Part 8 - SLIP Versus PPP

Serial Line Internet Protocol (SLIP) is an older protocol that was used to connect computers over phone or serial lines.  If you recall, the discussion on serial and parallel transmission, you will remember that computers have various serial ports.  Well, two computers can be connected through their serial ports for simple serial communication.  This serial connection can also occur over greater distance through a modem.  When two computers link through modems, this creates the same type of serial connection as if the computers were connected through their serial ports [1].

A serial connection exists on the Physical Layer of the OSI model.  As such there is no way to arrange the data for transmission.  So engineers created SLIP to help arrange the data for serial transmission.  This was an informal protocol that for a while became the de facto standard for serial transmission, but it was never adopted as a true TCP/IP protocol.  In fact, when the Internet Engineering Task Force (IETF) finally described it in RFC 1055 in 1988, they made it perfectly clear that SLIP was "nonstandard" [1].  The protocol only provides Layer 2 framing when needed and does so in a very simple manner, so it was deemed that there was not enough going on to standardize it.

The way that SLIP frames data is that it simply appends a character to the end of each datagram to define its boundaries.  It does not do much more than that, therefore error correction is not supported in SLIP [2].  Even before the IETF had issued RFC 1055, engineers were already working on a better protocol for directly connected computers and eventually came up with Point-to-Point Protocol (PPP), so SLIP is not used much anymore [1].

Point-to-Point Protocol provides services on both the Physical and Data Link Layer of the OSI model, which allows a modem to act as a network interface controller (NIC).  Using several sub-protocols, PPP provides full Layer 2 connectivity through physically linked devices.  These sub-protocols include, IP, IPX, and NetBEUI.  Error checking and data compression are also supported making PPP faster and more reliable than SLIP [2].  The Point-to-Point protocol is pretty complex and I am not going to delve too deep into it for the sake of simplicity here, but if you are interested in learning more about it please check out the links.  Suffice it to say that even though it is much more complex than SLIP, PPP has replaced SLIP for TCP/IP physical connections because of the much more robust service that it provides [2].

This concludes Part 8 of the the Data Transmission and Protocol series.  In Part 9 I will clarify the difference between Hypertext Transfer Protocol (HTTP) and File Transfer Protocol (FTP).  I know some of you are only a bit familiar with these two protocols.  Anyone who has ever typed in a URL manually will at least have seen HTTP in the address.  In Part 9 I will explain what it is and how it is used and give FTP the same considerations.

[1] Kozierok, C. M. (2005). The TCP/IP Guide. Retrieved from http://bit.ly/qMaZ62

[2] What is SLIP/PPP?. (n.d.). Retrieved from http://sunsite.nus.edu.sg/pub/slip-ppp/whatis.html (no longer active)

Part 7 - TCP Versus IP

In my last post I mentioned that there is a difference between TCP/IP, TCP, and IP.  The difference is this: TCP and IP are Internet protocols and TCP/IP is a suite of Internet protocols.  The TCP/IP suite includes not just the TCP and IP protocols, but also, dozens of other protocols such as, HTTP, POP, SMTP, and ISDN to name a few.  I'm not going to elaborate on the meaning of those acronyms nor am I going to explain all of the protocols in the Internet Protocol Suite in this series.  I will only be going over six of these starting with TCP and IP.  So what are they?

Transmissions Control Protocol (TCP) and Internet Protocol (IP) were created by the Department of Defense(DoD) during the infancy of the Internet [1].  In 1983 the United States military invaded the island of Grenada in an operation called "Operation Urgent Fury," which was an attempt to protect and rescue US medical students who lived there.  However, the military had a hard time finding the students due to lack of intelligence [2].  This lack of intelligence was caused in part by the inability of the different branches (Army, Navy, and Air Force) computer systems to communicate over the communications system (ARPANET) that the Defense Advanced Research Projects Agency (DARPA) had created and been using since the late 1960s [3].

The problem was that by law, defense contracts have to go to the lowest bidder.  When each branch launched bids to build their systems, they did so separately, so three different companies won three different bids to build systems; Digital Equipment Corporation won the Army's bid, International Business Machines won the Air Force's bid, and Unisys won the Navy's bid.  After the systems were built, each branch had an effective local area network (LAN), however, the LAN had trouble communicating to say the least, because there were no standards protocols set to allow them to.  Each network operated independently, which led to the trouble with the joint military operation in Grenada [1].

