π Lecture 5 β Error Detecting and Correcting Techniques (Part II)
π Overview: This lecture continues the study of error detection and correction in wireless networks, focusing on advanced block codes including Hamming, BCH, and Reed-Solomon codes. It then introduces Automatic Repeat Request (ARQ) mechanisms, particularly sliding window flow control and Go-Back-N ARQ, which are essential for reliable data transmission.
ποΈ Topics Covered
The lecture begins with a review of transmission errors, parity check (single-bit and 2D parity), cyclic redundancy check, and block error codes from the previous lecture. It then covers Hamming Code in detail, including its parameters, encoding and decoding processes with examples. Next, BCH codes are introduced as multiple error correcting codes generalizing Hamming codes, followed by Reed-Solomon codes which process data in symbols. The second half covers Automatic Repeat Request (ARQ), flow control using sliding window, error control requirements, and the Go-Back-N ARQ protocol including its timeout mechanism and transmitter/receiver operations.
π Lecture Summary
Review of Previous Lecture
The previous lecture covered fundamental error detection techniques: transmission errors, parity check (both single-bit parity and 2D parity), cyclic redundancy check (CRC), and block error codes. These form the foundation for understanding more advanced error correcting codes.
Hamming Code
Hamming Code is designed specifically to correct single bit errors. It belongs to a family of (n, k) block error-correcting codes with specific parameters:
- Block length: n = 2^m β 1
- Number of data bits: k = 2^m β m β 1
- Number of check bits: n β k = m
- Minimum distance: d_min = 3
This is classified as a single-error-correcting (SEC) code. It can also function as a SEC double-error-detecting (SEC-DED) code.
The encoding process takes k data bits and adds (n - k) check bits. The decoding process compares the received (n - k) bits with the calculated (n - k) bits using XOR operation. The resulting (n - k) bits form the syndrome word. The syndrome range is between 0 and 2^(n-k) - 1. Each bit of the syndrome indicates a match (0) or conflict (1) in that bit position.
π Definition β Syndrome Word: The result of XORing the received check bits with the calculated check bits, used to detect and locate errors in Hamming code decoding.
π Example of Hamming Encode: For data = 00111001, the encoding process calculates check bits and appends them to the data bits to form the transmitted codeword. The actual encoding involves determining parity bits at specific positions (powers of 2) within the codeword.
π Example of Decoding Hamming: The decoding process recalculates check bits from the received data, compares them with the received check bits using XOR to generate a syndrome word. A non-zero syndrome indicates an error, and its value points to the bit position that is in error.
π‘ Why this matters: Hamming codes provide a practical method for correcting single-bit errors in memory systems and simple communication links, making them fundamental to reliable data storage and transmission.
BCH Codes
BCH codes are named after their discoverers: Bose, Chaudhuri, and Hocquenghem. These are multiple error correcting codes and serve as a generalization of Hamming Code. They offer flexibility in choice of parameters including block length and code rate.
For a positive pair of integers m and t, a (n, k) BCH code has parameters:
- Block length: n = 2^m β 1
- Number of check bits: n β k β€ mt
- Minimum distance: d_min β₯ 2t + 1
BCH codes can correct combinations of t or fewer errors. The generator polynomial can be constructed from the factors of (X^(2m-1) + 1).
π‘ Why this matters: BCH codes provide powerful error correction for multiple errors, making them suitable for applications like satellite communications and data storage where burst errors are common.
Reed-Solomon Codes
Reed-Solomon (RS) codes are a subclass of nonbinary BCH codes. Data is processed in chunks of m bits, called symbols.
An (n, k) RS code has parameters:
- Symbol length: m bits per symbol
- Block length: n = 2^m β 1 symbols = m(2^m β 1) bits
- Data length: k symbols
- Size of check code: n β k = 2t symbols = m(2t) bits
- Minimum distance: d_min = 2t + 1 symbols
π Definition β Symbol: In Reed-Solomon codes, a symbol is a chunk of m bits that is processed as a single unit, allowing the code to correct errors at the symbol level rather than the bit level.
π‘ Why this matters: Reed-Solomon codes are extremely effective at correcting burst errors (consecutive errors) and are widely used in QR codes, CDs, DVDs, and deep-space communications.
Automatic Repeat Request
Automatic Repeat Request (ARQ) is a mechanism used in data link control and transport protocols. It relies on the use of an error detection code (such as CRC). ARQ provides both Flow Control and Error Control functions.
Flow Control
Flow Control assures that the transmitting entity does not overwhelm a receiving entity with data. Protocols with flow control mechanism allow multiple PDUs in transit at the same time. PDUs arrive in the same order they are sent.
Sliding-window flow control operates as follows:
- The transmitter maintains a list (window) of sequence numbers allowed to send
- The receiver maintains a list allowed to receive
Reasons for breaking up a block of data before transmitting include:
- Limited buffer size of receiver
- Retransmission of PDU due to error requires smaller amounts of data to be retransmitted
- On shared medium, larger PDUs occupy medium for extended period, causing delays at other sending stations
Error control mechanisms detect and correct transmission errors by handling:
- Lost PDU: a PDU fails to arrive
- Damaged PDU: PDU arrives with errors
Error Control Requirements
The requirements for effective error control include:
- Error detection: Receiver detects errors and discards PDUs
- Positive acknowledgement: Destination returns acknowledgment of received, error-free PDUs
- Retransmission after timeout: Source retransmits unacknowledged PDU
- Negative acknowledgement and retransmission: Destination returns negative acknowledgment to PDUs in error
Go-Back-N ARQ
Go-Back-N ARQ improves upon Stop-and-Wait by not waiting for each acknowledgment. It keeps the channel busy by continuing to send frames and allows a window of up to Ws outstanding frames. It uses m-bit sequence numbering.
