Computer Engineering is coming soon

We're still writing and reviewing the notes for every subject, and we'll open it when every subject is ready. Meanwhile you can practise aptitude, ask Aria about any subject, or see the branches that are ready.

Evolution from 1G to 5G and 6G

Clear, exam-focused notes tracing mobile cellular system evolution from 1G to 6G, explaining key technologies (FDMA/TDMA/CDMA/OFDMA, MIMO, mmWave, beamforming, network slicing), laws used (Shannon, Friis, reuse geometry), worked numerical examples, common mistakes and likely exam questions.

Drafted with Aria, reviewed by the AiCanCode.org team · Updated . How we write and check our notes. Spotted an error? Use Give Feedback at the bottom of the page.

Why it matters

Two or three sentences: Mobile cellular systems are the foundation of today’s wireless services. Engineers use the evolution from 1G → 5G (and ideas for 6G) to choose radio access methods, spectrum, and antenna techniques to meet capacity, latency and coverage requirements for applications such as voice, video, IoT and edge AI.

Key ideas

Start from first principles and simple definitions. Define terms before using them.

  • Cellular system: a radio network divided into geographic cells (small areas) each served by a base station; cells reuse frequencies so capacity scales with area. Define cell radius R as the distance from base station to a cell vertex (corner); this convention matches the reuse-distance formula D = R · sqrt(3N).

  • Generation (1G, 2G, ...): a labeled step in mobile system evolution where each generation denotes major changes in air interface, modulation, multiple access and services. Exams expect short definitions: 1G = analog voice, 2G = digital voice (GSM and IS-95/cdmaOne), 3G = mobile data (WCDMA/UMTS and CDMA2000), 4G = all-IP packet data (LTE: OFDMA downlink, SC-FDMA uplink), 5G = enhanced mobile broadband + low-latency + massive IoT (NR, mmWave, massive MIMO), 6G = research stage (terahertz, integrated sensing/communication, AI-native networks).

  • Multiple access: how users share spectrum. Define FDMA (frequency-division multiple access) — each user a separate frequency band; TDMA (time-division multiple access) — users share frequency in different time slots; CDMA (code-division multiple access) — users separated by pseudo-random orthogonal/non-orthogonal codes; OFDMA (orthogonal frequency-division multiple access) — dividing spectrum into orthogonal subcarriers assigned to users.

  • Spectral efficiency: bits/s/Hz — how many bits can be carried per second per Hz of spectrum. Higher spectral efficiency means more throughput for the same bandwidth.

  • Shannon–Hartley theorem (definition used in capacity reasoning): the maximum error-free data rate C of a band-limited AWGN (additive white Gaussian noise) channel is C = B · log2(1 + S/N) where B is bandwidth and S/N is signal-to-noise power ratio. Assumptions: ideal coding, Gaussian noise, single-input single-output (SISO) channel, no interference.

  • MIMO (multiple-input multiple-output): using multiple antennas at transmitter and receiver to create parallel spatial channels and increase capacity roughly by min(Nt, Nr) under favourable conditions. Define massive MIMO as tens to hundreds of antennas at base station.

  • Millimetre wave (mmWave): in 5G NR, FR2 spans 24.25–52.6 GHz; the broader engineering term mmWave is often 30–300 GHz. Higher frequency gives larger path loss but more available bandwidth.

  • Beamforming: forming a directional antenna pattern electronically to concentrate energy toward a user; reduces interference and increases effective S/N.

  • Network slicing: creating multiple virtual networks over same physical infrastructure, each tuned for different service requirements (eMBB, URLLC, mMTC). Define: eMBB = enhanced mobile broadband; URLLC = ultra-reliable low-latency communications; mMTC = massive machine-type communications.

  • Key performance indicators (KPIs): throughput (bits/s), latency (s), reliability (probability of success), device density (devices/km²), mobility (handover rates). Exams expect comparisons of these across generations; for 5G IMT-2020 targets include: peak rates ≈ 20 Gbps DL / 10 Gbps UL; user-experienced rates ≈ 100 Mbps DL / 50 Mbps UL; URLLC radio user-plane latency ≈ 1 ms with reliability ≈ 99.999% (1−10^-5); connection density ≈ 10^6 devices/km²; mobility up to ≈ 500 km/h; area traffic capacity ≈ 10 Mbps/m².

Diagrams to draw and label (exam answers often require these):

  • A timeline: draw a horizontal timeline labelled 1G, 2G, 2.5G (GPRS), 3G, 3.5G (HSPA), 4G (LTE), 4.5G (LTE-A), 5G NR, and 6G (research). Under each label, write the air interface (1G: analog FM; 2G: TDMA/FDMA for GSM and CDMA for IS-95; 3G: WCDMA/UMTS and CDMA2000; 4G: OFDMA downlink and SC-FDMA uplink; 5G: NR CP-OFDM, often TDD) and typical services (voice, SMS, mobile internet, IoT, URLLC).
  • Cell-reuse diagram for hexagonal cells: draw a hexagon cell, label R (cell radius), show cluster with reuse factor N and label D (reuse distance). Add base station at center.
  • Block diagram of a base station with multiple antennas showing beamforming vectors and users assigned subcarriers (OFDMA).

Derivation commonly asked: from Shannon to show effect of bandwidth and S/N on rate, and how MIMO multiplies capacity. Include assumptions and step-by-step algebra (below in Formulas and Worked examples).

Formulas

Each formula on its own line, with symbol meanings and units.

C = B · log2(1 + S/N)

  • C = channel capacity in bits per second (b/s); B = bandwidth in hertz (Hz); S/N = signal power divided by noise power (unitless). Applies for AWGN channel with ideal coding.

