TCP/IP separates a full network protocol suite into a number of tasks.

4 Layers of TCP/IP Stack:

1. Application: Allows programs that have been create4d for a particular OS to asscess the network. Provides applications for internet activities, network troubleshooting and file transfer. The following protocols apply to the Application Layer:
   • HTTP (HyperText Transfer Protocol)
   • FTP (File Transfer Protocol)
   • SMTP (Simple Mail Transfer Protocol)
   • TFTP (Trivial File Transfer Protocol)
   • NFS (Network File System)
2. Transport: Provides communication services directly to the application processes that run on network hosts. The Transport Layer contains the following protocols:
   • TCP (Transmission Control Protocol)
   • UDP (User Datagram Protocol)
3. Internet: Moves data between OSI data-link and transport layers. Defines the packet and addressing scheme and routes data packets to remote hosts or networks. Provides the routing data from the source to a destination. The protocols that apply to the Internet Layer are:
   • ICMP (Internet Control Message Protocol)
   • ARP (Address Resolution Protocol)
   • RARP (Reverse Address Resolution Protocol)
   • BOOTP (Bootstrap Protocol)
4. Network Access: Combines OSI Physical and Data-link layers. Physical layer defines specifications such as: physical connectors, maximum transmission distances, and physical data rates. Data-link layer specifices how network access s controlled and hot to format data for transmission. The network access layer support the following topologies:
   • Ethernet
   • FDDI (Fiber Distributed Data Interface
   • Token Ring
   • Other Topologies

TCP/IP consists of 2 separate protocol functions:

• TCP: uses network layer below to move packets between two communicating devices. The transfer of data between the two devices is transparent.
  • TCP Characteristics: Operates at a higher level. Breaks message packets into smaller packets before they are transported. These small packets are then sent across the internet and received at the destination. The TCP layer of the destination reassembles the small packets into the orginal message.
  • Transport Layer Protocol (OSI)
• IP: uses packets to transsmit information trhough the network. A packet is an independent unit of data that carries sufficient information to be routed from source to destination – this occurs without reliance on earlier exchanges.
  • IP Characteristics: Carried out at the lower level and deal with the addressing of packets. When each packet reaches one of th gateway computers on the network, the computer read the address to find out where to forward the packet.
  • Network Layer Protocol (OSI)

Benefits of TCP/IP:

• Open Standard
• Independant of hardware (will run over phone lines, token ring, ethernets, or x.25 networks)
• reliable and effecient
• routable protocol (can be transmitted over a specific route to reduce traffic on certain parts of the network.)
• common addressing scheme – easy for devices on the network to address one another

TCP/IP: Network Layer Functionality

OSI Network Layer routes packets for information from the source to the destination. The Network layer performs four tasks to carry out this routing procedure.

1. Defines the packet and addressing scheme
2. Moves data between the OSI Data-Link layer and the OSI Transport Layer
3. Routes information packets to remote hosts
4. Performs fragmentation and reassembly of information packets

OSI Network Layer protocols:

• IP (Internet Protocol): Ip provides best-effort delivery. It concentrates on the route each packet should take to get to its destination. IP does not read the content of the packets, because it is soles interested in routing the packet to its destination.
• ICMP (Internet Control Message Protocol): Control and messenging capabilities are provide by ICMP
• ARP (Address Resolution Protocol): When the IP address is known, ARP establishes the correstponding MAC address
• RARP (Reverse Address Resolution Protocol): When the MAC address is known, RARP establishes the corresponding IP address
• DHCP (Dynamic Host Configuration Protocol): provides a structure to enable IP hosts to be configured automatically

Characteristics of IP:

• Lost of corrupted packets are not recoverd: (Lost/Corrupted packets are taken care of the end system.)
• Packets are treated independently: Each packet carries the addresses of its sender and receiver
• Pacekt Delivery is not guaranteed: Packets can get lost, misdirected or duplicated on the way to their destination

IP Datagram Header feilds

TCP/IP common Protocols:

ICMP (Internet Control Message Protocol): Internet protocol that is used to respond to errors in TCP/IP messages. Specifies a small number of messages that are used for identification and management purposes. ICMP relies on IP to transport packets around the network.

Key characteristics of ICMP:

• announcing network errors: indicates if a host or portion of the network cannon be reached because of failure. Reports is a packet is directed at a port number that has no active receiver.
• announcing network congestion: when a router cannot transmit messages as fast as it is receiving them, the router generates source quench messages – asking the sender to reduce the rate at which it is transmitting packets onto the network.
• assisting troubleshooting: uses an echo function. Sends a special packet on a round trip between two hosts. PING (a network managment app based on the echo fucntion) transmits a series of packets, measures the average round-trip times, and computer the percentage of packets lost.
• announcing timeouts: a router discards packets if they have been in the ntework for too long. When using ICMP, the routers generates and ICMP packet announing that it is dicarding the packet. TRACEROUTE is a tool that sends out packets to map the routes in the network. These packets have a small time-to-live values and the Traceroute tool watches the ICMP timeout announcements.

