📡 GSM Cell Planning Assessment

Undergraduate Communication Engineering
Total Questions: 12 | Descriptive: 4 | Analytical: 4 | Quantitative: 4
Time Allowed: 90 Minutes | Total Marks: 100
Instructions: Answer all questions. For descriptive questions, provide detailed explanations with examples where applicable. For analytical questions, show your reasoning process. For quantitative questions, show all calculations step-by-step.
📝 Section A: Descriptive Questions (4 × 10 = 40 Marks)
1 Descriptive [10 Marks]
Explain the concept of frequency reuse in GSM cellular systems. Describe how the cluster size (N) is determined using the hexagonal cell geometry and the parameters i and j. Why is the cluster size critical for system capacity and interference management?
2 Descriptive [10 Marks]
Describe the co-channel interference mechanism in GSM networks. Explain the difference between interference experienced by a mobile station at the cell center versus the cell boundary. How does sectoring (120° or 60°) help reduce co-channel interference?
3 Descriptive [10 Marks]
Explain the cell splitting technique for capacity enhancement in GSM networks. Compare and contrast cell splitting with cell sectoring and the microcell zone concept. Discuss the trade-offs involved in each approach regarding infrastructure cost and interference.
4 Descriptive [10 Marks]
Describe the BCCH (Broadcast Control Channel) planning considerations in GSM. Why is the BCCH typically planned with a larger cluster size than TCH (Traffic Channel) frequencies? Explain the concept of frequency reuse patterns for BCCH and TCH layers.
🔍 Section B: Analytical Questions (4 × 10 = 40 Marks)
5 Analytical [10 Marks]
A GSM operator has a 5 MHz spectrum allocation. The operator plans to use different cluster sizes for BCCH and TCH layers. Given that GSM requires a minimum C/I of 9 dB for TCH and typically 15 dB for BCCH (with 6 dB margin), analyze:

a) What cluster sizes should be selected for BCCH and TCH layers? Justify your answer using the relationship between C/I and cluster size.
b) How does this dual-layer planning affect overall system capacity compared to using a single cluster size for all channels?
6 Analytical [10 Marks]
Analyze the impact of path loss exponent (γ) on frequency reuse planning. Compare the required cluster sizes for GSM (requiring C/I = 9 dB) in:

a) Urban environment (γ = 3)
b) Suburban environment (γ = 4)

Explain why the cluster size differs and discuss the practical implications for network deployment in different environments.
7 Analytical [10 Marks]
Consider a GSM network using omnidirectional antennas versus one using 120° sectoring. Analyze how sectoring affects:

a) The number of interfering cells in the first tier
b) The signal-to-interference ratio (S/I) improvement
c) The trunking efficiency and channel availability per sector

Discuss whether sectoring always results in better system performance.
8 Analytical [10 Marks]
A city has varying traffic density: high-density urban center (100 Erlangs/km²), medium-density suburbs (30 Erlangs/km²), and low-density rural areas (5 Erlangs/km²). Analyze and propose a heterogeneous cell planning strategy using:

a) Macro cells, micro cells, and picocells
b) Appropriate frequency reuse patterns for each layer
c) Handover considerations between different cell layers
🧮 Section C: Quantitative Questions (4 × 5 = 20 Marks)
9 Quantitative [5 Marks]
Calculate the frequency reuse distance (D) for a GSM system with:

• Cluster size N = 7
• Cell radius R = 2 km

Also determine the co-channel reuse ratio Q. Show all steps.
10 Quantitative [5 Marks]
A GSM operator has 395 total voice channels. Calculate:

a) The number of channels per cell for cluster sizes N = 4, 7, and 12
b) The theoretical S/I ratio for each cluster size assuming path loss exponent γ = 4 and omnidirectional antennas

Which cluster size would you recommend for GSM and why?
11 Quantitative [5 Marks]
Given a GSM system with the following parameters:

• Total bandwidth: 5 MHz (25 ARFCNs)
• BCCH cluster size: 7
• TCH cluster size: 3
• 1 BCCH TRX per cell (8 timeslots, 1 for BCCH, 7 for SDCCH/Traffic)

Calculate the total number of full-rate speech channels (TCH/F) available per cell. Show your channel allocation strategy.
12 Quantitative [5 Marks]
In a GSM network, the cell radius is reduced from 2 km to 1 km through cell splitting. Calculate:

a) The increase in the number of cells required to cover the same geographical area
b) The theoretical increase in system capacity (assuming same number of channels per cell)
c) The new co-channel reuse distance if the original D was 9.8 km
📚 Post-Test Answers & Explanations ▼
Question 1 Answer (Frequency Reuse & Cluster Size):
Frequency reuse is the fundamental concept of cellular systems where the same frequency channels are reused in spatially separated cells to maximize spectrum efficiency. The available spectrum is divided into channel groups, and each cell is assigned one group. The pattern of cells that collectively use all available frequencies once is called a cluster.

