Loading learning content...
The history of cellular communications is, in large part, the history of channelization techniques. From the first analog cellular systems using pure FDMA, through TDMA's digital revolution, to CDMA's spread spectrum breakthrough, and finally to the sophisticated OFDMA of 4G/5G—each generation has refined how we share the precious radio spectrum among millions of simultaneous users.
Understanding how these channelization techniques deploy in real cellular systems ties together everything we've learned. We'll see why certain technologies succeeded in their eras, how hybrid approaches combine the best of multiple techniques, and how modern systems achieve spectral efficiencies that early engineers would have deemed impossible.
This page serves as both a practical application summary and a bridge to understanding current and future wireless technologies.
By the end of this page, you will understand how FDMA, TDMA, and CDMA have been deployed across cellular generations, why GSM dominated 2G while CDMA powered 3G, how LTE's OFDMA combines frequency and time division elegantly, the spectral efficiency evolution from 1G to 5G, and future directions for multiple access in beyond-5G systems.
The first generation of cellular systems, deployed in the 1980s, used exclusively FDMA combined with FDD (Frequency Division Duplexing). These were analog systems—voice was transmitted as FM (Frequency Modulation) directly, without digitization.
Major 1G Systems:
| System | Region | Frequency Band | Channel Spacing | Year Deployed |
|---|---|---|---|---|
| AMPS | North America | 824-894 MHz | 30 kHz | 1983 |
| TACS | Europe (UK) | 890-960 MHz | 25 kHz | 1985 |
| NMT-450 | Nordic countries | 453-468 MHz | 25 kHz | 1981 |
| NMT-900 | Nordic countries | 890-960 MHz | 12.5 kHz | 1986 |
| C-450 | Germany | 450-465 MHz | 20 kHz | 1985 |
| J-TACS | Japan | 860-925 MHz | 25 kHz | 1988 |
AMPS Architecture:
AMPS (Advanced Mobile Phone System) exemplifies 1G design:
Cellular Concept:
To maximize capacity, 1G systems introduced the cellular concept—dividing coverage areas into cells, each served by a base station. The same frequencies could be reused in non-adjacent cells:
$$\text{Frequency Reuse Factor} = \frac{1}{N}$$
where $N$ typically equals 7 for AMPS (7-cell reuse pattern). This means each cell uses 1/7 of the total spectrum.
1G Capacity Calculation:
$$\text{Channels per cell} = \frac{\text{Total channels}}{N} = \frac{832}{7} \approx 119 \text{ channels}$$
With 21 control channels per cell, voice capacity ≈ 98 simultaneous calls per cell.
1G Limitations:
Despite its limitations, 1G proved cellular telephony viable. AMPS grew from 0 to 55 million subscribers before 2G deployment. The cellular concept—frequency reuse, handoff, roaming—established principles still fundamental in 5G. AMPS finally shut down in 2007 after 24 years of service.
2G marked the shift from analog to digital, introducing TDMA as the dominant multiple access technique (though CDMA also emerged). Digital transmission enabled voice compression, encryption, and data services impossible with analog.
GSM: The Global Success Story
GSM (Global System for Mobile Communications) combined TDMA with FDMA:
Capacity Improvement over AMPS:
$$\text{GSM channels per MHz} = \frac{1 \text{ MHz}}{200 \text{ kHz}} \times 8 = 40 \text{ time slots}$$
$$\text{AMPS channels per MHz} = \frac{1 \text{ MHz}}{30 \text{ kHz}} = 33 \text{ channels}$$
Direct improvement: 40/33 ≈ 1.2×. But with Half Rate codec (16 slots per carrier), GSM achieves 80 channels per MHz—2.4× AMPS.
| System | Access Method | Carrier BW | Voice Channels | Region |
|---|---|---|---|---|
| GSM | TDMA/FDMA | 200 kHz | 8/16 per carrier | Global |
| IS-136 (D-AMPS) | TDMA/FDMA | 30 kHz | 3 per carrier | Americas |
| IS-95 (cdmaOne) | CDMA | 1.25 MHz | ~40 per carrier | Americas, Asia |
| PDC | TDMA/FDMA | 25 kHz | 3 per carrier | Japan |
| PHS | TDMA/FDMA | 300 kHz | 4 per carrier | Japan |
GSM became the most successful mobile standard in history. At its peak, GSM served over 5 billion connections in 219 countries. The combination of TDMA efficiency, SIM card flexibility, and comprehensive specification (enabling multi-vendor equipment) created a virtuous cycle of adoption, manufacturing scale, and cost reduction.
3G systems were designed for high-speed data from the ground up. CDMA became the dominant air interface technology, chosen for its superior spectral efficiency and ability to support variable data rates.
