Basics of Wireless Local Area Networks

2.1 Networks Large and Small

This chapter introduces networking and reviews various wireless local network technologies,
emphasizing wireless data transmission. While the discussion is brief, it highlights practical applications
of networking concepts.

Data networks have evolved since the 19th century and now encompass a vast range of devices. The
Internet connects computers globally, facilitating communication.

Networks can be categorized by size:

 Personal Area Networks (PANs): Cover a few meters, connecting nearby devices. Wireless PANs
(e.g., Bluetooth) replace wired connections for data and sometimes voice.

 Local Area Networks (LANs): Connect computers within a single facility, extending from
hundreds of meters to a few kilometers. Typically owned by users, LANs are mostly used for data
transfer but now also support voice through VoIP. Ethernet is the most common LAN technology.

 Metropolitan Area Networks (MANs): Connect multiple buildings within a city. Unlike LANs,
MANs originated from the telephone network and typically support both voice and data. They
are managed by local phone companies, utilizing T-1 and fiber optic connections.

 Wide Area Networks (WANs): Span cities and countries, often managed by long-distance
telephone providers. Most long-distance traffic now uses fiber optics, supporting high data rates.
The Internet's growth has intertwined WANs and LANs.

In summary, small networks (PANs and LANs) focus on cost-sensitive, short-distance services, while
larger networks (MANs and WANs) prioritize reliable, long-distance communication. This book focuses
on transitioning small networks from wired to wireless.

All networks share common elements: addressing messages, formatting for understanding, accessing
physical media, and correcting transmission errors. These functions can be organized hierarchically,
supporting applications that route messages to destinations.

The Open Systems Interconnect (OSI) model provides a structured framework for these activities,
particularly highlighting the data link and physical layers crucial for wireless communication. The MAC
layer plays a vital role in efficiently using wireless media.
2.2 WLANs from LANs

Overview of LANs

LANs connect nearby computers for data exchange, with Ethernet being the most popular technology
since its invention in 1973. Ethernet operates asynchronously, allowing stations to share the medium
without fixed time slots or central control, utilizing carrier sensing and collision detection to manage
transmissions.

Ethernet Mechanisms

1. Carrier Sensing: Stations check if the medium is free before transmitting.

2. Collision Detection: Stations detect if another station is transmitting simultaneously.

3. Random Backoff: After a collision, stations wait a random time before trying again, preventing
repeated collisions.

Ethernet's design enables efficient medium sharing without central coordination, ensuring graceful
network performance even under heavy traffic.

Challenges in Wireless Networks

Wireless links face unique challenges compared to wired connections:

 Connection: Wireless stations must announce their presence and associate with the network
while managing power.

 Authentication: Unlike wired connections, wireless networks must verify user identities to
prevent unauthorized access.

 Medium Access Control: Protocols must efficiently manage who transmits when.

 Security: Wireless data needs encryption to protect against eavesdropping.

 Error Correction: Due to variable signal quality, packets may need acknowledgment and
fragmentation to ensure successful transmission.

 Packet Construction: Data packets include synchronization sequences for proper reception.

 Modulation: Packets are modulated for transmission and demodulated upon reception.

2.3 802.11 WLANs

In 1985, the FCC opened the 2.4-GHz band for unlicensed use, spurring the development of wireless data
links. Researchers adapted existing Ethernet technologies to create wireless links, addressing collision
detection challenges with modified MAC protocols.

The IEEE 802 working group oversees standards for LANs and WLANs. The 802.11 standard, released in
1997, introduced multiple physical layers, including infrared, frequency hopping, and direct-sequence
spread-spectrum (DSSS) links. While infrared has seen little use, DSSS became dominant, especially with
the higher data rates introduced in 802.11b in 1999.
The original 802.11 standard and its adaptations are collectively referred to as "802.11 classic," while the
term "802.11" encompasses all standards from the working group.

2.3.1 802.11 Architecture

The 802.11 standard supports both infrastructure and independent wireless networks. Infrastructure
networks connect to a wired network (typically Ethernet) via an access point, while independent
networks allow peer-to-peer connections without a wired backbone. This section focuses on
infrastructure networks, where wireless links serve as an alternative method for transmitting Ethernet
packets (or frames).

