The transponder works in PCM (pulse code modulation): how does it work in the transponder? I can find explanation of what PCM is, but not if the signal sent from the transponder antenna is digital or analog, and I do not understand which are the advantages of using PCM.
Type of modulation in term of ITU classification
In the context of radiocommunication, PCM is more accurately described as ITU class A1D:
- As bits are transmitted one by one, the carrier is sent or interrupted, according to the bit value.
- Phase and frequency are maintained constant.
It's like morse at high speed. It can also be seen as amplitude modulation (either the carrier has its maximum amplitude, or a null amplitude).
This form of modulation is also commonly known as binary amplitude shift keying (BASK) or on-off keying (OOK), and is a variant of pulse amplitude modulation (PAM). See this page for more details. Data communication, on the borders of multiple engineering disciplines is quite generous with synonyms.
There is a common confusion between A1D modulation and the PCM codec used in the context of audio files. This PCM is a near-lossless codec (coder-decoder) used in Compact Disc and legacy .wav files, and has nothing to do with a modulation.
But anyway... this produces this oscillogram (bottom: bits to transmit, top: signal transmitted by the antenna):
A binary 1 pulse generates a burst of the carrier wave.
A1D is often used to transmit information at low speed, because it's not very efficient in term of bandwidth, and this bandwidth increases with the data transmission rate. Sending the 12 bits of a transponder every now and then was okay in the past (where the 12 bits come from? see this answer). A1D is also used by GPS satellites which are actually slow data broadcasters.
Sort of signal: Analog vs digital
At the antenna the signal is the modulated carrier, and it's always an analog signal, regardless of the nature of information to carry. The carrier is generally a high frequency sine wave.
The carrier is modulated by the twelve bits of mode A and C information, which are digital (binary): Each bit, one after the other, modulates the carrier, that is the carrier is interrupted or transmitted according to the bit value, for a given period of time.
Modulation means changing something on the carrier to "imprint" the information into it, so that it can be demodulated at the receiver side. The "something" can be one (or more) of the three properties of a wave: Amplitude, phase or frequency. In A1D, this is the amplitude (as the "A" indicates).
The modulated signal at the antenna will still be an analog sine wave, so no strong interference will occur (though A1D is not the most friendly class of modulation in term of spectrum use).
But, why do we need a carrier in the first place? (baseband vs passband)
Transmission of the raw information without modulation is said a baseband transmission. This is used for instance in Ethernet over twisted pair (e.g. 100base-T, where base just means baseband).
The signal is ranging from 0 Hz to 100 MHz (assuming a random distribution of the data). Therefore if there were two channels at the same time on the wire, a receiver would receive data without being able to identify the channel (not mentioning the fact that each channel would destroy information of the other channel).
The only way to share the medium is to limit the transmission to a single channel at a time, and this is what is done in Ethernet (by collision detection and avoidance). However collision detection doesn't work when there is an unlimited number of transmitters, collision avoidance would leave no time for a useful transmission.
The other way is to add some fixed frequency to the second channel, e.g. by adding 100 MHz, the second channel frequency will now range from 100 MHz to 200 MHz, and a receiver can easily separate each channel (by frequency filtering) and receive both.
The second channel data will then be obtained by subtracting 100 MHz from the signal. This principle is just the one we know for radio, with a carrier (the fixed frequency the receiver is tuned for) and a modulation method. Technically this is called "passband" transmission.
The last step: Demodulation
The receiver receives the modulated carrier, usually very weak, and often distorted. The demodulation step is to extract the original information from the carrier by looking at the carrier property that was changed to carry the information.
For A1D modulation this is indeed immediate, carrier presence = bit of value 1, no carrier = bit of value 0 (this can be inverted if required, just a matter of agreement between the transmitter and the receiver).
Now you can look again at the previous oscillogram and think the top describes the signal received by the receiver antenna, and the bottom is the original information extracted by the demodulator stage. The binary signal received is not perfect, but it will be discriminated as 0 or 1 depending on whether the value is smaller or larger than a given threshold (so some errors can occur, there are solutions to prevent that).
The carrier is then discarded. Its role was only to carry the payload.
There are two kinds of transponders in use. Air Traffic Control Radio Beacon System (ATCRBS - pronounced at-crabs) is the older system that uses Mode A for ATC code and Mode C for altitude reporting. The newer Mode Select (Mode S) was introduced to support TCAS. It supports all three modes A, C, and S.
Mode A and C are pulse amplitude modulation (PAM). Both modes provide a 12 bit reply word. All pulses are transmitted between 2 framing pulses that are 20.3 micro-seconds apart and are nominally 0.45 micro-seconds long. Each "bit" has a location at a fixed time point after the first framing pulse. The presence of a pulse is a "1" and the lack of a pulse is a "0". There's actually room for a 13th bit, but it not currently used.
Mode A encodes the assigned ATC code - 4 numbers of 0 to 7. The 0 to 7 range aligns with 3 binary bits octal encoding. Thus 3 bits represent each of the 4 numbers - 12 bits total.
Mode C uses the 12 bits to encode the altitude in hundreds of feet using Gillham encoding (Gray code).
Mode S replies have much more data. They have a preamble of 4 pulses. Following the preamble there are either 56 or 112 bits of data encoded with pulse position modulation (PPM). In this coding, each 0.45 usec pulse is placed in the first or second half of a 1 usec slot. If the pulse is in the first half of the slot, it is a "1". If it's in the second half, it's a "0".
Mode S replies are quite numerous and contain multiple formats and data fields.
The transponder is sending digital data (baro, id, sq, etc.) and they are encoded into the reply of the transponder. The modulation of the digital signal to RF in Mode S into a block 56 bit or 112 bit block utilizing pulse position modulation (PPM). Also the data is aggregated with a CRC to assure data integrity of the fields.
The strategies for Mode A and C transponders are notably different from Mode S. Mode S is a datalink type system, and is designed to specifically address an aircraft. Mode A and C are not. Commonality however, includes the use of 1030 and 1090 MHz bands.
PCM is used to efficiently code data minimizing digital "bits" and pulse position modulation (PPM) is used to modulate the digital data into an RF signal to transport the digital data to a receiver. Strategies not utilizing PCM may take more data to accomplish the same effective data rates.
The digital data on that signal is a combination of the data fields for that transponder strategy, which is packaged and CRC added for error detection and tolerance. The digital data is PPM modulated for transmission via an 1090 MHz reply.