The operation principle: If an ACK for the oldest frame arrives before the window is exhausted, the transmitter can continue transmitting. If the window is exhausted, the transmitter must "pull back" and retransmit all outstanding frames. An alternative approach uses timeout.
In Go-Back-N ARQ:
- Frame transmissions are pipelined to keep the channel busy
- Frames with errors and subsequent out-of-sequence frames are ignored
- The transmitter is forced to go back when the window of 4 is exhausted
- The window size should be long enough to cover round trip time
π Definition β Go-Back-N ARQ: An ARQ protocol where the transmitter sends multiple frames without waiting for individual acknowledgments, and upon detecting an error, retransmits the errored frame and all subsequent frames.
Go-Back-N with Timeout
A problem with the basic Go-Back-N ARQ arises: If a frame is lost and the source does not have another frame to send, then the window will not be exhausted and recovery will not commence. To address this, a timeout is used with each frame. When a timeout expires, the transmitter resends all outstanding frames.
Go-Back-N Transmitter & Receiver
The Go-Back-N ARQ uses two types of acknowledgments:
- RR (Receive Ready): Indicates no errors occur
- REJ (Reject): Indicates error detected
Contingencies that must be handled include:
- Damaged PDU
- Damaged RR acknowledgment
- Damaged REJ acknowledgment
The Stop-and-Wait ARQ (simpler protocol) and Go-Back-N ARQ are compared, with Go-Back-N providing higher efficiency by pipelining frame transmissions.
β Key Takeaways
The most critical concepts from this lecture are: Hamming codes can correct single bit errors using syndrome calculation, while BCH codes generalize this to correct multiple errors with flexible parameters. Reed-Solomon codes process data in symbols rather than bits, making them powerful against burst errors. For reliable transmission, ARQ protocols combine error detection with retransmission strategies. Go-Back-N ARQ improves channel utilization by allowing multiple outstanding frames through sliding window flow control, but requires careful timeout management to handle lost frames when the transmitter has no data to send. The window size must be at least as large as the round-trip time's worth of frames to keep the channel fully utilized.
π§ Quick Revision Questions
- What are the parameters (n, k, m, d_min) for a Hamming code, and how many errors can it correct?
- How does the syndrome word in Hamming decoding indicate the location of an error?
- What is the relationship between BCH codes and Hamming codes, and what parameter determines how many errors a BCH code can correct?
- In Reed-Solomon codes, what is a "symbol" and how does processing symbols rather than bits affect error correction capability?
- Why does Go-Back-N ARQ need a timeout mechanism, and what happens when a timeout expires?
π Lecture 6 β Multiple Access Techniques
π Overview: This lecture covers the fundamental multiple access techniques used in wireless networks to allow multiple users to share a common communication channel. It explains FDMA, TDMA, CDMA, and random access methods, detailing their principles, advantages, disadvantages, and practical applications, which are essential for understanding how cellular and other wireless systems manage simultaneous user access.
ποΈ Topics Covered
The lecture begins with a review of previous topics on block codes (Hamming, BCH, Reed-Solomon), CDMA, and ARQ (sliding window, Go-back-N). It then introduces Multiple Access Techniques, covering Frequency Division Multiple Access (FDMA), Time Division Multiple Access (TDMA), and Random Access methods including ALOHA, Slotted ALOHA, and Reservation-based ALOHA. CDMA is also revisited with a detailed example. The lecture concludes with a summary and a preview of carrier-sense based random access and spread spectrum for the next lecture.
π Lecture Summary
Review of previous lecture #5
The lecture begins with a review of key concepts from Lecture 5, including Block Codes like Hamming, BCH, and Reed-Solomon codes. It also reviews CDMA and ARQ techniques, specifically the Sliding Window protocol and Go-back-N ARQ.
Multiple Access Techniques
The core techniques for multiple users to share a communication medium are introduced: Frequency Division Multiple Access (FDMA), Time Division Multiple Access (TDMA), Random Access, and Code Division Multiple Access (CDMA).
π Definition β Multiple Access Techniques: Methods that allow multiple users to share a finite amount of radio spectrum simultaneously or nearly simultaneously.
FDMA
Frequency Division Multiple Access (FDMA) was the initial multiple-access technique for cellular systems. It separates a large frequency band into smaller channels, each supporting a single user. Guard bands are used to separate channels to prevent co-channel interference. FDMA uses narrow bandwidth (30 kHz).
π‘ Why this matters: FDMA's simplicity made it the foundation for early cellular systems, but its dedicated channel per user leads to spectrum inefficiency when channels are idle.
π Advantages:
- Simple to implement in terms of hardware
- Fairly efficient with a small base population and constant traffic
π Disadvantages:
- Network and spectrum planning are intensive and time-consuming
- Channels are dedicated for a single user; idle channels add spectrum inefficiency
π Example: In an FDMA system, user 1 might be assigned frequency 880 MHz, user 2 frequency 880.03 MHz, and so on, with guard bands between them.
TDMA
Time Division Multiple Access (TDMA) makes the entire bandwidth available to each user, but only for a finite period of time. Users are allotted time slots for a channel, allowing sharing of a single channel. It requires time synchronization. Users take turns transmitting and receiving data in a round-robin fashion.
