In selective-reject ARQ, only the specific damaged or lost frame is __________.
- ARetransmitted
- BForwarded
- CSelected
- DRejected
Selective-reject resends only the frame that was damaged or lost.
Questions from past final papers.Pichhle final papers ke sawal.
160 MCQs · 88 short · 26 long questions · 160 MCQs from past papers160 MCQs · 88 short · 26 long sawal · 160 MCQs past papers se
In selective-reject ARQ, only the specific damaged or lost frame is __________.
Selective-reject resends only the frame that was damaged or lost.
YMODEM has a __________ byte data unit.
YMODEM uses 1024-byte data blocks (XMODEM uses 128).
Which of the following sublayers resolves the contention for the shared media?
The Media Access Control sublayer handles access to the shared medium.
Like 10Base5, 10Base2 is a __________ topology LAN.
Thick and thin Ethernet both use a bus topology.
At the CRC generator, __________ added to the data unit after the division process.
Zeros are appended before division; after division the remainder (CRC) replaces them.
Flow control is needed to prevent __________.
Flow control stops the sender from overwhelming the receiver's buffer.
In YMODEM, multiple files can be sent simultaneously.
YMODEM supports batch transfer of several files in one session.
In 56K modems the maximum uploading speed is ______ and the downloading speed is ______.
56K modems are asymmetric: up to 33.6 Kbps upstream and 56 Kbps downstream.
There are ________ basic categories of multiplexing.
The three are FDM, WDM and TDM.
EIA-232 allows for a maximum bit rate of __________ Kbps.
EIA-232 is specified for up to 20 Kbps.
In CRC, the quotient at the sender __________.
Only the remainder is used; the quotient is thrown away.
In line discipline, after the data transmission the sending system finishes with an __________ frame.
The sender ends the exchange with an EOT (end of transmission) frame.
Traditional modems are widespread now at a data rate of __________.
The common traditional dial-up modem standard is 56 Kbps (V.90/V.92).
A telephone network is an example of a __________ network.
The telephone network sets up a dedicated circuit for each call.
In case of uploading, at the switching station data is converted to a digital signal using __________.
The switching station digitizes the analog signal with pulse code modulation.
Category 5 UTP cable is used for data transmission of up to __________.
Cat 5 UTP supports data rates up to 100 Mbps.
Which of the following primarily uses guided media?
The local telephone loop uses twisted-pair cable, a guided medium.
Ultra high-frequency waves always use __________ propagation.
UHF signals travel by line-of-sight propagation.
Distortion occurs in a __________ signal.
Distortion arises because the frequency components of a composite signal travel at different speeds.
Which of the following are not used to measure the performance of transmission media?
Throughput, propagation speed and propagation time are all performance measures.
A portion of the path that carries transmission between a given pair of devices is known as __________.
A channel is the part of a link carrying one transmission between a device pair.
__________ takes data from one high-speed line and breaks it into portions.
Inverse multiplexing splits one high-speed stream over several lower-speed lines.
In primary-secondary communication, the __________ is always the initiator of a session.
In poll/select line discipline the primary device always starts the exchange.
__________ coordinates the amount of data that can be sent before receiving acknowledgment.
Flow control limits how much data the sender may send before an ACK.
If the ASCII character H is sent and the character I is received, what type of error is this?
H = 1001000 and I = 1001001 differ in only one bit.
In __________ ARQ, if a NAK is received, only the specific damaged or lost frame is retransmitted.
Selective-reject resends only the frame named in the NAK.
Data link protocols can be divided into __________ sub-groups.
Data link protocols are either asynchronous or synchronous.
If odd parity is used for ASCII error detection, the number of 0s per 8-bit symbol is __________.
With an odd number of 1s in 8 bits, the number of 0s (8 minus odd) is also odd.
The (LLC) PDU has no flag fields, no CRC, and no station address.
The IEEE 802.2 LLC PDU has only SAP addresses and control; framing and CRC are in the MAC frame.
The checksum method is used for __________ layers.
The Internet checksum is used by upper-layer protocols such as TCP/UDP, while the data link layer uses CRC.
A repeater is an amplifier, not a regenerator.
A repeater regenerates the original bit pattern rather than amplifying noise.
RARP stands for Reverse Address Resolution Protocol.
RARP is the Reverse Address Resolution Protocol, mapping physical to logical address.
The BNC-T connector is a T-shaped device with _______ ports.
A BNC-T has three ports: two for the cable and one for the NIC.
A periodic signal completes one cycle in 0.001 s. What is the frequency?
f = 1/T = 1/0.001 s = 1000 Hz = 1 kHz.
Zero-crossing bandwidth is also called equivalent noise bandwidth.
Zero-crossing bandwidth and equivalent noise bandwidth are different bandwidth definitions.
If FCC regulations are followed, the carrier frequencies of adjacent AM radio stations are __________ apart.
Each AM station gets a 10 kHz band.
There are __________ types of serial transmission.
Serial transmission is either asynchronous or synchronous.
Which of the following is an example of ITU-T modem standards?
ITU-T modem standards are the V-series (e.g. V.32, V.90).
In 56K modems, the maximum data rate in the uploading direction is still __________.
Upstream a 56K modem is limited to 33.6 Kbps.
All popular fiber-optic connectors are __________ shaped.
Popular fiber-optic connectors are barrel-shaped.
The VLF and LF bands use __________ propagation for communications.
VLF and LF waves follow the earth's curvature by ground propagation.
Multiplexing is the set of techniques that allows simultaneous transmission of multiple signals across a __________ data link.
Multiplexing shares one link among several signals.
The need for addressing makes asynchronous TDM inefficient for bit or byte __________.
Address overhead per slot makes asynchronous TDM wasteful for bit/byte interleaving.
A traditional telephone line has a bandwidth of __________.
Voice lines pass about 300-3300 Hz, a bandwidth of 3000 Hz.
Error control in the data link layer is based on __________.
Data link error control uses ARQ: retransmission of damaged or lost frames.
In line discipline, the initiator first transmits a frame called an __________.
The initiator sends ENQ to ask if the receiver is ready.
The primary device uses __________ to receive transmission from the secondary devices.
Polling is how the primary asks secondaries to send data.
For stop-and-wait ARQ, for N data packets sent, __________ acknowledgments are needed.
Each frame is acknowledged individually, so N frames need N ACKs.
HDLC is an acronym for __________.
HDLC stands for High-level Data Link Control.
Token Bus has no commercial application in data communications.
Token Bus (802.4) was used in factory automation, not commercial data networks.
Each station in a Token Ring regenerates the frame.
Every Token Ring station acts as a repeater, regenerating the frame.
A repeater works on the __________ layer.
A repeater only regenerates signals, a physical-layer function.
The RG number gives us information about __________.
Coaxial cables are categorized by RG (radio government) ratings.
The inner core of an optical fiber is __________ in composition.
Fiber cores are made of glass or plastic.
Which error detection method involves polynomials?
CRC treats data and divisor as polynomials.
Which error detection method involves the use of parity bits?
Both simple and two-dimensional parity checks add parity bits.
Which error detection method can detect a single-bit error?
Parity, 2-D parity and CRC all detect any single-bit error.
The Hamming code is a method of __________.
Hamming code locates the erroneous bit so it can be corrected.
Sliding window requires that data frames be transmitted __________.
Sliding window numbers frames and sends them in sequence.
IEEE divides the baseband category into ______ standards.
IEEE 802.3 baseband has five standards: 10Base5, 10Base2, 10Base-T, 1Base5 and 100Base-T.
Which of the following ___________ uses a series of filters to decompose a multiplexed signal into its constituent signals?
A demultiplexer uses filters to separate the combined signal back into its component signals.
In Fast Ethernet, data rate can be increased by _______________ collisions.
Fewer collisions mean less wasted time on the medium, so effective data rate rises.
In 4PSK each phase change represents ______ bits.
Four phases give log2(4) = 2 bits per phase change.
When a beam of light travels through media of two different densities, if the angle of incidence is greater than the critical angle, ____________ occurs.
Above the critical angle the light is totally reflected back into the denser medium.
When we talk about unguided media, usually we are referring to _________.
Unguided media transport signals through free space, usually the air.