Ethernet Frame Structure

The wireless link essentially moves Ethernet frames from one station to another, maintaining the
standard structure, which includes:

 Preamble

 Start Frame Delimiter

 Destination Address

 Source Address

 Length/Type

 Payload

 Error Check (Pad)

Basic Service Set (BSS) and Extended Service Set (ESS)

In an 802.11 network, stations associated with a specific access point form a Basic Service Set (BSS).
Multiple BSSs connected to a single wired network create an Extended Service Set (ESS).

 BSSID: Unique identifier for each BSS.

 ESSID: Unique identifier for the entire ESS, shared among its component BSSs.
Mobility and Roaming

The architecture accounts for mobile stations that may roam between access points within the same ESS.
Ideally, users should transition seamlessly from one access point to another without undergoing lengthy
reauthentication or obtaining new IP addresses.

To facilitate this, a distribution system manages the tracking of which stations belong to which BSSs and
routes their packets accordingly. However, the original 802.11 standards did not detail roaming
procedures. This was addressed with the introduction of the 802.11f Inter-Access Point Protocol (IAPP) in
2003, which provides a nonproprietary framework for roaming.

IAPP and Future Considerations

The IAPP utilizes TCP/IP and RADIUS servers for secure communication between access points, enabling
mobile clients to reassociate smoothly and receive forwarded packets as they move between BSSs. The
goal is for local roaming on 802.11 networks to be transparent to users, assuming compliant access
points and configurations become widely adopted.
2.3.2 MAC and CSMA/CA

While Ethernet was originally designed for wired networks, wireless networks face unique challenges
that necessitate different protocols. In a wired setup, all stations can hear each other's transmissions,
allowing for effective collision detection. However, in wireless networks, stations may not always receive
all transmissions due to the "hidden station problem." For instance, if Station A can communicate with
the access point but not with Station B, both stations could transmit simultaneously, leading to
undetected collisions.

Limitations of Collision Detection

Collision detection, as used in wired networks (CSMA/CD), is impractical in wireless networks. The
inability to consistently associate the absence of a signal with a free medium complicates the situation.
Therefore, 802.11 employs a different approach known as Carrier-Sense Multiple Access with Collision
Avoidance (CSMA/CA).

CSMA/CA Mechanism

In CSMA/CA, stations listen to the radio channel before transmitting. If the channel is busy, they defer
their transmission. A virtual carrier sensing mechanism, the Network Allocation Vector (NAV), is
implemented to further reduce collision risks. Each packet header provides an NAV value, which all
receiving stations note to determine when to defer their transmissions.

Timing and Access Control

To manage access to the medium, 802.11 defines various timing requirements known as interframe
spaces. A transmitting station captures the medium, and any station wishing to send a new packet must
ensure that both the physical carrier sensing mechanism and the NAV indicate that the medium is free
for a duration known as the Distributed Interframe Space (DIFS). This design allows ongoing exchanges to
be completed before new transmissions begin.

Acknowledgments and Retransmission

Since collisions cannot be detected, 802.11 relies on positive acknowledgment (ACK) from the receiving
station. If an ACK is not received, the sending station assumes a collision occurred. It will then wait for
the medium to be free, select a random time slot, and attempt to retransmit the packet.

Enhancements for Performance

To further improve reliability, two optional mechanisms can be utilized:

1. Request to Send / Clear to Send (RTS/CTS): The sending station can send an RTS packet to
inform the access point and other stations of an impending transmission, including the packet
length. The access point replies with a CTS packet, which reserves the medium for the
transmission duration. This method helps ensure that all stations hear the CTS, reducing the
likelihood of collisions.

2. Fragmentation: Large packets can be divided into smaller fragments, each of which is more likely
to be transmitted successfully. If a fragment encounters an error, only that specific fragment
 needs to be resent. Each fragment transmission also includes an NAV value to reserve the
medium for the entire exchange.

