**1. Introduction**

292 Real-Time Systems, Architecture, Scheduling, and Application

R. F. Hess, R. J Snowden. (1992). Temporal frequency filters in the human peripheral visual

R.J. Prokop, A.P. Reeves. (1992). A survey of moment-based techniques for unoccluded object representation and recognition. *Models Image Process*, vol. 54, pp. 438-460. Reinery L. (2000). Simulink-based Hw/Sw codesign of embedded neuron-fuzzy systems.

Rosin P.L. (1998). Thresholding for Change Detection, *Proceedings of the International* 

Simoncelli E. P., Adelson E. H., Heeger D. J. (1991). Probability Distributions of Optical Flow, *Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition*. Simoncelli E. P., Heeger D. J. (1998). A Model of Neuronal Responses in Visual Area MT,

Smith S. M. (1995). ASSET-2: Real-Time Motion Segmentation and Object Tracking. Technical Report TR95SMS2b available from www.fmrib.ox.ac.uk/~steve. Sobey P. and Srinivasan M. V. (1991). Measurement of Optical Flow by a Generalized Gradient Scheme, *Journal of the Optical Society of America A*, 8(9), pp. 1488-1498. Uras S., Girosi F., Verri A., and Torre V. (1998). A Computational Approach to Motion

V. Bruce, P.R. Green, M.A. Georgeson. (1996). *Visual Perception: Physiology, Psychology &* 

Van Hateren J. H., Ruderman D. L. (1998). Independent Component Analysis of Natural

Visual Cortex. *Proceedings of the Royal Society of London B 265*, pp. 2315- 2320. Verri A., Poggio T. (1989). Motion Field and Optical Flow: Qualitative Properties. *IEEE Transactions on Pattern Analysis and Machine Intelligence*, 11, pp. 490-498. Wallach H. (1976). *On Perceived Identity: 1. The Direction of Motion of Straight Lines*. In On

Weber K., Venkatesh S., Kieronska D. (1994). Insect Based Navigation and its Application to

Y.N. Zhang, Y. Zhang, C.Y. Wen. (2000). A new focus measure method using moments,

Yacoob. Y. and Davis. L. S. (1999). Temporal Multi-scale Models for Flow and Acceleration.

Yacoob. Y. and Davis. L. S. (1999). Temporal Multi-scale Models for Flow and Acceleration,

Young R. A., Lesperance R. M. (1993). A Physiological Model of Motion Analysis for Machine Vision. *Technical Report GMR 7878*, Warren, MI: General Motors Res. Lab. Zanker J. M. (1996). On the Elementary Mechanism Underlying Secondary Motion Processing, *Phil. Transactions of the Royal Society of London*, B 351, pp. 1725-1736. Zhang J. Z., Jonathan Wu. Q. M. (2001). A Pyramid Approach to Motion Tracking, *Real-Time* 

Zuloaga A., Bidarte U., Martin J., Ezquerra J. (1998). Optical Flow Estimator Using VHDL for Implementation in FPGA, *Proc. of the Int. Conf. on Image processing*, pp. 972-976.

Image Sequences Yield Spatio-Temporal Filters Similar to Simple Cells in Primary

the Autonomous Control of Mobile Robots, *Proceedings of the International Conference* 

*International Journal of Neural Systems*, Vol. 10, No. 3, pp. 211-226.

field, *Vision Research*, vol. 32, pp. 61-72.

*Conference of Computer Vision*, pp. 274-279.

Perception, *Biological Cybernetics*, 60, pp. 79-87.

*Ecology*. third ed., Laurence Erlbaum Associates, Hove.

*Vision Research*, 38 (5), pp. 743-761.

Perception, New York: Quadrangle.

*Image Vision Comput*. vol. 18, 959–965.

*Imaging*, 7, pp. 529-544.

*on Automation, Robotics and Computer Vision.*

*International Journal of Computer Vision*, 32 (2), pp.1-17.

*International Journal of Computer Vision*, 32 (2), pp.1-17.

Focus of this chapter is the description of the exposure systems used for data acquisition during the exposure to radiofrequency (RF) electromagnetic (EM) fields in bioelectromagnetic investigations. Such a kind of system will be referred to as real-time and can be defined as an EM structure (waveguide or antenna) able to generate and control a known and reproducible EM field and suitable to be used in experiments where data acquisition has to be carried out simultaneously with the exposure (Paffi et al., 2010).

Common real-time applications are usually based on programs that function within a time frame that the user senses as immediate and which require what is called real-time computing (RTC). In biomedicine the real-time concept is applied to both fast calculation of some parameters of biomedical significance (Seong et al., 2011; Wang et al., 2011) and the experimental acquisition of physiological data simultaneously with a correlated event (Li et al., 2010; Voyvodic et al., 2011).

In this context, real-time exposure systems are used to acquire fast biological responses, typically in the order of milliseconds, simultaneously with the exposure to EM fields, in order to study possible health effects due to EM exposure. Usually the responses to be recorded are electrophysiological signals as cellular currents (mA) or membrane potentials (mV); they need to be acquired through a sophisticated instrumentation made of microscopes, patch-clamp recording electrodes, temperature sensors, which fix strict and defined requirements to the design and optimization of the exposure system.

The introduction of this kind of systems has been made necessary due to the need to better investigate the coupling of RF EM fields with learning and memory in both animal models and humans. Neurons, which are at the basis of brain functioning, are electrically active cells. Their electric fields are maintained and controlled by a wide variety of biochemical and metabolic processes. In neurons, fundamental functions such as neurotransmitter release, enzyme activation, intracellular signal transduction, and gene expression are critically dependent on electrical signals. Therefore it has been postulated several times a possible coupling of their electrical activity with a RF EM field.