Delivery of ICMP packets is UNRELIABLE

IGMP (Internet Group Management Protocol): is used by IP Hosts between themselves and their immediate neighbor multicast agents to regiester which group they are a member of, add or delete memeber of a group, and regularly confirm their membership of a group. It is also used by routers that are connected to find out the members of the groups.

ARP (Address Resolution Protocol): Maps IP addresses to the MAC addresses. Address Resolution is the process of binding the network-layer IP address of a remote comptuer in an Ehternet network to its data-link layer MAC address.

ARP Cache: a table of IP addresses and their corresponding MAC addresses.

2 Types of messesages sent by the protocol are “ARP request” and “ARP reply”

RARP (Reverse Address Resolution Protocol): translates physical addresses (is MAC addresses and hardware interface addresses into IP addresses.) When a host, for example a diskless workstation, is turned on it may know its physical address but not its IP address. The host must learn its IP address from an external source. Normally this external source is the RARP server.

2 Types of messesages sent by the protocol are “RARP request” and “RARP reply”

DHCP (Dynamic Host Configuration Protocol): use to automatically asign IP addresse and to deliever TCP/IP stack configureation parameters (ie the default gateway and subnet mask). It can also be used to provide other configuration information (ie addresses for printers, news and time servers.)

DHCP has 2 components:

1. A protocol that delivers host-specific configuration parameters from the DHCP server to a host
2. a method of allocating a network addresses to hosts

TCP/IP: Transport Layer Functionality

Characteristics of TCP:

• connection-oriented protocol: to exchange data two computers set up a connection between them. They synchronize the end systems so that they can control the flow of packets and adapts to congestion on the network.
• Full-Duplex Operation: consists of a pair of virtual circuits – one in each direction. Only the two end stystems that are synchronized can use the connection.
• Error Checking: is provided by a checksum technique to verify that packets are not corrupt.
• Sequencing: Packets are sequenced and numbered. Using the sequence numbers the destination device can put the packets in order and determine if any packets are missing.
• Acknowledgements: The reciever sends and ackowledgment to the sender when it recieves one or more packets. If the sender does not recieve and acknowledgment then it may resend the packets or terminate the connection if it believes the receiver is no longer on the connection.
• Flow Control: occurs if the sender is transmitting too quickly and the buffer of the recievers overflows. If this happens the reciever drops packets. The sender learns that it should slow down or stop sending packets through the failed acknowledgements.
• Packet Recovery: The reciever can use the packet recovery services and request a packet to be retransmitted. The sender will retransmit the packets if the receipt of a packet is not acknowledged

There are 12 Feilds in the TCP header.

Port Numbers have the following assigned ranges:

• numbers below 1024 are assigne to well-known port
• from 1025 to 49151 there are registered ports that are listed by the IANA (Internet Assigned Numbers Authority)
• from numbers 49152 to 65535 they are assigned to dynamically assigned ports

End systems select the proper applications by using the port numbers. Port number identify the upper-layer protocol that uses the transport layer.

TCP connection setup and termination

When using TCP, a connect must be established between the two end systems before data transfer can begin. This connection is established using synchronization (SYN) and postive acknowledgment (ACK) segments betweent the two devices.

Another important function performed during the connection establishemtn is that the other device is informed of the Initial Sequence Number (ISN). This is used to track data byes on the connection.

3 step Connection Setup Procedure:

1. Host A: Send SYN: The device requesting a connection sends a SYN segment to the receiving device. The SYN contains the prot number and the first sequence number otherr device that the sender wants to connect to. The SYN control bit is set during the connection setup phase, in order to send the ISN.
2. Host B: Send SYN, ACK – Ack = 101: The receiving device replies with an ACK segment and a SYN. The receiving device indicate the sequence number it expects from the next byte of data to be received. This sequence number is the sender’s ISN increased by one, because the SYN uses one byte of the sequence space.
3. Host A: Send ACK -  Ack = 301: The initiating device acknowledges the SYN segment from the reciver. The SYNn bit is unset in the TCP header.

TCP sequences segments and uses a forward reference acknowledgment. Each segment is numbered before it is transmitted. When the segments are recived by the other device, TCP reassembles them in the correct order to create a complete message.

If there is a sequence number missing in the series, that segment and all of the segments sent after it are retransmitted. If a segment is not acknowledged within a specific period of time, it is retransmitted.

When all the data has been transmitted, the conection is terminated.

3 Step Connection Terminatio Procedure:

1. Host A: Send FIN – seq = 100 – ctl = FIN: The sending device send a segment to the reciving device with the finishing control code, FIN set to one
2. Host B: Send FIN ACK – seq = 300 Ack = 101 Ctl = FIN ACK: Receipt of the segment is acknowledged by the receiving device. The control code in the TCP header is set with FIN=1 in the ACK
3. Host A: Send ACK: The connection is terminated after the sending device acknowledges the acknowlegdment from the receiving device.

TCP/IP: Common TCP/IP Protocols

A VLAN is a network of computers tha have different physical locations, but they communicate as if they are connect to the same segment of a LAN. The computers share the same IP network number. Network components are grouped logically into broadcast domains in a virtual LAN. All devices can send and receive broadcast frames from another device within that particular domain.