Half-Duplex Operation

A significant characteristic of the MAC layer in 802.11 is that the radio can either transmit or receive at
any given time but not both simultaneously. This half-duplex operation simplifies the radio design by
eliminating the need to distinguish between a strong transmitted signal and a weak received signal.

Continuity of Packet Exchanges

The MAC layer of the original 802.11 standard remains largely unchanged in enhanced versions like
802.11b, a, and g. While the MAC is more complex than the original Ethernet MAC, it still lacks
centralized coordination for contending stations and provides no guarantees of performance outside of
ongoing packet exchanges.

2.3.3 802.11 Classic Direct-Sequence PHY

To optimize the sharing of unlicensed bands while minimizing interference, the FCC mandated that
systems using these bands spread their transmission energy across a significant portion of the frequency
range. This requirement was established to reduce interference among collocated systems.

Frequency Hopping vs. Direct-Sequence Spread Spectrum (DSSS)

One method to achieve this is frequency hopping, where systems transmit on many channels in a
pseudo-random sequence. This approach, however, requires narrow channels to prevent overlap,
limiting data transmission rates in low-cost wireless links.

A more sophisticated method is the Direct-Sequence Spread Spectrum (DSSS), initially developed for
military applications and utilized in CDMA-based cellular phones. DSSS works by multiplying slower data
bits with a much faster pseudo-random sequence, creating a transmitted signal with a wider bandwidth
than the original data. This method ensures compliance with the FCC's spreading requirement while also
enhancing interference resistance.

In 802.11 classic DSSS, each data bit is multiplied by an 11-chip Barker sequence, which provides good
autocorrelation properties. This enables easy synchronization with the transmitter. DSSS can also
accommodate multiple users in the same frequency band with minimal interference by using different
orthogonal spreading codes, although this technique is not employed in the 802.11 standards.

Technical Details

The basic symbol rate is set at 1 Mbps, with an 11 Mbps chip rate resulting in a spreading gain of 11:1
(approximately 10.4 dB), satisfying the FCC's requirement for a minimum spreading gain of 10 dB. The
actual bandwidth required exceeds 11 MHz, with a defined spectral mask limiting the transmitter's
radiated power at various distances from the nominal frequency.

802.11 classic operates on specific channel frequencies spaced 5 MHz apart, as shown in Table 2.1. Due
to the bandwidth of the transmitted signal, adjacent channels can interfere with one another, leading to
the conclusion that only three non-overlapping channels are practical in the U.S. (channels 1, 6, and 11).

Data Rates and Modulation

Two data rates are supported in 802.11 classic: 1 Mbps and 2 Mbps, both utilizing the same chip rate of
11 Mbps. The difference arises from the modulation techniques used:

 DBPSK (Differential BPSK) is employed for the basic rate of 1 Mbps.

 DQPSK (Differential Quaternary PSK) is used for the extended rate of 2 Mbps, effectively
doubling the data rate without increasing bandwidth.

The modulations are differential, meaning each symbol's phase is defined relative to the previous
symbol. This design simplifies synchronization and reduces hardware requirements, enhancing the
robustness of the transmission.

Link Budget Considerations

The maximum transmit power of an 802.11 radio is capped by regulatory standards, typically allowing
less than 1 W, with most access points using about 30 to 100 mW. The standard mandates that receivers
achieve a frame error rate of less than 8% for 1024-byte frames at a signal power of 80 dBm.

A link budget calculation illustrates the potential range and performance of the system. For example,
with a transmit power of 100 mW (20 dBm) and a gain from a directional antenna, the allowed path loss
can be substantial, leading to effective indoor ranges of around 63 meters, depending on environmental
factors.

The standard's receiver sensitivity requirements, coupled with the averaging of QPSK chips, indicate that
reasonably priced components can achieve the desired performance levels in an indoor LAN
environment, where 10 to 100 meters of range is typical.

Overall, the 802.11 classic DSSS PHY offers a balanced approach to wireless communication,
accommodating multiple users while ensuring robust performance against interference.

