15
\$\begingroup\$

Like many people I'm often annoyed by ugly cable clutter, especially when it's long distance. In my opinion it would be ideal if cables could be eliminated altogether, so that you'd just have two corresponding plugs without the cable, which you plugin at either end and transmit their information wirelessly.

Maybe that would be a little futuristic for now, but what about transporting data over a wireless computer network (or two paired bluetooth transmitters)? That's perfectly possible, and all you'd need is a little box that can convert data to a format that can be transmitted over a wireless network, and a box that converts it back to the appropriate format at the other end. Over time this technology would surely be miniaturized until it fits in a plug. (you could use this for HDMI, audio or other information for example)

Why doesn't this technology exist already? Am I overlooking something, or are there some difficulties with it that have to be solved first?

\$\endgroup\$
  • 5
    \$\begingroup\$ Have you thought about the isolation of signals from each other, how do you expect to pair the ends in such a way as to allow for many different pairs, and yet not just allow anyone into your network? \$\endgroup\$ – Martin Jul 25 '12 at 10:53
  • 1
    \$\begingroup\$ There are commercial solutions already, eg Serial over Bluetooth, CAN over Bluetooth \$\endgroup\$ – Martin Jul 25 '12 at 11:16
  • 1
    \$\begingroup\$ Bandwidth and management are the big issues. Long long ago people like Wang Labs (remember them)(Wang who?) provided wideband systems that basically carried the whole MF frequency spectrum in a cable and used bandpass filters to acess selected parts of the band. Cable TV also offers very wide bandwidth. ... (Wang -> DEC -> Compaq -> HP -> ???) \$\endgroup\$ – Russell McMahon Jul 25 '12 at 12:05
  • 2
    \$\begingroup\$ Every computer tech knows that, when setting up a computer network, WiFi is almost always inferior. The main issue is that only one computer can be transmitting on the same frequency at the same time. If two computers try to transmit at the same time, they stop, and wait a random period of time before attempting to retransmit. This makes wireless extremely slow with as little as two or three active computers, and is why it's almost unusable at coffee shops and hotels. Also, the signal is severely affected by microwaves. \$\endgroup\$ – BlueRaja - Danny Pflughoeft Jul 25 '12 at 16:47
  • 1
    \$\begingroup\$ Essentially, what we are comparing here is whispering (wired) vs shouting (wireless). Choose which one best fits the situation... and choose wisely. Shouting, apparently, is a largely complicated affair and involves a lot of effort. ;-) \$\endgroup\$ – shimofuri Jul 25 '12 at 16:57
24
\$\begingroup\$

Wireless technology is great and can be used in all sorts of scenarios but it's complex and hard to design for. Wires are in fact superior in many ways.

( Taken from: Essentials of Short Range Wireless Standards )

With wires:

  • Range is not an issue - just add more cable
  • Latency is excellent - what goes in one end appears immediately at the other
  • They're transparent to data protocols and formats
  • Throughput is excellent
  • No issue with security - you know what you plug it into
  • Interoperability is excellent, At most, you only need to change the plug
  • Power consumption may be higher, but the cable can carry power
  • They can be specified on a single page
  • Topology is simple - it's typically one-to-one
  • Robustness to interference is generally a minor issue
  • Backwards compatibility is normally no more difficult than changing a plug
  • There's generally no license agreement, no qualification requirements and export controls
| improve this answer | |
\$\endgroup\$
  • 15
    \$\begingroup\$ Actually even with a cable range and latency are both issues at high frequencies - cables start acting like capacitors and that needs to be worked around because what goes at one end starts charging the capacitor (the cable). \$\endgroup\$ – sharptooth Jul 25 '12 at 10:25
  • 4
    \$\begingroup\$ Range is an issue with cables, all cables have transmission limits which is why long distance cables have repeaters. \$\endgroup\$ – user5957 Jul 25 '12 at 14:05
  • 3
    \$\begingroup\$ Both electromagnetic signal in wireless and electric signal in wires travels at the speed of light; neither are instantaneous. And it is just plain misleading to say that cables are transparent to data protocols, signals that are carried in cables have to be encoded in a predetermined manner just like signals in wireless, and both sides do have to agree on a common coding otherwise you'd just get garbage. Saying that any cable can carry any electric signals is just as useful as saying that any air can carry any electromagnetic wave; both are missing the point. \$\endgroup\$ – Lie Ryan Jul 25 '12 at 17:25
  • 2
    \$\begingroup\$ I'm not sure what they meant that backward compatibility by just "changing a plug". You can't use a NIC designed for Gigabit Ethernet to talk to 10 Gigabit Ethernet NIC at the higher speed, by just "changing the plug", just as you can't use router designed for 802.11g to talk to 802.11n at the higher speed. Both schemes allow the higher speed NIC/router to fallback to the slower speed. \$\endgroup\$ – Lie Ryan Jul 25 '12 at 17:42
  • 3
    \$\begingroup\$ Many comments here may not realize that much of what Joby posted are major simplifications of the situation and imply a relation to a wireless protocol. \$\endgroup\$ – Kortuk Jul 26 '12 at 10:44
22
\$\begingroup\$

Wired connections have some properties that wireless connections don't:

  • Robustness: a wireless connection can be subject to various forms of interference (think to microwave ovens) and obstacles, that can affect the quality of the received signal.