π How it works:
- User presses Push-to-Talk (PTT) button
- A control channel registers the radio to the closest base station (BS)
- The BS assigns an available pair of channels
- Unlike FDMA, a TDMA system also assigns an available time slot within the channel
- Data transmission is not continuous but sent and received in bursts
- The bursts are reassembled and appear like continuous transmission
π Advantages:
- Extended battery life and talk time
- More efficient use of spectrum compared to FDMA
- Will accommodate more users in the same spectrum space than an FDMA system
π Disadvantages:
- Network and spectrum planning are intensive
- Multipath interference affects call quality
- Dropped calls are possible when users switch in and out of different cells
- Too few users result in idle channels (rural versus urban environment)
- Higher costs due to greater equipment sophistication
π Example: In a TDMA system with 3 time slots, user 1 transmits in slot 1, user 2 in slot 2, user 3 in slot 3, then the cycle repeats.
CDMA
Code Division Multiple Access (CDMA) is a spread spectrum technique used to increase spectrum efficiency. It has been used in military applications due to anti-jamming and security.
π Basic Principles of CDMA:
- D = rate of data signal
- Break each bit into k chips (chips are a user-specific fixed pattern)
- Chip data rate of new channel = kD
π Formula:
- For a '1' bit, sender A sends code as chip pattern:
<c1, c2, c3, c4, c5, c6> - For a '0' bit, sender A sends complement of code:
<-c1, -c2, -c3, -c4, -c5, -c6> - Receiver knows sender's code and performs an electronic decode function:
<d1, d2, d3, d4, d5, d6>= received chip pattern<c1, c2, c3, c4, c5, c6>= sender's codeS = (d1 Γ c1) + (d2 Γ c2) + (d3 Γ c3) + (d4 Γ c4) + (d5 Γ c5) + (d6 Γ c6)
π Example:
- User A code =
<1, β1, β1, 1, β1, 1>- To send a 1 bit:
<1, β1, β1, 1, β1, 1> - To send a 0 bit:
<β1, 1, 1, β1, 1, β1>
- To send a 1 bit:
- User B code =
<1, 1, β1, β1, 1, 1>- To send a 1 bit:
<1, 1, β1, β1, 1, 1>
- To send a 1 bit:
- Receiver receiving with A's code: (A's code) Γ (received chip pattern)
- User A '1' bit: 6 β 1
- User A '0' bit: -6 β 0
- User B '1' bit: 0 β unwanted signal ignored
π Advantages:
- Greatest spectrum efficiency
- CDMA improves call quality by filtering out background noise, cross-talk, and interference
- Simplified frequency planning - all users on a CDMA system use the same radio frequency spectrum
- Random Walsh codes enhance user privacy (a spread-spectrum advantage)
- Precise power control increases talk time and battery size for mobile phones
π Disadvantages:
- Backwards compatibility techniques are costly
- Currently, base station equipment is expensive
- Low traffic areas lead to inefficient use of spectrum and equipment resources
Random Access
Random Access methods are a more efficient way of managing medium access for communicating short, bursty messages. In contrast to fixed-access schemes, each user gains access to the medium only when needed (has data to send). The drawback is that users must compete to access the medium ('random access'), leading to collision of contending transmissions.
Random access methods in wireless networks are divided into two groups:
- ALOHA-based: no coordination between users
- Carrier-sense based: indirect coordinationβusers sense availability of medium before transmitting
ALOHA-based Random Access
In Pure ALOHA, a user accesses the medium as soon as it has a packet ready to transmit. After transmission, the user waits a length of time > round-trip delay in the network for an ACK (acknowledgment) from the receiver. If no ACK arrives, the user waits a random interval of time (to avoid repeated collision) and retransmits.
π Advantages:
- Simple
- No synchronization among users required
π Disadvantages:
- Low throughput under heavy load conditions
- Probability of collision increases as number of users increases
- Max throughput = 18% of channel capacity
Slotted ALOHA
In Slotted ALOHA, time is divided into equal time slots. When a user has a packet to transmit, the packet is buffered and transmitted at the start of the next time slot. The BS transmits a beacon signal for timing, and all users must synchronize their clocks.
π Advantages:
- Partial packet collision is avoided
π Disadvantages:
- Throughput is still quite low
- There is either no collision or a complete collision
- Max throughput = 36% of channel capacity
Slotted ALOHA in GSM
In GSM, there are two types of channels:
- Traffic channels (TCH): used for transmission of user dataβbased on FDMA/TDMA
- Signalling channels: used for control and management of a cellular network
- Random Access Channel (RACH): a signalling channel for establishing access to the network (BS)βemploys Slotted ALOHAβonly channel in GSM where contention can occur
Reservation ALOHA
In Reservation ALOHA (R-ALOHA), time slots are divided into reservation and transmission slots/periods. During the reservation period, stations can reserve future slots in the transmission period. The reservation slot size is much smaller than the transmission slot size. Collisions occur only in reservation slots.
π Advantages:
- Higher throughput under heavy loads
- Max throughput up to 80% of channel capacity
π Disadvantages:
- More demanding on users as they have to obtain/keep a 'reservation list' up-to-date
R-ALOHA is most commonly used in satellite systems. The satellite collects requests, compiles a 'reservation list', and finally sends the list back to users.
