Full duplex or half duplex?

Ok, I'll admit it up front. I never learned RF. Indistiguishable from magic, without even needing smoke, or mirrors.

But after dropping my traditional Ma Bell twisted pair landline, and trying to assess how well its replacment (Ooma) is doing, I've been trying to notice which conversations are effectively half duplex and which are full duplex. And that has led to the more general question of how all RF devices operate in this regard.

My landline was full duplex. Even though both sides of the conversation were on the same twisted pair at the same time, circuits at both ends were able to subtract what they were sending from the combined signal, and end up with the received signal, at the same time. But that landline didn't involve transmitting with RF.

The replacement Ooma VOIP device, when used with a corded phone, seems to be full duplex, but cell phones and cordless phones at the other end often seem to be half duplex.

It seems to me that a device cannot both send and receive at the same time over the same antenna. Well, at least not at anywhere close to the same frequency, and maybe not at all - because the transmitted energy would completely swamp the much weaker received signal. But if that's true, then it would appear that the wireless devices we use all the time are actually half duplex devices. And that would even be true of a cablemodem, which transmits and receives modulated carriers - over coax instead of over the air.

Generally, what would be the actual duplex status of these devices:

Cell phone Cell tower Cordless phone Bluetooth headset Cable modem VOIP phone via a cable modem Wi-Fi

And those half-duplex devices which in practice appear to function as full- duplex - how do they do that?

Reply to
Peabody
Loading thread data ...

Bwuahahahahahaha!

Funny post.

Reply to
FatBytestard

Look up "Hybrid coupler". You might be surprised.

You can get around 40dB of isolation between ports at the same frequency, just using pieces of coax. There's a 3dB penalty, though.

Much more with sophisticated designs.

--
"For a successful technology, reality must take precedence  
over public relations, for nature cannot be fooled." 
                                       (Richard Feynman)
Reply to
Fred Abse

-- "For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." (Richard Feynman)

Reply to
Fred Abse

RF is magic. Mirrors are fine, but smoke is very bad.

Presumably, you have either a DSL or cable internet connection. If it's a satellite internet connection, you will hear about a 1 second delay as the signal goes up and down to the satellite. This may sound like half duplex, but it's really full duplex. It would be helpful if you disclose how your Ooma box connects to the internet.

I already answered that. RF is magic. For those that believe in RF, no explanation is necessary. For those that fail to put their faith in RF, no explanation is possible.

Actually, it did if you don't care whether the RF is going over a metallic conductor, through an optical fiber, or between antennas. However, for the purposes of answering your question, we'll assume that no RF or magic was involved in a POTS (plain old telephone service) line.

Cell phones and cordless phones come in two flavors. The full duplex variety use two different frequencies at the same time, also known as FDM (frequency division multiplex). One frequency for transmit and one for receive. Both can operate simultaneously.

There is also a TDD (time division duplex) which works with one frequency. The audio is digitized and compressed, which allows the radios to switch direction rapidly, simulating a full duplex connection. This is rather uncommon.

Actually, it's quite easy and common using either a diplexer or duplexer. This is a filter that separates the two frequencies so that the transmitter doesn't overload the receiver, and so that they can coexist on the same antenna. There are other devices, such as isolators and circulators involved.

Well, that's very logical, but also not very correct. It really depends on the modulation. For example, if you compress the audio going in both directions, you gain additional time during which to rapidly switch the direction of the transmission as in TDD.

Far more complexicated is MIMO (multiple-input and multiple-output), which transits two or more different signals, on the same frequency. In spatial multiplexing, the various signals can be received separately, recombined into a single data stream, and produce much faster throughput.

There are three possible methods.

- FDM also known as full duplex, where the transmitter and receiver are on different frequencies and operate simultaneously in both directions.

- TDD, where fast direction switching on a single frequency allows simulated full duplex, with a small added delay. This requires audio compression.

- Half-Duplex, which uses two frequencies, and really means that the transmitter in one direction is on full time, while the transmitter in the other direction is keyed on as required.

Voice: full duplex FDM Data: full duplex FDM and simulated full duplex using TDD methods.

Same as cell phone.

Just about every flavor possible. Worst case is DECT 6.0, which can be FDMA at the RF layer and simultaneously TDD to allow for multiple users on the same RF channel.

BlueGoof and Wi-Fi are both TDD. Transmission occurs only in one direction at a time.

FDM. The data you receive arrives on one channel. The return path is via a different channel.

VoIP generally means SIP (session initialization protocol) which requires either a full duplex (FDM) or simulated full duplex (TDD) data path. From your perspective, it's full duplex, even though the service provider may switch to half duplex audio to improve audio quality.