This led to the DoD setting out to create protocols that would allow these independent networks to communicate as if they were a part of one big LAN.  They wanted to create a language that would allow multiple LAN to communicate with one another creating a wide area network (WAN).  Not only did this language make things like file transfers and email possible between branches, it provided a fail safe so that if any one LAN or phone line went down or was destroyed in battle, the WAN could still communicate without having to be rebuilt.  Essentially, the WAN would self-recover [1].

Now with the history lesson out of the way, here's what each protocol does.  On LAN, there are various protocols used to assign addresses to individual computers depending on what vendor is used to build the network.  These protocols are generally not compatible with one another.  In other words, LAN using Apple computers use Appletalk and LAN using Novell machines use IPX to communicate, but these two LAN could not communicate without a common protocol.  Internet Protocol (IP) is that protocol.  Internet Protocol is the main protocol used in Internet communications. Unique 4-byte (32-bit) addresses (with each byte having a value of 0-254) are provided for all computers on the network using IP and when I say unique, I mean unique.  The intent of IP is to assign every computer in the world that is connected to the Internet a different IP address [1].  That is how the cops can find you if you download that illegal copy of Photoshop (I kid of course. I know none of you would do anything illegal).  This is done to ensure that the data is being sent to the right computer.

"Wait!  Every computer in the world? Surely there are more computers in the world than can be provided for with a 4-byte address."  Well, each byte has a value of 0-254 or 255 possible combinations for just the one byte.  Since we are combining four bytes that makes 255⁴ or 4,228,250,625 unique addresses.  Although some of these addresses are reserved, 4.2 billion is a lot of computers.  Despite that, IP addresses are in fact running out [4].  The IP we have been using is referred to as "IPv4" or Internet Protocol version 4, and most of us do have a IPv4 address, but since around 2005 or so IPv6 has been rolling out all over the world.  IPv6 provides 128-bit addresses [5].  Since as we know, a bit can only be a one or a zero, that would make 2¹²⁸ or 3.4x10³⁸ different combinations.

But I digress. We were discussing how IP works.  In addition to assigning each computer a unique identifying address, transmitted data is organized into packets with a header that contains source and destination IP addresses as well as other information about the data. This helps TCP do its job which we will get to in a minute.  Internet Protocol operates on Layer 3 of the Open Systems Interconnection (OSI) model, which allows it to function on different mediums such as Ethernet or Wi-Fi [6].  I do not want to delve into a discussion of the OSI model as it is beyond the scope of this series, but just so you know, Layer 3 is the the level on which LAN communication and routing occurs.  In other words, we just have LAN only transmissions and preparation for WAN transmission occurring on this Layer, no interpretation or application of data goes on [7].

Transmissions Control Protocol (TCP) provides guaranteed client/server transmission by supporting timeouts and re-transmissions when data is lost or errors occur. It is used when applications require guaranteed delivery of data. This protocol uses full-duplex transmission and each client is identified with an IP address and a TCP port number. Transmissions Control Protocol also provides basic data transfer, reliability, flow control, and multiplexing [6].  Essentially, TCP is the protocol that ensures that the illegal copy of Photoshop that you are downloading gets to your computer intact with no bits of the file missing. That way the cops have evidence that you did in fact download an illegal copy of Photoshop.  Transmissions Control Protocol operates on Layer 4 of the OSI model also called the Transport layer.  This is the Layer that is responsible for transmission and error checking [7].

This concludes Part 7 of the the Data Transmission and Protocol series.  In Part 8 I will explain Serial Line Internet Protocol and Point-To-Point Protocol.  After reading Part 8 you will understand why PPP is so widely used while SLIP is barely used at all.

[1] Gilbert, H. (1995). Introduction to TCP/IP. Retrieved from http://bit.ly/pSChlt

[2] Military Advantage. (2011). Invasion of Granada: Operation Urgent Fury. Retrieved from http://bit.ly/nNfhaA

[3] Howe, W. (2010). A brief history of the Internet. Retrieved from http://bit.ly/qUfkcb

[4] Huston, G. (2011). IPv4 Address Report. Retrieved from August 4, 2011 http://bit.ly/n4OZ78

[5] Das, K. (2008). IPv6 deployment around the world. Retrieved from http://bit.ly/oIH9lK

[6] Mitchell, B. (2010). IP - Internet Protocol. Retrieved from http://bit.ly/qTodlE

