Air Time calculation in (Micro)Python

[rcolistete]

As it is a common question how to calculate the air time of LoRa packets, so I post a Python 3/MicroPython script to calculate the data rate and airtime. Links about the calculations :
LoRa data rate calculator
LoRa air time calculator
Spreadsheet for LoRa airtime calculation

LoRa.py (needed for Python3, not needed by Pycom MicroPython as it already has in 'network.LoRa' submodule)

BW_125KHZ = 0
BW_250KHZ = 1
BW_500KHZ = 2

CODING_4_5 = 1
CODING_4_6 = 2
CODING_4_7 = 3
CODING_4_8 = 4

LoRaAirTimeCalc.py :

# LoRaAirTimeCalc.py v0.1 (26/04/2018) for normal Python 3 (for PC) or Pycom MicroPython
# By Roberto Colistete Jr (roberto.colistete at gmail.com)

# from network import WLAN, LoRa   # For Pycom MicroPython
import LoRa   # For Python3
import struct
from math import ceil

# Using Pycom constants for bw and cr, calculates the DR (datarate) in bps
def dataratetheoretical(sf, bw, cr):
   if bw == LoRa.BW_125KHZ:
       bwv = 125
   if bw == LoRa.BW_250KHZ:
       bwv = 250
   if bw == LoRa.BW_500KHZ:
       bwv = 500
   crv = 4/(4 + cr) 
   return 1000*sf*bwv*(crv)/2**sf

# Air time (in seconds) theoretical calculation for LoRa-RAW, where there is a default preamble of 8 bytes plus 5 bytes of CRC, etc
def airtimetheoretical(payloadsize, sf, bw, cr):
   if bw == LoRa.BW_125KHZ:
       bwv = 125
   if bw == LoRa.BW_250KHZ:
       bwv = 250
   if bw == LoRa.BW_500KHZ:
       bwv = 500
   if sf in [11,12]:
       lowDRopt = 1
   else:
       lowDRopt = 0
   tsym = (2**sf)/(bwv*1000)
   tpreamble = (8 + 4.25)*tsym # Preamble with 8 bytes
   numbersymbolspayload = 8 + max(ceil((8*payloadsize - 4*sf + 28 + 16)/(4*(sf - 2*lowDRopt)))*(4 + cr),0)
   tpayload = numbersymbolspayload*tsym
   tpacket = tpreamble + tpayload
   return tpacket, tpreamble, tpayload, tsym, numbersymbolspayload

Usage example (within ipython3), the first number is the airtime in seconds :

In [1]: from LoRaAirTimeCalc import *
In [2]: airtimetheoretical(20, 10, LoRa.BW_125KHZ, LoRa.CODING_4_5)
Out[2]: (0.37068799999999996, 0.100352, 0.27033599999999997, 0.008192, 33)

Another example, showing the airtime in ms (miliseconds) :

In [1]: from LoRaAirTimeCalc import *
In [2]:  [(payloadsize, airtimetheoretical(payloadsize, 8, LoRa.BW_125KHZ, LoRa.CODING_4_5)[0]*1000) for payloadsize in range(1, 33)]
Out[2]:
[(1, 51.711999999999996),
 (2, 51.711999999999996),
 (3, 61.952),
 (4, 61.952),
 (5, 61.952),
 (6, 61.952),
 (7, 72.19200000000001),
 (8, 72.19200000000001),
 (9, 72.19200000000001),
 (10, 72.19200000000001),
 (11, 82.432),
 (12, 82.432),
 (13, 82.432),
 (14, 82.432),
 (15, 92.672),
 (16, 92.672),
 (17, 92.672),
 (18, 92.672),
 (19, 102.91199999999999),
 (20, 102.91199999999999),
 (21, 102.91199999999999),
 (22, 102.91199999999999),
 (23, 113.15199999999999),
 (24, 113.15199999999999),
 (25, 113.15199999999999),
 (26, 113.15199999999999),
 (27, 123.392),
 (28, 123.392),
 (29, 123.392),
 (30, 123.392),
 (31, 133.632),
 (32, 133.632)

In this GitLab repository/folder "LoRaAirTimeCalc" it has better documentation and examples.

[jmarcelino]

@rcolistete
That's very useful. Thank you for for posting.

[lloureiro]

@rcolistete very nice and usefull, values equal to TTN Data page. Thanks a lot

[kjm]

How do your calcs compare with

m=lora.stats()
print('TxTime(ms)',m[7])