Statistical Tools and 

Optoelectronic Measuring Instruments 

Ionel Sabin and Ionel Ioana 

Universitatea “Politehnica” Timişoara 

Romania

1. Introduction 

In the frame of European research projects, several air quality measuring campaigns in cross 

roads, streets, parks as well in a non ecological waste deposit were realized. The analyzed 

signals, representing CO, NO2, O3, SO2 and HC concentrations, were measured with several 

optoelectronic instruments. Two of the utilized optoelectronic devices are shortly presented 

at the beginning of the chapter. 

Due to their random character, pollutant concentrations signals can be analysed using 

statistical processing methods. The main statistical functions and parameters taken into 

account within this chapter are histograms, correlation coefficients, correlation and 

covariance functions (Ionel et al., 2009). Actually, statistical tools are usually utilized in 

analysing ecological data (Zuur et al., 2007) but, as far we know, it is not common to imply 

statistics in a comparative analysis of optoelectronic devices (Ionel et al., 2007). 

Specific pre-processing procedures must be used for signal conditioning. Thus, „ideal“ low-

pass filtering based on fast Fourier transform can be implemented for the rejection of 

measurement noise and artefacts from the pollutant concentration signals. On the other 

hand, „ideal“ high-pass filtering allows the extraction of the variable component of the 

pollutant level signals. In order to avoid redundant measurements, one can use 

interpolation for increasing the number of samples, especially in the case of slowly varying 

meteorological parameters. 

Computer experiments with real pollutant concentration signals lead to some practical 

recommendations concerning acquisition parameters like data size and sampling 

frequency. The most important practical rules are as follow: assure the temporal length of 

the measured signal, assure the necessary resolution on the time axis, and make 

interactive verifications of the acquisition parameters during the measuring campaign. 

Guidance on MATLAB software for calculating statistical functions an

d parameters are 

provided. 

As a particular application, the correlative comparison of two carbon monoxide (CO) 

measuring instruments is presented. The point source device and the open path optical 

remote sensing instrument do actually not measure the same quantity but a statistical 

comparison of the two instruments is still possible. The correlative analysis leads to the 

expected conclusion that the open path instrument is more suitable for monitoring the 

pollution level in a large area than the classical point source device. 

www.intechopen.com

412 Optoelectronic Devices and Properties 

2. Pollutant concentrations measured with optoelectronic instruments 

2.1 The optoelectronic measuring instruments 

One of the utilized instruments was the specialized HORIBA APMA-350E CO monitor, 

which furnishes the local pollution level. Fig. 1 presents a bloc diagram of this instrument 

working on the classical Non-Disperse Infrared (NDIR) method. The APMA-350E

represents a generation of ambient CO monitors designed to eliminate routine calibration 

cycles and to provide long-term stable measurements and unattended continuous 

operation. It features a newly developed cross-flow modulation (CFM) technique which 

results in remarkable improved zero drift performance and sensitivity. The cross-flow 

modulated analyzer incorporates the basic design features of the conventional NDIR 

analyzer.

The essential new element in this design, according to Fig. 1, is a rotary valve that

alternately directs the sample gas and a reference gas to the one cell of the analyzer. By this 

method, the distinction between the sample and the reference optical path is eliminated and 

each path alternately functions as a reference and a sample path. The requirement for an 

optical chopper to modulate the detector output is thereby eliminated. In the cross-flow 

analyzer design, sensitivity is inherently increased because the amount of IR (infrared) 

energy absorbed and translated into the output signal is theoretically doubled for any 

concentration at the given modulation frequency. In addition, the signal-to-noise ratio is 

significantly better because the optical chopper which tends to introduce noise in the 

conventional NDIR instrument is removed in this CFM design. In the CFM scheme, gas flow 

rates and cell configuration can be selecting providing very smooth modulation. To 

minimize interference, dual detector system employing a compensating detector located 

behind the main detector is adopted in this instrument. The two detectors are charged in 

such a way that response to the interference gas in the second detector is compared to that 

of the measured gas. The signal from this detector is amplified and subtracted from the 

main detector signal, in the electronic part of the analyzer.

The second utilized instrument was an IR HAWK system from Siemens Environmental 

Systems, with the schematic diagram presented in Fig. 2. This instrument is an IR DOAS 

(Differential Optical Absorption Spectroscopy) apparatus, which can be configured to detect 

several species of pollutants including carbon monoxide. The beam path can be up to 400m 

and detection is typically better than 50ppb. The HAWK system works by measuring the 

absorption of infrared radiation passing along the instrument beam path by the gas to be 

measured. The system consists of a monitor, which contains the source and the detector 

unit, and a reflector. The total path length is therefore twice the distance between monitor 

and reflector. The source emits over a range of wavelengths and the beam is modulated after 

generation. The beam is reflected back to the monitor where it is filtered at a wavelength 

specific to gas of interest. The filtered beam is focused onto a detector which compares 

filtered and unfiltered reflected light in order to measure the concentration of the target gas. 

Open path techniques have an advantage over the point source detectors: the sample volume 

is mach greater, the non-uniformity of the sample is eliminated and a more representative 

value of the concentration to be measured is obtained. Under field conditions, the degree of 

mixing is affected by the local environment, primarily, wind and thermal gradients. 

Fig. 3 shows a typical relative setup for the HAWK and HORIBA analysers. One should 

observe also the meteorological mast, which continuously sent data (15 minutes mean 

values) to the general data acquisition system.

sumber : Hoffmann, J. & Quint, F. (2007). Signalverarbeitung mit MATLAB® und SIMULINK®. 

Anwendungsorientierte Simulationen, Oldenbourg Verlag, ISBN 978-3-486-58427-1, 

München