How fast can LTE actually send data? The answer comes down to counting and multiplying.
Core idea: Count how many symbols you send per second, then figure out how many bits each symbol carries.
The Building Blocks
These are fixed constants from the LTE specification:
| Component | Value | Why |
|---|---|---|
| Subcarriers per RB | 12 | Defined by 3GPP |
| Slots per subframe | 2 | Subframe = 1ms, Slot = 0.5ms |
| Subframes per second | 1000 | Subframe = 1ms |
| Symbols per slot | 7 (Normal CP) | or 6 with Extended CP |
Note: Symbols per slot depends on the Cyclic Prefix. Normal CP = 7 symbols, Extended CP = 6 symbols.
And bandwidth determines how many Resource Blocks you have:
| Bandwidth | Resource Blocks |
|---|---|
| 5 MHz | 25 RBs |
| 10 MHz | 50 RBs |
| 15 MHz | 75 RBs |
| 20 MHz | 100 RBs |
Step 1: Count Symbols Per Second
Every RB gives you 12 subcarriers × 7 symbols per slot. You get 2 slots per subframe, and 1000 subframes per second.
Symbols/sec = RBs × 12 × 7 × 2 × 1000
Example for 20 MHz:
That’s 16.8 million symbols flying through the air every second.
Step 2: Bits Per Symbol (Modulation)
Each symbol carries a certain number of bits, depending on the modulation scheme:
| Modulation | Bits/Symbol | When to Use |
|---|---|---|
| QPSK | 2 | Poor signal, cell edge |
| 16-QAM | 4 | Medium signal |
| 32-QAM | 5 | Good signal |
| 64-QAM | 6 | Excellent signal, close to tower |
Higher QAM packs more bits per symbol, but needs a cleaner signal to decode correctly.
Step 3: Raw Bit Rate
Multiply symbols by bits per symbol:
Raw bits/sec = Symbols/sec × bits/symbol
Example: 20 MHz with 16-QAM:
But this is the theoretical maximum. Real systems have overhead.
Step 4: Apply Real-World Factors
Two things eat into your raw rate:
Coding Rate
Error correction adds redundancy. A 2/3 coding rate means for every 3 bits transmitted, only 2 are actual data.
Bandwidth Efficiency
Not all spectrum carries user data. Guard bands, control channels, and reference signals take up space. Typical efficiency is 90%.
Step 5: MIMO Multiplication
MIMO (Multiple-Input Multiple-Output) sends multiple data streams simultaneously using multiple antennas.
| MIMO Config | Spatial Layers | Multiplier |
|---|---|---|
| 1×1 (SISO) | 1 | ×1 |
| 2×2 | 2 | ×2 |
| 4×4 | 4 | ×4 |
| 8×8 | 8 | ×8 |
Final throughput = Effective rate × MIMO layers
This is why MIMO is so powerful. It literally multiplies your capacity.
The Complete Formula
Putting it all together:
Worked Example 1
Calculate the maximum achievable throughput of the FDD LTE system in downlink under the following assumptions:
- Normal Cyclic Prefix (CP), 10 MHz Bandwidth, 16-QAM Modulation, 4×4 MIMO
Hint: In Normal CP, there are 7 symbols and 12 subcarriers in a slot of 0.5 ms, 10 MHz contains 50 Resource Blocks, one modulation symbol carries 4 bits in 16-QAM.
Step 1: Resource Blocks for 10 MHz = 50 RBs
Step 2: Symbols per second:
Step 3: Bits per symbol for 16-QAM = 4 bits
Step 4: Raw bit rate:
Step 5: With 4×4 MIMO:
No coding rate or efficiency given, so we use raw throughput.
Worked Example 2
For an LTE release 8 system, consider an FDD system configured to use 20 MHz of spectrum. Consider a Type 1 frame with normal Cyclic Prefix length of seven (7) symbols, and a transmission effective bandwidth of 90%.
(i) What is the effective downlink data rate using 16-QAM and 2/3 encoding rate? (10 marks)
(ii) What is the impact on the system capacity when a 2×2 MIMO antenna configuration is used? (2 marks)
Part (i):
Step 1: Resource Blocks for 20 MHz = 100 RBs
Step 2: Symbols per second:
Step 3: Bits per symbol for 16-QAM = 4 bits
Step 4: Raw bit rate:
Step 5: Apply coding rate:
Step 6: Apply bandwidth efficiency:
Part (ii):
With 2×2 MIMO, we get 2 spatial layers, so capacity doubles:
Worked Example 3
For an LTE release 8 system, consider an FDD system configured to use 20 MHz of spectrum. Consider a Type 1 frame with normal Cyclic Prefix length of seven (7) symbols, and a transmission effective bandwidth of 90%.
(i) What is the effective downlink data rate using 32-QAM and 2/3 encoding rate? (10 marks)
(ii) What is the impact on the system capacity when an 8×8 MIMO antenna configuration is used? (2 marks)
Part (i):
Step 1: Resource Blocks for 20 MHz = 100 RBs
Step 2: Symbols per second:
Step 3: Bits per symbol for 32-QAM = 5 bits
Step 4: Raw bit rate:
Step 5: Apply coding rate:
Step 6: Apply bandwidth efficiency:
Part (ii):
With 8×8 MIMO, we get 8 spatial layers:
Quick Reference
| Parameter | Values to Memorize |
|---|---|
| 10 MHz | 50 RBs |
| 20 MHz | 100 RBs |
| QPSK | 2 bits/symbol |
| 16-QAM | 4 bits/symbol |
| 32-QAM | 5 bits/symbol |
| 64-QAM | 6 bits/symbol |
| Symbols formula | RBs × 12 × 7 × 2 × 1000 |
The calculation is just multiplication. Count symbols, convert to bits, apply penalties, multiply by MIMO. That’s it.