β Key Takeaways
The most critical concepts from this lecture are the fundamental differences between the four multiple access techniques: FDMA divides bandwidth into dedicated frequency channels for each user but suffers from inefficiency with idle channels; TDMA divides time into slots for shared use of the full bandwidth, requiring synchronization but improving spectrum efficiency; CDMA uses spread spectrum with unique codes per user, allowing all users to transmit simultaneously on the same frequency with high spectral efficiency; and Random Access methods like ALOHA, Slotted ALOHA, and Reservation ALOHA are essential for bursty traffic, with throughput increasing from 18% (Pure ALOHA) to 36% (Slotted ALOHA) and up to 80% (R-ALOHA) by introducing reservation mechanisms. Understanding the trade-offs in throughput, complexity, synchronization requirements, and implementation cost is key for choosing the right technique for different wireless network scenarios.
π§ Quick Revision Questions
- What is the maximum throughput of Pure ALOHA and Slotted ALOHA as a percentage of channel capacity?
- What is the key difference between how FDMA and TDMA allocate the communication channel to multiple users?
- In the CDMA example, if User A's code is
<1, -1, -1, 1, -1, 1>, what chip pattern is sent to represent a '0' bit? - In a GSM network, which channel type employs Slotted ALOHA and is the only channel where contention can occur?
- How does Reservation ALOHA achieve a significantly higher maximum throughput (up to 80%) compared to Pure ALOHA (18%)?
π Lecture 7 β CSMA and Spread Spectrum
π Overview: This lecture covers Carrier Sense Multiple Access (CSMA) protocols, which improve upon ALOHA by requiring nodes to listen before transmitting. It then introduces spread spectrum techniques, specifically Frequency Hopping Spread Spectrum (FHSS) and Direct Sequence Spread Spectrum (DSSS), which are used to combat interference and jamming in wireless networks.
ποΈ Topics Covered
This lecture begins with a review of ALOHA, Slotted ALOHA, and Reservation-based ALOHA. It then covers Carrier Sense Multiple Access (CSMA), including its non-persistent, 1-persistent, and p-persistent versions, as well as CSMA/CA (Collision Avoidance) with its algorithm details. The second half covers Spread Spectrum, focusing on Frequency Hopping Spread Spectrum (FHSS) using MFSK, including slow-hop and fast-hop variants, and performance considerations, followed by an introduction to Direct Sequence Spread Spectrum (DSSS).
π Lecture Summary
Last Lecture Review
The lecture reviews multiple access methods from the previous lecture: FDMA, TDMA, CDMA, and Random Access. It specifically revisits ALOHA, Slotted ALOHA, and Reservation-based ALOHA as random access protocols.
Carrier Sense Multiple Access (CSMA)
The section begins by outlining the disadvantages of ALOHA: users do not listen to the channel before or while transmitting, making it suitable only for networks with long propagation delays. CSMA is introduced as a "polite version of ALOHA" where the node listens to the channel before transmitting. If the channel is sensed busy, the node backs off (defers transmission) and senses the channel again after a random amount of time. If the channel is idle, the node transmits the entire frame.
π Definition β CSMA (Carrier Sense Multiple Access): A medium access control protocol where a node listens to the channel (carrier sense) before transmitting to avoid collisions.
Versions of CSMA
Different node behaviors are employed when the channel is found busy:
- Non-persistent CSMA: After sensing a busy channel, the node waits the entire back-off period before sensing again.
- Persistent CSMA: After sensing a busy channel, the node continues sensing until the channel becomes free. Then:
- 1-persistent CSMA: The node transmits immediately with probability 1.
- p-persistent CSMA: The node transmits with probability p, or defers transmission with probability (1-p).
CSMA / Collision Avoidance
CSMA/CA is used where CSMA/CD cannot be used, particularly in wireless mediums where collisions cannot be easily detected because the power of the transmitting antenna overwhelms the receiving antenna. It is designed to reduce collision probability at points where collisions would most likely occur, such as when the medium has become idle after a busy state, as several users could have been waiting. Key elements include:
- IFS (Interframe Spacing): A priority mechanism where the shorter the IFS, the higher the priority for transmission.
- CW intervals (Contention Window): Intervals used for contention and transmission of packet frames.
- Backoff counter: Used only if two or more stations compete for transmission.
π‘ Why this matters: CSMA/CA is the foundation of Wi-Fi (IEEE 802.11) medium access, making it essential for understanding how wireless local area networks avoid data collisions.
CSMA/CA Algorithm
The lecture presents a flowchart or algorithm for CSMA/CA (detailed in the source text as steps and decision points), which outlines the process of sensing the medium, waiting for IFS, managing the contention window, and decrementing the backoff counter to determine when to transmit.
Spread Spectrum
This section addresses the problem of radio transmission: frequency-dependent fading can wipe out narrowband signals for the duration of interference. The solution is to spread the narrowband signal into a broad band signal using a special code. This technique was initially developed for military use to combat jamming and interception. The power of the spread signal is the same as the narrowband signal, resulting in a lower power spectral density due to the larger bandwidth.
π Definition β Spread Spectrum: A technique where a narrowband signal is spread across a wider frequency band using a special code, making it resistant to interference and jamming.
Types of Spreading
Two main types are introduced:
- Direct Sequence Spread Spectrum (DSSS)
- Frequency Hopping Spread Spectrum (FHSS)
Frequency Hopping Spread Spectrum (FHSS)
In FHSS, the signal is broadcast over a seemingly random series of radio frequencies. A number of channels are allocated for the FH signal, and the width of each channel corresponds to the bandwidth of the input signal. The signal hops from frequency to frequency at fixed intervals: the transmitter operates in one channel at a time, bits are transmitted using some encoding scheme, and at each successive interval, a new carrier frequency is selected. The channel sequence is dictated by a spreading code. The receiver, hopping between frequencies in synchronization with the transmitter, picks up the message.