[7] Tomsho, G., Tittel, E., & Johnson, G. (2004). Guide to networking essentials. Thomson Course Technology: Boston, MA. ISBN-13: 978-1-4188-3718-1

Part 6 - Serial and Parallel Transmission

Serial and parallel transmissions are another way that computers communicate and are usually used in peripheral communication.  With serial communication bits are sent one at a time over the medium, but since processors handle data in parallel, communication controllers are used for conversion.  This conversion is done using a shift register.  The way a shift register works is it takes bits in parallel chunks and with the use of a clock sends each bit out one at a time to the serial device.  The same is done in reverse with serial data coming back from the device.  The shift register takes one bit at a time until it fills up the register then sends the parallel chunk to the processor [1].  Serial communication is used between your computer and monitor.  That somewhat rectangular port with nine pins that you plug your monitor into on the back of your computer is a serial port.  Another type of serial port you might use are the small round ones with six pins that you plug your keyboard and mouse into.  These are called PS/2 ports.  However, some of you might use a USB mouse, but guess what?  That's serial communication too.  USB stands for Universal Serial Bus.  Anything you plug into a USB port is using serial communication.

Parallel transmissions are used to communicate with devices capable of parallel communication.  With this mode of transmission a communication controller is not needed.  Instead, the processor transmits bits simultaneously through a parallel medium like a ribbon cable.  This mode of transmission is obviously faster, but it is subject to interference because the wires run in parallel and signal leakage (crosstalk) can cause corruption of the data especially at higher speeds [1].  If you are as old as me you might remember the old serial printers with the big fat ribbon cable.  These aren't used too much any more as USB printers have become more common, but you probably still have a serial port on the back of your computer.  It's that long, somewhat rectangular one with 25 pins.  Aside from a few specialized devices, printers are about the only things that use that port, but that does not mean that your computer doesn't use parallel communication if don't have anything plugged into that port.  Inside your computer you may find at least one parallel device; your hard drive or CD/DVD drive.  Many internal hard drives use a 40-pin Parallel Advanced Technology Attachment (PATA) connection.  Some optical drives also uses PATA connections, but I think Serial ATA (SATA) connections are becoming the standard for these devices.  That said, there are still a lot of older computers out there using the PATA drives.  I see them all the time.  In fact, someone reading this right now probably has one in the computer they are reading it on.

There is also a form of parallel transmission that is done through a single line.  This is similar to broadband in that the line’s bandwidth is divide into channels of differing frequency and each bit is sent down a channel [1].  This type of parallel transmission is often used in fiber optics, which is a huge topic on its own, so I will leave that for discussion another day.

To control data flow in serial or parallel communication Xon/Xoff protocol is used.  To illustrate how Xon/Xoff works, computer/printer communication can be used as an example.  When a computer sends data to a printer, the data often comes faster than the printer can print it.  Therefore, most printers have a small buffer to store incoming data until it can be printed.  When this buffer fills up, the printer sends an Xoff signal to the computer which tells it to stop transmitting data.  When the printer catches up and clears some of the buffer it sends an Xon signal telling the computer to resume sending data.  Modems may also use Xon/Xoff in a similar fashion when communicating with a computer’s operating system ("Xon/Xoff", 2010).

This concludes Part 6 of the the Data Transmission and Protocol series.  Part 7 begins the protocol section of the series.  It seems fitting that the first set of protocols I will discuss were be the earliest protocols defined for Internet transmission; Transmissions Control Protocol (TCP) and Internet Protocol (IP).  I have run into some that say, "Don't you mean TCP/IP?"  My answer?  "No."  TCP, IP, and TCP/IP are three completely different things.  If you don't understand, you will after reading Part 7.

[1] Kioskea. (2008). Data transmission - Transmission modes. Retrieved from http://bit.ly/pu9GeN

[2] Xon/Xoff. (2010). In Computer, Telephony & Electronics Industry Glossary. Retrieved from http://bit.ly/nTRyCj

Part 5 - Baseband and Broadband Transmission

Baseband and broadband are two ways to send signals over cables.  Baseband transmission is most often used in networking transmission.  Baseband takes data, encodes it digitally, and sends the digital signal at a fixed frequency over the medium as electrical or light pulses.  The signal uses the entire bandwidth of the medium so is automatically half-duplex unless a second line is used.  Repeaters are often used in baseband transmissions to strengthen signals and clean up signal degradation over longer lengths of line [1].

