Study of the tactical and technical characteristics of the walk-through multi-zone metal detector Intelliscan 12000.

Study of the tactical and technical characteristics of the Intelliscan 12000 walk-through multi-zone metal detector.

Material received in April 1999.

The purpose of this study was to determine the probabilistic characteristics of the walk-through multi-zone metal detector «Intelliscan 12000» when carrying through a controlled passage simulators: a pistol (IP) and personal use items (PLP).

The metal detector «Intelliscan 12000» is manufactured by the American company RANGER.

Technical specifications:
(according to the operating documentation)

The tests were carried out using the following method.

1. In the controlled passage of the product, nine points were identified, the location of which is shown in Fig. 1.
2. The sought-after search objects were carried through each of the identified points, located in space sequentially in three mutually perpendicular positions, as shown in Fig. 2.

The IP corresponded in characteristics to a PSM-type pistol, and the PLP corresponded to a set of personal metal items with a total weight of 180 g.

The number of passes through each point in each position was ten, and the total number of passes of each object through the metal detector was 270. For each point, the response frequency p* of the product was calculated for 30 passes of the IP and PLP in all three positions.

3. The product operating modes were selected as follows:

• sensitivity modes: 10, 25, 50, 75;
• frequency mode 1.

4. The obtained research results are presented in Table 1.

Table 1

5. From the analysis of the obtained characteristics of the metal detector it follows:

5.1 Of the nine points considered in the area controlled by the metal detector, the worst results for detecting the IP were obtained for point 8. Therefore, further, the detection characteristics of the product are assessed based on the results obtained at this point.

The worst results for false alarms when carrying the PLP were found at points 7 and 9. Therefore, the selective characteristics of the product are estimated at these points.

According to the Poisson theorem, with an increase in the number of experiments, the frequency p* converges in probability to the arithmetic mean of the probabilities in each experiment. Therefore, we will consider p* to be an approximate value of the detection probabilities of the IP (Pbn.) and PLP (Rl.t.).

For point 8 with a sensitivity of 25 conventional units, where RBn. = 0.73, the probability that the error in replacing p* with RBn. does not exceed 20% in 30 experiments is 0.93. For point 9 with a sensitivity of 25 conventional units, where RBn. = 0.67, the probability that the error in replacing p* with RBn. does not exceed 20% in 30 experiments is 0.88.

Figure 3 shows graphs of the dependence on sensitivity of RBn. and RBn. at the worst points in the controlled area.

5.2 Increasing the sensitivity from 10 to 50 conventional units ensures an increase in the probability of detecting the IP from 0.67 to 1.0, while the probability of triggering from the PLP increases from 0.4 to 0.7.

5.3 When excluding from the analysis the least probable points of PLP carry (points 1, 3, 7, 9 in Fig. 1) for a sensitivity of 30 conventional units, the probability of detecting the IP will be 0.93 with a probability of triggering from the PLP of 0.6.

5.4 When a TV set (line scan frequency of 16 kHz) or a source of powerful low-frequency interference (50 Hz) were located near the product racks, false alarms did not occur. When a source of impulse interference with a frequency of 400 — 500 Hz or a personal computer monitor (line scan frequency of 30 kHz), the product gave false alarms at any sensitivity. The impact of changing the generator operating frequency (switching is carried out with the product unit casing removed) on noise immunity was not assessed during the studies.