Chapter 6 is the first and longest of the seven chapters about OSI model layers. The objectives important to this chapter are on page 6-1:
Concepts:The short answer is on page 6-2: the Physical layer of the ISO-OSI model tells us how to transmit bits. This layer tells us how the network is physically set up. The topic chart on page 6-3 (you get a relevant chart in each chapter)
shows the topics and methods for this chapter. A topic is a thing
this layer does. A method is a way to do it.
Also note, on page 6-3, that this layer is associated with certain hardware:
Connection types come in two varieties: point-to-point and multipoint. Basically, a point-to-point connection is from one device to one other device, a direct channel, such as the cable from my computer at home to my printer at home. (Note that this could also be through a wireless medium.) A multipoint connection exists when several devices share the channel connecting them to a resource, such as the customers of a cable TV provider who share the cable lines leading to the server. The capacity of the channel on a multipoint network is an issue, since it is shared by all users. Physical topology is the way the network is wired (or wireless-ed?)
together. Five methods are discussed in your book:
Physical topologies can be compared on four factors, listed on page 6-7:
In a physical bus network (we'll talk about a logical bus in the next chapter), think of the bus as being one long network access channel that all the devices connected to it have to tap into. The bus can be called a backbone, and the node connections can be connected with drop cables and cable taps (like the vampire tap we discussed for Thick Ethernet). A concept is presented on page 6-9 about a device being downstream of another device. This phrase applies to devices that are on a backbone in which devices send data (primarily) in one direction. Comparison factors for a bus LAN:
A physical ring passes information from one node on the network to the next until it gets back to the unit that sent it. Signals are received and retransmitted by each station. Information passes in one direction only, unless there is a second ring for fault tolerance. Comparison factors for a ring LAN:
A star topology is called a star because it uses a central device (a concentrator or hub), from which drop cables radiate to all nodes. Comparison factors for a star LAN:
A mesh topology is characterized by multiple redundant connections between nodes. Comparison factors for a mesh LAN:
A cellular topology is what cell phone companies use to provide customers with mobile phone service. The illustration on page 6-20 shows the basic idea of radio towers (a cell phone is a radio, right?) that are arranged so that they serve overlapping areas. As long as a mobile user is in an area, a cell, served by a transceiver (tower) for the net, the user can be connected to the net. Comparison factors for a cellular LAN:
Digital Signaling is the next topic. There is a good discussion in the text about the difference between digital and analog. It may help to think of anything digital as being like a light switch: either on or off, with no state in between. Things that are analog in nature are more like a stream of water from a faucet: the stream could be any volume per second, depending on how far you turn the control knob. Infinite range of values. A digital signal on a network will be a pulse of light or electricity. We read the information in the signal either by its current state or watching for a state transition. Consider the graph of the signal on page 6-24. It shows time across the X axis and signal voltage on the Y axis. (If that means nothing to you, do a web search on Rene Descartes, or Cartesian graphs before going on). Current state Digital systems represent different messages with different voltages. In the graph on page 6-25 (IGNORE the dotted line values!), one voltage could represent a 0 and another represent a 1. The signal is measured at set intervals, and the receiver "sees" the bit that the voltage being sent represents. State Transition Digital systems watch voltages, too, but they use a trick. The graph on page 6-25 is confusing unless you understand the trick. The trick is that these systems agree on a scheme that says if I send the same voltage that I sent a moment ago, it means one thing, and if I send a different voltage, it means the opposite. (NOW you can read the dotted line values!) In the illustration, it shows the first measurement was a constant voltage, so it means 0. The second measurement showed a voltage change (a state transition), so it means a 1. The third measurement shows the same voltage as the last, so it means a 0 (no transition). The fourth measurement shows a change back to the first voltage (a state transition), so this time the change to a low voltage means a 1. Any change means 1, and no change means 0. As a memorization tool for the above schemes, note that all schemes named with one word (EXCEPT Manchester) are Current State schemes. All schemes named with two or more words (EXCEPT Return-to-Zero) are State Transition schemes. Analog Signaling assumes that signals are constantly changing
waves. Waves have three properties, listed on page 6-27, which
we can vary and measure to pass data:
Now, the next part is a little confusing, so stay with me: there are Current State schemes and State Transition schemes for Analog signaling, different from the ones discussed above. Current State schemes for Analog signaling are:
We are given one State Transition scheme for Analog signaling:
The next topic (wondered if we'd ever get to it, right?) is Bit Synchronization. This is any method used to let the receiver know when to measure the channel for the next bit (1 or 0). Essentially, two types of methods are used:
The Bandwidth topic is split into two methods for using the bandwidth of a medium.
Multiplexing concerns the methods used to place more than one signal on the medium at once. Three methods are discussed:
|