Cluster Size Determination: For hexagonal cell geometry, valid cluster sizes follow the formula:
N = i² + ij + j²
where i and j are non-negative integers that determine the co-channel cell location. Common values include:
N = 1
i=0, j=1
N = 3
i=1, j=1
N = 4
i=2, j=0
N = 7
i=2, j=1
N = 9
i=3, j=0
N = 12
i=2, j=2
The parameters i and j determine the relative location of co-channel cells: move i cells along any chain of hexagons, turn 60° counter-clockwise, then move j cells.

Importance: Cluster size is critical because it represents the trade-off between capacity and quality. Smaller N increases capacity (more channels per area) but reduces reuse distance, increasing co-channel interference. GSM typically uses N = 3, 4, or 7 depending on interference conditions.
Question 2 Answer (Co-channel Interference):
Co-channel interference (CCI) occurs when a mobile station receives signals on the same frequency from multiple base stations simultaneously. This is the primary limiting factor in cellular system capacity.

Cell Center vs. Boundary:
  • Cell Center: Mobile is closest to serving BS and equidistant (distance D) from all 6 interfering co-channel cells. S/I ratio is best here: S/I = (√3N)^γ / 6
  • Cell Boundary: Mobile is at distance R from serving BS but at varying distances from interferers (D-R from two nearest, D from two, D+R from two). Worst-case S/I is significantly lower.
Sectoring: Using directional antennas (120° or 60° sectors) reduces the number of interfering cells in the first tier:
  • 120° sectoring (3 sectors): Reduces interferers from 6 to 2
  • 60° sectoring (6 sectors): Reduces interferers from 6 to 1
This improves S/I by approximately 3-5 dB, allowing smaller cluster sizes. However, sectoring reduces trunking efficiency due to channel partitioning.
Question 3 Answer (Cell Splitting & Capacity Techniques):
Cell Splitting: Dividing large cells into smaller cells to increase capacity. When traffic density increases, cells are split (typically radius halved), increasing the number of cells per area by 4x. New base stations are placed at corners or centers with reduced transmit power to maintain S/I ratios. This is a rescaling technique that preserves the frequency reuse plan.

Comparison:
Technique Mechanism Infrastructure Cost Interference Impact
Cell Splitting Reduce cell radius, add new BS High (new towers) Maintains S/I ratio
Cell Sectoring Divide cell into sectors with directional antennas Medium (antenna upgrade) Reduces CCI but increases handovers
Microcell Zone Multiple zones connected to one BS Low (remote antennas) Better than sectoring, flexible channel assignment
Trade-offs: Cell splitting provides the highest capacity increase but requires significant infrastructure investment. Sectoring is cheaper but reduces trunking efficiency. Microcell zones offer the best balance for in-building coverage.
Question 4 Answer (BCCH Planning):
BCCH (Broadcast Control Channel) is the most critical control channel in GSM, carrying frequency correction (FCCH), synchronization (SCH), and system information. Every mobile must decode BCCH to camp on a cell.

Why Larger Cluster Size for BCCH:
  • BCCH must be reliably decoded at the cell edge for proper cell selection and reselection
  • Poor BCCH coverage leads to dropped calls, failed handovers, and network unavailability
  • GSM specifications require C/I > 9 dB for TCH but operators typically plan BCCH with 6 dB additional margin (15 dB total)
  • Higher C/I requirement necessitates larger reuse distance → larger cluster size
Typical Planning:
TCH Layer
Cluster size: 3 or 4
More aggressive reuse
BCCH Layer
Cluster size: 7 or 9
Conservative reuse
This multiple reuse rate approach optimizes spectrum efficiency while ensuring control channel reliability. BCCH frequencies are carefully chosen to avoid interference with neighboring cells' BCCH, while TCH frequencies can be more tightly packed.
Question 5 Answer (Dual-Layer Planning Analysis):
a) Cluster Size Selection:
Using the relationship between C/I and cluster size for interference-limited systems:
C/I ≈ (√3N)^γ / C(α)
Where C(α) ≈ 6 for omnidirectional cells (number of interferers).