TDD. Same as BlueGoof.

Magic... Let's pretend you have a single channel data path, that can pass data at say 4Mbits/sec but only in one direction at a time. If I switch the direction with a 50% duty cycle every 1/2 second, I can send 4Mbits/sec through the pipe, but only half the time. The resultant throughput looks like a 4Mbit/sec burst, every 1/2 second for an average throughput of 2Mbits/sec. If I now compress the data at each end by at least 50%, and uncompress it at the destination end, the 2Mbit/sec bursts, will expand into 4Mbits/sec continuous data. If you didn't know what magic was going on with the compression and direction switching, you would think it was a full duplex 4Mbit/sec data path. That's TDD.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558
Reply to
Jeff Liebermann

Most modern communications things digitize voice signals, compress them a whole lot, and occasionally shoot out a data packet. The receiver uncompresses the data and drives a DAC or equivalent. Many such systems are HDX but can sound full duplex.

--

John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    

Precision electronic instrumentation 
Picosecond-resolution Digital Delay and Pulse generators 
Custom timing and laser controllers 
Photonics and fiberoptic TTL data links 
VME  analog, thermocouple, LVDT, synchro, tachometer 
Multichannel arbitrary waveform generators
Reply to
John Larkin

At my PPoE the major product used ten slots for five full-duplex channels over a single frequency (at a time, it was frequency-hopping spread-spectrum in the ISM bands). The base mixed and assembled all of the packets for each recipient. It sounded full duplex but the digital data was all half-duplex.

Reply to
krw

So at least I was right about that.

I appreciate your response, and will study it to see if some of the magic will rub off on me.

Reply to
Peabody

trying

notice

duplex.

operate

were

able to

with the

to be

seem

time

then

actually

the

full-

For the devices mentioned they are mostly half duplex, but due to data compression and time division (or code division) multiple access they act full duplex (you can get into some interesting issues with quality of service (QoS) and traffic priorities versus dropped packets and voice quality). The actual data transmission rates on these devices are much higher than is required for voice (approximately 9600 baud for decent voice fidelity) and most of these is in the high 100 kilobit/s range to

11 Mbit/s (or even more). The cable modem in particular uses some variants of Frequency division multiplexing as well, with the back channels being in a different frequency region than the incoming data (including the TV channels).

This is just a general description, others here can fill in details better than myself usually. Also many "speaker phones" are internally half duplex to avoid having to compensate for room acoustics.

?-)

Reply to
josephkk

That would be worth the effort. Jeff is really spot on at this level. 'E wisely left out many more details and variations.

?-)

Reply to
josephkk

I hate to split hairs, but they're not half-duplex. Rather, they're TDD (time division duplex). Half-duplex is exactly what the name implies. One end of the link is full-duplex, while the other end is essentially simplex. The full-duplex end can transmit and receive at the same time, while the other end cannot. An example of a real half-duplex system is a conventional land mobile or public safety repeater system. Since cell phone users want their phones to operate exactly like a POTS land line, they are either full-duplex (as in the old AMPS cell phones), or TDD as in most every other cell phone.

Much less than 9.6Kbits/sec. EVRC-B is the codec de jure for CDMA. It's a variable rate encoder that can do 8, 4, and 0.8Kbits/sec. The average rate for speech is about 5 to 6Kbits/sec. However, you did qualify your 9600 baud as the minimum for "decent voice fidelity". Should that ever happen in cellular telephony, the service providers will immediately reduce the bit rate until the quality deteriorates to the lowest possible level of usability, in order to cram in more users per channel. The quest for a better codec is not based on quality, but on density.

Incidentally, the open source CODEC2 works at 1.2Kbits/sec and is usable over an HF SSB radio. I keep wanting to try it but can't find the time: I expect to see it added to VoIP software, SIP phones, and possibly cell phones. Try the sound clips at: for what your next cell phone might sound like. I'm having more difficulties with the Australian accent than the speech quality.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558
Reply to
Jeff Liebermann

act

This is rather different from the traditional definition. e.g. Ham radio, CB, and FRS radio is half duplex, each side can transmit or receive but not both at the same time. Thus two way connection, but only one direction at a time.

See:

formatting link
formatting link
formatting link
formatting link

Digital cell phones can be described as TDD full duplex, old AMPS cell phones were (are) FD full duplex.

11

variants

being

TV

Cool beans. At 1200 baud codec2 voice quality is showing noticeable deterioration, though not so noticeable at a mere 1400 baud. All other codecs except g.729 show various artifacts, though LPC-10 ain't half bad.

?-)

Reply to
josephkk

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.