Optical fibers are defined by the ratio of the ___________ of their core to the diameter of their cladding.
Fiber types are named by core diameter to cladding diameter ratio, e.g. 62.5/125.
Radio wave transmission utilizes ___________ different types of propagation.
The course lists five types: surface, tropospheric, ionospheric, line-of-sight and space propagation.
dB is ____________ if a signal is amplified.
Amplification makes P2 > P1, so 10 log10(P2/P1) is positive.
A prism can deflect the light depending upon the angle of ____________ and the frequency.
A prism bends light according to its angle of incidence and its frequency.
Asynchronous TDM is efficient only when the size of the time slot is kept relatively _______________
Each asynchronous TDM slot carries address overhead, so large slots keep that overhead proportionally small.
The local loop has ___________ cable that connects the subscriber telephone to the nearest end office.
The local loop is a twisted-pair line from the subscriber to the end office.
Data from a computer are ____________; the local loop handles _________ signals.
Computers produce digital data while the traditional local loop carries analog signals, hence modems.
Which error detection method uses ones complement arithmetic?
The checksum adds segments in ones complement arithmetic and complements the sum.
Error control is both error _________ and error ___________
Error control covers detecting errors and correcting them, usually by retransmission.
Addressing is not needed in _______________ configuration.
With only two devices on a point-to-point link there is no need to identify the receiver.
In a Go-Back-N ARQ, if the window size is 63, what is the range of sequence numbers?
Window size 2^m - 1 = 63 gives m = 6, so sequence numbers run 0 to 63.
A timer is set when ___________ is sent out.
The sender starts a timer for each data frame so it can retransmit if no acknowledgment arrives.
BLAST stands for:
BLAST is Blocked Asynchronous Transmission.
The HDLC ___________ field defines the beginning and end of a frame.
The 01111110 flag marks the start and end of an HDLC frame.
_________ is the access protocol used by traditional Ethernet.
Traditional Ethernet uses CSMA/CD.
Bridges can divide a large ________ into smaller segments.
Bridges split a large network into smaller segments to isolate traffic.
Like VRC, LRC and CRC, checksum is also based on _______________.
All these error-detection methods add redundant bits to the data.
Data must be converted into __________ before transmission.
Data must be transformed into signals to travel over a medium.
A parabolic dish antenna is a(n) ______________ antenna.
A parabolic dish focuses signals in a single direction.
Which multiplexing technique involves signals composed of light beams?
Wavelength-division multiplexing combines light signals of different wavelengths.
Multiplexing has long been used as an essential tool in the ____________.
Telephone companies have long multiplexed many calls over shared links.
In CRC there is no error if the remainder at the receiver is ____________
A zero remainder at the checker means the data unit is accepted.
Which error detection method can detect a burst error?
Both two-dimensional parity and CRC can detect burst errors; simple parity cannot reliably.
At the CRC checker, ________ means that the data unit is damaged.
Any nonzero remainder at the checker indicates an error.
XMODEM is a ______________ protocol designed for telephone-line communication between PCs.
XMODEM is a half-duplex stop-and-wait file transfer protocol.
Which of the following combines features of the other two?
ZMODEM combines features of XMODEM and YMODEM.
Which one of the following uses full-duplex operation with sliding window flow control?
BLAST is full duplex with sliding window flow control, unlike the stop-and-wait MODEM protocols.
The DSAP and SSAP are addresses used by _________ to identify the protocol stacks.
DSAP and SSAP in the LLC PDU identify the upper-layer protocols.
Which of the following is a bus topology LAN that uses baseband signaling and has a maximum segment length of 500 meters?
10Base5 (thick Ethernet) is a baseband bus with 500 m segments.
When a collision is detected in a network using CSMA/CD, _________
On detecting a collision the station sends a jam signal so all stations know about it.
When systems share a point-to-point link, the protocols used are called ___________ protocols.
Communication over a dedicated point-to-point link uses direct protocols.
An unauthorized user is a network ___________ issue.
Protecting against unauthorized access is a security concern.
Physical layer defines characteristics of interface between device and _________
The physical layer specifies the interface between devices and the transmission medium.
________ layer deals with syntax and semantics of information exchange.
The presentation layer handles the syntax and semantics of exchanged information.
To allow access to network resources is the function of ___________
The application layer gives users access to network resources.
EIA-449 provides much better functionality than EIA-________
EIA-449 was designed to improve on the speed and distance limits of EIA-232.
The _________ is an association that sponsors the use of infrared waves.
The Infrared Data Association (IrDA) promotes infrared communication.
The section of EM spectrum defined as Radio Communication is divided into _____ ranges called BANDS.
Radio communication (3 kHz-300 GHz) is split into eight bands, VLF through EHF.
In __________ propagation, low-frequency radio waves hug the earth.
In ground propagation low-frequency waves travel close to and follow the earth's surface.
In bit ____________, MUX adds extra bits to a device.
Bit stuffing (pulse stuffing) adds dummy bits so a slower source matches the TDM slot rate.
FTTC stands for _____________
FTTC means fiber to the curb.
In data link layer, communication requires at least ___________ devices working together.
Data link communication needs at least a sender and a receiver.
Data link control is composed of ____________ important functions.
The three functions are line discipline, flow control and error control.
YMODEM uses ITU-T CRC-_____ for error checking.
YMODEM replaced XMODEM's checksum with ITU-T CRC-16.
What is present in all HDLC control fields?
I, S and U frame control fields all contain the poll/final bit.
In FDDI, token passing is used as access method.
FDDI uses token passing on a dual fiber ring.
Headers are added at layers 1 and 7 of OSI model.
Headers are added at layers 6 to 2, not at the physical layer.
Transmission media are usually categorized as _____________
Media are classed as guided (cables) or unguided (wireless).
Radio wave and microwave frequencies range from _________.
The radio communication spectrum spans VLF (3 kHz) to EHF (300 GHz).
In ______________ method a signal can be directed in a straight line from antenna to antenna.
Line-of-sight propagation sends very high frequency signals straight between antennas.
If a T-1 carries 8000 frames, the data rate is ___________
8000 frames/s x 193 bits per frame = 1.544 Mbps.
Optical signals are multiplexed using _________ at switching office to create wider BW optical signals.
Wavelength division multiplexing combines light signals on a fiber.
Which error detection method consists of a parity bit for each unit as well as an entire data unit of parity bits?
Two-dimensional parity adds a parity bit per unit plus a row of column parity bits.
ENQ/ACK stands for ______________
ENQ is enquiry and ACK is acknowledgment.
The shortest frame in HDLC protocol is usually the _________ frame.
Supervisory frames carry no information field, so they are usually the shortest.
In a time-domain plot, the vertical axis is the measure of ________.
A time-domain plot shows amplitude on the vertical axis against time.
Which topology requires a multipoint connection?
All devices share one bus cable, a multipoint link.
If the data unit is 111111, the divisor 1010, and the remainder 110, what is the dividend at the receiver?
The receiver gets the data followed by the CRC remainder: 111111 + 110 = 111111110.
OSI standard is practically implemented in the form of internet.
The Internet is built on the TCP/IP suite, not the OSI model.
TCP/IP application layers combine the functions of OSI Application, Session and Transport layers.
The TCP/IP application layer combines OSI application, presentation and session layers; transport stays separate.
Unipolar uses one voltage level.
Unipolar encoding uses one nonzero voltage level, with zero for the other bit.
The material of core and cladding in fiber optic cable is same.
Core and cladding have different densities (refractive indices) so light reflects at the boundary.
Burst error is more easily corrected than bit error.
Burst errors change several bits, so they are harder to correct than single-bit errors.
Set of rules governing communication is known as
A protocol is the set of rules that governs data communication.
WDM (wave division multiplexing) deals with
WDM multiplexes light signals of different wavelengths on fiber.
A null modem is a
A null modem crosses interface lines to connect two DTEs directly without modems.
Router works on
Routers forward packets using network-layer addresses.
Poll/Select is a technique related with
Poll/select is the line discipline method for primary-secondary links.
In cyclic redundancy checking, what is the CRC?
The CRC is the remainder of the binary division appended to the data.
In fiber optics the signal source is ____________ waves.
Optical fiber carries signals as light.
Each device has a dedicated point-to-point line configuration only with the two devices on either side of it.