For γ = 4:
For TCH (C/I = 9 dB = 7.943 linear):
7.943 = (√3N)⁴ / 6
(√3N)⁴ = 47.66
3N = 47.66^(1/4) ≈ 2.62
N ≈ 2.29 → Select N = 3 (nearest valid cluster size)

For BCCH (C/I = 15 dB = 31.623 linear):
31.623 = (√3N)⁴ / 6
(√3N)⁴ = 189.74
3N = 189.74^(1/4) ≈ 3.71
N ≈ 4.59 → Select N = 7 (nearest valid cluster size)
b) Capacity Impact:
With 5 MHz (25 ARFCNs):
  • Single layer (N=7): 25/7 ≈ 3.6 ARFCNs/cell → Low capacity but uniform quality
  • Dual layer: 7 ARFCNs for BCCH (N=7) + 18 ARFCNs for TCH (N=3) = 6 TCH TRX/cell
Dual-layer planning increases TCH capacity by approximately 75% while maintaining BCCH reliability. This is the standard approach in modern GSM networks.
Question 6 Answer (Path Loss Exponent Impact):
The path loss exponent γ determines how quickly signal strength decreases with distance, directly affecting interference levels.

Cluster Size Calculation: Using N = (1/3)[6×(C/I)]^(2/γ) for GSM (C/I = 9 dB = 7.943):

a) Urban (γ = 3):
N = (1/3)[6 × 7.943]^(2/3) = (1/3)[47.66]^0.667 = (1/3) × 13.1 ≈ 4.37 → N = 7

b) Suburban (γ = 4):
N = (1/3)[6 × 7.943]^(2/4) = (1/3)[47.66]^0.5 = (1/3) × 6.9 ≈ 2.30 → N = 3 or 4
Implications:
  • Urban environments (γ=3): Signals decay slower, so interference travels further. Requires larger cluster size (N=7) to maintain acceptable C/I.
  • Suburban/Rural (γ=4): Faster signal decay means less interference. Can use tighter reuse (N=3 or 4) for higher capacity.
Practical Deployment: Urban areas with high-rise buildings often use N=7 or 9, while suburban areas may use N=4. Some operators use N=3 in rural areas with sectoring to maximize coverage with limited spectrum.
Question 7 Answer (Sectoring Analysis):
a) Interfering Cells Reduction:
Configuration First Tier Interferers Active Interferers
Omnidirectional 6 cells 6 (all contribute)
120° Sectoring (3 sectors) 6 cells 2 (only those in antenna pattern)
60° Sectoring (6 sectors) 6 cells 1 (only one in narrow beam)
b) S/I Improvement:
S/I improvement comes from reducing the number of interferers in the denominator:
Omnidirectional: S/I = (√3N)^γ / 6
120° Sectoring: S/I = (√3N)^γ / 2 → Improvement: 10×log(6/2) = 4.77 dB
60° Sectoring: S/I = (√3N)^γ / 1 → Improvement: 10×log(6/1) = 7.78 dB
c) Trunking Efficiency:
Sectoring partitions the cell's channel pool. With 42 channels per cell:
  • Omnidirectional: 42 channels in one pool → High trunking efficiency, low blocking
  • 3 sectors: 14 channels per sector → Lower efficiency per sector, more handovers between sectors
  • 6 sectors: 7 channels per sector → Poor trunking efficiency, frequent sector handovers
Conclusion: Sectoring doesn't always improve performance. While it improves S/I allowing smaller N, the loss of trunking efficiency and increased handover load may offset gains. 120° sectoring is the most common compromise.
Question 8 Answer (Heterogeneous Planning):
Strategy for Multi-Density Environment:

a) Cell Hierarchy:
Urban Center
Picocells (200-500m)
Microcells (500m-1km)
Suburbs
Microcells (1-2km)
Macro cells (2-5km)
Rural
Macro cells (5-15km)
Repeaters for coverage
b) Frequency Reuse Patterns:
  • Picocells (indoor/urban): N=1 or 3 with aggressive power control, indoor penetration loss provides isolation
  • Microcells: N=3 or 4, line-of-sight street canyon propagation allows tight reuse
  • Macro cells: N=7 or 9 for suburban, N=12 for rural overlaid coverage
c) Handover Considerations:
  • Layered handover: Prioritize lower layers (pico → micro → macro) to offload traffic
  • Umbrella cells: Macro cells cover micro/pico areas for fast-moving users to prevent excessive handovers
  • Cell reselection: C1/C2 criteria adjusted so high-capacity layers are preferred when signal is adequate
  • Inter-layer interference: Careful frequency planning to avoid micro-to-macro interference (different bands or guard bands)
This cellular hierarchy maximizes spectrum efficiency while ensuring seamless mobility across environments.
Question 9 Answer (Reuse Distance Calculation):
Given: N = 7, R = 2 km