"Wait I thought you said, 'digital signals were not subject to signal degradation'?"  This is true as long as the medium is also digital, like a CD.  However, notice that the digital signal is sent "at a fixed frequency."  This indicates an analog medium.  Think of it like this: One bit of data is either a one or a zero.  To signal a one let's say that the pulse has to be between 3-5 kHz and anything lower is a zero.  That one bit alone is an analog signal.  However, all the bits as a whole is a digital signal.  As this bit travels over the medium is is being subjected to degradation of the frequency.  The further it travels, the more it degrades to the point where it will eventually be interpreted as a zero instead of a one.  Each bit is subjected to the same degradation.  Some might experience less, some might experience more, but all degrade which is why repeaters are used over longer lines.  It is also worth mentioning that repeaters are also used for wireless signals as well and they work in much the same way.  Before things start getting confused, let me point something out.  Data that is sent out from your computer is always digital and the medium that it travels on is almost always analog.

Now, with broadband transmission, data is transmitted as a continuous electromagnetic or optical wave (see an analog wave instead of a digital pulse) [1].  This is a form of analog signaling and as mentioned in Part 3, can be divided into channels, over which different data can be transmitted.  This is called multiplexing, or more specifically, frequency division multiplexing (FDM).  Briefly, FDM divides the medium into two or more channels of different frequencies.  These channels are combined into an aggregate frequency that can be sent over the medium.  Each channel is separated from one another by an unused frequency so that data from one frequency does not bleed into the one next to it.  This is called "crosstalk"  and the unused frequencies are called "guard bands" [2].  Falling back to the busy street example you could think of the guard bands as concrete barriers between the lanes.  They keep all cars (data) in their proper lane (channel).

Now, the receivers must have some means to dial into the right channel, like the cable box you have for your TV and the cable modem you use for your computer.  Cable companies use broadband to deliver hundreds of channels and high-speed Internet to customers.  Broadband transmissions are also half-duplex. However, two methods are used to create full-duplex broadband.  One is called mid-split broadband which splits bandwidth into two groups of channels, one for sending and one for receiving.  The other method is called dual-cable broadband, which just as its name implies, uses a transmission cable and receiver cable.  Similarly to baseband, broadband has a way to improve transmission distances.  Amplifiers detect and amplify weak signals then rebroadcast then to the next section of line [1].

Since, baseband signals are of very low frequencies, they are not good for transmitting over large distances  because degradation is much more pronounced in low frequency waves.  However, baseband is very efficient over short distances and is therefore commonly used in local area network (LAN) communication.  In addition, baseband does not have to be modulated into an analog signal, so less equipment is needed to communicate between devices.  The Ethernet cable that you have going from your modem to your computer or router is a baseband line.  Most office networks are set up on a baseband network with a broadband connection to the Internet.

Broadband uses much higher frequencies and therefore is much better for transmitting over long distances.  Wide area networks (WAN) use broadband to communicate.  In fact, the whole Internet is really just one big WAN.  Broadband is generally faster than baseband, but only because of the multiple signals factor.  The data is really traveling in general at about the same rate.  It's just that more signals can be passed through broadband and therefore, the resulting data transfer rate is higher.  However, broadband is more expensive to implement due to the equipment involved and cost of service.  Since broadband travels over cables that are buried underground and are usually owned by another entity, you have to pay to use them. Since baseband does not require much special equipment and because you can run the lines above ground,  it is much less expensive to implement [3], not to mention that you own the lines.

This concludes the Part 5 of the the Data Transmission and Protocol series.  In Part 6 I will go over the final set of transmission methods, serial and parallel.  These transmission methods are generally used in peripheral communication.  How does your computer communicate with your monitor?  How about the printer?  You will know after reading Part 6.  Then in Part 7 I will begin contrasting and explaining protocols starting with Transmission Control Protocol and Internet Protocol.

[1] Tomsho, G., Tittel, E., & Johnson, G. (2004). Guide to networking essentials. Thomson Course Technology: Boston, MA. ISBN-13: 978-1-4188-3718-1

[2] General Telecom.  (1998). Multiplexing Techniques. Retrieved from http://bit.ly/mTPngG
[3] James, E. (2010). Broadband. Retrieved from http://bit.ly/oZRc2t

Part 4 - Simplex and Duplex Transmission

In any transmission, data can flow in one of three ways, simplex, half-duplex, and full-duplex.  When a simplex transmission is referred to this means data only flows in one direction over the medium [1].  A simple example of simplex communication is that which occurs between a mouse and the computer.  The mouse has no need for any information from the computer, so it can just constantly transmit data to the computer on its movement and button actions, which the computer can translate and convert into cursor movements and other functions.