In a ring each device links point-to-point only to its two neighbors.
_______ measures the relative strengths of two signals.
The decibel expresses the ratio of two signal powers.
Which multiplexing technique shifts each signal to a different carrier frequency?
FDM modulates each signal onto its own carrier frequency.
The _____________ of a signal is the width of the frequency spectrum.
Bandwidth is the range (width) of frequencies a signal contains.
Ethernet LANs can support data rates between _____________.
The IEEE 802.3 family in the course spans 1 Mbps (1Base5) to 100 Mbps (Fast Ethernet).
Two devices have started their communications that are attached with each other through coaxial cable. What function will the coaxial cable perform here?
The cable is the transmission medium carrying signals between the devices.
___________ protocols are used in shared point to point link.
When entities share a point-to-point (or multipoint) link, they communicate with a direct protocol.
Standards restricted within an organization fall in the category of ___________ standards.
Proprietary (closed) standards are owned by and restricted to one organization.
Protocol Data Unit (PDU) at network layer is specifically named as _________.
The network-layer PDU is a packet.
EIA 449 uses __________ standards to define its electrical specifications.
EIA-449 uses RS-423 (unbalanced) and RS-422 (balanced) for its electrical specifications.
The speed of light is __________ in vacuum.
Light travels at about 3 x 10^8 m/s, i.e. 300,000 km/s.
___________ is popular for supporting higher bandwidth and higher data rate.
Optical fiber offers far higher bandwidth and data rates than copper.
_________ is a technique which specifically works on light beams.
Wave-division multiplexing combines light beams of different wavelengths on a fiber.
CSMA/CD stands for ___________.
CSMA/CD is carrier sense multiple access with collision detection.
________ is a bus topology LAN that uses baseband signaling and has a max. segment length of 500 meters.
10Base5 (thick Ethernet) is a baseband bus with 500 m segments.
In FDDI, __________ passing is used as access method.
FDDI uses token passing.
RARP stands for _____________.
RARP is the Reverse Address Resolution Protocol.
In the process of CRC, _________ has no importance and is not attached with data at sender/receiver side.
Only the remainder matters in CRC; the quotient is discarded.
At the CRC checker, ________ means that the data unit is not damaged.
A zero remainder at the checker means the data arrived intact.
At the beginning of transmission, the receiver window contains ________ spaces for frames in sliding window mechanism.
With sequence numbers 0 to n-1 the window initially holds n-1 frames.
One Multistation Access Unit can support up to ____ stations.
A Token Ring MAU supports up to eight stations.
In fiber optics, signals are transmitted in the form of ____________.
Optical fiber carries signals as light.
BNC connectors are used by ___________ cables.
BNC (bayonet) connectors are used with coaxial cable.
________ measures the relative strength of two signals or a signal at two different points.
The decibel expresses the relative power of two signals or points.
The Hamming code is used for __________.
Hamming code detects and corrects single-bit errors.
No MCQs match this filter.Is filter mein koi MCQ nahi.
Is Hamming code a technique used for error detection or error correction?
Hamming code is an error-correction technique. Redundant bits are placed at positions 1, 2, 4, 8, ... and each covers a set of bit positions; at the receiver the recomputed parity bits form a number that shows the position of a single-bit error, which is then corrected by inverting that bit. (It also detects the error.)
What are guard bands in multiplexing?
In FDM, guard bands are unused strips of bandwidth left between adjacent channels. They keep the modulated signals from overlapping and prevent interference (crosstalk) between channels.
Which types of frames are present in BSC?
BSC (Binary Synchronous Communication) uses two types of frames: data frames (carry user data, with SYN, STX/ETX and BCC fields) and control frames (carry control information such as ENQ, ACK, NAK, EOT for connection set-up, flow and error control).
What are fractional T lines?
Many users do not need the full 1.544 Mbps of a T-1 line. A fractional T-1 lets several subscribers share one T-1 line (or one subscriber rent part of it) by using some of its 24 DS-0 channels of 64 kbps each, e.g. 4 x 64 = 256 kbps. A DSU/CSU at the customer side multiplexes and demultiplexes these channels, so each user pays only for the fraction used.
What are the light sources used for optical fiber?
The light source must be a focused source: 1) LED (light-emitting diode) - cheap, but light is unfocused and spreads, so it is used for short distances; 2) ILD (injection laser diode) - gives a narrow, highly focused beam that keeps its character over long distances. A photodiode at the receiver converts the light back to electrical signals.
Write the names of the Link Access Protocols developed by ITU-T.
ITU-T developed the LAP family, based on HDLC: LAPB (Link Access Procedure, Balanced - used in X.25), LAPD (Link Access Procedure for D-channel - used in ISDN), LAPM (Link Access Procedure for Modems), and LAPX (half-duplex LAPB used in teletex).
What is stop-and-wait ARQ in error control?
Stop-and-wait ARQ is stop-and-wait flow control extended with retransmission. The sender sends one frame and keeps a copy until it is acknowledged; it starts a timer. Frames and ACKs are numbered alternately 0 and 1. If the receiver gets a damaged frame it discards it (or sends NAK); if the timer expires because the frame or its ACK was lost, the sender resends the copy. Duplicate frames are recognized by their sequence number and discarded.
What is interleaving?
Interleaving is the way synchronous TDM builds frames: the multiplexer acts like a rotating switch that takes one unit (bit, byte or block) from each input device in turn and places it in the frame, then repeats. At the receiver the demultiplexer distributes the units to the outputs in the same order.
What is DSU in terms of digital services?
A DSU (Digital Service Unit) is a device that connects a user's digital equipment to a digital line (e.g. switched/56 or DDS). It changes the user's digital data rate to the rate of the service and encodes the data in the format used by the carrier, in place of a modem; it is often combined with a CSU.
What is a spike in noise terms?
A spike is impulse noise: a sudden, short, very high-energy signal (from e.g. power lines or lightning) that lasts a very short time and can corrupt many bits of data.
What is an even parity generator in the VRC error detection mechanism?
In VRC a parity bit is added to each data unit. The even parity generator counts the 1s in the data unit; if the number is odd it adds a parity bit of 1, if even it adds 0, so the total number of 1s (including parity) is even. E.g. 1100001 -> parity 1 -> 11000011. The receiver's checker counts 1s and rejects the unit if the count is odd.
What is the formula to calculate the number of redundancy bits required to correct a bit error in a given number of data bits?
For m data bits and r redundancy bits (Hamming code), r must satisfy 2^r >= m + r + 1. The r bits must be able to point to any of the m + r bit positions plus the 'no error' state. Example: m = 7 gives r = 4, since 2^4 = 16 >= 7 + 4 + 1 = 12.
What is the RG rating of coaxial cable?
Coaxial cables are classified by Radio Government (RG) ratings. Each RG number describes a set of physical specifications (inner conductor gauge, insulation, shield, jacket) suited to a particular use. Examples: RG-8, RG-9 and RG-11 are used in thick Ethernet; RG-58 is used in thin Ethernet; RG-59 is used for cable TV.
What are the advantages of thin Ethernet?
Thin Ethernet (10Base2, Cheapernet) costs less than thick Ethernet and is easier to install. Its thin coaxial cable is lighter and more flexible, so it bends easily, and stations connect with simple BNC-T connectors instead of transceivers and drop cables.
What is the difference between a unicast, multicast, and broadcast address?
A unicast address identifies one single destination, so the frame or packet goes from one sender to exactly one receiver. A multicast address identifies a group of stations; one copy is sent and every member of that group receives it. A broadcast address (all 1s, e.g. FF:FF:FF:FF:FF:FF) means every station on the network receives the frame.
T lines are designed for digital data. How can they be used for analog transmission?
The analog signal (for example voice) is first sampled and digitized with PCM: 8000 samples per second at 8 bits each gives a 64 Kbps DS-0 stream. These digital streams are then combined by TDM onto the T line (24 channels on a T-1). At the receiving end the samples are demultiplexed and converted back to analog.
What are the three types of guided media?
Guided media give signals a physical path from one device to another. The three types are: (1) twisted-pair cable, which carries electrical signals; (2) coaxial cable, which carries electrical signals; (3) fiber-optic cable, which carries signals as light.
