Shannon's Channel Capacity
Calculate the maximum data rate of a noisy communication channel using Shannon's theorem.
The Formula
Shannon's Channel Capacity theorem, published by Claude Shannon in 1948, establishes the absolute maximum rate at which information can be reliably transmitted over a communication channel in the presence of noise. This result is considered one of the most important theorems in all of engineering and information science. It set the foundation for modern digital communications, from Wi-Fi and cellular networks to deep-space probes.
The theorem states that for any channel with a given bandwidth and signal-to-noise ratio, there exists a theoretical upper limit on data throughput. Below this limit, it is possible (in principle) to transmit data with an arbitrarily low error rate using sufficiently clever encoding. Above this limit, reliable communication is mathematically impossible regardless of the encoding scheme used.
What makes this result remarkable is that Shannon proved the limit exists without specifying how to achieve it. It took decades for practical coding schemes (like turbo codes in 1993 and LDPC codes) to approach the Shannon limit. Modern 5G networks and satellite communications operate within a fraction of a decibel of this theoretical bound.
The signal-to-noise ratio (S/N or SNR) is typically expressed in linear terms for this formula, not in decibels. If you have SNR in decibels, you must convert first: S/N (linear) = 10SNR(dB)/10. Bandwidth B is measured in hertz, and the resulting capacity C is in bits per second.
Shannon's theorem explains why increasing bandwidth or reducing noise both improve data rates. It also reveals the diminishing returns of boosting signal power: doubling S/N does not double capacity because of the logarithmic relationship. This insight guides engineers in making cost-effective design tradeoffs between antenna size, transmit power, bandwidth allocation, and error-correction complexity.
Variables
| Symbol | Meaning |
|---|---|
| C | Channel capacity — maximum data rate (bits per second) |
| B | Bandwidth of the channel (hertz, Hz) |
| S | Average signal power (watts) |
| N | Average noise power (watts) |
| S/N | Signal-to-noise ratio (linear, not dB) |
Example 1
A telephone line has a bandwidth of 3,000 Hz and an SNR of 30 dB. What is the channel capacity?
Convert SNR from dB to linear: S/N = 1030/10 = 1000
C = 3000 × log2(1 + 1000)
C = 3000 × log2(1001) ≈ 3000 × 9.97
C ≈ 29,901 bits per second (about 30 kbps)
Example 2
A Wi-Fi channel has 20 MHz bandwidth and an SNR of 20 dB. What is the theoretical maximum throughput?
Convert SNR: S/N = 1020/10 = 100
C = 20,000,000 × log2(1 + 100)
C = 20,000,000 × log2(101) ≈ 20,000,000 × 6.66
C ≈ 133,200,000 bps ≈ 133 Mbps theoretical maximum
When to Use It
Shannon's theorem is used whenever you need to understand or design communication systems.
- Designing wireless networks (Wi-Fi, 4G/5G, satellite links)
- Evaluating whether a channel can support a required data rate
- Choosing between increasing bandwidth vs. improving signal quality
- Setting upper bounds for data compression and error-correction codes
- Comparing real-world system performance against the theoretical limit