Example data table
| Scenario | Known inputs | Assumptions | Estimated net (per direction) |
|---|---|---|---|
| Short-range RF link | 20 MHz, 2.5 bits/s/Hz | Overhead 10%, coding 0.85, 1 stream | ~38.25 Mbps |
| Backhaul microwave | 56 MHz, 4.2 bits/s/Hz | Overhead 7%, coding 0.90, 2 streams | ~393.62 Mbps |
| Serial bus | 1.5 GBd, 1 bit/symbol | Overhead 20%, coding 1.00, 1 lane | ~1.20 Gbps |
| File transfer check | 500 MiB in 30 s | Overhead 8%, retrans 1.05 | ~119.70 Mbps |
| Ethernet payload | 1 GB in 12 s | Overhead 6%, retrans 1.00 | ~626.67 Mbps |
Formula used
- R_base = SymbolRate × BitsPerSymbol
- R_base = Bandwidth × SpectralEfficiency
- R_base = PayloadBits ÷ TimeSeconds
R_net = (R_gross × CodingRate × (1 − Overhead)) ÷ RetransFactor
How to use this calculator
- Select the method that matches your available measurements.
- Enter the primary values (symbol rate, bandwidth, or payload and time).
- Set streams, duplex, and efficiency assumptions (coding rate, overhead, retransmissions).
- Click Calculate data rate. Results display above the form.
- Use CSV or PDF to export recent results for documentation.
Why net throughput matters
Channel capacity planning starts with a realistic target throughput, not the advertised physical layer rate. In this calculator, gross rate is computed first and then adjusted for coding, overhead, and retransmissions to estimate application‑usable throughput. Engineers can compare designs quickly by keeping assumptions consistent across scenarios.
Symbol-rate method examples
Symbol-rate driven links are common in high-speed serial, microwave, and modem chains. If a lane runs at 1.5 GBd with 1 bit per symbol, gross is 1.5 Gbps. With 20% framing overhead, the net becomes 1.2 Gbps before other losses. Moving to 4 bits per symbol raises gross fourfold, but SNR and linearity constraints may reduce achievable coding rate.
Bandwidth method examples
Bandwidth-based estimates suit radio channels and shared spectrum systems. A 20 MHz channel at 2.5 bits/s/Hz yields 50 Mbps gross. With coding rate 0.85 and 10% overhead, net is 38.25 Mbps, matching typical small-cell backhaul budgets. Pushing spectral efficiency from 2.5 to 4.0 bits/s/Hz increases gross by 60%, but may require tighter EVM and higher transmit power.
Payload timing validation
Payload-and-time measurements validate end-to-end performance. Transferring 500 MiB in 30 seconds equals about 139.81 Mbps gross. After 8% overhead and a 1.05 retransmission factor, net is roughly 119.70 Mbps, which is useful for verifying transport tuning. Use binary units (MiB) when comparing to file systems, and decimal units (MB) when comparing to link specifications.
Interpreting streams, duplex, and efficiency
Streams and duplex settings model parallelism and directionality. Two spatial streams or lanes double per-direction throughput when streams are independent. Full duplex doubles aggregate capacity because both directions carry traffic simultaneously, while half duplex shares airtime and should be treated as one direction at a time. For planning, keep per-direction net as the conservative baseline and treat aggregate as a peak figure.
Efficiency inputs represent controllable design tradeoffs. Overhead captures pilots, guards, preambles, headers, and idle patterns; even 6% overhead matters on multi‑gigabit links. Coding rate reflects redundancy and can be adjusted with FEC profiles. Retransmission factor represents loss and ARQ behavior; values above 1.10 often signal interference, buffer drops, or suboptimal MTU sizing. When documenting results, export CSV for batch comparisons and PDF for approvals. Record modulation, channel width, and environment notes so future tests can reproduce the same assumptions accurately with confidence.
FAQs
What is the difference between gross and net data rate?
Gross is the raw computed throughput before losses. Net applies coding rate, overhead reduction, and retransmission effects to estimate payload‑usable throughput for applications and planning.
How do I choose bits per symbol for modulation?
Use the modulation order: BPSK=1, QPSK=2, 16‑QAM=4, 64‑QAM=6, 256‑QAM=8. Higher values increase gross rate but require higher SNR and better linearity.
What should I enter for spectral efficiency?
Spectral efficiency is bits/s/Hz after practical limits. Typical ranges: 1–3 for robust links, 3–6 for good RF conditions, and 6+ for very clean channels with strong coding and hardware.
Does full duplex always double my throughput?
It doubles aggregate capacity only when both directions can transmit simultaneously without sharing the same time resource. Many systems are effectively half duplex due to spectrum sharing or self‑interference limits.
How do overhead and retransmissions affect results?
Overhead removes capacity used by headers, pilots, guards, and gaps. Retransmission factor inflates required throughput to account for retries. Together, they often explain why measured throughput is below link rate.
Which method should I use for real-world testing?
Use payload and time when you have measured transfers. Use bandwidth with spectral efficiency for RF planning. Use symbol rate when you know the clocking and modulation on a serial or modem interface.