OpportunistiKapacity

A python library to turn mobility traces into opportunistic contact capacity. This library aims to simplify the calculation of total network data capacity, more precisely in the context of contact networks.

Basic run using wrapper

You must provide at least 4 parameters for the software to run:

1. The trace kind
2. The trace path
3. A propagation model
4. A modulation scheme

If these arguments are known, you can run the script: ./OKwrapper.py trace_kind dataset propagation_model modulation_scheme

Supported traces

This library only considers two types of datasets. Both are text-based.

Mobility trace

The mobility trace kind is a format where all existing nodes have their geographical coordinates updated at each snapshot. For example:

0.0 	a	 50479.67 	54248.598
0.0 	b 	47721.375 	54930.71
0.0 	c 	48519.473 	55673.586
0.0 	d 	48798.016 	54503.477
0.0 	e 	50233.27 	55252.617
1.0 	a 	50479.684 	54248.62
1.0 	b 	47722.023 	54930.7
1.0 	c 	48518.676 	55675.266
1.0 	d 	48798.703 	54505.332
1.0 	e 	50235.68 	55251.816 

The above snippet has four columns, from left to right: time, node id, x coordinate, y coordinate. Publishing traces in such a format is a common practice within mobility simulation studies (e.g., TAPASCologne).

Contact trace

The contact trace kind is a format describing a contact duration between two nodes, regardless of when the contact actually happened.

1	2	1156084891	1156084895	
2	3	1156085092	1156085159	
4	2	1156085110	1156085118	
5	8	1156085190	1156085191	
4	2	1156085219	1156085302	

The above snippet has four columns, from left to right: id node 1, id node 2, start contact, end contact.

Propagation models

The propagation models, found in the communications module, calculate the signal loss according to the distance between the receiver and transmitter. So far we include the following path loss models:

• Freespace (freespace_loss)
• Log-distance (logDistance_loss)
• Two-Ray ground reflection (twoRay_loss)

Modulation schemes

The modulation schemes, found in the communications module, set a data rate for a given RSSI between two nodes. So far, we propose the following path loss models (explained thoroughly in our paper):

• Wifi5_empirical_goodput
• Wifi5_stepwise_max
• Wifi5_stepwise_linear_adjusted
• Wifi5_stepwise_fit

Module and classes

The module is separated in 3 files:

• communications everything function to do with the translation of distance into a throughput estimation.
• datasetparser: provides two classes, (ContactParser, MobilityParser) to iterate over a dataset file.
• contactcalculator, provides the two classes (GeographicTrace, ContactTrace) to calculate the contact capacity.