PIT Tag Technology
The following text describes the general principles of operation of low frequency PIT tag systems. It has been written with particular emphasis on application to fisheries. It is intended as a guide only and figures quoted should not be interpreted as part of specifications for products offered.
A definition of a transponder is: "A combined receiver and transmitter whose function is to transmit signals automatically when triggered by an interrogator".
The interrogator generates an alternating voltage at a specific frequency to drive an antenna used to energise the PIT tag. It also receives the signal transmitted by the PIT tag via the antenna, filters, amplifies and decodes it and formats it appropriately for the user.
The antenna generates and radiates an alternating magnetic field from the alternating voltage generated by the interrogator. It also receives the alternating magnetic field re-radiated by the PIT tag converting this to an alternating voltage to feed the interrogator.
Read-only PIT tags are programmed at manufacture with a unique number. This may be achieved by electrically programming an on-board Read Only Memory (ROM) device or by laser fusing polysilicon links corresponding to logic levels of the binary ID code. Write-once-read-many devices may or may not have a unique ID code programmed at manufacture, but allow one-time access, to program the device with an ID code. The device can then be read an unlimited number of times. Read-write devices again may have a factory programmed ID field and allow users to program additional data into the device at a later date.
The number of data bits governs
how much and the type of data that can be stored in the device. The minimum
number of data bits that will normally be used is 64. This number of bits
permits a very large number of unique ID numbers to exist whilst leaving
enough bits for error detection and framing purposes. This may be extended
to thousands of bits if other information needs to be stored. The greater
the number of data bits, the longer the period required to read the data
from the PIT tag and therefore the lower the maximum PIT tag detection
As frequency increases, absorption of electromagnetic energy by water and body tissue increases which tends to rule out the use of passive tags at UHF and VHF frequencies for fish studies. As frequency reduces, antenna size generally increases. The use of ferrite-based antennae to reduce antenna size is generally restricted at frequencies in excess of several MHz due to the poor performance of ferrite material at these frequencies. For these reasons, the 125kHz to 135kHz band has been widely adopted for PIT tags for animal and fish use. This is generally referred to as the 'low frequency' (LF) band.
The return modulation frequency (i.e. the frequency at which the PIT tag transmits data back to the interrogator) for LF PIT tags is always less or equal to the energisation frequency. It is simpler electronically to generate new frequencies that are sub-multiples of the energisation frequency using division, than multiples using multiplication techniques.
The term 'modulation' refers to the method used to attach data onto the return frequency generated by the PIT tag. PIT tags use 'backscatter modulation'. The RF link between interrogator and PIT tag behaves essentially as a transformer. As the secondary winding (PIT tag coil) is momentarily shunted, the primary winding (interrogator antenna coil) experiences a momentary voltage drop in response to the increased load offered by the secondary winding. By repeatedly shunting the PIT tag coil in sympathy with the binary data stream representing the data using a transistor, data from the PIT tag can be conveyed to the interrogator antenna coil. The interrogator must be able to detect this change. The data bits can then be encoded or further modulated in a number of ways. In 'direct modulation', the amplitude modulation of the backscatter signal is the only modulation used. A high in the envelope is a '1' and a low is a '0'. Direct modulation can provide a high data rate but low noise immunity. With 'frequency shift keying' (FSK), the amplitude of the backscatter signal is varied at two different rates to convey data. The most common FSK mode uses fc/8 (where fc is energisation carrier frequency) to represent a '0' and fc/10 represents a '1' . The amplitude modulation of the carrier thus switches from fc/8 to fc/10 corresponding to '0''s and '1's in the bit stream. This type of modulation allows for a simple receiver design, provides strong noise immunity, but results in a lower data rate than some forms of modulation. 'Phase shift keying' (PSK), is similar to FSK except only one frequency is used and the shift between '1's and '0's is accomplished by shifting the phase of the backscatter signal by 180deg. in sympathy with the bitstream.
A 'full duplex' (FDX) system receives and transmits simultaneously as opposed to a 'half duplex' (HDX) system that transmits then receives. FDX systems allow maximised data transfer but may be more complex because the receiving system needs to be able to separate the small signal received from the PIT tag (mV's) from the high energisation voltage (100V's). The difference in voltage levels can be in the order of 100dB. A HDX system usually has an average lower data rate, as there is idle time whist the interrogator is transmitting the energisation signal after which the PIT tag responds with data. HDX does have advantages in that energy can be built up in the PIT tag over a period of time and used to transmit its data after excitation has been completed. The resultant receiver circuitry can be far simpler because there is no energisation voltage to filter out and in consequence greater read range can be achieved.