What are the conditions for the polynomial used by the CRC generator?
A good CRC generator polynomial must (1) not be divisible by x, which guarantees detection of all burst errors whose length is equal to or less than the degree of the polynomial; and (2) be divisible by (x + 1), which guarantees detection of all burst errors that affect an odd number of bits.
What are intelligent modems?
Intelligent (smart) modems contain software that lets them accept and carry out commands, for example the Hayes AT command set, as well as transfer data. They can do tasks such as automatic dialing, automatic answering, redialing, choosing a speed and error-correction mode, and reporting status, without manual operation.
What is the basic purpose of a router?
A router works at the network layer and connects separate networks (for example a LAN to a WAN or the Internet). It reads the destination logical (IP) address of each packet, chooses the best path using its routing table, and forwards the packet to the next router or host on that path.
What is a Multistation Access Unit (MAU) in Token Ring?
In Token Ring, stations are connected through a MAU instead of directly to each other. The MAU is a hub that contains individual automatic switches, one per station. If a station fails or is disconnected, its relay closes and bypasses it, so the ring stays complete. One MAU supports up to 8 stations. The cabling looks like a star, but logically the network is still a ring.
What kind of error is undetectable by the checksum?
Errors that cancel each other out in the one's complement sum are not detected. For example, if a bit in one segment changes from 1 to 0 and the bit in the same position of another segment changes from 0 to 1, the sum stays the same. More generally, any combination of errors that keeps the sum unchanged (for example, segments swapped in order) is not detected.
What are the properties of signals?
A periodic signal is described by three properties: amplitude (the signal's strength or height, in volts or watts), frequency (cycles per second in Hz, equal to 1/period), and phase (the position of the waveform relative to time zero, in degrees or radians).
In asynchronous or synchronous TDM, where is addressing used?
Addressing is used in asynchronous (statistical) TDM. Its slots are not tied to fixed input lines, so each slot carries an address that tells the demultiplexer where to deliver it. Synchronous TDM needs no address because each slot position always belongs to the same device.
Why do we need a null modem?
A null modem connects two DTEs (for example two computers) directly without modems or a telephone line. Both DTEs transmit on the same pin and receive on the same pin, so a straight cable would connect transmit to transmit. The null modem crosses the transmit and receive lines and the control signals (such as RTS/CTS and DTR/DSR), so each DTE sees the other as a DCE. It is used for short-distance links such as file transfer between nearby PCs.
Count LRC for the following bits: 10011010 10100101 11010110
Arrange the bytes in rows and compute even parity for each column (XOR of the column): 10011010 XOR 10100101 = 00111111; 00111111 XOR 11010110 = 11101001. LRC (even parity) = 11101001, so the block sent is 10011010 10100101 11010110 11101001. (If odd parity were required, the LRC would be the complement, 00010110.)
What are the categories of multiplexing?
There are three basic categories. (1) Frequency-division multiplexing (FDM): analog; each signal gets its own frequency band. (2) Wave-division multiplexing (WDM): analog; used on fiber, each signal gets its own light wavelength. (3) Time-division multiplexing (TDM): digital; each signal gets time slots. TDM is further divided into synchronous TDM and asynchronous (statistical) TDM.
What are the three purposes of control frames at the data link layer?
In BSC, control frames are used for (1) establishing connections (line discipline, e.g. ENQ/ACK or poll/select); (2) maintaining flow and error control during data transfer (ACK/NAK and similar responses); and (3) terminating connections (e.g. EOT).
What are the advantages of a multipoint connection over a point-to-point connection?
In a multipoint connection more than two devices share one link. Advantages: fewer cables and I/O ports are needed, so it is cheaper and easier to install. The link capacity is shared and better used, because idle devices do not tie up a dedicated line. New devices can be added without new links to every other device.
What is the name of the telephone service in which there is no need of dialing?
A leased (dedicated) line service, in which the line is permanently connected between two customers, so no dialing is needed. Examples: the analog leased line service, and its digital version, Digital Data Service (DDS).
What methods of line discipline are used for peer-to-peer and primary-secondary communication?
ENQ/ACK (enquiry/acknowledgment) is used for peer-to-peer communication, where either device can start a session. Poll/Select is used for primary-secondary communication, where the primary controls the link by polling or selecting secondaries.
How does the checksum checker know that the received data unit is undamaged?
The checker divides the received unit, including the checksum, into n-bit segments and adds them using one's complement arithmetic. It then complements the sum. If the result is all 0s (the sum before complementing is all 1s), the data unit is accepted as undamaged. Any non-zero result means an error, and the unit is rejected.
Which one has more overhead, a repeater or a bridge? Explain your answer.
A bridge has more overhead. A repeater works only at the physical layer and just regenerates the signal bit by bit. A bridge works at both the physical and data link layers. It must regenerate the signal and also read each frame's physical (MAC) address, look it up in its table, and decide whether and to which port to forward it. So a bridge does everything a repeater does and more.
Write down disadvantages of ring topology.
(1) Traffic is unidirectional, so data may pass through many stations. (2) A break in the ring or a disabled station can disable the whole network (fixed by a dual ring or a switch/MAU that closes off the break). (3) Adding or removing a device disturbs the ring, and the number of devices and the ring length are limited. (4) The token or frame must pass through every station, which adds delay.
How are parity bits counted in VRC error detection in the case of an odd parity generator?
With odd parity the generator counts the 1s in the data unit and adds a parity bit so that the total number of 1s, including the parity bit, is odd. If the count is already odd, the parity bit is 0; if it is even, the parity bit is 1. Example: data 1100001 has three 1s (odd), so the parity bit is 0 and 11000010 is sent. Data 1100011 has four 1s (even), so the parity bit is 1 and 11000111 is sent. The receiver counts the 1s in all 8 bits. An odd count is accepted; an even count shows an error, and the unit is rejected.
What is hybrid topology?
A hybrid topology combines two or more basic topologies in one network. Example: a main star topology in which each branch connects several stations in a bus, or several star networks joined by a bus backbone. It gives flexibility to join networks of different types.
What is a combined station in HDLC?
A combined station can act as both a primary and a secondary station: it can send commands and responses. Combined stations are used in balanced (peer-to-peer) configurations, such as asynchronous balanced mode (ABM).
How are bit rate and baud rate related?
Bit rate is the number of bits sent per second; baud rate is the number of signal elements (signal changes) per second. Bit rate = baud rate x number of bits carried by each signal element (log2 L for L signal levels). So bit rate >= baud rate. Example: 8-QAM at 1000 baud carries 3 bits per baud, so the bit rate is 3000 bps; in binary ASK/FSK, bit rate = baud rate.
What do the following abbreviations stand for: ARP, RARP, ICMP?
ARP = Address Resolution Protocol (finds the physical address for a known IP address). RARP = Reverse Address Resolution Protocol (finds the IP address for a known physical address). ICMP = Internet Control Message Protocol (sends error and control messages, e.g. echo request/reply used by ping).
Differentiate between polling and selecting.
Both are used in primary-secondary (poll/select) line discipline. Polling: the primary wants to receive data. It asks each secondary in turn whether it has anything to send; the secondary replies with data or with NAK if it has none. Selecting: the primary wants to send data. It sends a select (SEL) frame with the intended secondary's address, waits for an ACK showing the secondary is ready, and then sends its data. Polling moves data from secondary to primary; selecting moves data from primary to secondary.
In 1Base5 (StarLAN), how can the range be increased?
StarLAN (1Base5) links each station to a hub with twisted pair over a limited distance, about 250-500 m. The range can be increased by daisy-chaining: hubs (and stations) are linked in a chain or tree, with hubs connected to higher-level hubs, so the network covers a larger area. The limit is a few levels, e.g. up to 5 hubs in a path.
For the two groups of bits, write the polynomials: (a) 10111 (b) 00101
Each bit is the coefficient of a power of x, with the leftmost bit as the highest power. (a) 10111 = x^4 + x^2 + x + 1. (b) 00101 = x^2 + 1.
Using Shannon's capacity formula, find the capacity of a channel whose bandwidth is 3000 Hz when the signal-to-noise ratio is zero.
