self-adaptation to the environment, it is to note that the performances of the system also depend on the changing environment. As an example, a cell-phone system may communicate differently when it is operating in-door or out-door, when there are obstacles in the measurement environment that could disturb the reception, etc. Here again, the components, provided with a dedicated ‘Self-Adaptation Circuitry’ (SAC), may be able to autonomously adapt their electrical characteristics to the changing environment, in order to optimize the system performances. This second case is more complex than the adaptation to the application that is typically performed only once when the components are integrated into the system. In addition, the range of possible applications is known at the time when the components are being designed. On the other hand, adaptation to the environment has to be performed continuously with the operation. In addition, the environment modifications or evolutions are intrinsically unpredictable. This is why we tend to qualify as ‘static’ the adaptation to the application and ‘dynamic’ the adaptation to the environment.
Although the methodology proposed in the project is generic and can be virtually applied to any high performance system, it will be demonstrated in the context of e-health. To this end, the project focuses on e-monitoring where the patient is equipped with one ingested sensor which communicates with an external electronic device. This device plays the role of a gateway to a network infrastructure. In this case, the two main performances that must be guaranteed are i) the quality of the communication and ii) the power consumption of the sensor.
In this project, we plan to develop generic solutions for static and dynamic adaptation for medical devices deployed in e-monitoring systems. We identified three critical cases in classical medical devices: electrical device with passive front-end, electrical device with active frontend, and power management of the application. The objective of this project is to address these three cases. All the developments will be validated on two prototypes: a Near Field Communication (NFC) circuit for the passive front-end example and an electrical medical pill for the active front-end example.