Benefits of VLANs:

• Efficient utilization of bandwidth: Reduce the need to have routers control broadcast traffic. Flooding of a packet is limited to the switchports in the VLAN
• Enforcement of Network Security Policy: Broadcast domain is confined to VLAN – devices are isolated from listening to or recieving broadcast that are not intend for them. Devices of one VLAN cannot communicate with the device on other VLANs if a router is not connected between them.
• Reduced Administration Costs: Devices that are physically scattered can be logically grouped on the network. If a user moves you don’t need to reconfigure the device. If a user changes job function you can change the device’s VLAN membership.
• Reduced Network Traffic: traffic is reduced because of the confinment of the broadcast domains on the network.

Frames are forwarded between VLANs in 3 steps:

1. The IP address of the packets destination is not on the smae VLAN as the host
2. The host sends the packet to its IP default gateway’s MAC address
3. The packet is forwarded to this MAC address by a router.

IEEE 802.1Q Standard establishes the method for inserting VLAN membership information in the Ethernet frames using trunking encapsulation

VLAN components:

• Switches: Entry point into the network for end-station devices
• Trucks: Create a point-to-point link between multiple switch ports
• Routers: Enable communication between VLANs

VLAN membership can be configured using

• port-based: each port is assigned to a VLAN
• MAC address-based: switch does not need to be reconfigured if a user moves to a different port. Can be time consuming. single MAC addresses can not easily be a member of multiple VLANs. (This makes it difficult to share server resources between multiple VLANs.)

LANs have 3 Communication Methods:

• Multicast: Destination of transmitted frames is a group of clients
• Broadcast:  Frames are sent to all other devices
• Unicast: Frames are sent from a sigle host to a single destination

Collision Domain: A group of devices on a network that are directly connected by hubs.

Layer 2 Network devices (Switches and Bridges) use a process called segmentation to dedicate bandwith to users and reduce the size of collision domains.

CAM (Content Accessable Memory) Table: Same as a MAC Table – a record of MAC addresses used by that switch

Data-Link Layer Opperations: MAC Addresses

The Header (and trailer) of the MAC Sublayer of the Data-Link layer contains:

• MAC Address of the sending computer
• MAC Address of the destination computer
• A type feild containing a SAP (Service Access Point) to identify the type of protocol that is being carried and may give the length of the data part.

MAC Address are written in one of two ways:

• MM-MM-MM-SS-SS-SS
• MM:MM:MM:SS:SS:SS

(Hexidecimal numbers 12 Digits long – 42 bits in binary)

MAC Addresses made up of two parts:

1. Manufacturer’s OUI (Organiztionally Unique Identifyer) nubmer.
2. Adapter’s serial number

Data-Link Layer Opperations: Swithcing Fundamentals

Both Switches and Bridges:

• Connect different LAN segments
• Determine the network segment of a frame needs to be transmitted on by using a table of MAC address
• reduce network trafffic

Switches use Application-Specific Integrated Ciruits (ASICs). Optimized and embedded in the hardware. Results in shorted exectution time.

Switches support new functions like VLANs (Virtual LANs) that operate at higher speeds and at lower costs with greater port density than bridges.

Switches Support:

• Dedicated communication between devices. (Miscrosegmentation is a method of creating private or dedicated segments each of which only has one user.)
• Full-Duplex Communication: Only possible on dedicated connections
• Media Rate Allocation: Switches can translate between 10 and 100 Mbps
• Multiple Simotaneous conversations

Bridgeds loops can occurs when there are more than one path between the source and the destination. Frames travel the loop continuously.

The Spanning Tree Algorithm is used in bridged networks to help prevent bridge loops.

Two Switching Techniques are

1. Store-and-forward switching: Error-Checking
   • Switch must receive the complete frame before forwarding takes place.
   • Error checking is performed and frames that contain errors are discarded.
   • The entire frame is copied into the buffers onboard the LAN switch and the CRC (Cyclic Redundancy Check) is calculated.
   • If the frame contains a CRC error or its size is outside the rand of 64 byes to 1518 bytes, it is discarded.
   • If the frame contains no errors and is an acceptable size, the switch determines the outgoing port by looking up the destination address in the MAC table.
   • The switch then forward the frame to it final destination.
   • A frame less than 64 bytes is called a runt
   • A frame larger than 1518 bytes is called a giant
2. Cut-through switching: Reduces latency – begins forwarding frame as soon as header is received and destination interface is determined.
   • the LAN switch waits until the destination address of the frame is copied into the onboard buffers of the switch.
   • The destination address is indicated in the first six bytes following the preamble in the frame
   • The switch uses the MAC table to look up the destination address and determine the outgoing port.
   • It then forward the frame to its destination

Switches use 6 Steps to forward frames:

1. Switch receives a frame from a source on the network
2. The switch enters the switch port that received the frame and the MAC address of the source into the MAC table
3. If the switch does not know which port the destination address of the frame is on it floods the frame to all ports
4. When the destination device receives the broadcast, it replies. The destination device is the only device that replies
5. When the switch receives the reply from the destination device, it enters the switch port it received the reply on and the MAC address of the destination device into the MAC table
6. Now the switch can switch frames between the source and destination devices without broadcasting them on to the entire network.