802.11 Standards Overview

 Early Products: The late 1990s saw the introduction of products based on the original 802.11
standard, but the 2 Mbps data rate was inadequate compared to wired Ethernet (10 Mbps).

 802.11b: Launched as a significant improvement, it maintained compatibility with the original

802.11 but offered data rates of up to 11 Mbps using new encoding techniques (CCK and PBCC).
It quickly gained popularity due to its balance of performance, cost, and simplicity.

 802.11a: Introduced later, this standard operated at 5 GHz with a maximum data rate of 54
Mbps. While initially viewed as a future reserve, companies like Atheros began implementing it,
emphasizing the need for backward compatibility with 802.11b.

 802.11g: This standard combined the high data rates of 802.11a with the compatibility of
802.11b, leading to its rapid adoption.

Key Technical Concepts

 Physical Layer Modulation:

 o CCK (Complementary Code Keying): A block coding method that allows for error
detection and correction by using a limited set of code words.

o PBCC (Packet Binary Convolutional Coding): A convolutional coding technique providing

higher performance but with increased decoding complexity.

 Quality of Service (QoS): As demand for time-sensitive applications like voice and video grew,
task groups began defining standards to ensure timely data delivery.

Market Dynamics

 Commercial Success: By 2000, 802.11b became a commercial success due to its effective blend
of speed and affordability, leading to widespread deployment and a decrease in equipment
prices.

Security and Compatibility Efforts

 The formation of the Wireless Ethernet Compatibility Alliance (WECA) aimed to certify the
interoperability of 802.11b products, enhancing consumer confidence and market growth.

The 802.11a PHY represents a significant evolution in wireless communication technology, particularly
through its adoption of the 5 GHz UNII band, which addresses the interference issues prevalent in the
crowded 2.4 GHz ISM band. This transition allows for a wider bandwidth, enabling higher data rates and
improved performance. Here's a concise summary of the key points:

1. Frequency Band: Unlike its predecessors, 802.11a operates in the 5.15–5.825 GHz UNII band,
providing a substantial increase in available bandwidth (300 MHz compared to 80 MHz in the
ISM band).

2. Regulatory Changes: The FCC has since expanded the spectrum available for 802.11a,
introducing additional bandwidth and enforcing restrictions on power output to mitigate
interference, particularly requiring transmit power control and dynamic frequency selection for
certain bands.

3. Channel Availability: The UNII band allows for numerous nonoverlapping channels (up to 19),
enhancing network design flexibility and reducing interference in densely populated areas.

4. Modulation Techniques: 802.11a employs orthogonal frequency-division multiplexing (OFDM),

utilizing multiple subcarriers to transmit data simultaneously. This method improves data rates
and resilience against multipath interference.

5. Coding Schemes: Data is encoded using convolutional encoding with various coding rates (1/2,
2/3, and 3/4) to balance throughput and error correction.

6. Efficiency and Complexity: OFDM frames include a preamble for synchronization, and the overall
system relies on advanced digital signal processing techniques, implemented efficiently with
modern CMOS technology.

7. Challenges: While OFDM offers advantages, it also presents challenges, such as a high peak-to-
average power ratio, necessitating precise signal handling and amplification.
These advancements position 802.11a as a robust standard for high-speed wireless networking,
particularly in environments where interference is a critical concern.

2.3.7 802.11g PHY

The primary limitation of the 802.11a physical layer (PHY) is its incompatibility with older 802.11 and
802.11b systems. To address this issue while maintaining the high data rates of 802.11a, task group "G"
introduced enhancements to the Wi-Fi PHY. One significant change involved transferring the 802.11a
Orthogonal Frequency-Division Multiplexing (OFDM) physical layer to the ISM band. This was feasible
because the bandwidth of the 802.11a symbols is approximately 16 MHz, similar to that of the classic
symbols.

However, this redesign poses challenges: older systems cannot decode the complex OFDM symbols and
interpret them as noise. The Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) MAC
layer relies on all stations being able to communicate with the access point, making invisible preambles a
significant issue. Receiving stations must recognize the packet preambles to set their Network Allocation
Vector (NAV) and reserve the medium.