C = B log2(1 + SNR) = 3000 x log2(1 + 0) = 3000 x log2(1) = 3000 x 0 = 0 bps. When the signal is so weak that the SNR is (almost) zero, the channel cannot carry any data, whatever its bandwidth.
What are the types of asynchronous protocols used by modems?
Common asynchronous data link protocols for modem file transfer are XMODEM, YMODEM, ZMODEM, BLAST (Blocked Asynchronous Transmission) and Kermit.
What are the sub-layers of the data link layer?
In the IEEE 802 model the data link layer is split into two sub-layers: Logical Link Control (LLC), which handles flow and error control and the interface to the network layer, and Media Access Control (MAC), which handles access to the shared medium (e.g. CSMA/CD, token passing) and framing with physical addresses.
Explain the voice channels of T-1 and T-2.
T-1 (DS-1) multiplexes 24 voice channels of 64 Kbps each; with framing, its rate is 1.544 Mbps. T-2 (DS-2) multiplexes 96 voice channels (four DS-1s); with overhead, its rate is 6.312 Mbps.
Explain baseband and broadband.
Baseband: the digital signal is sent directly on the medium without modulation, and one signal uses the whole bandwidth of the cable. IEEE baseband standards include 10Base5, 10Base2, 10BaseT, 1Base5 and 100BaseT. Broadband: the signal is modulated onto a carrier, so the transmission is analog, and the bandwidth can be divided by FDM into several channels. IEEE defines one broadband standard, 10Broad36 (10 Mbps over 3600 m).
What are the segments, types and speed of Fast Ethernet?
Fast Ethernet (IEEE 802.3u) runs at 100 Mbps using CSMA/CD with a smaller collision domain. Implementations: 100Base-TX (two pairs of Cat 5 UTP or STP), 100Base-FX (two optical fibers), and 100Base-T4 (four pairs of Cat 3 or higher UTP). Twisted-pair segments are limited to 100 m from station to hub/switch; fiber segments can be longer, e.g. about 2 km in full duplex.
Explain sliding window.
Sliding window flow control lets the sender transmit several frames before it needs an acknowledgment. Frames are numbered modulo 2^m. The sender keeps a window of frames it may send (at most 2^m - 1 for go-back-N). Each frame sent shrinks the window from the left; each ACK received slides it to the right by the number of frames acknowledged. The receiver window shows which frames it can accept. It shrinks as frames arrive and grows when ACKs are sent. An ACK carries the number of the next frame expected and can acknowledge several frames at once. This makes much better use of the link than stop-and-wait.
What types of Ethernet cables/wires connect with a computer?
Thick coaxial cable (10Base5, attached through a transceiver/AUI drop cable); thin coaxial cable (10Base2, with BNC-T connectors); unshielded twisted pair (10BaseT/100Base-TX, with RJ-45 connectors to a hub or switch); and fiber-optic cable (10Base-F/100Base-FX).
In 56K modems, what are the maximum uploading and downloading data rates? What is the limiting factor in downloading?
Upload: up to 33.6 Kbps. Download: up to 56 Kbps. Upload is lower because the analog signal is sampled at the telephone switching office, which adds quantization noise. In download the data are already digital, so there is no quantization noise. The limit comes from the telephone company's PCM digital channel: 8000 samples/s with 7 usable bits per sample (one bit is used for control) = 8000 x 7 = 56 Kbps.
List the types of TDM.
Synchronous TDM (each input has a fixed slot in every frame) and asynchronous or statistical TDM (slots are allocated dynamically to active inputs and carry addresses).
Fill up the table for HDLC NRM and ARM: station type and initiator.
NRM (normal response mode): station types are one primary and one or more secondaries; only the primary can start a transmission, and a secondary sends only after the primary gives permission (polls it). ARM (asynchronous response mode): station types are again a primary and secondaries; either can start, because a secondary may start a transmission without permission, but the primary remains responsible for the line.
Find the polynomials for the data groups 10111 and 0010.
10111 = x^4 + x^2 + x + 1. 0010 = x (only the x^1 position is 1).
A company is employing synchronous TDM. (a) Why are time slots wasted? (b) What is the solution to this problem?
(a) In synchronous TDM every device gets a fixed slot in every frame, whether or not it has data. When a device is idle its slot is sent empty, so link capacity is wasted. (b) Use asynchronous (statistical) TDM: slots are given dynamically only to devices that have data, with an address in each slot, so few slots go empty. The link can even have fewer slots than devices.
In a 50/125 optical fiber, which value represents the diameter of the core and which the cladding?
50 is the core diameter and 125 is the cladding diameter, both in micrometres (microns).
If five channels, each of 100 kHz, are to be multiplexed with guard bands of 10 kHz between them, what is the minimum bandwidth required?
Five channels need 4 guard bands between them. Bandwidth = 5 x 100 kHz + 4 x 10 kHz = 500 + 40 = 540 kHz.
From the electromagnetic spectrum, write the band range of radio communication.
Radio communication covers 3 kHz to 300 GHz, divided into eight bands: VLF 3-30 kHz, LF 30-300 kHz, MF 300 kHz-3 MHz, HF 3-30 MHz, VHF 30-300 MHz, UHF 300 MHz-3 GHz, SHF 3-30 GHz, EHF 30-300 GHz.
Suppose a company ABC needs fast and reliable data communication in its network. The company is willing to spend as much as a fast and reliable network demands. Considering the situation, suggest the suitable guided medium.
Fiber-optic cable. It has the highest bandwidth and data rates of the guided media, very low attenuation over long distances, immunity to electromagnetic interference and noise, and better security. Its higher cost and harder installation are acceptable here.
How many bits can Hamming code correct at a time?
One bit. The standard Hamming code locates and corrects a single-bit error in a data unit (with an extra parity bit it can also detect, but not correct, double-bit errors).
The section of the electromagnetic spectrum is divided into several ranges called BANDS. Write the complete names of only 3 of those bands which are associated with radio communication.
Any three of: Very Low Frequency (VLF), Low Frequency (LF), Middle Frequency (MF), High Frequency (HF), Very High Frequency (VHF), Ultra High Frequency (UHF), Super High Frequency (SHF), Extremely High Frequency (EHF).
During data transmission in a network, there is a need of error control. At which OSI layer will it be implemented? Also mention the names of subcategories of sliding window ARQ error control mechanism.
Error control (with flow control) is implemented at the data link layer (layer 2), and also end-to-end at the transport layer. Sliding window ARQ has two subcategories: Go-Back-N ARQ and Selective-Reject (Selective Repeat) ARQ.
Suppose Saim Electronics Company wants a Gigabit Ethernet computer network for better bandwidth, less congestion and overall better performance. Write the names of common implementations of Gigabit Ethernet which can be used for this company.
1000Base-SX (short-wave laser over multimode fiber), 1000Base-LX (long-wave laser over single-mode or multimode fiber), 1000Base-CX (shielded copper twisted-pair jumpers), and 1000Base-T (four pairs of Category 5 UTP).
What is the difference between guided and unguided media?
Guided (bounded/wired) media provide a physical conduit from one device to another, so the signal is confined to and directed along a specific path, e.g. twisted-pair, coaxial and fibre-optic cable. Unguided (wireless/unbounded) media carry electromagnetic waves through air, water or vacuum without any physical conductor to guide them, e.g. radio waves, microwaves, infrared and satellite links; the signal is broadcast and available to anyone with a suitable receiver.
What is the difference between FDM and TDM?
FDM (Frequency Division Multiplexing) is an analog technique: the link bandwidth is divided into frequency bands (channels), each signal modulates a different carrier, and all signals are sent simultaneously, separated by guard bands. TDM (Time Division Multiplexing) is a digital technique: the full link capacity is shared in time; each source gets time slots in turn and data units are interleaved into frames. FDM needs link bandwidth greater than the combined signal bandwidths; TDM needs link data rate greater than the combined data rates. FDM example: radio/TV broadcasting, 1G analog cellular; TDM example: T-1/E-1 lines, 2G digital cellular.
What is power bandwidth?
In the course (signal spectrum context), power bandwidth is the frequency band in which 99% of the total power of the signal resides. (In amplifier terms it is the frequency range over which the amplifier can deliver at least half of its rated output power without excessive distortion.)
What does the CRC generator append to the data unit?
