Networks Exam 2 - Tao Shu

Ethernet (IEEE 802.3) uses CSMA/CD as its MAC protocol.

True

What does CSMA/CD stand for?

Carrier Sense Multiple Access with Collision Detection

What does CSMA/CA stand for?

Carrier Sense Multiple Access with Collision Avoidance

The IEEE standard number for Ethernet is 802.11.

False — Ethernet is IEEE 802.3; 802.11 is Wi-Fi (WLAN).

In Ethernet the minimum frame size (excluding the preamble) is 512 bits (64 bytes).

True

What is the purpose of the 64-bit preamble in an Ethernet frame?

It allows the receiver to synchronize with the incoming signal using a sequence of alternating 0s and 1s.

Each Ethernet address (MAC address) is 32 bits long.

False — Ethernet MAC addresses are 48 bits long.

An Ethernet address in which all 48 bits are set to 1 is a broadcast address.

True

An Ethernet address whose first bit is 1 (but is not all 1s) is a multicast address.

True

What is the maximum data payload size in a standard Ethernet frame?

1500 bytes

What is the minimum data payload size in a standard Ethernet frame and why?

46 bytes — the frame must be long enough (512 bits total minimum) for collision detection to work.

The maximum number of repeaters between any two hosts in classic Ethernet is 4 (giving 5 segments).

True

In classic Ethernet each segment can be at most 500 meters long so the maximum network span is 2500 meters.

True

The Ethernet CRC field is 16 bits long.

False — the Ethernet CRC (FCS) field is 32 bits long.

The Packet Type field in an Ethernet frame is 16 bits and serves as a demultiplexing key to identify the higher-level protocol.

True

The Ethernet frame format is implemented in software at the operating system level.

False — it is implemented in hardware on the network adaptor.

The minimum distance between Ethernet transceivers is 2.5 meters.

True

The maximum number of hosts supported per classic Ethernet segment is 1024.

True

In Ethernet every adaptor on the segment receives every transmitted frame regardless of the destination address.

True

An Ethernet adaptor accepts frames addressed to its own address and frames addressed to the broadcast address and frames addressed to an instructed multicast address.

True

Every Ethernet manufacturer is assigned a unique 24-bit prefix for MAC addresses to ensure global uniqueness.

True

Ethernet was originally based on a bus topology and later evolved to hub-based and then switch-based topologies.

True

A hub-based Ethernet creates separate collision domains for each connected device.

False — a hub-based Ethernet keeps all connected devices in the same collision domain; switches create separate collision domains.

A switch-based Ethernet uses a tree topology and connections between devices are dynamically configured.

True

In nonpersistent CSMA a station that finds the medium busy will continuously listen until the medium becomes idle then transmit immediately.

False — in nonpersistent CSMA the station backs off for a random time and then re-senses. Continuous listening and immediate transmission describes 1-persistent CSMA.

Which CSMA variant transmits immediately with probability 1 as soon as the medium becomes idle? (A) Nonpersistent CSMA (B) 1-Persistent CSMA (C) p-Persistent CSMA (D) CSMA/CA

(B) 1-Persistent CSMA

In 1-persistent CSMA if two or more stations become ready while the medium is busy a collision is guaranteed when the medium becomes idle.

True

In p-persistent CSMA time is divided into slots where each slot duration typically equals the maximum propagation delay.

True

In p-persistent CSMA a station that finds the medium idle transmits with probability p or defers one slot with probability (1 − p).

True

Nonpersistent CSMA stations are considered selfish because they immediately transmit when the medium becomes idle.

False — nonpersistent stations are considered deferential; 1-persistent stations are selfish.

Which CSMA variant wastes bandwidth because its random backoff may keep the channel idle even when stations have frames to send? (A) 1-persistent (B) p-persistent (C) Nonpersistent (D) All of the above

(C) Nonpersistent CSMA

p-persistent CSMA combines the advantages of both nonpersistent (reducing collisions) and 1-persistent (reducing idle time).

True

What happens when a collision is detected in Ethernet (CSMA/CD)?

The adapter stops transmitting immediately and uses an exponential backoff algorithm — it picks a random delay and retries.

In Ethernet exponential backoff for the nth retry the adapter selects a random k between 0 and 2^n − 1 and waits k × 51.2 μs.

True

On the 1st retry attempt in Ethernet exponential backoff the station randomly chooses either 0 or 1 slots.

True

On the 2nd retry attempt in Ethernet exponential backoff the station randomly chooses from 0 to 3 slots.