In half-duplex transmissions, data can flow in both directions over the medium, but only in one direction at a time [1].  A classic example of half-duplex communication is two-way radios.  When using a two-way radio, only one person can speak at a time.  If signals are sent simultaneously, they collide and neither receiver gets the information.  Half-duplex transmissions are usually synchronous [2] because in order for the information to be exchanged reliably, the transmissions must be timed and in sync.  Half-duplex transmission also use the full bandwidth of the medium [3].

Finally, when transmissions are full-duplex, this means that information can be sent in both directions simultaneously [1].  Telephone communication is the most common example of full-duplex communication.  Full-duplex transmissions are usually asynchronous [2].  Since data can flow in both directions there is less need to time and synchronize the transmission, data can be sent and received "on-demand".  In addition, with full-duplex transmissions bandwidth is divided so that data can flow in both directions over the same medium [3].

Regarding bandwidth . . . In general, full-duplex transmissions are going to require more than twice the bandwidth of simplex and half-duplex transmission.  I say in general because there are cases where transmissions require very little bandwidth in the first place, so full-duplex transmissions experience very little hit in performance in these cases.  However, in cases where large amounts of data are being sent in both directions, low bandwidth mediums are going to prove inadequate.  I fall back to online gaming as being a good example of this, which is why most popular online games require a broadband connection to play, although they may not require a full-duplex connection.  You will understand this more after reading Part 5 on Baseband and Broadband Transmission. 

Think of it like this: Let's say we have a four-lane, one-way street in a busy city.  All four lanes are consistently full and travelling in one direction.  This is how simplex transmission works.  The four lanes are your bandwidth and the cars are your data.  

Now take that same four-lane street and make a law that says that every other day the traffic can travel in the other direction.  This is a physical example of half-duplex transmission.  Again, the lanes are your bandwidth, cars are your data, but this time we have a law, which represents a protocol (more on protocols in future posts).  Notice what happened though.  Even though our cars can still use the full four lanes available to travel in one direction, the amount of traffic that can can travel in one direction has been reduced by the law.  Likewise, half-duplex is going to reduce the amount of data that can travel over the medium when compared to simplex communication.  So even though the same amount of data is being transferred over all, only half the data is being transferred in either direction.

Now again, let's take that same four-lane street and say, "okay we are going to make this a two-way street."  Now cars can travel either way at the same time and we are going to discard the old law and create a new one that says, "cars must stay in the right two lanes relative to the direction they are travelling" to avoid collisions.  This is a physical example of full-duplex transmission.  Things  all remain the same, except the type of transmission and the protocol have changed.  Now notice how the traffic or amount of data being exchanged has changed.  Essentially the same amount of data is being transferred in either direction, i.e in a two-day period, four lanes in one direction with a one day wait = two lanes in one direction with no wait.  However, there are more things to consider.  Even though we are transferring at the same rate, look at the congestion we have created by making the street two-way.  What was a busy, but efficient one-way-only or one-way-every-other-day street, is now a clogged-up two-way street.  And even though we have a new protocol intended to prevent collisions, due to congestion, collisions are highly likely to occur as traffic tries to get to where it is supposed to be going.  Therefore, even though half and full-duplex can send at the same rate over a period of time, the rate at any given instant is at least half and maybe even more due to congestion.  So in situations where you will be switching to full-duplex transmission, bandwidth is definitely plays a factor.

This concludes the Part 4 of the the Data Transmission and Protocol series.  I spoke a bit about bandwidth in this part, so in Part 5 I be discussing baseband and broadband transmissions.  Baseband and broadband are two sides of the same coin and often can dictate the way in which you set up a network.  After reading Part 5 you will know the difference between the two as well as their capabilities and limitations.