Step 1: Calculate co-channel reuse ratio Q:
Q = D/R = √(3N)
Step 2: Calculate Q value:
Q = √(3 × 7) = √21 ≈ 4.583
Step 3: Calculate reuse distance D:
D = Q × R = 4.583 × 2 km = 9.166 km
Verification: The frequency reuse distance is approximately 9.17 km. This means co-channel cells using the same frequencies must be separated by at least this distance to maintain acceptable interference levels.
Question 10 Answer (Channels per Cell & S/I):
Given: Total channels = 395, γ = 4

a) Channels per cell:
N = 4: 395/4 = 98.75 ≈ 99 channels/cell
N = 7: 395/7 = 56.43 ≈ 56 channels/cell
N = 12: 395/12 = 32.92 ≈ 33 channels/cell
b) S/I Calculation: Using S/I = (√3N)^γ / 6
For N = 4:
S/I = (√12)^4 / 6 = (3.464)^4 / 6 = 144 / 6 = 24
S/I (dB) = 10log(24) = 13.8 dB

For N = 7:
S/I = (√21)^4 / 6 = (4.583)^4 / 6 = 441 / 6 = 73.5
S/I (dB) = 10log(73.5) = 18.7 dB

For N = 12:
S/I = (√36)^4 / 6 = 6^4 / 6 = 1296 / 6 = 216
S/I (dB) = 10log(216) = 23.3 dB
Recommendation: N = 7 is recommended for GSM. N=4 provides only 13.8 dB S/I, which is marginal for GSM's 9 dB requirement (especially at cell edges). N=12 provides excellent quality (23.3 dB) but wastes capacity. N=7 offers 18.7 dB S/I, providing adequate margin (9.7 dB above minimum) while maintaining reasonable capacity (56 channels/cell).
Question 11 Answer (Channel Allocation):
Given: 25 ARFCNs (5 MHz), BCCH N=7, TCH N=3

Step 1: BCCH Layer Allocation:
BCCH ARFCNs needed = 25 / 7 = 3.57 → 4 ARFCNs for BCCH layer
(Conservative planning: 1 BCCH TRX per cell using 7 frequencies across cluster)
Step 2: TCH Layer Allocation:
Remaining ARFCNs = 25 - 7 = 18 ARFCNs for TCH
With N=3 for TCH: 18 / 3 = 6 TCH TRX per cell
Step 3: Channel Calculation per Cell:
BCCH TRX:
• 1 timeslot for BCCH (control)
• 1 timeslot for CCCH (common control)
• 6 timeslots for SDCCH/8 or TCH
→ 6 TCH/F channels (if SDCCH combined) or 7 TCH/F (if SDCCH on different TS)

TCH TRX (6 TRX × 8 TS):
• 6 × 8 = 48 timeslots
• Assuming 1 TS per TRX for SACCH (signaling): 48 - 6 = 42
→ 42 TCH/F channels

Total TCH/F per cell: 7 (from BCCH) + 42 = 49 full-rate speech channels
Note: If half-rate (TCH/H) is used, capacity doubles to 98 channels.
Question 12 Answer (Cell Splitting Quantitative):
Given: Original R = 2 km, New R = 1 km, Original D = 9.8 km

a) Number of Cells Increase:
Cell area is proportional to R² (hexagonal area = 2.598 × R²)
Area ratio = (R_original / R_new)² = (2/1)² = 4
4 times more cells required to cover same area
b) Capacity Increase:
Assuming same channels per cell:
Original capacity: C (for area A)
New capacity: 4 cells × C = 4C (400% increase or 4× capacity)

If original had 56 channels/cell (N=7):
New system: 4 cells × 56 = 224 channels in same area vs. original 56
→ Theoretical capacity increase: 300%
c) New Reuse Distance:
Reuse distance scales with cell radius if cluster size is maintained:
Original: D = 9.8 km, R = 2 km, Q = 4.9
New R = 1 km, maintaining Q = 4.9:
D_new = Q × R_new = 4.9 × 1 = 4.9 km

Or using ratio: D_new = D_original × (R_new/R_original) = 9.8 × 0.5 = 4.9 km
Key Insight: Cell splitting maintains the co-channel reuse ratio Q, preserving S/I quality while increasing capacity through spatial reuse. The reuse distance halves, allowing frequencies to be reused twice as often in the same geographical area.