- Advantages: Eavesdroppers hear only unintelligible blips, and attempts to jam the signal on one frequency succeed only at knocking out a few bits.
π Definition β FHSS (Frequency Hopping Spread Spectrum): A spread spectrum technique where the signal is broadcast over a seemingly random series of radio frequencies, hopping between frequencies at fixed intervals as dictated by a spreading code.
FHSS Using MFSK
In this implementation, an MFSK (Multiple Frequency Shift Keying) signal is translated to a new frequency every Tc seconds by modulating the MFSK signal with the FHSS carrier signal. For a data rate of R:
- Duration of a bit: T = 1/R seconds
- Duration of signal element: Ts = L T seconds
Two variants are defined based on the relationship between Tc and Ts:
- Tc β₯ Ts: Slow-frequency-hop spread spectrum
- Tc < Ts: Fast-frequency-hop spread spectrum
π Definition β Slow-hop FHSS: A frequency hopping variant where the hop rate is slower than or equal to the signal element rate (Tc β₯ Ts). π Definition β Fast-hop FHSS: A frequency hopping variant where the hop rate is faster than the signal element rate (Tc < Ts).
π Formula: For FHSS with MFSK, bit duration T = 1/R, and signal element duration Ts = L T. π Example: If data rate R = 1000 bps, then bit duration T = 1/1000 = 1 ms. If L = 2 bits per signal element, then Ts = 2 * 1 ms = 2 ms. If Tc = 1 ms, then Tc < Ts, resulting in fast-hop FHSS.
FHSS Performance Considerations
Using a large number of frequencies results in a system that is quite resistant to jamming. The jammer must jam all frequencies, and with fixed power, this reduces the jamming power in any one frequency band.
Direct Sequence Spread Spectrum (DSSS)
In DSSS, each bit in the original signal is represented by multiple bits in the transmitted signal. The spreading code spreads the signal across a wider frequency band, and the spread is in direct proportion to the number of bits used. One technique combines the digital information stream with the spreading code bit stream using an exclusive-OR (XOR) operation.
π Definition β DSSS (Direct Sequence Spread Spectrum): A spread spectrum technique where each bit in the original signal is represented by multiple bits (chips) in the transmitted signal, spreading it across a wider frequency band in proportion to the number of chips used.
β Key Takeaways
CSMA improves on ALOHA by requiring nodes to listen (carrier sense) before transmitting, with variants like non-persistent, 1-persistent, and p-persistent defining different behaviors when the channel is busy. CSMA/CA is essential for wireless networks because collision detection (CSMA/CD) is impractical due to the overwhelming transmit power at the receiving antenna. Spread Spectrum techniques, FHSS and DSSS, solve the problem of frequency-dependent fading and jamming by spreading a narrowband signal across a wider bandwidth using a special code. FHSS hops frequencies at fixed intervals using a spreading code, while DSSS represents each bit with multiple chips using XOR with a spreading code.
π§ Quick Revision Questions
- What are the two main disadvantages of ALOHA that CSMA addresses?
- In p-persistent CSMA, what happens if a node senses the channel is busy? If idle, what is the probability of immediate transmission?
- Why is CSMA/CA preferred over CSMA/CD in wireless networks?
- What is the difference between slow-frequency-hop and fast-frequency-hop spread spectrum in terms of Tc and Ts?
- How does Direct Sequence Spread Spectrum (DSSS) use a spreading code to spread the signal?
π Lecture 8 β Evolution of Wireless Networks
π Overview: This lecture traces the evolution of wireless networks from pre-cellular high-power systems to first-generation (1G) analog cellular standards and second-generation (2G) digital cellular standards. It covers the key technological shifts that enabled higher capacity, better security, and new data services, laying the foundation for modern mobile communications.
ποΈ Topics Covered
The lecture reviews previous content on CSMA and 2G systems, then details the evolution of wireless systems from pre-cellular high-power systems to cellular architecture. It thoroughly examines first-generation analog cellular standards including NMT, AMPS, and TACS along with their technical specifications and limitations. The lecture then introduces second-generation digital cellular networks, explaining the advantages of digital over analog, and provides detailed coverage of GSM, IS-136 (D-AMPS), PDC, and IS-95 (cdmaOne) standards.
π Lecture Summary
Today Goals
The lecture aims to review previous lecture #7 and cover 1G wireless cellular networks including NMT, AMPS, and TACS, as well as spread spectrum techniques of frequency hopping and direct sequence.
Last Lecture Review
Previous lecture covered CSMA, its versions and CSMA/CA with examples, and 2G cellular systems including GSM, IS-136, PDC, and IS-95.
Evolution of Wireless Systems
The worldwide success of cellular telephone led to development of newer wireless systems for other types of communications beyond mobile voice, including high-speed data traffic, replacing fiber optics and copper lines over several kilometers, and replacing wires within homes and offices (leading to Bluetooth evolution).
Prior to cellular phones, mobile telephone service used a high power transmitter/receiver supporting about 25 channels with an effective radius of about 80 km. The way to increase system capacity is to use lower-power systems with shorter radius and numerous transmitters/receivers. Cellular systems evolved to provide organization of these transmitters/receivers and further improve system capacity.