Read range is governed by antenna dimensions (PIT tag and interrogator), energisation power, modulation type, duplex mode of operation and semiconductor fabrication technology utilised on chip (IC efficiency). It is the combination of these factors that will determine the overall read range of the system.
Read time is governed by,
interrogator antenna type (tube, panel), energisation power and by number
of data bits, modulation type and duplex mode of the PIT tag..
The message sent by a PIT tag consists of a sequence of binary bits. To enable the message to be understood by the receiving system, the message needs to have a pre-determined format. As the bit sequence (with a FDX) PIT tag is repeatedly transmitted without breaks, the receiving system needs to know where the start of the message is. It is usual to transmit a fixed pattern of bits, referred to as the 'header' at the start of the sequence. The code structure must be designed such that the header pattern cannot possibly occur anywhere else in the message sequence. Once the header can be identified, ID data bits can follow. Because the number of data bits in a message is fixed, it is unnecessary to have an 'end of message' sequence. To ensure data integrity, it is usual to include error-checking bits in the message. These may be in the form of a checksum or parity bits. Some structures may enable error correction of simple errors. Manufacturers using identical parameters such as frequency, modulation, number of bits etc, need to ensure that the code structure used does not result in misinterpretation of the code by a multi-reading system e.g. a "Company A" PIT tag must not be capable of giving a valid decoded PIT tag ID if read using the decoding algorithm for "Company B" devices. Once decoded, the ID number may be displayed in hexadecimal format e.g. D17FA219B0 or denary 127126034028096
Structure of PIT Tag
An antenna coil consisting of several hundred turns of fine enamelled copper wire is wound onto a ferrite rod. The purpose of the rod is to concentrate the magnetic flux within the antenna coil, thus increasing its efficiency compared with an air-cored coil. The antenna coil connections are bonded onto an integrated circuit and a capacitor is used to bring the antenna coil to resonance at typically 125kHz to 142kHz. Tuning the antenna coil to the energisation frequency, enables maximum energy to be captured from the interrogating device. This same coil is used as an untuned radiator for the signal transmitted by the PIT tag. The IC is attached to the end of the antenna ferrite and connections made between the IC and coil.. With a HDX device, a charge store capacitor is required to build up and store energy during the excitation period. This energy is then used to power the device and transmit its ID code when the energisation field has collapsed. It is the relatively large size of the charge store capacitor that governs the minimum size of HDX devices. A FDX device uses power derived from the excitation field to simultaneuosly transmit its ID code and does not therefore require an energy store capacitor. The assembly is encapsulated within a bio-stable glass envelope, mechanically stabilised by the addition of a small amount of silicon encapsulant.
The purpose of the antenna is to generate an alternating magnetic field from the alternating output voltage from the interrogator and to receive the signal back from the PIT tag. Antennae used in low frequency PIT tag systems are always based on loops of wire. The simplest antenna can consist of a single coil of wire used for both energisation and reception. Some antennae use separate energise and receive coils. When an alternating current is passed through a coil, a magnetic field is produced in which the field strength is at a maximum in the plane of the loop. The magnetic field strength is directly proportional to the current, number of turns and surface area of the loop and decays with distance from the loop 1/r3 (where r is distance from loop). It is this decay that is the main limiting factor of read range.
Simple Swim-Through Coil
The figure above shows best orientation of the PIT tag at various positions relative to the coil. The coil is tuned to the energisation frequency; this is essential in order to maximise energisation current flow through the coil. This may result in an energisation voltage being developed across the coil of.1000V p-p. The same coil is also used to receive the signal back from the PIT tag. Towards the limits of detection, the signal induced into the antenna coil by the transponder may be as small as 1mV p-p. The interrogator receiver circuitry needs to be capable of separating the small, PIT tag signal from the much greater energisation voltage. The greatest read range is achieved with the transponder positioned along the axis of the coil.
As tag orientation in a fish tends to be along its length, the orientation offered to the antenna is close to optimum in most cases. This type of antenna arrangement is suitable for high detection efficiency swim-through systems where maximum antenna diameter is not required and swim speeds are <5ms-1.
Single coil - swim over
When mounted horizontally on the bed of a stream or flume, the best orientation is achieved when a fish passes over the periphery of the coil, with worst orientation when the fish is over the centre of the antenna as the magnetic field will be close to 90 degrees to the PIT tag. Since the detection path length is now very short (c.5cm) the detection speed is limited to <1.25ms-1. As tag detection over the antenna is unpredictable due to orientation effects, this type of arrangement is best suited to sampling fish locations such as in habitat studies rather than fish movements.