It appends the CRC: a string of redundant bits equal to the remainder of a modulo-2 (binary) division. The generator first appends n zeros to the data (n = one less than the number of bits in the divisor), divides by the predetermined divisor, and the n-bit remainder replaces those zeros, so the resulting bit sequence is exactly divisible by the divisor.
How much bandwidth is required for a modem in the case of FSK?
Minimum bandwidth for FSK = baud rate + frequency shift, i.e. BW = (f_c1 - f_c0) + N_baud. On a voice-grade telephone line this must fit in about 3 kHz (300-3300 Hz), and half-duplex vs full-duplex modems divide it accordingly.
What is the difference between the angle of incidence and the angle of reflection?
Both are measured from the normal (the line perpendicular to the boundary). The angle of incidence is the angle between the incoming (incident) ray and the normal; the angle of reflection is the angle between the reflected ray and the normal. By the law of reflection they are equal. (Refraction, by contrast, bends the ray when it crosses into a medium of different density.)
What is daisy chaining in 1Base5 (StarLAN)?
1Base5 StarLAN (1 Mbps, twisted pair, star topology) allows daisy chaining: instead of connecting every station to the hub, several stations (up to about ten) are linked one after another in a chain, and only the first station of the chain is connected to the hub. This reduces cabling and lets more stations share a hub port; data passes from station to station along the chain.
What is the responsibility of the Application layer?
The application layer enables the user, whether human or software, to access the network. It provides the user interface and support for services such as electronic mail, remote file access and transfer (FTAM), shared database management, directory services and network virtual terminal.
What is the critical angle?
When light moves from a denser to a less dense medium, increasing the angle of incidence increases the angle of refraction. The critical angle is the angle of incidence at which the refracted ray lies along the boundary (angle of refraction = 90 degrees). For incidence angles greater than the critical angle the light is totally reflected back into the denser medium, which is the principle used in optical fibre.
Is the VRC error detection method used for single-bit errors or burst errors?
VRC (parity check) detects all single-bit errors. It can also detect burst errors only when the total number of changed bits is odd; it cannot detect errors where an even number of bits are changed.
Which modem was first developed commercially in the 1970s?
The Bell modems, developed by the Bell Telephone Company in the early 1970s, were the first commercial modems (e.g. Bell 103/113, 202, 212).
A major telecom company uses RZ encoding for its signal conversion. What will be the major problem faced with this type of encoding?
RZ uses three levels and returns to zero in the middle of every bit, so each bit needs two signal changes. It solves the synchronisation problem (long runs of 0s or 1s no longer cause loss of timing), but it occupies more bandwidth; this higher bandwidth requirement is the main drawback.
What are geosynchronous satellites?
Line-of-sight propagation requires sending and receiving antennas to stay locked on each other, so for continuous communication a satellite must move at the same angular speed as the Earth's rotation, so that it appears fixed relative to a point on Earth. Such satellites are called geosynchronous (geostationary) satellites; they orbit at about 36,000 km above the equator, and three of them can cover almost the whole Earth.
Which LAN architecture was developed by ITU-T and ANSI?
FDDI (Fiber Distributed Data Interface), a 100 Mbps token-passing dual-ring LAN standardised by ANSI and ITU-T.
Why is addressing required in the Poll/Select method and not in the ENQ/ACK method?
Poll/Select is used on multipoint links with one primary and several secondaries; for the primary to identify and communicate with a specific secondary (and for a secondary to know a frame is for it), each frame must carry an address. ENQ/ACK is used on point-to-point links with only two devices, so any transmission can only be for the other device and no addressing is needed.
What do you know about ITU-T modems?
ITU-T (formerly CCITT) modem standards are the V-series, today's most popular modem standards. Some are compatible with Bell modems (e.g. V.21 like Bell 103, V.22 like Bell 212); others include V.22bis (2400 bps), V.32 (9600 bps, trellis-coded QAM), V.32bis (14.4 Kbps), V.33, V.34 (28.8/33.6 Kbps) and V.90 (56 Kbps downstream). Other V-series modems listed in the course include V.23, V.26, V.27 and V.29.
What do the abbreviations ARP, RARP and ICMP stand for?
ARP: Address Resolution Protocol. RARP: Reverse Address Resolution Protocol. ICMP: Internet Control Message Protocol.
What are the two major classes of synchronous protocols at the data link layer?
Character-oriented (byte-oriented) protocols and bit-oriented protocols.
Define multiplexing. What is its advantage?
Multiplexing is the set of techniques that allows simultaneous transmission of multiple signals across a single data link. Advantage: multiple users share the total capacity of one transmission medium, reducing cost and using the link bandwidth efficiently.
What is the purpose of a dual ring?
In a ring, traffic is unidirectional and a single break can disable the network. A dual ring adds a second ring carrying traffic in the opposite direction, so if one ring or station fails the other ring can be used (or the rings wrapped) to keep the network working.
Write any two functions of the physical layer.
Any two of: it defines the characteristics of the interface between the device and the transmission medium; it defines the type of transmission medium; it handles representation of bits (encoding into signals) and the data rate; it is concerned with line configuration (point-to-point or multipoint), physical topology and transmission mode (simplex, half/full duplex); bit synchronisation.
Define high frequency (HF) and super high frequency (SHF). Which devices use these frequencies?
HF (3-30 MHz) uses ionospheric (sky) propagation: waves move up into the ionosphere, where the density difference reflects them back to Earth. Used for citizens band radio, international broadcasting, military communication, long-distance aircraft/ship communication, telephone, telegraph and fax. SHF (3-30 GHz) is transmitted mostly by line-of-sight and some space propagation. Used for terrestrial and satellite microwave and radar.
Write all steps of the checksum method.
Sender: (1) divide the data unit into k equal segments of n bits (usually 16); (2) add the segments using one's complement arithmetic; (3) complement the sum - this is the checksum; (4) append the checksum to the data and send. Receiver: (5) divide the received unit into n-bit segments and add all of them (including the checksum) using one's complement; (6) complement the result; (7) if the result is zero the data is accepted, otherwise it contains an error and is rejected.
Differentiate between an internet and the Internet.
An internet (lowercase) is a generic term for any interconnection of individual networks, built with connecting devices such as routers and gateways. The Internet (uppercase) is the name of one specific worldwide network of networks that uses the TCP/IP protocol suite.
Is addressing used in asynchronous or synchronous TDM?
Addressing is used only in asynchronous (statistical) TDM, because slots are allocated dynamically; in synchronous TDM each device's position in the frame identifies it, so no address is needed.
What are the troposphere and the ionosphere?
Troposphere: the lowest layer of the atmosphere, from the earth's surface up to about 30 miles (45 km); it contains air, clouds and weather and is where line-of-sight and low-level sky propagation take place. Ionosphere: the layer above the troposphere and below space; it contains free electrically charged particles (ions). Sky-propagated signals (e.g. MF/HF) are sent upward and are reflected back to earth by the ionosphere, allowing long-distance communication with low power.
What is CSMA/CD? Why is it needed?
CSMA/CD (Carrier Sense Multiple Access with Collision Detection) is the access method of traditional Ethernet. Need: on a shared bus many stations use one medium; if two send at once their signals collide and the data are destroyed, so a method is needed to control access and handle collisions. Working: 1) Carrier sense - a station listens to the medium before sending; if it is busy it waits. 2) Multiple access - if idle, any station may send. 3) Collision detection - while sending it keeps listening; because of propagation delay two stations may start at nearly the same time and collide. 4) On detecting a collision the station stops and sends a short jam signal so all stations know. 5) Each station waits a random backoff time (binary exponential backoff) and tries again. The minimum frame length must be such that a collision is detected before the frame ends. This reduces wasted capacity and gives fair access without a central controller.
What are the asynchronous protocols in the data link layer?
Asynchronous protocols treat each character as independent; they are used mainly in modems and are slow but simple. Main ones: 1) XMODEM - half-duplex stop-and-wait file transfer; frame = SOH, 2 sequence-number bytes (number and its complement), 128-byte data, 1-byte checksum/CRC; ACK, NAK and CAN control. 2) YMODEM - like XMODEM but 1024-byte data unit, 2-byte CRC, and several files can be sent in one session. 3) ZMODEM - combines features of XMODEM and YMODEM (streaming, crash recovery). 4) BLAST (Blocked Asynchronous Transmission) - full duplex with sliding-window flow control; transfers data and binary files. 5) Kermit (Columbia University) - file transfer protocol that turns control characters into printable characters so they pass through the network; can also work as terminal emulation.