First-Generation Cellular Networks
First-generation (1G) networks were analog systems with several standards:
- NMT (Nordic Mobile Telephone) β used in Nordic countries, Switzerland, Netherlands, Eastern Europe, and Russia
- AMPS (Advanced Mobile Phone System) β used in the United States
- TACS (Total Access Communications System) β used in the United Kingdom
- C-450 β in West Germany, Portugal, and South Africa
- Radiocom 2000 β in France
- RTMI β in Italy
- Japan had multiple systems with three standards: TZ-801, TZ-802, and TZ-803
NMT (Nordic Mobile Telephone)
NMT was the first fully-automatic cellular phone system, started in 1970 and in service by 1981. It had two standards: NMT-450 and NMT-900, corresponding to different frequency bands, with NMT-900 using higher bands.
π Definition β Cell size range: Cell sizes range from 2 km to 30 km, using smaller sizes in urban areas for better quality and larger sizes in less-populated areas.
π Definition β Handset power: Handsets use 1 watt, while car phones use 6-15 watts.
NMT featured automatic switching (dialing) and handover. It had no specification for voice traffic encryption β anyone could buy a scanner, tune to the desired channel, and intercept calls. NMT also supported a simple data transfer mode called DMS (Data and Messaging Service) or NMT-Text. Using DMS, text messaging was possible between two NMT handsets before SMS service started in GSM, but this feature was never commercially available except in Russian and Polish NMT networks.
NMT was suspended in Finland (TeliasSonera, December 31, 2002), Norway (December 31, 2004), and Sweden (TeliasSonera, December 31, 2007).
AMPS (Advanced Mobile Phone System)
AMPS was the 1G cellular phone system used in the US, which uses FDMA (Frequency Division Multiple Access). It operates in the 800 MHz band.
π Formula: Total of 832 channels β 416 in 824β849 MHz for transmissions from mobile to base, and 416 in 869β894 MHz for transmissions from base to mobile. Each channel is 30 KHz wide.
AMPS required large bandwidth for large base population. It offered no protection against eavesdroppers β the ESN (Electronic Serial Number) was cloned in the 1990s to make free calls from different cells. AMPS was replaced with D-AMPS, GSM, and CDMA for better security and capacity.
TACS (Total Access Communications System)
TACS is a variant of AMPS developed by Motorola. It was used in some European countries (including the UK and Ireland), as well as Japan and Hong Kong. ETACS was an extended version of TACS with more channels. The last ETACS service operated by Vodafone was discontinued on May 31, 2001.
π‘ Why this matters: 1G systems were purely analog with limited capacity, no encryption (easy eavesdropping), and no data services. Their vulnerabilities motivated the shift to digital 2G systems.
Second-Generation Cellular Networks
Second-generation (2G) networks are digital systems β voice is digitized. Unlike 1G which relied on FDMA/FDD, 2G uses digital modulation formats and TDMA/FDD or CDMA/FDD multiple access techniques. They can be divided into two standards: TDMA-based and CDMA-based.
The main 2G standards are:
- GSM (TDMA-based), originally from Europe but used worldwide
- IS-136 aka D-AMPS (TDMA-based), used in the Americas
- IS-95 aka cdmaOne (CDMA-based), used in the Americas and parts of Asia
- PDC (TDMA-based), used exclusively in Japan
Using digital signals between handsets and towers increases system capacity in two key ways:
- Digital voice data can be compressed and multiplexed much more effectively than analog voice encodings through various CODECs, allowing more calls in the same radio bandwidth.
- Digital systems emit less radio power from handsets, allowing smaller cells so more cells can be placed in the same space, made possible by less expensive cell towers and equipment.
π Definition β 2G Advantages:
- Lower powered radio signals require less battery power, so phones last longer between charges with smaller batteries
- Digital voice encoding allows digital error checking, improving sound quality by reducing dynamic range and lowering the noise floor
- Going all-digital allows digital data services such as SMS and email
- Better security β harder to be scanned
GSM (Global System for Mobile Communications)
GSM has 2.27 billion subscribers across more than 212 countries, representing 81% of the global mobile market. Its ubiquity provides very common international roaming.
π Formula: 8-slots TDMA with 200 KHz radio channel, with frame duration of 4.615 ms. The channel data rate is 270.833 kbit/s.
GSM operates in four different bands: mostly 900 MHz or 1800 MHz, while the US and Canada use 850 MHz and 1900 MHz. Each band has 25 MHz bandwidth subdivided into 124 channels. For example, in 900 MHz: uplink 890-915 MHz, downlink 935-960 MHz.
Other Systems
IS-136 or D-AMPS:
- 3-slot TDMA, used in North and South America, Australia
- Channel bandwidth is 30 KHz
- Frequency bands: 824-849 MHz and 869-894 MHz
Pacific Digital Cellular (PDC):
- Japanese standard similar to IS-136
- 25 KHz channel
- Data rates: 11.2 kbps at 3-slot and 5.6 kbps at 6-slot
- Operates in 800 MHz (downlink 810-888 MHz, uplink 893-958 MHz) and 1.5 GHz (downlink 1477-1501 MHz, uplink 1429-1453 MHz)
IS-95 or cdmaOne:
- Supports up to 64 users that are orthogonally coded
- Channel bandwidth is 1.25 MHz
- Widely deployed in North America, Korea, Japan, China, South America, Australia
- Channel data rate is 1.2288 Mchips/s (Mega Chips per second)
β Key Takeaways
The evolution from 1G to 2G represents a fundamental shift from analog to digital cellular technology, bringing dramatic improvements in capacity, security, battery life, and new data services like SMS. 1G systems like NMT, AMPS, and TACS were analog with FDMA, no encryption (allowing eavesdropping and cloning), and limited capacity determined by high-power transmitters with large cells. 2G systems introduced digital modulation with TDMA (GSM, IS-136, PDC) or CDMA (IS-95), enabling voice compression, error checking, smaller cells, better security, and international roaming. Students must know the specific technical parameters of each standard including channel bandwidth, frequency bands, number of slots, and data rates for exam comparison questions.