  • Latency: wireless connections make a large use of acknowledge signals and error checking codes, due to their lower reliability. This means that you need to wait more for valid data.

  • Parallelization: while you can run 10 cables to transfer 10x signals, with wireless links it's harder because you will have interference and a limited amount of channels. This applies also when you have many networks in the same place.

  • Security: with a wire you (almost) always know where you're sending the signal, and getting into a wired network requires at least to get access to the cable. With wireless connection the thing is much harder, because signals are broadcast in the air. Encryption systems are evolving, but with GPGPUs and parallelized computing it's becoming much quicker to make brute force attacks.

  • Range: you know the range of a wired connection if you have the specification and a long enough cable; it's harder with wireless networks because they are subject to reflection and all the obstacles on the path, and estimating the range is mostly by trial and error.

  • Safety: this is a minor point, but there is research about the effects of RF on the human body

As a side note, there are some applications (think to smartphones) which are only possible with wireless connection. So the odds are that both communication media will survive also in the long term.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ On robustness and latency, much of the complexity of wireless protocols are designed to mitigate the unreliability of the wireless data path: the Two Generals' Problem. en.wikipedia.org/wiki/Two_Generals%27_Problem \$\endgroup\$ – shimofuri Jul 25 '12 at 16:20
  • \$\begingroup\$ @shimofuri right, I've mentioned it in the bullet about latency :) \$\endgroup\$ – clabacchio Jul 25 '12 at 17:32
12
\$\begingroup\$

What you're talking about I use every day: my PC is wireless connected with my router and can transceive data at 300 Mbps. Works fine. Ten years ago I used a CAT-5 cable for that (a bit lower speed back then).

But there's cost. A few meter of wire will probably be cheaper, especially at low speeds, when there aren't that strict requirements for the cable.

Also, a cable has the advantage of kind of isolating the data from the mean world outside. Data is inside the cable, noise stays outside. If you use wireless all kinds of signals may interfere with one another.

| improve this answer | |
\$\endgroup\$
3
\$\begingroup\$

Things are gradually going to wireless but at the moment cables are way better in several ways.

1) Speed

Thunderbolt 10Gbps and USB 3 4.8Gbps vs ~150Mbps for wireless n. Then if you look at things like HDMI and Displayport you have 10.2Gbps and 17.28Gbps wireless just simply can't push data that quick.

2) Interference: things are bad enough now to get your wireless router working nicely through your house. Imagine every TV, game console, keyboard, mouse, external drive etc all trying to find a frequency to chat on. You can try routing everything through your internet router but then it is all on one frequency which has limitations on bandwidth.

3) Ability to transmit power:

Particularly for USB devices you can power them with the cable you use if you don't have a cable for data transmition you'd likely need one to plug it into the wall anyways.

4) Range

For longer distances like corporate campuses or telecom wires have the advantage of having a long range (1,000s of kilometers for fiber optics with repeaters for example) where as the atmosphere and interference will attenuate a wireless signal quickly.

| improve this answer | |
\$\endgroup\$
1
\$\begingroup\$

All the answers given already are great from a design stand point. But there is a political regulatory side to this story.

In the U.S., almost anything wireless is within the jurisdiction of the FCC (not sure about other countries). Typically, this means that to stay in compliance, this adds an extra layer of complexity to design specifications. As a result, it becomes more expensive either by complying, or being fined for not complying if you get caught...

| improve this answer | |
\$\endgroup\$
  • 2
    \$\begingroup\$ I'd call that regulatory, not political. \$\endgroup\$ – stevenvh Jul 25 '12 at 13:55
  • \$\begingroup\$ @stevenvh, unfortunately the entities which should regulate often turn to politics, e.g. LTE spectrum battles. \$\endgroup\$ – whitequark Jul 25 '12 at 14:08
  • \$\begingroup\$ @stevenvh In all fairness, you are correct. I just tend to put the two together because at the end of the day, it's people that sets the regulations. Thx for pointing that out, and I will correct accordingly. \$\endgroup\$ – Chad Harrison Jul 25 '12 at 14:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy