Part of the appeal of the PIR sensor is that it very reliably distinguishes moving bodies from stationary ones, and from stationary and moving objects. Its basic mode of operation is to detect the difference in heat signature between two segments in its field of view. The model of the internal structure of a PIR sensor (Figure 1) is the best way to understand how the device works.To avoid triggering on normal temperature variations or airflows, in a dual-element sensor the two elements are connected in pairs and inverted to each other in terms of polarisation. When the two inverted elements are exposed to the same infrared radiation level, they cancel each other out, providing a zero output signal as a result. his means that the detected body will have to move into or between the two elements' field of view to make the sensor generate an output signal. This makes a dual- or quad-element sensor good at rejecting false detections. By using a multi-element sensor it is also possible to detect the direction in which the object is moving. The quad-element sensor in Figure 1 has two outputs: this means that it can indicate in which direction the movement is taking place (i.e. in the horizontal or vertical plane). The dual sensor can only indicate movement in one axis. When specifying a sensor, the number of elements is only the first consideration. Other important parameters that vary between sensors are frequency response, which determines the sensor's ability to detect low- and high-speed movements; the angle of the field of view, which will affect the size of the sensor's coverage area; and immunity to RF and background noise.

Configuring the lens

The sensor itself is inefficient if it does not have a lens to focus the radiation. The most commonly used lens type is the Fresnel lens, due to its low losses and small form factor. A Fresnel lens is a compressed plano-convex lens that is built of a set of discontinued surfaces. The grooves on the lens face the PIR sensor, which leaves a flat, dust-/weather-proof surface facing the outside and protecting the sensitive sensor.

Signal conditioning

To be usable, the signal from the sensor has to be amplified and then converted into a digital value for further analysis in software - a typical block schematic for this application is shown in Figure 2. There are several ways to design a circuit to realise this schematic: the two preferred approaches are to use either discrete analogue components, or - in a more integrated implementation - a mixed-signal programmable array (such as a Cypress PSoC device).

Discrete solutions

The most common approach to PIR-sensor signal conditioning is to design the amplifier and signal-conditioning stage by using discrete components such as operational amplifiers, comparators, diodes, resistors and capacitors; followed by a microcontroller with an integrated ADC for signal identification, and communications interfaces such as a radio. This traditional approach requires a large PCB footprint. But a PIR sensor produces a very low signal level, so it is essential to keep PCB traces short and the design compact to avoid creating unwanted antennas. These can pick up background noise and RF signals, which can cause the device to trigger falsely. If the PIR sensor is connected to a wireless network (for instance, as part of an intruder-alarm system), this danger is particularly high.

PSoC solutions

A second approach, which offers a more compact system, is to use a PSoC (programmable mixed-signal controller). Each one has an 8-bit core and a set of analogue and digital blocks that can be used to create the functionality needed. Analogue blocks that can be realised in PSoCs include ADCs, DACs, filters, amplifiers and comparators. On the digital side, you will find timers, counters, UARTs, SPI and PWMs. The first step in implementing a PIR sensor with a PSoC device should be to identify what analogue and digital functionality is required. When the block diagram is defined, an appropriate device with the right number of programmable blocks can be selected.

But why use a PIR sensor?

Other technologies for motion detection exist, including ultrasonic and microwave-radiation sensors, but the PIR sensor is the most popular. This is due to the combination of ease of set-up and performance. In addition, PIR sensors are cheap and draw little power. Future Electronics expects the rate of adoption of PIR sensors to grow fast, with applications such as surveillance and alarm systems, as well as power-saving devices, driving increased levels of usage.