To facilitate coexistence between “g” stations and older “b” stations, several strategies can be employed:

1. RTS/CTS Mechanism: "g" stations can send a Request to Send/Clear to Send (RTS/CTS) using
802.11b packets to reserve the medium before transmitting faster 802.11g packets. This method
is straightforward but incurs additional overhead.

2. CTS-to-Self: A station can send a CTS packet to itself, which is particularly useful for access
points, assuming all associated stations can hear it.

3. Mixed Frame Format: "g" stations can send the preamble in 802.11b format and then append
OFDM data. This approach adds overhead due to the low data rate of the older preamble, but it
avoids the need for extra packet exchanges.

4. PBCC Coding System: “g” defines a PHY using an extension of the PBCC coding system, where
8PSK replaces QPSK modulation at the highest data rates.

Trade-offs in enhancing the PHY include the necessity for better Signal-to-Noise Ratios (S/N) for higher
data rates, which can reduce communication range. OFDM's peak-to-average power ratios allow for
lower transmission power from amplifiers to prevent distortion. Coexistence with legacy systems often
leads to reduced data rates due to the overhead of coexistence protocols. Despite these challenges,
802.11g is likely to dominate the WLAN market until the ISM band becomes overcrowded, prompting a
shift toward dual-band devices.

2.3.8 802.11 (In)Security

802.11 wireless networks face significant security challenges. In enterprise LANs, internal authentication
is minimal because physical security typically protects Ethernet ports. External access is usually
controlled by firewalls, and remote users may employ Virtual Private Networks (VPNs) for secure access.

Recognizing that wireless communication is more susceptible to interception than wired connections,
developers introduced Wired Equivalent Privacy (WEP) to provide a comparable level of security. WEP
uses symmetric key encryption, where the same key encrypts and decrypts data. The encryption process
involves generating a pseudo-random binary string (cipher stream) of the same length as the plaintext
data.

However, WEP has several vulnerabilities:

1. Initialization Vector (IV) Weakness: The 24-bit IV is sent in clear text, and due to its limited size,
it can be reused in busy networks, exposing identical cipher streams to potential attackers.

2. Authentication Weakness: If an attacker intercepts the challenge and response during

authentication, they can determine the cipher stream.

3. Key Management Issues: Users typically enter shared keys manually, leading to infrequent key
changes and increased vulnerability to attacks.

Attempts to enhance WEP by increasing key lengths have been made, but many implementations remain
susceptible to known vulnerabilities, such as those identified by Fluhrer, Mantin, and Shamir (FMS),
which exploit weaknesses in the RC4 random number generator used in WEP.

To improve wireless security, Wi-Fi Protected Access (WPA) was introduced as a backward-compatible
enhancement of WEP, utilizing the IEEE 802.1x standard for authentication and the Temporal Key
Integrity Protocol for encryption. WPA creates unique keys for each client by mixing the initial encryption
key with the client’s MAC address and a sequence number, significantly reducing the risk of IV reuse
attacks.

WPA addresses the vulnerabilities of WEP, making it suitable for most home and enterprise applications.
As of mid-2004, the 802.11i task group approved an enhanced standard based on the Advanced
Encryption Standard (AES) for even greater security.

For highly sensitive data, VPNs and SSL are recommended, as they secure communications across all
stages, not just the wireless link. However, they can be complex to set up and may not facilitate seamless
roaming between access points.

2.4 HiperLAN and HiperLAN 2

Researchers involved in defining the IEEE 802.11a standard were also part of the European efforts to
develop HiperLAN standards. HiperLAN 2's physical layer closely resembles that of 802.11a, utilizing an
OFDM system in the 5-GHz band, with similar modulation schemes and data rates. It meets ETSI
requirements for dynamic frequency selection and power control.

However, the MAC layer differs significantly, as it is based on asynchronous transfer mode (ATM) rather
than the Ethernet-based approach. ATM was created by telecommunications providers in the 1980s and
1990s to facilitate efficient transport of video, data, and voice while ensuring quality of service and traffic
capacity control. HiperLAN's MAC layer is designed to interface seamlessly with ATM networks, allocating
fixed time slots to stations via a central access point.