Switching Frames Procedure: the process of a switch searching for a destination port and address before forwarding a frame.

LAN standards specify signaling and cabling at the physical and data-link layers of the OSI model

IEEE divided the OSI data-link layer into two separate sublayers:

• Logical Link Control (LLC)
  • operates independently of the technology it is workin in
  • interfaces between the network layer above it and the MAC sublayer below it
  • involved int he encapsulation process – an LLC header on a packet instructs the data-link layer what to do with the packet.
• Medium Access Control (MAC)
  • uniquely idenifies multiple devices at the data-link layer
  • for a device to operate on a network it must have  a MAC address

Ethernet (IEEE 802.3 standard) features:

• baseband signaling
  • when a network signal uses all of the available signal frequencies (or the entire bandwidth) to transmit data
• Carrier Sense Multiple Access with Collision Detection (CSMA/CD)
  • used in ethernet and wireless ethernet networks
  • listens to the signal (Carrier Sensing) – only transmits when line is free
  • Listens to see if a collision occurs (Collision Detection) – if so, both devices wait a random amount of time and resend the signal.
  • More than one stantion can be on the network at the same time (multiple access)

802.2 – defines LLC (data-link sublayer) only

802.3 – defines physical layer and MAC (data-link sublayer)

Steps in the CSMA/CD Process

1. Host wants to transmit
2. Is carrier sensed? (if yes > 1 if no >3)
3. Assemble Frame
4. Transmit Data Frame
5. If collision detected? (if yes > 9 if no > 6)
6. Keep transmitting
7. Is transmission done? (if yes > 8 if no > 6)
8. Transmission is complete – media is idle
9. Jam signal is broadcast if collision is detected
10. Attempts +1 (transmission attempt counter increates by one)
11. Attemps too many? (if yes > 12 if no > 13)
12. Too many attempts and transmission is aborted
13. a backoff algorithm calculates “t” (a random length of time that must pass before transmission is attempted)
14. host waits ‘t’ microseconds as calculated by the algorithm then returns to step 4 to try transmitting again.

4 Categories of Ehternet:

10Mbps -

• uses fiber, coax, or TP cables
  • 10Base5 – Coax – 10Mbps – 300 meters per segment
  • 10Base2 – Coax – 10Mbps – 185 meters per segment
  • 10BaseT – UTP – 10 Mbps – 100 meters per segment
  • 10BaseF – Fiber – 10Mbps – 500 – 2000 meters

100 Mbps -

• Uses Fiber, STP or UTP
  • 100BaseT – UTP – 100Mbps – 100 Meters per segment
  • 100BaseVG – UTP – 100Mbps – 213 Meters (Cat 5) or 100 Meters (Cat 3)
  • 100BasesT4 – UTP – 100Mbps  100 Meters per segment
    • uses 4 pairs of Cat 3, Cat 4 or Cat UTP
  • 100BaseTX – UTP – 100Mbps  100 Meters per segment
    • uses 2 pairs of Cat 5 UTP or Type 1 STP
  • 100BaseFX – Fiber – 100Mbps  2000 Meters per segment
  • FDDI (Fiber Distributed Data Interface) – Multimode Fiber – 100Mbps – 10 Km per segment

Gigabit Ethernet -

• uses Multimode  Fiber, UTP or copper Cables
• IEEE 802.3z and 802.3ab relate to Gigabit Ethernet using Fiber optic and TP cables
  • 1000BaseT – copper – 1Gbps – 100 meters per segment
  • 1000BaseTx – Cat 5 – 1Gbps – 100 meters per segment
  • 1000BaseCx – copper STP – 1Gbps – 25 meters
  • 1000BaseSx – multimode Fiber – 1Gbps – 550 meters
  • 1000BaseLx – multimode Fiber – 1Gbps – 550 meters

10 Gigabit -

• 802.3ae standard relates to 10 Gigabit Ethernet
  • 10GBaseCX4 – twin-axial copper cables – 10Gbps – 15 meters per segment
  • 10GBaseSR – FDDI-grade multimode fiber – 10Gbps – 300 meters
  • 10GBaseLX4 – FDDI-grade multimode fiber or single mode fiber – 10Gbps – 300 m or 10 km
  • 10GBaseLR – Single mode fiber – 10Gbps – 10 km
  • 10GBaseER – single mode fiber – 10Gbps – 40 km

Types of LANs: Token Ring and FDDI

802.5 – IEEE toekn ring standard (almost identical to IBM token ring standard)

Uses Star topology

all NIC are connected to a Multistation Access Unit (MAU or MSAU)

Token Ring Advantages:

• Performs Regeneration – reduces degradation because each data signal transmitted on the network is read and repeated by each of the devices on the network that it passes
• Performance “degrades w/ Grace” – as traffic increases the network gets slower because there is only one token. It does not crash. TR networks are very reliable
• Communicates with IBM mainframes – IBM mainframes are still used
• Reliable deterministic – able to calculate maximum amount of wait time
• Uses sophisticated priority system – allows “high-priority” nodes

Token Ring Management Mechanisms:

• Active Monitors – any station can be an active monitor. becomes the centralized source of timing information for other stations in the ring
• MAUs (Multistation Access Units or MSAUs) – in star topology MSUs see all connections on the network, can check devices for faults and remove faulted stations from the network.
• Beaconing - when a station detects a fault (for example a cable break) it sends out a beacon frame. The beacon frame defines the failure domain: the station reporting the failure, the nearest active upstream neighbor (NAUN) and everything inbetween. Activates autoreconfiguration – MSAUs can do this using electrical reconfiguration.