What is frequency division multiplexing?
FDM is an analog multiplexing technique used when the bandwidth of the link is greater than the combined bandwidths of the signals. Each input signal modulates a different carrier frequency, so each occupies its own channel (frequency range). The modulated signals are added into one composite signal and sent over the link. Channels are separated by guard bands to avoid overlap. At the receiver, band-pass filters split the composite signal and each signal is demodulated back. Example: AM/FM radio and cable TV, telephone analog hierarchy (groups, supergroups).
Find the binary equivalent of the polynomial x^8 + x^3 + x + 1.
Write a bit for each power from x^8 down to x^0: x^8=1, x^7=0, x^6=0, x^5=0, x^4=0, x^3=1, x^2=0, x^1=1, x^0=1. Binary equivalent = 100001011 (9 bits).
How does the Go-Back-N ARQ method of error control work 1) when a data frame is lost, 2) when an acknowledgement is lost? Explain with the help of a simple diagram.
Go-Back-N: the sender may send several frames (up to the window size) before an ACK; it keeps copies and a timer. The receiver accepts frames only in order. 1) Lost (or damaged) data frame: suppose frames 0,1,2,3 are sent and frame 2 is lost. The receiver gets 0 and 1 and ACKs them; when frame 3 arrives out of order it discards it (and may send NAK 2). The sender, on NAK 2 or timer expiry for frame 2, goes back and resends frame 2 and every frame sent after it (2, 3, ...). Diagram: S: 0 1 2x 3 | R: ACK1, ACK2(expect 2), discard 3 | S: resend 2 3. 2) Lost acknowledgement: ACKs are cumulative, so if ACK for frame 1 is lost but ACK 3 arrives later, frames 1-2 are also confirmed and nothing is resent. If no later ACK arrives before the timer expires, the sender goes back and resends all frames from the oldest unacknowledged frame; the receiver discards the duplicates (recognized by sequence number) and re-sends the ACK.
Apply the Hamming code algorithm to transmit the seven-bit data 1101010. Invert the 6th bit of the transmitted data and prove it at the receiving end. Bits used for each r value: r1: bits 1, 3, 5, 7, 9, 11; r2: bits 2, 3, 6, 7, 10, 11; r4: bits 4, 5, 6, 7; r8: bits 8, 9, 10, 11.
Place data in positions 11,10,9,7,6,5,3 = 1,1,0,1,0,1,0 (r at 8,4,2,1). Even parity: r1 (3,5,7,9,11 = 0,1,1,0,1, three 1s) = 1; r2 (3,6,7,10,11 = 0,0,1,1,1, three 1s) = 1; r4 (5,6,7 = 1,0,1, two 1s) = 0; r8 (9,10,11 = 0,1,1, two 1s) = 0. Transmitted (11..1): 1 1 0 0 1 0 1 0 0 1 1 = 11001010011. Invert bit 6 (0->1): received 11001110011. Receiver check: r1 over 1,3,5,7,9,11 = 1,0,1,1,0,1 -> four 1s -> 0; r2 over 2,3,6,7,10,11 = 1,0,1,1,1,1 -> five -> 1; r4 over 4,5,6,7 = 0,1,1,1 -> three -> 1; r8 over 8,9,10,11 = 0,0,1,1 -> two -> 0. Syndrome r8 r4 r2 r1 = 0110 = 6, so bit 6 is in error; inverting it gives back 11001010011 and data 1101010.
An analog signal carries 7 bits in each signal element. If 1550 signal elements are sent per second, find the baud rate and bit rate.
Baud rate = number of signal elements per second = 1550 baud. Bit rate = baud rate x bits per signal element = 1550 x 7 = 10850 bps.
Apply the checksum error detection technique at the sending end to the following two bit streams. If there is no error, prove at the receiving end that the pattern is OK. a) 1110101 b) 1010010
Sender (7-bit one's complement): 1110101 + 1010010 = 11000111; wrap the carry: 1000111 + 1 = 1001000. Checksum = complement = 0110111. Sent: 1110101, 1010010, 0110111. Receiver: 1110101 + 1010010 + 0110111 = 11111110; wrap carry: 1111110 + 1 = 1111111. Complement = 0000000, so the data are accepted (no error).
a) Encode the bit stream 1 0 1 0 1 0 1 0 1 0 using Return to Zero (RZ) encoding. b) Encode the bit stream 1 1 0 0 1 1 0 0 1 1 using Alternate Mark Inversion (AMI) encoding.
a) RZ uses +V, 0 and -V; the signal returns to zero in the middle of every bit. Bit 1 = +V for the first half then 0; bit 0 = -V for the first half then 0. For 1010101010: (+,0)(-,0)(+,0)(-,0)(+,0)(-,0)(+,0)(-,0)(+,0)(-,0). b) AMI (bipolar): 0 = zero voltage; each 1 is a pulse of opposite polarity to the previous 1. For 1100110011 (first 1 taken as +): +V, -V, 0, 0, +V, -V, 0, 0, +V, -V.
Why do we need inverse multiplexing?
Inverse multiplexing is the opposite of multiplexing: it takes the data stream from one high-speed line and splits it into portions that are sent at the same time over several lower-speed lines; the receiver puts them back together. It is needed because an organization's bandwidth demand varies. For example, it may sometimes need to send voice, sometimes data and sometimes video. Leasing one large line (such as a 1.544 Mbps T-1) for the peak need wastes money most of the time, and a single low-rate line is too slow for video. With an inverse multiplexer the organization can lease several separate low-rate channels and use as many as each job needs: one channel for voice, several channels together for data or video. This gives bandwidth on demand at lower cost, and the remaining links keep working if one fails.
Describe the method of checksum briefly.
Sender: the data unit is divided into k segments of n bits (usually 16). The segments are added using one's complement arithmetic, so any end-around carry is added back into the sum. The sum is complemented and the result is the checksum. The checksum is sent along with the data. Receiver: the data and checksum are again divided into n-bit segments and added using one's complement arithmetic. The result is then complemented. If the result is all 0s, the data is accepted; otherwise it contains an error and is rejected.
Explain Asynchronous Time Division Multiplexing in detail. Also discuss its advantages over synchronous TDM.
Synchronous TDM gives every input device a fixed slot in every frame, so a frame has as many slots as there are devices. If a device has nothing to send, its slot goes out empty and capacity is wasted. Asynchronous (statistical) TDM allocates slots dynamically. The number of slots in a frame (m) is less than the number of input lines (n). The multiplexer scans the input lines in turn and puts data only from lines that have data into the next free slot until the frame is full. If fewer lines are active, the frame is sent partly filled or shortened. Because a slot is no longer tied to a fixed line, each slot must carry an address that tells the demultiplexer where its data belongs. Buffers are needed to hold data that arrives in bursts. Advantages over synchronous TDM: (1) no slots are wasted on idle devices, so link utilization is much higher; (2) the link capacity can be lower than the sum of all input rates, because not all devices transmit at once, which lowers cost; (3) more devices can share a link of the same capacity; (4) variable-length slots can be used. Trade-offs: the address overhead makes it inefficient for bit or byte interleaving, so slots should be relatively large, and it needs buffering and more complex multiplexers.
Give characteristics of dual ring; if necessary, draw the diagram.
A dual ring uses two rings that connect every node: a primary ring and a secondary ring. Characteristics: (1) The two rings carry data in opposite directions. (2) Normally the primary ring carries traffic and the secondary ring is a backup. (3) If a link or station fails, the nodes next to the fault wrap the primary ring onto the secondary ring, forming one loop. Traffic keeps flowing, so the network tolerates a single break. (4) This removes the main weaknesses of a single ring: one-way traffic and one break stopping the whole network. (5) It needs twice the cabling and extra hardware. FDDI is an example. Diagram: nodes arranged in a circle, with an inner and an outer ring linking each pair of neighbours and arrows in opposite directions on the two rings.
What will the receiver receive if the checksum method is applied to the following bits: 10101001 00111001?