Major 3G Standards:
WCDMA (UMTS):
CDMA2000:
TD-SCDMA:
| Parameter | WCDMA (UMTS) | CDMA2000 1x | TD-SCDMA |
|---|---|---|---|
| Chip rate | 3.84 Mcps | 1.2288 Mcps | 1.28 Mcps |
| Channel BW | 5 MHz | 1.25 MHz | 1.6 MHz |
| Duplexing | FDD | FDD | TDD |
| Max SF | 512 (DL) / 256 (UL) | 256 / 64 | 16 |
| Voice capacity /5MHz | ~100 users | ~40 users/1.25MHz | ~50 users |
| Peak DL (enhanced) | 14.4 Mbps (HSPA) | 3.1 Mbps (EV-DO) | 2.8 Mbps |
Why CDMA for 3G?
CDMA offered several advantages crucial for 3G:
Soft Capacity — Data users have bursty traffic. CDMA's interference-based capacity naturally accommodates varying loads.
Variable Rate Support — OVSF codes enable seamless rate switching from 12 kbps voice to 2 Mbps data.
Soft Handoff — Mobile data sessions benefit from continuous connectivity during handoff.
Universal Frequency Reuse — Critical for data services where frequency planning complexity multiplies.
Power Efficiency — Digital signal processing advances made CDMA receivers practical in mobile devices.
HSPA Enhancement:
High-Speed Packet Access (HSPA) dramatically improved 3G performance:
Key HSPA Innovations:
3G transformed mobile phones from voice devices to pocket computers. App stores, mobile video, social networking—all became possible with 3G data speeds. HSPA's peak rates of 42 Mbps rivaled many home broadband connections, enabling smartphones to replace laptops for many tasks.
4G/LTE marked a radical departure from CDMA: the adoption of OFDMA (Orthogonal Frequency Division Multiple Access). This hybrid technique combines FDMA's frequency separation with digital signal processing innovations to achieve unprecedented spectral efficiency.
OFDMA Concept:
OFDMA divides the channel into many narrowband subcarriers (typically 15 kHz in LTE), then allocates groups of subcarriers (called Resource Blocks) to different users:
Why OFDMA over CDMA?
| Parameter | Value | Notes |
|---|---|---|
| Channel bandwidth | 1.4–20 MHz | Scalable in 1.4, 3, 5, 10, 15, 20 MHz |
| Subcarrier spacing | 15 kHz | Uniform across bandwidth |
| Resource Block | 12 subcarriers × 7 symbols | Smallest allocation unit |
| Slot duration | 0.5 ms | 7 OFDM symbols |
| Subframe duration | 1 ms | 2 slots, scheduling granularity |
| Frame duration | 10 ms | 10 subframes |
| Peak DL rate | 150 Mbps (Cat 4) | 20 MHz, 2×2 MIMO, 64-QAM |
| Peak UL rate | 50 Mbps (Cat 4) | 20 MHz, 16-QAM |
LTE-Advanced (4G+):
LTE-Advanced enhanced performance through:
Spectral Efficiency Evolution:
| Generation | Peak Spectral Efficiency |
|---|---|
| 2G (GSM) | 0.1 bps/Hz |
| 3G (WCDMA R99) | 0.5 bps/Hz |
| 3G (HSPA+) | 5.5 bps/Hz |
| 4G (LTE) | 15 bps/Hz |
| 4G (LTE-A) | 30 bps/Hz |
OFDMA's combination of frequency and time multiplexing with per-subcarrier scheduling achieves near-Shannon-limit efficiency. By assigning each user the subcarriers where their channel is best, LTE exploits frequency-selective fading rather than fighting it. This channel-aware scheduling is impossible with CDMA's spread-spectrum approach.
5G New Radio (NR) builds on LTE's OFDMA foundation while introducing flexible numerology to serve diverse use cases—from ultra-reliable low-latency communication (URLLC) to massive machine-type communication (mMTC).
5G NR Numerologies:
5G defines multiple subcarrier spacings (SCS) to optimize for different scenarios:
| Numerology (µ) | Subcarrier Spacing | Symbol Duration | Use Case |
|---|---|---|---|
| 0 | 15 kHz | 66.7 µs | Sub-6 GHz, low mobility |
| 1 | 30 kHz | 33.3 µs | Sub-6 GHz, general use |
| 2 | 60 kHz | 16.7 µs | Sub-6 GHz, low latency |
| 3 | 120 kHz | 8.3 µs | mmWave, high data rate |
| 4 | 240 kHz | 4.2 µs | mmWave, experimental |
Why Flexible SCS?
| Feature | 5G NR | Comparison to LTE |
|---|---|---|
| Frequency range | Sub-6 GHz + mmWave (24–100 GHz) | LTE: sub-6 GHz only |
| Max bandwidth | 400 MHz (mmWave) | LTE: 20 MHz (100 MHz with CA) |
| Subcarrier spacing | 15–240 kHz | LTE: 15 kHz only |
| Max MIMO layers | 8 DL / 4 UL | LTE: 8 DL / 4 UL |
| Peak DL rate | 20 Gbps | LTE-A: 3 Gbps |
| Latency (user plane) | 1 ms | LTE: 10 ms |
Massive MIMO in 5G:
5G exploits Massive MIMO with 64–256 antenna elements at base stations:
Spectral Efficiency:
5G NR achieves 30+ bps/Hz peak spectral efficiency through:
5G's flexible architecture serves three distinct use cases: (1) eMBB (enhanced Mobile Broadband): 20 Gbps for video streaming, AR/VR; (2) URLLC (Ultra-Reliable Low Latency): 1 ms latency for autonomous vehicles, remote surgery; (3) mMTC (massive Machine-Type Communication): 1 million devices/km² for IoT sensors. Different numerologies and configurations optimize for each.