π§ Quick Revision Questions
- What are the key differences between 1G and 2G cellular systems in terms of modulation, multiple access technique, and security?
- What is the channel bandwidth, frequency band, and number of channels for AMPS?
- How does GSM achieve capacity improvements through its TDMA structure (number of slots, channel bandwidth, frame duration, and data rate)?
- Compare IS-136 (D-AMPS) and PDC in terms of channel bandwidth, number of TDMA slots, and data rates.
- What is the channel bandwidth and chip rate for IS-95 (cdmaOne), and how many orthogonal users does it support?
π Lecture 9 β Evolution of Wireless Networks (Part II)
π Overview: This lecture continues the evolution of wireless networks, covering the transition from 2G to 2.5G technologies and the emergence of 3G standards. It explains the upgrade paths for GSM and CDMA networks, detailing how data rates and services improved without completely replacing existing infrastructure.
ποΈ Topics Covered
This lecture begins with a review of 1G and 2G systems from the previous lecture, then introduces 2.5G technologies including HSCSD, GPRS, and EDGE, along with IS-95B. It then covers 3G evolution, focusing on the two main camps: UMTS/W-CDMA from the GSM side and CDMA2000 from the CDMA side. The lecture concludes with a brief overview of TD-SCDMA, China's 3G standard.
π Lecture Summary
Review of last lecture #8
The review covers 1G wireless cellular networks including NMT, AMPS, and TACS, as well as key 2G technologies such as GSM, IS-136, PDC, and IS-95. These provide the foundation for understanding the upgrade paths discussed in this lecture.
Evolution to 2.5G
2.5G upgrades must be compatible with existing 2G technology. Three different upgrade paths were developed for GSM, with two also supporting IS-136: High Speed Circuit Switched Data (HSCSD), General Packet Radio Service (GPRS), and Enhanced Data rates for GSM Evolution (EDGE). GPRS and EDGE support IS-136, while IS-95B serves as the upgrade for IS-95.
HSCSD
HSCSD works in circuit switch mode. Speed is increased by allowing a single user to use consecutive time slots in the GSM standard. It relaxes error control coding algorithms specified in GSM, increasing data rate from 9.600 to 14.400 Kbps per slot. By using 4 slots, a raw data rate of up to 57.6 kbps is available to an individual user. This technology is ideal for dedicated streaming or real-time interactive web sessions.
π Definition β HSCSD: High Speed Circuit Switched Data β a 2.5G upgrade that increases data rates by allowing a single user to use multiple consecutive GSM time slots in circuit-switched mode.
π Formula β Data rate calculation: 4 slots Γ 14.400 Kbps = 57.6 kbps
π Example: A user streaming a video would be assigned 4 consecutive GSM time slots, achieving a data rate of 57.6 kbps using relaxed error control coding.
GPRS
GPRS is packet-based and well-suited for non-real-time traffic like email, faxes, and web browsing. Unlike HSCSD, GPRS allows multi-user channel sharing of individual radio channels and time slots, supporting many more users. GPRS units are automatically instructed to tune to dedicated GPRS channels and particular time slots for always-on access. When all 8 slots are dedicated, the data rate reaches 171.2 kbps (8 Γ 21.4 kbps of raw un-coded data).
π‘ Why this matters: GPRS introduced packet-switching to cellular networks, enabling efficient "always-on" data connections and better spectrum utilization compared to circuit-switched HSCSD.
EDGE (2.75G)
EDGE is a more advanced upgrade to 2G that requires new hardware and software. It was developed as a path to eventual 3G high-speed data access. EDGE introduces 8-PSK modulation in addition to the GSM standard GMSK. It allows nine different formats known as Multiple Modulation and Coding Scheme (MCS). Each MCS state may use either GMSK (low rate) or 8-PSK (high rate). Users start first with maximum error protection and maximum data rate until the link has unacceptable outage or delay. By combining different channels using multi-carrier transmission, EDGE provides up to several megabits per second data throughput.
π Definition β EDGE: Enhanced Data rates for GSM Evolution β a 2.75G technology that uses 8-PSK modulation and MCS to achieve higher data rates than GPRS, providing a path to 3G.
IS-95B or cdmaOne
IS-95/CDMA has a single upgrade path IS-95B for eventual 3G operation. It dedicates multiple orthogonal user channels for specific users. IS-95A supports 64 users with a data rate of 14,400 Kbps. The medium data rate service allows a user to command up to 8 Walsh codes, achieving a raw data rate of 8 Γ 14,400 = 115.2 kbps. IS-95B supports hard handoff procedure, allowing units to search different radio channels without instruction from the switch, enabling rapid tuning to different base stations.