FDDI – combines fault tolerance of Token Ring with high-speed cababilities of Fiber Optic. Opperate at 100Mbps

Supports real-time allocation of network bandwith

• Synchronous: used for Voice and Video – allocated to stations that need continuous transmission capability
• Asynchronous: The bandwitch left after the synchronous allocation is allocated to the asynchronous traffic. And 8 level priority scheme is used – higher level priority stations can lock out other lower-level stations

FDDI  use a dual-righ architecture that is counter rotating. The secondary ring is used when the primary ring fails.

Stations on a FDDI network

SAS (Single Attached Stations): attached to only one ring

DAS (Dual Attached Stations): attached to both rings

Types of LANs: Wireless Transmission

1. Spread Spectrum: the frequency of the transmitted signal is deliberately varied over a range or frequencies causing the signal to become noise-like and harder to intercept

-much more resistant to interference vs. conventional narrowband wireless signals.

802.11 – spread spectrum standard for IEEE

DSSS (Direct Sequence Spread Spectrum)

• Chipping: Uses a spreading code, called a PN (pseudorandom noise code) a sequence of chips or bit of information. Each “0″ or “1″ bit in the signal is represented by the code sequences so the signal is represented by a long code instead of the signal itself
• Signal Modification: the encoded representation of teh signal then modulates the carrier signal – spreading it over the range of reequencies being used (bandwidth)
  • There is a peak in the signal’s power at the main boradcast fequency and on the either side of this peack there are gradulally dissipating peakes called “side lobes”.
  • The width and number of side lobes depends on hte spreading code used and the signal itself.

FHSS (Frequency Hopping Spread Spectrum)

• Frequency Hopping: most widely used – if intercepted it is only for a moment before the frequency hops again.
• Signal Modulation:
  • Step 1: signal modulates the frequency of the carrier wave – results in a regular narrowband signal
  • Step 2: Spredding code is applied to modulate the carrier wave causing it to hop between frequencies – the spreading code provides a list of frequencies for the wave to hop to as well as the length of time is should stay at each frequency

2. Bluetooth - short range radio technology that operates on the 2.4 GHz ISM (Industrial Scientific Medical) band.

Uses:

• Automatic Synchronization: allows automatic communication between devices such as cellphone and computers
• Internet Bridge: Allows cellphone or modem to act as a wireless modem to dial up to the internet or receive data calls
• Bluetooth Headset: cellphone headset.

Bluetooth Integrated Security Features:

• Challenge-Response authentication
• Encryption
• session key generation (session keys can be changed at any time during a connection)

Two bluetooth ranges:

10 Meters at 1 mW (milliWatt)

100 Meters at 100 mW (milliwatts)

Not suitable for LAN or WAN applications – not desinged to carry heavy traffic loads.

3. Infrared (IR) – a form of electromagnetic (EM) radiation that operates at an extremely high frequency.

- in the EM spectrum IR is located between microwaves and visible light – the most useful band is the band just below visible light

LED IR has a Typical range of 3 M

Diffuse IR emits beams in an arc – the beams can bounce off obstructions and find at least one path tot he reciving device. Diffuse IR has a range of 270 square meters

IR Lasers can travel 5 km – suceptable to interference from other light sources and requires a filter at recieving end.

Types of LANs: Wireless LANs

802.11 standard uses

DSSS in noisy environments at 1 Mbps

FHSS in less noisy environments at 2 Mbps

802.11 uses CSMA/CA

802.11 a – 54 Mbps, 5 GHz, 25 – 75 ft.

802.11 b – 11 Mbps, 2.4 GHz, 100 – 150 ft.

802.11 g – 54 Mbps, 2.4 GHz, 100 – 150 ft.

Devices on a wireless network are either basestations or clietents. (Basestations are commonly called access points)

Basestations are required to provide: association, distribution, integration and reassociation services to clients on the wireless network.

Wireless LAN modes:

• Ad hoc: no basestaions – all clients can access each other and communicate directly. All nodes have equal rights and responsibilities. Separate networks may coexists on the same frequencey using different SSIDs (Service Set Identifiers)
• Infrastructure: basestation is used – clients only communicate with basestation and do not directly communicate with each other. Multiple basesations can be used to broaden the coverage area.

3 Main Components in a Wireless LAN:

1. Wireless NICs: All devices or nodes need a NIC. A wireless NIC has a fixed or internal antenna.
2. Access Points: Used to connect existing wired or wireless networks or to extend the range or a wireless network. Uses an omnidirectional antenna, a wired NIC, and bridging software (forwards data between the LAN and the wireless nodes)
3. Wireless Bridges: Used to connect two LANs. Use unlicensed Spread Spectrum Radio Frequency (RF) or Lazer IR.