S/N = P_r / (N_0 · B)

  • P_r = received signal power in watts (W); N_0 = noise power spectral density in W/Hz; B = bandwidth in Hz. Used to compute S/N when thermal noise dominates.

P_r = P_t · G_t · G_r · (λ / (4πd))^2 (Friis transmission equation, free-space)

  • P_r = received power (W); P_t = transmit power (W); G_t, G_r = transmit and receive antenna gains (unitless); λ = wavelength (m); d = distance (m). Applies in free-space, line-of-sight, far-field.

D = R · sqrt(3N)

  • D = reuse distance between cells using same frequency (m); R = cell radius (m); N = frequency reuse cluster size (integer). Used for hexagonal reuse geometry.

C_MIMO ≈ M · B · log2(1 + S/N)

  • C_MIMO = approximate capacity with M parallel spatial streams (b/s); M = min(Nt, Nr) number of spatial streams (unitless); other symbols as above. Applies under ideal rich-scattering conditions and independent channels.

Worked examples

Example 1 — Shannon capacity comparison (university numerical standard) Problem: Compare the maximum theoretical data rate (capacity) for a single 20 MHz channel at S/N = 10 dB for (a) a SISO link and (b) a 4×4 MIMO link assuming 4 independent spatial streams. Use Shannon formula. Steps:

  1. Write formula: C = B · log2(1 + S/N).
  2. Convert S/N from dB to linear: 10 dB → 10.
  3. For SISO: B = 20 MHz = 20×10^6 Hz.
  4. Compute C_SISO = 20×10^6 · log2(1 + 10) = 20×10^6 · log2(11).
  5. Compute log2(11) ≈ ln(11)/ln(2) ≈ 2.3979/0.6931 ≈ 3.459.
  6. C_SISO ≈ 20×10^6 · 3.459 = 69.18×10^6 b/s = 69.18 Mbps.
  7. For 4×4 MIMO with 4 spatial streams: C_MIMO ≈ 4 · C_SISO = 4 · 69.18 Mbps = 276.7 Mbps. Answer: SISO ≈ 69.2 Mbps, 4×4 MIMO ≈ 276.7 Mbps. Notes: This is ideal maximum capacity; real systems have overheads and imperfect channel conditions so practical rates are lower.

Example 2 — Frequency reuse distance (hexagonal geometry) Problem: For a hexagonal cellular layout with cell radius R = 1 km and reuse cluster size N = 7 (common in older designs), compute the reuse distance D. Steps:

  1. Formula: D = R · sqrt(3N).
  2. Plug values: R = 1 km = 1000 m; N = 7.
  3. sqrt(3N) = sqrt(21) ≈ 4.5826.
  4. D = 1000 m · 4.5826 = 4582.6 m = 4.583 km. Answer: Reuse distance D ≈ 4.584 km. Diagram to draw: hexagonal cells showing same-colour cells separated by distance D; label R and D.

Example 3 — (Theory model answer) Explain why 5G uses mmWave and massive MIMO together. Model answer (exam style): mmWave provides large contiguous bandwidths (tens to hundreds of MHz to several GHz) needed for high peak rates. However, free-space path loss grows with frequency (Friis: for fixed distance and antenna gains, path loss ∝ f^2, equivalently P_r ∝ 1/f^2), so mmWave suffers higher path loss and blockages. Massive MIMO and beamforming compensate: antenna arrays create narrow beams, increasing directional gain (G_t, G_r) and restoring S/N. Therefore mmWave (for bandwidth) + massive MIMO (for gain and spatial multiplexing) are complementary. State assumptions: line-of-sight or beam-tracked links, dense small-cell deployments.

Common mistakes

  • Applying Shannon without checking S/N or bandwidth units (mixing Hz and MHz leads to errors).
  • Assuming MIMO always multiplies capacity by Nt or Nr — this requires independent spatial channels (rich-scattering or resolved angles) and sufficient S/N.
  • Quoting peak theoretical 5G user rates as typical; exams expect distinction between theoretical peak and practical observed rates.
  • Using Friis equation in non-line-of-sight or near-field conditions — wrong assumptions.

For your exam

Kinds of questions and what full-marks answers include:

  • Definitions: define generation, OFDMA, MIMO, mmWave, beamforming. Include short examples (e.g., LTE uses OFDMA downlink and SC-FDMA uplink).
  • Short notes: compare 2G, 3G, 4G, 5G in terms of air interface, services, and KPIs. Use a small table in answer if allowed.
  • Derivations: derive Shannon capacity formula usage and show effect of doubling bandwidth vs doubling S/N on capacity. State assumptions clearly.
  • Numericals: compute capacity using Shannon, reuse distances, path loss using Friis. Show unit conversions and steps.
  • Diagrams: draw hexagonal cell reuse and annotate R and D; draw base station antenna array illustrating beamforming.
  • Essay/theory: explain why 5G adopted OFDMA, massive MIMO, and mmWave; discuss challenges for 6G. A full answer states trade-offs, gives examples and mentions research directions (terahertz, integrated sensing and communication, native AI).

Quick check

  1. What multiple access method does LTE use on the downlink?
  2. State Shannon–Hartley theorem in words (one line).
  3. For a 10 MHz channel with S/N = 0 dB, approximate capacity using Shannon (give numeric value in Mbps). Answers: OFDMA; maximum rate = bandwidth × log2(1 + S/N) (AWGN, ideal coding); for Q3 S/N = 0 dB = 1, C = 10×10^6 · log2(2) = 10×10^6 · 1 = 10 Mbps.

Finished this topic? Mark it so your progress, study plan and readiness keep up.

Stuck on something here?