Saturday, February 11, 2006



This documentation is intended to be used as a sensor selection reference during the design and planning of collision detection and avoidance systems. The documentation contains compendium of sensor technologies that can be used to enhance collision detection and avoidance in both permanent and temporary installations and facilities.
Before the IR sensor most important aspect to be discussed is Infrared (IR)Radiation. Infrared radiation is electromagnetic radiation of a wavelength longer than that of visible light, but shorter than that of microwave radiation. The name means “below red” (from the Latin infra, “below”),red being the color of visible light of longest wavelength. Infrared radiation spans three orders of magnitude and has a wavelengths between approximately 750nm and 1mm.
Different regions in the infrared
IR is often subdivided into:
near infrared NIR , IR-A DIN,0.75-1.4 µm in wavelength,defined by the water absorption,and commaonly used in fiber optic telecommunication because of low attenuation losses in the SiO2
short wavelength (shortwave)IR SWIR,IR-B DIN 1.4-3 µm,water absorption increases significantly at 1450 nm
mid wavelength IR MWIR,IR-C DIN,also intermediate-IR(IIR),3-8 µm
long wavelength IR LWIR,IR-c DIN,8-15 µm
far infrared FIR,15-1000 µm
However,these terms are not precise,and are used differently in various studies ie.e near (0.75-5 µm)/mid(5-30 µm)./long(30-1000 µm).Especially at the telecom-wavelengths the spectrum is further subdivided into individual bands, due to limitations of detectors,amplifiers and sources. Infrared radiation is often linked to heat,since objects at room temperature or above will emit radiation mostly concentrated in the mid-infrared band.


The performance of autonomous systems executing complex tasks or solving problems is highly dependent on their perception of the environment. This means that the more precisely the systems can recognize the environment, the more complex problems can be solved. These circumstances apply to almost all automation applications, where sensors detect the environment in order t allow actions in accordance with the current situation. Despite the advancement in vision systems,active infrared sensors, are still, and especially,being used in autonomous systems, because they require less computational power and entail lower costs. They consist o emitting and receiving elements , which , in most cases, are infrared LEDs and photodiodes or phototransistors.
While the emitter illuminates the surroundings,the receiver measures the amount of light returned by, e.g., being reflected by a nearby object this work focuses on the efficient and flexible simulation of various active sensor configurations and also shows how the results can be evaluated in order to improve the recognition abilities. Being able to predict what a certain sensor system can recognize is mandatory for the design of new systems for specific applications and can help in the analysis of limitations of existing systems in executing certain tasks. In this case,simulation can be a very efficient tool, as real measurements involving a variety of sensor configurations can be very costly and time consuming.


The integration of sensors and systems is a major design consideration and is best accomplished as part of an overall system/installation/facility security screen. Although sensors are designed primarily for either interior or exterior applications, many sensors can be used in both environments. Exterior detection sensors are used to detect unauthorized entry into clear areas or isolation zones that constitute the perimeter of a protected area, a building or a fixed site facility. Interior detection sensors are used to detect penetration into a structure, movement within a structure or to provide knowledge of intruder contact with a critical or sensitive item.


Six factors typically affect the Probability of Detection (Pd) of most area surveillance (volumetric) sensors, although to varying degrees. These are the:1) amount and pattern of emitted energy; 2) size of the object; 3) distance to the object; 4) speed of the object; 5) direction of movement and 6) reflection/absorption characteristics of the energy waves by the intruder and the environment (e.g. open area, shrubbery, or wooded). Theoretically, the more definitive the energy pattern, the better. Likewise, the larger the intruder/moving object the higher the probability of detection. Similarly, the shorter the distance from the sensor to the intruder/object, and the faster the movement of the intruder/object, the higher the probability of detection. A lateral movement that is fast typically has a higher probability of detection than a slow straight-on movement. Lastly, the greater the contrast between the intruder/moving object and the overall reflection/absorption characteristics of the environment (area under surveillance), the greater the probability of detection.


Exterior sensors detect intruders crossing a particular boundary or entering a protected zone. The sensors can be placed in clear zones, e.g. open fields, around buildings or along fence lines. Exterior sensors must be resilient enough not only to withstand outdoor weather conditions, such as extreme heat, cold, dust, rain, sleet and snow, but also reliable enough to detect intrusion during such harsh environmental conditions.

Exterior sensors have a lower probability of detecting intruders and a higher false alarm rate than their interior counterparts. This is due largely to many uncontrollable factors such as: wind, rain, ice, standing water, blowing debris, random animals and human activity, as well as other sources to include electronic interference. These factors often require the use of two or more sensors to ensure an effective intrusion detection screen.

Interior sensors are used to detect intrusion into a building or facility or a specified area inside a building or facility. Many of these sensors are designed for indoor use only, and should not be exposed to weather elements. Interior sensors perform one of three functions: (1) detection of an intruder approaching or penetrating a secured boundary, such as a door, wall, roof, floor, vent or window, (2) detection of an intruder moving within a secured area, such as a room or hallway and, (3) detection of an intruder moving, lifting, or touching a particular object. Interior sensors are also susceptible to false and nuisance alarms, however not to the extent of their exterior counterparts. This is due to the more controlled nature of the environment in which the sensors are employed.