Wireless Topologies:

Bus: all nodes are within point-to-point coverage and they communicate with each other forming a BSS (Basic Service Set.) No access point is needed.

Star: Allow nodes to communicate beyond BSS to ESS (Extended Service Set). Use access points to extend network coverage by 400% creating and ESA (Extended Service Area.)

Factors that affect performance:

Interference: Caused when a signal, other than the desired signal, is transmitted on the same or nearyby frequency.

Surrounding Environment: Walls, Concrete floos, electrical equipment, building structural elements, and natrual obstructions (trees, mountains, etc…) negatively affect performance by obstructing wireless signals.

Types of Antennae used: Can Omnidirectional or Point to Point.

Noise: the less noisy the conditions the better the performance of the wireless LAN

Types of media

Twisted Pair

• STP – Sheilded twisted pair -
• UTP – unsheilded twisted pair – small diameter, reduced interference.
  • - 22 or 24 gauge wire, 100 Ohms
  • Cat 1 – 56 K (telephone comm only)
  • Cat 2 – 4 mbps – token ring
  • Cat 3 – 10 mbps
  • Cat 4 – 16 mbps
  • Cat 5 – 100 mbps
  • Cat 5e – 1 gbps
  • Cat 6 – 1 gbps, 24 gauge wire

N<signal>X

N = speed in mbps (ie “100″ in 100baseT)

signal = signaling type – base or broad (baseband or broadband)

x = cabling scheme – either distance or type (ie “5″ in 10base5 represents 500 meters in thicknet coaxial cable. “T” in 10baseT represent “twisted pair”)

Coax -

10base2 – thinnet – more flexible, needed to be grounded, 3.5 mm

10base5 – Thicknet – very rigid, used as backbone cabling, specific purpose.

Fiber Optics -

support speeds of 100 Gbps

Parts of Fiber Optic Cable:

Core Fiber – made of pure glass or high grade plastic.

Cladding – made of plastic or glass, prevents light from escaping from the core in a process called “total internal refraction.”

Modes of Fiber Optics:

Single Mode: Uses lasers, alows one mode of light to pass at one time. greater bandwidth and greater distances than multi mode. more expensive. max cable length of 60 kilometers.

Multi Mode: uses LED, allows multiple light modes to pass through, max cable length of 2 km.

Physical Media: Network Cable Connectors

Twisted Pair:

• RJ-11
• RJ-45

Coax:

• AUI/DB-15 – Thicknet
• BNC – Thinnet
• F-type – “normal”

Fiber:

• SC – Subscriber Connector – duplex
• ST – Straight Tip
• LC – Local Connector
• MT-RJ – (replaced SC)

IEEE 1394 – Firewire/i-Link

Physical Media: Network Installation Tools

Patch Cable > Wall Jack > Drop Cable > Patch Pannel > Switch/Hub

Tools -

• Punch Down
• Crimper
• Wire Map Tester
• Continuity Tester – Checks for opens, shorts and crossed pairs
• Toner
• OLTS (Optical Loss Test Set) – Tests Fiber Optic Cable
• Multifunction Test

• Components used by a computer to enable network connectivity:
• CPU
• bus
• drives
• memory components
• port
• cards

As well as drivers and memory the NIC needs:

• I/O address – to read or write data to the computer. The I/O address identifies a part of memory that is assigned for use by the NIC
• IRQ signal – identifies which device in the computer is requestingthe CPU to perform a funcion. For example, when data arrives into a NIC its IRQ indicates to the CPU that is had data that needs to be processed for the NIC.

To install a NIC you need to understand:

• EPROM – Erasable Programmable Read-Only Memory
• Jumpers and switch settings
• Pulg and Play software

If there is a problem with the NIC you should be able to:

• Use Card diagnostics – such as loop back tests and diagnostic procedures supplied by the vendor
• fix hardware resource conflicts with IRQ and DMA (direct memory access).

Network Devices and Topologies:  Network Devices

OSI L1 – (Physical Layer/Bit Layer)

• Repeaters – regenerate and retransmit electrical signals so they can travel greater distances without deteriorating. The operate at the bit level. A multiport repeater is called a Hub
• Hub – a multiport repeater receives on one port and transmits on all the other ports

OSI L2 – (Data-Link/Frame Layer – MAC address)

• NIC – each NIC has a unique MAC address
• Bridge – create LAN segments. Provides extra bandwidth for data exhcange by
  • filtering local LAN traffic
  • Maintaining connectivity between network segments directly connected to it
  • Bridges (unlike hubs) can effeciently manage data transmission between connected segments based on MAC addresses and filter out unnecessary traffic from reaching a segment
  • Gathers and manages MAC address in a table. Each port created a separate collision domain – so a collision on one segment does not affect other segments.
• Switch – create multiple bridge connections
  • conect LAN segments into single networks and use MAC addresses to decide where to forward the traffic. Faster than bridges because switching decisions are performed using specialized hardware instead of software.
  • each port behaves like a separate bridge (sometimes called “microsegmentation”)

OSI L3 (Network Layer/Packet Layer – IP Addres)

Router – uses L3 addresses (IP Addresses) to transmit data packets between networks. Uses IP instead of MAC to determine path.