Despite HiperLAN's advantages, it has not seen significant commercial success, with few products
available. The 802.11a standard's economies of scale make it more attractive, particularly as major
vendors shift focus toward ultrawideband (UWB) technologies for video delivery.

2.5 From LANs to PANs

While WLANs extend computer network services wirelessly, WPANs aim to replace physical cables rather
than integrate them into a larger network. WPANs target small personal areas, replacing connections like
printer cables and USB interfaces with a single wireless link. As such, they need to be cost-effective and
power-efficient, especially for portable devices.

2.5.1 Bluetooth: Skip the Danegeld, Keep the Dane

Bluetooth emerged as the first notable WPAN standard, initiated in 1998 by Ericsson with backing from
Intel, IBM, Nokia, and Toshiba. Unlike the IEEE 802.11, Bluetooth operates independently, resulting in a
standard that lacks compatibility with 802.11 networks. The Bluetooth PAN is designed for minimal user
configuration, employing ad hoc device discovery and network formation.

Bluetooth uses frequency hopping to minimize interference, with a master device managing time slots
for communication. A piconet, consisting of one master and up to seven active slave devices, utilizes a
hopping pattern that allows overlapping piconets to coexist.

The Bluetooth PHY layer operates in the 2.4-GHz ISM band, divided into 79 channels. Devices hop
frequencies 1600 times per second, using Gaussian minimum-shift keying (GMSK) for modulation.
Bluetooth devices are classified by transmission power, with class 1 (100 mW), class 2 (2.5 mW), and
class 3 (1 mW) devices.

Security in Bluetooth links benefits from lower transmission power and the context of use, but they still
incorporate authentication and encryption mechanisms. However, weaknesses arise from user practices,
particularly with PIN codes. Secure communications require careful key exchange to prevent
interception.

Bluetooth's design focuses on affordability and efficiency, suitable for a range of applications, despite
some security vulnerabilities. As WPAN technology continues to evolve, Bluetooth remains a prominent
player in wireless communication.

802.15.3 Enhancements

 Objective: Enhance 802.15.1 (Bluetooth) while maintaining low cost, voice support, and range.

 Data Rates: Targeted data rates up to 55 Mbps for applications like streaming audio/video and
high-resolution image printing.

 Physical Layer (PHY): Aimed for compatibility with 802.11 devices, adopting an 11 Msps symbol
rate and 15 MHz bandwidth to allow more nonoverlapping channels than 802.11b/g.

 Coding Techniques: Introduced trellis-coded modulation (TCM) to mix coding and modulation,
improving performance in noise conditions.

 Channel Capacity: Allowed for five nonoverlapping channels in the 15 MHz window, reducing
interference.

UWB PANs and Future Proposals

 Higher Data Rates: The 802.15.3a task group aims for data rates of at least 100 Mbps, targeting
400 Mbps for high-definition applications.
  Ultrawideband (UWB): UWB allows transmission over a wide spectrum, with restrictions to
minimize interference with existing systems. This approach supports short-range, high-rate
communications.

 Link Performance: Various proposals are under consideration, including direct-sequence CDMA
and multiband OFDM, both designed to support high data rates at short distances.

 Trade-offs: UWB's high data rate capabilities come with a limited range due to power
restrictions.

Technical Comparisons

 Link Performance: Compared to two approaches (DS-CDMA and OFDM) for 200 Mbps data
rates. OFDM shows slight advantages in noise tolerance and sensitivity.

 Implementation Parameters: Details on data rates, modulation types, coding rates, and noise
figures highlight the technical specifications necessary for achieving robust wireless
communication.

Future Directions

 Standardization and Market Viability: While proposals for these high-rate technologies are still
under review, their development points to future trends in wireless communication,
emphasizing short-range, high-capacity needs.

This summary encapsulates the evolution and technical advancements of PANs through the lens of IEEE
standards, emphasizing the balance between increased data rates and practical implementation
challenges.