Add the two segments using one's complement arithmetic: 10101001 + 00111001 = 11100010 (no carry out). Checksum = complement of the sum = 00011101. Transmitted (and received, if there is no error) data: 10101001 00111001 00011101. Receiver check: 10101001 + 00111001 + 00011101 = 11111111, and its complement is 00000000, so the data is accepted.
What is refraction in terms of optical fiber? Give one example.
Light travels in a straight line through a uniform medium. When it passes into a medium of different density, its speed changes and it bends; this is refraction. Going from a denser medium (core) to a less dense medium (cladding), light bends away from the normal. As the angle of incidence increases, it reaches the critical angle, where the refracted ray runs along the boundary. Beyond it, the light is totally reflected. Optical fiber uses this: a glass core of higher refractive index surrounded by cladding of lower index keeps light inside by reflection whenever it hits the boundary above the critical angle. Example: a straw in a glass of water looks bent at the water surface because light refracts there.
What are the asynchronous protocols in data link layer? Discuss in detail with examples.
Asynchronous data link protocols treat each character as an independent unit framed by start and stop bits; they are used mainly for modem file transfer and are being replaced by faster synchronous protocols. Main examples: (1) XMODEM (Ward Christensen, 1979): half-duplex stop-and-wait ARQ. Frame = 1-byte SOH, 2-byte header (sequence number and its 1's complement), 128 bytes of data, 1-byte checksum (later CRC). ACK, NAK and CAN control the transfer. (2) YMODEM: like XMODEM but uses 1024-byte data units, two CAN characters to abort, ITU-T CRC-16 for error checking, and can send multiple files in one session. (3) ZMODEM: combines features of XMODEM and YMODEM and adds streaming with sliding windows, crash recovery and better speed. (4) BLAST (Blocked Asynchronous Transmission): full-duplex with sliding-window flow control, more powerful than XMODEM. (5) Kermit (Columbia University): widely used file transfer protocol that turns control characters into printable characters so they can pass through any system; also a terminal emulation program.
Compare the line discipline methods ENQ/ACK and Poll/Select.
ENQ/ACK: used for peer-to-peer communication on a dedicated point-to-point link where both stations have equal rank. Either station can start a session. The initiator sends an ENQ asking whether the receiver is ready. The receiver replies with ACK (ready) or NAK (not ready). If ready, the initiator sends its data and ends with EOT. No addresses are needed. Poll/Select: used in primary-secondary (multipoint) configurations where one station is the primary and the others are secondaries. Only the primary can start a session. Poll: the primary asks each secondary in turn whether it has data. The secondary sends data or a NAK. Select: when the primary has data to send, it sends a SEL frame with the secondary's address and waits for an ACK before sending. Addresses are required because many devices share the link. Summary: ENQ/ACK = peer stations, point-to-point, either side initiates; Poll/Select = primary controls the link, multipoint, addressing needed.
What is the difference between character-oriented and bit-oriented protocols?
Character-oriented (byte-oriented) protocols treat a frame as a sequence of characters (usually 8-bit ASCII). Control information such as start of text, end of text and acknowledgments is sent as special control characters, and character stuffing is used if a control character appears in the data. Example: BSC. They are less efficient and depend on a particular character code. Bit-oriented protocols treat a frame as a sequence of bits. Control information is carried in bit fields of the header and trailer, with a flag (01111110) marking frame boundaries. Bit stuffing (inserting a 0 after five consecutive 1s) keeps the data transparent. They can pack more control information into fewer bits, are code-independent and more efficient. Examples: HDLC, SDLC, LAPB, LAPD, PPP.
How are a lost acknowledgment and a lost frame handled at the sender site?
The sender keeps a copy of each sent frame and starts a timer when it sends the frame. Lost frame: the receiver gets nothing, so it sends no ACK (or, in sliding window, it sends a NAK when the next frame arrives out of order). When the timer expires (or the NAK arrives), the sender retransmits the frame: only that frame in stop-and-wait or selective-reject, and that frame plus all later outstanding frames in go-back-N. Lost acknowledgment: the frame arrived, but the sender never sees the ACK, so its timer expires and it resends the frame. The receiver recognizes the duplicate by its sequence number, discards it, and acknowledges again. In sliding window a later ACK can also cover the lost one, because ACKs are cumulative. Sequence numbers on frames and timers at the sender let both cases be recovered.
Explain Protocol Data Unit (PDU).
A protocol data unit is the unit of data exchanged between peer layers of a protocol. It is the layer's header (and trailer, if any) plus the data received from the layer above. At the transport layer it is a segment, at the network layer a packet, at the data link layer a frame, and at the physical layer bits. Each layer's PDU is read only by its peer layer at the receiver, which removes its header and passes the data up. In the IEEE 802 project the LLC sublayer's PDU contains DSAP (destination service access point), SSAP (source service access point), a control field (HDLC-like) and the information (data) field. It has no flags, CRC or station addresses, because those are added by the MAC frame in which it is carried.
Write down the function of primary-secondary communication in line discipline.
In primary-secondary (poll/select) line discipline, one device is the primary and the rest are secondaries, usually on a multipoint link. The primary controls the link and always starts a session; secondaries send only when told to. Its two functions are: Poll (the primary receives data): the primary sends a poll frame to each secondary in turn. A secondary with data responds with its data frame, which the primary acknowledges. A secondary with nothing to send replies with NAK, and the primary moves on to the next one. Select (the primary sends data): the primary sends a SEL frame containing the target secondary's address. The secondary replies with ACK if it is ready. The primary then sends its data and waits for an acknowledgment. Addresses are needed in every frame because many secondaries share the link. This avoids collisions and gives orderly access to the shared line.
Using odd parity, find the missing bits in the LRC for the data 10110111, 11010010, 00101100, where the received LRC is 1 ? ? ? 0 ? 1 ?.
Odd parity per column: the LRC bit makes the number of 1s in each column odd. Columns: 1,1,0 (two 1s) -> 1; 0,1,0 (one) -> 0; 1,0,1 (two) -> 1; 1,1,0 (two) -> 1; 0,0,1 (one) -> 0; 1,0,1 (two) -> 1; 1,1,0 (two) -> 1; 1,0,0 (one) -> 0. LRC = 1 0 1 1 0 1 1 0, so the missing bits (positions 2, 3, 4, 6, 8) are 0, 1, 1, 1, 0. The given bits 1, 0 and 1 match.
From baseband and broadband, which one specifies an analog signal? Also write any three baseband standards.
Broadband specifies an analog signal: data are modulated onto a carrier and the cable bandwidth can be split into several channels (IEEE 10Broad36). Baseband sends digital signals directly. Any three baseband standards: 10Base5 (thick Ethernet), 10Base2 (thin Ethernet), 10BaseT (twisted-pair Ethernet), 1Base5 (StarLAN), 100BaseT (Fast Ethernet). In each name, the first number is the data rate in Mbps and the last number or letter is the maximum segment length in hundreds of metres, or the cable type (T = twisted pair).
Explain redundancy bits in Hamming code.
Hamming code adds r redundancy bits to m data bits so that a single-bit error can be located and corrected. The r bits must indicate m + r + 1 states (no error, or an error in any of the m + r positions), so 2^r >= m + r + 1. Example: for m = 7 (ASCII), r = 4 since 2^4 = 16 >= 7 + 4 + 1 = 12 (r = 3 gives 8 < 11). The redundancy bits are placed at positions that are powers of 2 (1, 2, 4, 8). Each bit is a parity bit over a set of positions: r1 covers positions whose binary index has bit 1 set (1,3,5,7,9,11), r2 covers 2,3,6,7,10,11, r4 covers 4,5,6,7, r8 covers 8,9,10,11. The receiver recomputes the parities; the resulting binary number gives the position of the erroneous bit, which is then inverted.
What are the advantages of tree topology?
A tree is a variation of star: nodes connect to secondary hubs that connect to a central (active) hub. Advantages: more devices can be attached to the central hub through secondary hubs; active hubs repeat the signal, increasing the distance a signal can travel; it allows the network to isolate and prioritise communication from different computers (e.g. a secondary hub's computers can get priority), so time-sensitive data need not wait; other advantages are the same as star (easy installation, fault isolation, easy addition of nodes).