The evolution of channelization techniques has driven remarkable improvements in spectral efficiency—the amount of data transmitted per Hz of bandwidth. Let's quantify this progression.
| Generation | Technology | Spectral Efficiency (bps/Hz) | Improvement vs. 1G |
|---|---|---|---|
| 1G | AMPS (FDMA) | 0.03 (voice) | 1× |
| 2G | GSM (TDMA) | 0.13 (voice) | 4× |
| 2G | IS-95 (CDMA) | 0.20 (voice) | 7× |
| 2.5G | EDGE | 0.47 (data) | 16× |
| 3G | WCDMA R99 | 0.5 (data) | 17× |
| 3G | HSPA | 1.5 (data) | 50× |
| 3G | HSPA+ | 5.5 (data) | 180× |
| 4G | LTE | 5-7 (average) | 200× |
| 4G | LTE-A | 15 (peak) | 500× |
| 5G | NR (sub-6 GHz) | 10-15 (average) | 400× |
| 5G | NR (mmWave) | 30+ (peak) | 1000× |
Efficiency Drivers by Generation:
1G → 2G (4× improvement):
2G → 3G (10× improvement):
3G → 4G (10× improvement):
4G → 5G (3× improvement):
With 5G achieving 30+ bps/Hz, we're approaching Shannon's theoretical limit of ~40 bps/Hz at typical SNR values. Future capacity growth must come from: (1) more spectrum (mmWave, spectrum sharing), (2) more antennas (Massive MIMO, reconfigurable intelligent surfaces), (3) cell densification. Pure modulation/coding gains are nearly exhausted.
Modern and future systems increasingly combine multiple access techniques to optimize for different scenarios. Understanding these hybrids reveals the ongoing evolution of channelization.
| Technique | Domain | Orthogonal? | Complexity | Best For |
|---|---|---|---|---|
| FDMA | Frequency | Yes | Low | Analog, broadcast |
| TDMA | Time | Yes | Medium | Digital voice |
| CDMA | Code | Quasi | High | 3G, GPS |
| OFDMA | Freq + Time | Yes | Medium | 4G/5G broadband |
| NOMA | Power | No | High | Future IoT |
| SCMA | Code (sparse) | No | Very High | 6G mMTC |
6G Research Directions:
Beyond 5G, researchers are exploring:
Terahertz Communications: 0.1–10 THz bands with 100+ GHz bandwidth, requiring new multiple access designs for extreme directional beams.
Reconfigurable Intelligent Surfaces (RIS): Programmable reflectors that shape propagation environment, creating new interference management possibilities.
Semantic Communications: Transmit meaning rather than bits, potentially revolutionizing what "capacity" means.
AI-Native Multiple Access: Machine learning-optimized resource allocation that adapts to traffic patterns and user behavior in real-time.
While OFDMA dominates current systems, the multiple access story is far from over. Non-orthogonal techniques promise to break capacity barriers by allowing controlled overlap. AI/ML will optimize allocations in ways impossible to analyze mathematically. The principles you've learned—orthogonality, interference, efficiency—remain fundamental even as implementations evolve.
We've traced channelization from 1G's analog FDMA through 5G's sophisticated OFDMA. Let's consolidate the essential concepts from this page and the entire module:
Module Conclusion:
Channelization represents a fundamental engineering challenge: sharing finite spectrum among unlimited users. The solutions—frequency division, time division, code division, and their hybrids—demonstrate how mathematical orthogonality translates into practical communication systems.
From the simple elegance of FDMA to the sophisticated signal processing of CDMA to the adaptive scheduling of OFDMA, each technique offers distinct trade-offs. Understanding these trade-offs enables network architects to choose appropriate technologies and communications engineers to optimize system performance.
As wireless demand continues to grow exponentially—driven by video, IoT, autonomous vehicles, and applications yet unimagined—the principles you've learned in this module will remain relevant even as specific technologies evolve.
Congratulations! You've completed the Channelization module with a comprehensive understanding of FDMA, TDMA, CDMA, code orthogonality, and their cellular applications. These structured multiple access techniques form the foundation of all modern wireless communication systems. You now understand not just how they work, but why each was chosen for its era and application—knowledge that will serve you well as wireless technology continues to evolve.