Real Time Radio Frequency Exposure for Bio-Physical Data Acquisition 295

the appropriate directions for the development of exposure systems (Carlo, 1998). In 1996, the EMF Project of the World Health Organization (WHO) fixed these concepts and emphasized the importance of an accurate dosimetry in all scientific studies (WHO, 1996). Such items, together with a deep discussion on quality assurance, were the main arguments of two European Cooperation in Science and Technology (COST) workshops: "Exposure systems and their dosimetry", in Zurich, in February 1999 (Schönborn et al., 1999; Bitz et al., 1999) and "Forum on future European research on mobile communication and health", in

In 2000, recommended minimal requirements for exposure systems, in order to obtain reproducible and scientifically valuable results, were synthesized in (Kuster and Schönborn, 2000). Basically, two classes of requirements can be identified: biological and EM ones. Biological requirements are dictated by the laboratory equipment, the experimental procedures, and the environment; the EM ones define the exposure parameters (frequency, amplitude, modulation scheme of the signal), the characteristics of the induced EM field (polarization, intensity and homogeneity), and the "dose" at the location of the biological

More in detail, to meet the biological requirements, the exposure system must allow for the exposure of the required number of samples or animals; in the meantime, the environmental conditions required by the specific experiment must be guaranteed (Kuster and Schönborn, 2000). As for the EM requirements, (i) the delivered signal must be precisely defined in terms of frequency, amplitude, and modulation scheme; (ii) the electric and magnetic field strength and polarization must be known at the location of the exposed biological target; (iii) the fields inside the sample should be homogeneous; (iv) any electromagnetic interference (EMI) and/or electromagnetic compatibility (EMC) issue must be avoided (Kuster and Schönborn, 2000). Moreover, the system should guarantee the monitoring of relevant parameters, such as the temperature and the delivered power, during the experiment and the possibility of carrying out sham exposures and blind experiments. In the sham exposure, a number of cell cultures or animals are subjected to environmental conditions identical to those of the group of the exposed subjects, except for the exposure. Data collected by the sham group are thus used as negative control. This is an unavoidable procedure to prevent the incorrect attribution to the RF exposure of an observed effect, which might be due to other factors, e.g. to the stress of the animals (Samaras et al., 2005). In blind experiments the exposure setup provides an automated procedure to assign the real and the sham exposure to two different groups of samples. In this way, experimenter polarization errors are

removed, since he/she does not know which samples are really exposed.

sample per unitary input power feeding the EM structure.

Finally, the exposure system should be easy to be handled even by non-engineering personnel and its cost should be reasonable (Kuster and Schönborn, 2000). To maintain low the cost of the whole exposure setup, in particular of the signal amplifier of the generation chain, the EM structure should be designed to have the power efficiency as high as possible. The power efficiency is defined as the mean EM power absorbed by the unitary mass of

Biological requirements are usually the most limiting ones on the exposure setups. As example, the kind of experiment may dictate the equipment needed and the environment; the overall duration of the experiment and the number of samples or animals necessary for statistical significance may strongly influence the choice of the EM structure employed as

Bordeaux, in April 1999 (Kuster and Schönborn, 1999).

sample.

In the past, several studies were carried out suggesting a possible interference of the EM fields with neurons, but with controversial results and unable to clearly state the molecular basis of this interaction (Sienkiewicz et al., 2000; Wang and Lai, 2000; Dubreuil et al., 2003; Preece et al., 2005). For a review on EM field effects on cognitive functions see (D'Andrea et al., 2003). This is not surprising: experimental investigation of the EM coupling with neurons, in fact, is rather complicated since it involves measurements of the electrical activity of neurons, i.e. the acquisition of electrophysiological recordings of the transmembrane voltage and of the ionic currents. Therefore, a prerequisite of well-posed experiments involving neuronal activity is the possibility to acquire the useful signals simultaneously with the exposure to the RF EM field. Hence, in the last years, there has been a pressing need to design real-time exposure systems.

Some of the most recent results achieved with the aid of ad hoc designed real-time systems state that there is no significant coupling between low intensity RF EM fields and specific membrane channels, i.e. biological membrane proteins which allow the passive movement of ions from the external to the internal of the cell and vice-versa (Marchionni et al., 2006; Platano et al., 2007). Other studies, investigating the effects of low intensity RF EM fields on electrical activity in rat hippocampus, one of the major component of brain, (Tattersall et al., 2001) described an effect on synaptic transmission; however the effect reported has been later explained by localised heating produced by interaction of the RF fields with the recording and stimulating electrodes. This contradiction emphasizes the fundamental issue related to the proper design of real-time systems.

Although real-time investigations have been gaining increasing interest in the last ten years, a complete and systematic framework on specific requirements of RF exposure systems when applied to real-time data acquisition is still lacking.

Aim of this chapter is to merge the well-assessed design procedure of RF exposure systems (Kuster and Schönborn, 2000; Paffi et al., 2010) with the requirements emerging from realtime investigations, thus providing a reference work on this segment of knowledge.

In the chapter, at first a description of what is an exposure system and which are the guidelines to optimally design it are given. Then, it is provided how to adapt the general rules for exposure system design to real-time systems ones, which, of course, have very specific requirements. Finally, a complete and updated review of the real-time systems available in literature is provided.