Single Coil Solenoid - Swim Through
If a coil is stretched to form a solenoid, the detection path length is extended enabling the detection of fish moving through the antenna at greater speed. To maintain conditions for good antenna efficiency, length tends to be limited to c.30cm and diameter to c.40cm. Since detection is still possible at c.10cm beyond the ends of the solenoid, the total detection path length may be up to 50cm. This will result in a maximum readable fish velocity of 12.5ms-1. This type of arrangement offers high detection efficiency at high swim speeds.
In a single coil antenna, the coil is tuned to facilitate maximum energisation performance. By introducing an additional coil or coils that are tuned to the signal frequency emitted by the PIT tag, it is possible to enhance the detection efficiency. Receive coils can also be positioned to reduce tag orientation effects Most configurations described previously can be implemented using separate receive coils.
The flatbed antenna is mounted horizontally on a stream or flume bed and consists of an outer energisation coil and multiple ferrite-cored receive coils. The ferrite-cored receive coils are positioned close to one edge of the antenna so that the preferred horizontal orientation of the tag over the ferrites correlates closely with the preferred orientation over the periphery of the energisation coil. Similarly, towards the centre of the energisation coil the preferred vertical orientation of the energisation coil correlates with the preferred orientation of the receive ferrites. In this way, the orientation sensitivity for a PIT tagged fish swimming over the antenna is decreased and a detection path length of c.20cm is achieved over the leading edge of the antenna for horizontally orientated PIT tags with a detection distance of c.20cm. This antenna is suitable for efficiently detecting the movement of fish at a distance of <20cm from a stream or flume bed and at swimming speeds <5ms-1.
Interrogator or Reader
PIT tags interrogators can
tend to be categorised as hand-held/stationary and non-multiplexed/multiplexed.
Hand-held interrogators tend to offer limited read range due to battery
and antenna size limitations. Most have internal storage for tag numbers
to enable tagged fish to be read, possibly time/date stamped and data
saved for download to a computer at a later date. This type of interrogator
tends to be used during the tag application process and in studies where
tagged fish are regularly caught, read and released. i.e. handling is
required. Stationary or fixed installations will normally use a mains
power supply and be used for automatic tracking of pre-tagged fish. There
are two basic different modes of operation of stationary interrogator
- non-multiplexed and multiplexed.
Non-Multiplexed FDX Interogator
Most standard interrogators operate non-multiplexed. i.e. each antenna requires a separate interrogator. With this type of arrangement, an antenna can be scanned more or less continually for the presence of a tag. Other processor functions such as scanning for inputs on serial ports or keypads may prevent scanning for tags for 100% of the time, but missed tag events will usually be minimal
The 134kHz oscillator and
amplifier generates a sinusoidal voltage that is then amplified to sufficient
level to drive the antenna. The drive power output from this stage may
be typically c.10w rms. The output from this stage should have minimum
distortion and low noise levels at PIT signal frequencies to maximise
detection efficiency. PIT tag signals picked up by the antenna (c.1mV
to 1V) are filtered to select the band of frequencies of interest and
amplified. The encoded data is demodulated and decoded to extract the
ID number. Parity checks are performed to validate the data. The ID number
may be stored in memory or transmitted to an external device via the serial
A multiplexed interrogator permits several antennae to be physically connected simultaneously. A Time Division Multiplexing (TDM) technique then enables each antenna to be electronically connected in sequence. Each antenna is then polled by the system where an antenna is energised and tag data looked for, for a given period of time before progressing onto the next antenna in the sequence. Due to the high voltages that are developed across an antenna coil and the type of signals that need to be switched, electromechanical relays are often still used as the switching element. A period of c.10ms is typically allowed as sufficient settling time for relay contacts (due to contact bounce) prior to antenna energisation. Relays of this type have a mechanical operating life typically of 5x10^7 operations, resulting in an expected mechanical life of 1.5 years if switched once per second. 200ms is typically required to perform an antenna switch and scan an antenna. A 16-channel multiplexed system would therefore require 3.2s to poll round all 16 antennae. Because each antenna is only scanned for 0.2s in every 3.2s, this type of system is not suitable for applications where rapid fish movement is expected and high detection efficiency is required. It is however suitable for scanning arrays of antennae where fish are moving around slowly or are stationary such as may be the case in habitat studies.