With the advent of modern day electronics, the flexibility to integrate a variety of equipment and capabilities greatly enhances the potential to design an collision avoidence and detection system to meet specific needs. The main elements of an collision avoidence and detection system include: a) the collision Detection Sensor(s), b) the Alarm Processor, c) the detection/avoidence Monitoring Station, and d) the communications structure that connects these elements and connects the system to the reaction elements. However, all systems also include people and procedures, both of which are of equal and possibly greater importance than the individual technology aspects of the system. In order to effectively utilize an installed security system, personnel are required to operate, monitor and maintain the system, while an equally professional team is needed to assess and respond to possible collisions.Collisions detection sensors discussed in this Handbook have been designed to provide collision detection and include sensors for use in the ground, open areas, inside rooms and buildings, doors and windows. They can be used as standalone devices or in conjunction with other sensors to enhance the probability of detection. In the majority of applications, intrusion detection sensors are used in conjunction with a set of physical barriers and personnel/vehicles access control systems. Determining which sensor(s) are to be employed begins with a determination of what has to be protected, its current vulnerabilities, and the potential threat. All of these factors are elements of a Risk Assessment, which is the first set in the design process.


In the process of evaluating individual collision detection sensors, there are at least three performance characteristics which should be considered: Probability of Detection (PD), False Alarm Rate (FAR), and Vulnerability to Defeat (i.e. typical measures used to defeat or circumvent the sensor). A major goal of the security planner is to field an integrated collision Detection System (ICS), which exhibits a low FAR and a high PD and is not susceptible to defeat. Probability of Detection provides an indication of sensor performance in detecting movement within a zone covered by the sensor. Probability of detection involves not only the characteristics of the sensor, but also the environment, the method of installation and adjustment, and the assumed behavior of an intruder. False Alarm Rate indicates the expected rate of occurrence of alarms high is not attributable to intrusion activity. For purposes of this Handbook, "false alarms" and "nuisance alarms" are included under the overall term "False Alarm Rate", although technically, there is a distinction between the two terms. A nuisance alarm is an alarm event which the reason is known or suspected (e.g. animal movement/electric disturbance) was probably not caused by an intruder. A false alarm is an alarm when the cause is unknown and an collision is therefore possible, but a determination after the fact indicates no collision was attempted. However, since the cause of most alarms (both nuisance/false) usually cannot be assessed immediately, all must be responded to as if there is a valid intrusion attempt.

Vulnerability to Defeat is another measure of the effectiveness of sensors. Since there is presently no single sensor which can reliably detect all collisions, and still have an acceptably low FAR, the potential for "defeat" can be reduced by designing sensor coverage using multiple units of the same sensor, and/or including more than one type of sensor, to provide overlapping of the coverage area and mutual protection for each sensor.


From a technology perspective, the integration of sensors into a coherent security system has become relatively easy. Typically, most sensor systems have an alarm relay, from points a, b or c, and may have an additional relay to indicate a tamper condition. This relay is connected to field panels via four wires, two for the alarm relay and two for the tamper relay, or two wires, with a resistive network installed to differentiate between an alarm and tamper condition. Most monitoring systems will also provide a means of monitoring the status of the wiring to each device. This is called line supervision. This monitoring of the wiring provides the user with additional security by indicating if circuits have been cut or bypassed.

Additionally, different sensors can be integrated to reduce false alarm rates, and/or increase the probability of intrusion detection. Sensor alarm and tamper circuits can be joined together by installing a logic "and" circuit. This "and" system then requires multiple sensors to indicate an alarm condition prior to the field unit sending an alarm indication. Usage of the logic "and" circuit can reduce false alarm rates but it may decrease the probability of detection because two or more sensors are required to detect an alarm condition prior to initiating an alarm.


Communications between the front-end computer and the field elements (sensors, processors) usually employ a variety of standard communications protocols. RS-485, RS-232, Frequency Shift Keying (FSK), and Dual Tone Multi Frequency (DTMF) dial are the most common, although occasionally manufacturers will use their own proprietary communications protocol which can limit the option for future upgrades and additions. In order to reduce the tasks required to be handled by the computer, some systems require a preprocessing unit located between the computer and the field processing elements. This preprocessor acts as the communications coordinator to "talk" to the field elements thus relieving the computer of these responsibilities.


Regardless of how well designed and installed, all collision detection systems are vulnerable to power losses, and many do not have an automatic restart capability without human intervention. Potential intruders are aware of this vulnerability and may seek to "cut" power if they cannot circumvent the system via other means. It is critical that all elements of the system have power backups incorporated into the design and operation to guarantee uninterrupted integrity of the sensor field, alarm reporting, situation assessment, and response force reaction.


The costs of an Collision Detection System are easy to underestimate. Sensor manufacturers often quote a cost per meter, cost per protected volume, for the sensor system. Often this figure is representative of the hardware cost only, and does not include the costs of installation, any associated construction or maintenance. Normally, the costs associated with procuring the sensor components are outweighed by the costs associated with acquiring and installing the assessment and alarm reporting systems.


Most sensors have been designed with a specific application in mind. The environment categorizes these applications where they are most commonly employed. The two basic environments or categories are Exterior and Interior. Each of the two basic categories has a number of sub-sets, such as fence, door, window, hallway, and room. The first two of the following set of graphics show a "family tree" illustration of the sensors most applicable to the these two environments (exterior/interior). As mentioned previously, some of the technologies can be used in both environments and consequently are shown on both graphics.


Post a Comment

<< Home