True

On the 3rd retry attempt in Ethernet exponential backoff the station randomly selects from 0 to 7 slots.

True

What is the worst-case time for a node to detect a collision in Ethernet?

2 × t_prop (twice the maximum one-way propagation delay between the two farthest stations).

The vulnerable time for collision detection in Ethernet equals the one-way maximum propagation delay.

False — it equals twice the maximum propagation delay (2 × t_prop).

In a 10 Mbps Ethernet the slot time is 51.2 μs which corresponds to 512 bits.

True

The minimum frame size of 512 bits exists to ensure that a transmitting station can always detect a collision before it finishes sending the frame.

True

In the worst-case collision scenario station A begins transmitting and A's frame arrives at distant station B at time t_prop. Station B begins transmitting just before receiving A's frame and B's collision signal arrives back at A at time 2 × t_prop.

True

Why is the maximum Ethernet segment length limited to 2500 meters?

Because the farther apart two nodes are the longer the propagation delay and the network is vulnerable to undetected collisions. The 512-bit minimum frame at 10 Mbps requires the round-trip time not to exceed 51.2 μs.

Collision detection is performed only while the node is actively transmitting.

True

What is the IEEE standard number for wireless LAN (Wi-Fi)?

IEEE 802.11

IEEE 802.11 networks operate over licensed spectrum bands.

False — 802.11 operates over unlicensed ISM (Industrial Scientific and Medical) spectrum bands.

Which two frequency bands are most commonly used by 802.11 Wi-Fi? (A) 900 MHz and 1.8 GHz (B) 2.4 GHz and 5 GHz (C) 700 MHz and 3.5 GHz (D) 5 GHz and 28 GHz

(B) 2.4 GHz and 5 GHz

What are the two modes of operation for IEEE 802.11 networks?

Infrastructure mode (using an Access Point within a BSS) and Ad Hoc mode (IBSS — Independent Basic Service Set with no AP).

What does BSS stand for in 802.11?

Basic Service Set

What does IBSS stand for?

Independent Basic Service Set — the technical name for an 802.11 ad hoc network.

In infrastructure mode the Access Point broadcasts beacons containing its SSID.

True

The Point Coordination Function (PCF) in 802.11 can be used in both ad hoc and infrastructure modes.

False — PCF can only be used in infrastructure mode. DCF can be used in both modes.

Which 802.11 MAC function is contention-free and uses polling by the access point? (A) DCF (B) PCF (C) CSMA/CA (D) RTS/CTS

(B) PCF — Point Coordination Function

PCF is intended for time-sensitive traffic such as voice and video.

True

The Distributed Coordination Function (DCF) is a contention-based protocol based on CSMA/CA.

True

DCF is appropriate for delay-tolerant applications such as file transfer and web browsing.

True

In 802.11 CSMA/CA a node detects collisions by monitoring its own transmission on the medium just like Ethernet CD.

False — in wireless CSMA/CA collision detection during transmission is infeasible due to high signal attenuation; 802.11 relies on ACKs to determine if a collision occurred.

Why is collision detection (CD) infeasible in wireless networks?

The high attenuation of the colliding signal relative to the node's own strong transmission renders the collision undetectable at the transmitting node.

In 802.11 CSMA/CA a node's backoff timer is decremented while the channel is idle and frozen while the channel is busy.

True

In 802.11 CSMA/CA after a successful transmission the contention window is doubled.

False — after a successful transmission the contention window is reset to CW_min. It is doubled only after a failed transmission.

After a collision in 802.11 CSMA/CA the contention window is doubled up to CW_max.

True

A node in CSMA/CA must wait for a DIFS period after the channel becomes idle before starting its backoff countdown.

True

What does DIFS stand for in 802.11?

DCF Inter-Frame Space — the mandatory waiting period after the channel becomes idle before a station begins its backoff countdown.

What does SIFS stand for in 802.11?

Short Inter-Frame Space — a shorter interval used between control/response frames like RTS→CTS and DATA→ACK.

SIFS is longer than DIFS in 802.11.

False — SIFS is shorter than DIFS which gives higher priority to response frames (CTS ACK) over new data transmissions.

The correct frame exchange ordering in 802.11 with virtual carrier sensing is: RTS → SIFS → CTS → SIFS → DATA → SIFS → ACK.

True

What is the hidden terminal problem?

A node X cannot hear another transmitting node A because they are out of range but both reach a common receiver B. X senses the medium as free and transmits causing a collision at B.