• can connect differnt L2 technologies (ie – Token Ring and Ethernet)
• can examine L4 (Transport Layer) information when sending data between devices or networks.

Multilayer Switches – can use L3 as well as L2 addresses. Cand deliever L3 functions with the same spped as L2 functions.

Multilayer Devices:

Gateways: A complex network device that connects disparate network environments. (ie a LAn to a mainframe)

• Uses a combination of hardware and software and can carry out translations at various layers of the OSI model
• E-Mail programs typically use gateways to communicate with internet mail servers. They use gateways to translate LAN-based mail mesage into SMTP (Simple Mail Transfer Protocol) format.

Firewalls:  Positioned on the edge of private networks to protect them from any unwanted traffic or attacks attempting to access the internal network.

• Examines all traffic and packets to ensure that they are legitimate dropping any that fail to conform to its rules of entry.

AAA Server: Authentication, Authorization and Accounting. Processs requests from users to gain access to network resources.

• only allows authenticated users onto the network
• gives usuer access only to the resources they are authroized to use
• keeps an account of user behavior.

Network Devices and Topologies:  Network Topologies

A Topologies describes how a network is laid out and how data is transmitted on it.

• Logical Topology - describes structure and path connection types between different parts of the network. It defines how data flows in the network.
• Physical Topology – Describes the physical arrangement of devices on the network

Bus Topology - all devices are connected to a  common cable. A signal terminator is used to absorb the signal at the end of the cable so that is does not reflect back.

Star Topology – all devices are on the network are linked to a common device, such as a switch.

Ring Topology – network is arranged in a logical circle. data is passed on a “token” around the circle. Eliminates the need for a termination device. There are no packet collisions on a ring topology network.

• Single Ring – one ring, data travels in one direction only
• Dual Ring – uses two rings, data travels in both directions. incorporates fault tolerance. FDDI (Fiber Distributed Data Interface) typically uses a dual ring topology.

Mesh topology – each device is connected to the others in the network

• Full - every host is connected to every other host. Allows for redundancy and high level of fault tolerance. Complex and expensive to implement.
• Partial - One host is connected to a number of other hosts. The other hosts are connected to some, but not all of the other hosts.

LAN (Local Area Network)

WAN (Wide Area Network)

Server - Provides resources to other computers on the network

Workstation - any computer that can request resources and be used to do work. Workstations and Clients are different

Client - any device that can request resources

Host - any network device that has a TCP/IP address

Peer to Peer – no centralized authority. access rights are managed by computer holding the resource

Client/Server – managed from a central point. some computers are dedicated to serving others. each computer is either a client or a server.

Advantages of Client/Server vs P2P:

• More organized
• More efficient
• files and resources easier to locate
• More secure – require password and username
• Can be scaled to infinite size

Networking Basics: Types of Networks

LAN - high speed, low-error, data network confine to a small area

Type of LANs

• Ethernet
• Fast Ethernet
• Gigabit Ethernet

Network Standars are defined by the Institute for Electrical and Electronics Engineers (IEEE)

Standard Ethernet 802.3 is commonly referred to as “Ethernet”

WAN - covers a large geographic area. The Internet (public) and Private Global Corportate networks are examples of WANs.

WAN vs LAN

• WAN covers larger distance
• WAN speeds are slower
• WANs are more expensive
• LANs are usually private while WANs can be public or private

WANS require several core device to function:

• Routers - used to direct traffic on the network to its correct destination. A router is connected to at least 2 networks.
• WAN switches – used to connect routers on the WAN using virtual circuits.
• Modems - provide remote access to networks by converting digital signals to analog ones so that data can be transmitted over analog communication lines (such as telephone lines)

MAN - (Metropolitan Area Network) Covers a metropolitan area such as a City or Suburb. Larger than a LAN, smaller than a WAN

Intranet - private network consisting of interlinked LANs.

Extranet - when part of a company’s intranet becomes available to customers, suppliers or anyone outside the company.  Extranets use IP and a Firewall is used to ensure secruity.

Networking Basics: The OSI Model

OSI – Open System Interconnection model – developed in 1984 by ISO (International Organiztion for Standardization)

A standard or technology can be either proprietary or open

Proprietary - One company or a small group of companies controls the technology rights and usage

Open - available for free usage to the public

OSI Model 7 layers:

