4G Turbo-V Intel® FPGA IP User Guide

ID 683882
Date 4/01/2024
Public
Document Table of Contents

4.1.5. Throughput Calculations

The downlink accelerator has buffers between two neighboring subcomponents, so that all sub-components can process in parallel at the same time.

The overall throughput can be calculated as

Throughput= K/max(L(crc), L(encoder), L(subblock-interleaver), L(pruning))*freq. (bits per second).

Example 1: K = 6144, Kπ = 6176, E = 18444, freq. = 300 MHz

L(crc) = 776 cycles

L(encoder) = 782 cycles

L(subblock-interleaver) = 814 cycles

L(pruning) = 786 cycles

L(output) = 815 cycles

Throughput(downlink) = 6144/815*300M = 2262 Mbps

Example 2: K = 40, Kπ= 64, E = 132, freq. = 300 MHz

L(crc) = 13 cycles

L(encoder) = 19 cycles

L(subblock-interleaver) = 50 cycles

L(pruning) = 22 cycles

L(output) = 52 cycles

Throughput(downlink) = 40/52*300M = 231 Mbps