What is the exposed terminal problem?

A node X hears a nearby transmitter A and defers its own transmission even though X's intended receiver Y is out of range of A and the transmission would not cause interference.

In the hidden terminal problem no carrier sensed means it is safe to transmit.

False — the hidden terminal problem exists precisely because no carrier does NOT mean it is safe to transmit.

In the exposed terminal problem the presence of carrier always means transmission should be deferred.

False — in the exposed terminal problem presence of carrier does NOT necessarily mean transmission should be held off.

What does RTS stand for in 802.11?

Request to Send

What does CTS stand for in 802.11?

Clear to Send

What does NAV stand for in 802.11?

Network Allocation Vector — a duration value embedded in RTS and CTS frames telling other nodes how long to defer.

The RTS/CTS mechanism in 802.11 is mandatory for all transmissions.

False — RTS/CTS virtual carrier sensing is optional and used only for unicast transmissions.

How does RTS/CTS help with the hidden terminal problem?

The receiver's CTS is heard by hidden nodes near the receiver causing them to defer for the NAV duration preventing collisions at the receiver.

RTS/CTS completely eliminates the hidden terminal problem.

False — hidden terminals can persist if a node does not successfully receive the CTS (e.g. it is out of range of the receiver).

Any node that hears an RTS defers medium access.

True

Any node that hears a CTS defers medium access.

True

In 802.11 the RTS/CTS exchange helps address the exposed terminal problem because a node that hears an RTS but not the corresponding CTS knows it is safe to transmit.

True

What was the Busy-Tone solution proposed by Kleinrock (1975)?

The receiver transmits a busy tone on a separate channel while receiving. All nodes hearing the busy tone remain silent. It fixes both hidden and exposed terminal problems but requires a separate out-of-band channel.

What is ZigZag decoding?

A technique to recover packets from repeated collisions between two hidden terminals by exploiting different random timing offsets in 802.11 retransmissions to iteratively decode interference-free bits.

ZigZag exploits the fact that Δ1 ≠ Δ2 (timing offsets differ between collisions) so each collision begins with some interference-free bits.

True

The 802.11b standard operates at 2.4 GHz with a maximum data rate of 11 Mbps using DSSS modulation.

True

The 802.11a standard operates at 2.4 GHz.

False — 802.11a operates at 5 GHz using OFDM modulation.

802.11g operates at 2.4 GHz with a maximum data rate of 54 Mbps.

True

Which 802.11 standard introduced operation at both 2.4 GHz and 5 GHz with speeds up to 540 Mbps? (A) 802.11a (B) 802.11g (C) 802.11n (D) 802.11b

(C) 802.11n

Wireless networks are standardized by the IEEE 802 LAN and MAN standards committee.

True

The 802.11 MAC and PHY correspond to the Data Link and Physical layers of the OSI model.

True

The 802.11 DCF uses CSMA/CA with optional extensions for RTS/CTS.

True

In 802.11 after the channel becomes idle a station waits DIFS then counts down its random backoff. If the channel becomes busy the station freezes its counter and resumes after another DIFS idle.

True

Ethernet uses acknowledgments (ACKs) to confirm successful frame delivery.

False — standard Ethernet does not use ACKs at the MAC layer. 802.11 Wi-Fi uses ACKs.

What is the key difference between datagram (connectionless) switching and virtual circuit (connection-oriented) switching?

In datagram switching each packet carries the full destination address and is independently routed. In VC switching a connection is established first and data packets follow the path using short VCI labels.

In datagram switching all packets of the same flow are guaranteed to follow the same path.

False — packets may follow independent paths and can arrive out of order.

Datagram switching is more resilient to switch failures than VC switching because alternative paths can be used without re-establishing a connection.

True

In virtual circuit switching the VCI has global scope and is the same on every link along the path.

False — the VCI has link-local scope and typically changes at each switch.

What are the four fields in a virtual circuit (VC) table entry?

Incoming Interface — Incoming VCI — Outgoing Interface — Outgoing VCI.

What is a Permanent Virtual Circuit (PVC)?

A virtual circuit set up manually by an administrator with a permanent VC table configuration.

What is a Switched Virtual Circuit (SVC)?

A virtual circuit established dynamically through signaling and torn down after data transmission completes.

In SVC setup the initial setup message contains the full destination address but subsequent data packets need only the short VCI number.

True

During SVC setup the outgoing VCI values at each switch are filled in during the ACK phase (reverse direction) not during the forward setup phase.

True