• Application
  • File, Print, message, and applacation database services. It provides network service to applications that require access to the network. It controls how these services are advertised and made available. Deals with User authentication and privacy.
  • Does not provide services to any other layer
• Presentation
  • Manages data representation. Transforms data into a mutually agreed format that each application can understand. It formats and structures data, ensuring it is readable between two hosts.
  • Examples include: ASCII (American Standard Code for Information Exchange) and EBCDIC (Extended Binary-Coded Decimal Interchange Code).
  • Organized the sytax of data transfer for the application layer. Data compression and encryption take place in this layer.
• Session
  • Provides communication between hosts. It does this by connection establishment, data transfer, and connection release.
  • Includes authentication, creation management, and termination of sessions between different applications.
  • Provides services to the Presentation Layer.
• Transport
  • aids point to point communications
  • provides reliability in the transportation of data between hosts and ensures complete data transfer.
  • Uses error detection and recovery information flow control to establish, maintain, and terminate all virutal circuits.
• Network
  • ensures data delivery by providing connectivity and path selection between two host systems
  • selects the most appropriate path for sending data and routes data packets
  • Logical addressing and resovling names to host physical address is carried out
  • works with IP addresses assigned to hosts
  • provide logical LAN to LAN communications by supporitng the routing of data between different networks
• Data-Link
  • arranges bits from the physical layer into logical chunks of data known as frames.
  • A frame is a contiguous series of data witha common function
  • framing enables the network to organize bits into logical data format and send them to the correct computer
  • controls how data is formatted and how data on the network is controlled
• Physical
  • responsible for providing the most basic element of data transport – binary transmission
  • outlines the functional, procedural, electrical, and mechanical specifications for controlling physical links
  • specifications relate to the activation, maintenance, and deactivation of physical links.
  • controlls transmission of data on to physical media

OSI has advantages

• enables compatability between different types of networks
• provides standards for users
• facillitates and understanding of how data travels within a network

Adantages of layering network functions:

• accelerates evolution: supports updates and improments to individual componenants without affecting other componenants or needing to rewrite and entire protocol.
• ensures interoperable technology: prevents changes in one layer from affecting another layer
• facillitates modular engineering: allows different types of network hardware and software to communicate with each other.
• reduces complexity: breaks network communcation into smaller, simpler components
• standarizes interfaces: provides vendors a set of standards that ensure greater compatability between the various network technologies.
• simplies teaching and learning: breaks up the task of networking into distict layers

Encapsulation - wraps the data with the required protocol information before transmitting data to the network. As moves through the layers of the OSI model each layer adds a Header (and a Trailer if applicable.)

8 Steps in the Encapsulation Process:

1. user data is sent from an application to the 7th layer of the OSI model – the Application layer
2. The Application Layer adds the header “L7″ to the user data. the L7 and the original user data become the data that is passed down to the Presentation Layer.
3. The Presentation Layer adds the Presentation Layer Header “L6″ This now become the data that is passed down to the Session Layer.
4. The Session Layer adds the Session Layer header “L5.” This now become the data that is passed down to the Transport Layer.
5. The Transport Layer adds the Transport Layer header “L4″ This now become the data that is passed down to the Network Layer.
6. The network Layer adds the Network Layer header “L3″ This now becomes the data that is passed down to the Data-Link Layer.
7. The Data-Link Layer adds the Data-Link Layer header “L2″ This now becomes the data that is passed down to the Physical Layer.
8. The Physical Layer transmits the bits onto the network media.

De-encapsulation occurs when the data is recieved. First it is check for errors then it is striped of its header.

4 Tasks of De-encapsulation

1. Checks the Data-Link trailer to see of there are any errors
2. If the data contains errors it may be discarded. The Data-Link layer may ask that the data be retransmitted
3. If the data contains NO errors then the Data-Link layer reads and interprets the control information in the data-link header.
4. The Data-Link layer strips the data of its header and trailer and then move the data up to the network layer based on the control information found in the data-link header.

During the encapsulation process the protocols at each layer exchange information called PDUs (Protocol Data Units)  between the peer layers.

PDUs are named differently depending on which layer they are going through:

• Transport = Segments
• Network = Packets
• Data-Link = Frames
• Physical = Bits

Peer to Peer communication: each layer of the OSI Model communicates with its peer layer at the destination.

TCP/IP Stack

Comprises 4 layers:

• Application
  • Mangage High Level Protocols
  • Includes aspects related to dialog control, ecoding, and resprentation
  • TCP/IP Stack groups all application aspects into one layer and ensure that application-related data is properly packaged for the layer below
• Transport
  • Handles quality of service issues such as reliability, flow-control, and any error correction
  • TCP protocol provides reliable network communication
• Internet
  • provide packet delievery and hierarchical addressing services.
  • Sends packets from any network on the internetwork and esures they arrive at the correct destination no matter what path they have taken to get there.
• Network Access (Host-to-Network Layer)
  • Looks after all the issues handled by the OSI Physical and Data-Link layers.
  • Includes the LAN and WAN protocols

Similarities between OSI Model and TCP/IP Stack:

1. Both have Application Layers (Functions of the App Layer in each model is different)
2. Both use Packet-Switched Technology. (As opposed to Circuit Switched)
3. The Transport and Network Layers are comporable in both models

Differeneces between OSI and TCP/IP Stack:

1. OSI Data-Link and Physical Layers are combined into the Network Access Layer of TCP/IP Stack.
2. OSI Presentation and Session Layers are combined into the Application Layer of TCP/IP Stack
3. TCP/IP is the standard around which the internet was developed. The OSI Model is generally only used as a guide.