[1] Tomsho, G., Tittel, E., & Johnson, G. (2004). Guide to networking essentials. Thomson Course Technology: Boston, MA. ISBN-13: 978-1-4188-3718-1
[2] InetDaemon Enterprises. (2010). Asynchronous vs. synchronous. Retrieved from http://bit.ly/pOhOl8
[3] Kioskea. (2008). Data transmission - Transmission modes. Retrieved from http://bit.ly/pu9GeN

Part 3 - Analog and Digital Signals

Telecommunication, whether it is data, voice, or some other type of transmission, occurs over two types of signals. The earliest transmission signaling is analog. Analog technology takes the data (or audio, video, etc.) and converts it into electronic pulses or radio waves depending on the medium over which the information is travelling.  On the contrary, digital technology takes information and converts it into binary data. Binary data consists of just zeroes and ones so much more data can be crammed into a digital signal than an analog signal [1].

In the early years of computing, there was no way to transmit a digital signal directly over an analog medium. Therefore, engineers designed the MODEM which takes a digital signal from a computer and converts it to an analog signal (modulates; “MO”) that can be sent over the phone lines. The modem on the receiving end takes the analog signal and converts it back to a digital signal (demodulates; “DEM”) that the computer can understand [1]. Although analog transmission is still used, many devices are switching to digital signals; telephones and television being two examples. However, analog is still important in modern telecommunications. In fact, anytime frequency is mentioned in relating to data transmission, such as 2.4 GHz wireless signal, analog signals are being used.

The easiest way to illustrate the difference between digital and analog signals is by using sound as an example.  When you hold a note with your voice, you produce a sound wave that can be recorded.  When you record that sound wave on an analog medium, such as a cassette tape, the wave is duplicated exactly as it is.  That is to say that every point in that sound wave is captured and recorded to an infinite degree, so when played back the wave is reproduced exactly as it was created.  This creates a richness in sound that has audiophiles pining for the days of vinyl records.

Now if you were to do the same thing and record the sound wave digitally, the wave has to be turned into a series of ones and zeros called bits.  In order to do this, the recording device has to take a number of quick snapshots of the wave with each snapshot representing a one or a zero.  This is called a sampling rate.  Since we are taking a certain number of snapshots at timed intervals, some of the sound wave is lost.  The higher your sampling rate, the less space between snapshots and the more of the wave you will capture, therefore, the richer the wave will sound.  Compact discs typically have a sampling rate of 44.1 kHz or around 44,000 bits per second.  For most humans, this is a high enough threshold that the ear cannot tell the difference, but some folks insist that they can.

One of the biggest advantages of digital signaling is that there is no degradation of the signal due to signal noise.  Since a digital signal is just a static series of numbers, as long as those numbers can be read, the signal remains intact regardless of any noise introduced into the transmission medium. Since, each one and each zero only has to be within a certain electronic threshold, noise spikes are very unlikely to affect the signal.  However, with analog, noise introduced into the transmission medium can corrupt the signal because each noise spike disrupts the natural wave of the signal.  The more distance an analog signal has to cover, the more noise that can get introduced in to the signal.  Think about when you travel out of range of a radio station.  The further you get from the station's broadcast tower the more static or even other signals you start to hear.  The same thing holds true for any other analog signal.  For example, many of you probably use WiFi on your computers.  You may notice that sometimes your Internet connection is faster than it is at other times.  This is often because of signal degradation.  Just about every electronic device in your home gives off some sort of electromagnetic frequency that can interfere with your WiFi signal, thus degrading it.  While analog may give you a more precise representation of the signal, digital is less prone to degradation.  So, if an analog medium is being used to transmit digital data, it is important to protect this data from degradation which is why protocols are so important.  I will be talking about protocols more specifically starting in Part 7 (and yes I did miscalculate how many parts this series will be. The new count is 9).

However, analog signals do have one important advantage of digital signals.  Since analog signals can are sent as a frequency, we can adjust that frequency so that it can fit within a single medium along with other signals.  By dividing the medium into channels, we can allow analog signals to travel in ether direction at the same time.  This is called "full-duplex" signaling.  Digital signals cannot do this because they have to use the entire bandwidth of the medium to send pulses.  So this means that if you send a digital signal it has to be either simplex or half-duplex and it can only travel in one direction at a time.  I will be talking about simplex and duplex transmissions in Part 4.  Until then I will continue explaining transmission methods.

This concludes the Part 3 of the the Data Transmission and Protocol series. In Part 4 I will go over simplex and duplex transmission.  These types of transmissions are not as complicated as they sound, and once explained you will have no trouble telling the difference between simplex, half-duplex and full-duplex, which, for engineers, should increase your understanding of which type should be used in a given situation.

[1] Wotel, P. (2004). Analog. Digital. What’s the difference?. Retrieved from http://bit.ly/p8R7i3