π Formula β IS-95B data rate: 8 Walsh codes Γ 14,400 bps = 115.2 kbps
Evolution to 3G
Third generation mobile phone standards are based on the International Telecommunication Union (ITU) family of standards under the International Mobile Telecommunications programme, "IMT-2000". 3G technologies enable network operators to offer a wider range of advanced services while achieving greater network capacity through improved spectral efficiency. Services include broadband wireless data in a mobile environment, typically at 5-10 Mb per second. The most significant feature of 3G is that it supports greater numbers of voice and data customers at higher data rates at lower incremental cost than 2G.
3G Evolution
The community remains split into two camps. The GSM/IS-136/PDC camp's 3G evolution is Wideband CDMA (W-CDMA), also known as UMTS. The IS-95B or CDMA camp's evolution path is cdma2000, with several variants all based on IS-95B. ITU-2000 standards are separated into two major organizations reflecting these two camps: 3GPP (3G Partnership Project) for W-CDMA and 3GPP2 (3G Partnership Project 2) for cdma2000.
π Definition β 3GPP: 3G Partnership Project β the organization responsible for standardizing W-CDMA (UMTS) as the 3G evolution for GSM-based networks.
π Definition β 3GPP2: 3G Partnership Project 2 β the organization responsible for standardizing cdma2000 as the 3G evolution for CDMA-based networks.
3G W-CDMA (UMTS)
This standard evolved under the European Telecom. Standards Institute (ETSI). It is backward compatible with 2G standards GSM, IS-136, and PDC technologies, as well as 2.5G. Bit-level packaging of GSM data is retained, with additional capacity and bandwidth provided by a new CDMA air interface. It provides always-on packet-based service for computers, entertainment devices, and telephones. W-CDMA requires expensive new base station equipment, making installation slow and gradual. It supports data rates up to 2.048 Mbps per user, allowing high-quality data, multimedia, and streaming audio (for stationary users). Future versions will support data rates in excess of 8 Mbps. The minimum spectral allocation is 5 MHz. Data rates from as low as 8 kbps to as high as 2 Mbps can be carried simultaneously on a single radio channel. Each channel can support between 100 and 350 voice calls simultaneously, depending on propagation conditions.
π Definition β W-CDMA/UMTS: Wideband Code Division Multiple Access/Universal Mobile Telecommunications System β the 3G standard for GSM-based networks, using CDMA air interface with 5 MHz minimum spectral allocation and data rates up to 2.048 Mbps.
3G CDMA 2000
cdma2000 provides a seamless and evolutionary upgrade path for 2G and 2.5G CDMA technology. It centers on the original 1.25 MHz radio channel. CDMA operators may seamlessly and selectively upgrade without changing entire base station equipment. The first 3G CDMA standard is cdma2000 1xRTT, using a single channel (1x meaning multi-carrier). cdma2000 1x supports data rates up to 307 kbps in packet mode and can support up to twice as many users as 2G CDMA, requiring no additional equipmentβsimply software and new channel cards at the base station. cdma2000 1xEV is a proprietary high data rate packet standard by Qualcomm, overlaid on existing CDMA. CDMA 1xEV-DO dedicates the channel strictly to data users and supports 2.4 Mbps per channel. cdma2000 3xRTT is the ultimate 3G solution, relying on multicarrier that gangs adjacent channels together into 3.75 MHz. Three non-adjacent channels may be operated simultaneously and in parallel, providing data rates in excess of 2 Mbps. Advocates of cdma2000 claim their standard provides a much more seamless and less expensive upgrade path compared to W-CDMA.
π Definition β cdma2000 1xRTT: The first 3G CDMA standard using single 1.25 MHz channel, supporting up to 307 kbps in packet mode and up to twice as many users as 2G CDMA.
π Definition β cdma2000 1xEV-DO: Evolution Data Optimized β a cdma2000 variant that dedicates the channel strictly to data users, supporting up to 2.4 Mbps per channel.
3G TD-SCDMA
In China, where more than 8 million GSM subscribers were added in just one month, China's desire to craft its own wireless vision led to the Time Division Synchronous Code Division Multiple Access (TD-SCDMA) standard. Chinese CATT and Siemens jointly submitted this IMT-2000 3G standard. TD-SCDMA relies on existing GSM infrastructure, uses 1.6 MHz channels and smart antennas to yield more spectral efficiency. It uses 5 ms frames divided into 7 slots allocated to a single data-only user or several slow users. TD-SCDMA allows easy upgrade from GSM.
β Key Takeaways
The lecture shows that 2.5G was a necessary stepping stone using HSCSD, GPRS, and EDGE to improve data rates without full 3G deployment. The cellular world split into two 3G camps: UMTS/W-CDMA (backward compatible with GSM, requiring new BS equipment, 2 Mbps data) and cdma2000 (seamless CDMA upgrade using 1.25 MHz channels, with 1xRTT at 307 kbps and 1xEV-DO at 2.4 Mbps). IS-95B was the CDMA upgrade using Walsh codes for 115.2 kbps. GPRS introduced packet-switched always-on data, while EDGE added 8-PSK modulation and MCS for higher rates. China developed TD-SCDMA as its own 3G standard using 1.6 MHz channels and smart antennas with GSM compatibility.
π§ Quick Revision Questions
- What are the three upgrade paths for GSM to 2.5G, and which two also support IS-136?
- How does GPRS differ from HSCSD in terms of channel sharing and suitability for different traffic types?
- What modulation scheme did EDGE add beyond GSM's GMSK, and what is the MCS framework?
- What are the data rates and spectral allocation differences between UMTS/W-CDMA (5 MHz) and cdma2000 (1.25 MHz)?
- Which organization (3GPP or 3GPP2) is responsible for W-CDMA, and which for cdma2000?