summaryrefslogtreecommitdiff
path: root/validation/test/P2P_throughput/P2P_throughput.rst
blob: ca68a89f5319da0020068a9c17ee3d4d455b0a76 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
P2P throughput report (Iperf)
+++++++++++++++++++++++++++++
.. _MSTAR: http://www.mstarsemi.com

:Company:
 MSTAR_
:Plug model/version (host A):
 {plug_model_a} - {plug_version_a}
:Plug model/version (host B):
 {plug_model_b} - {plug_version_b}

.. sectnum::

.. |date| date::

.. contents:: Table of contents

.. footer::
   ###Title### - {plug_model_a} vs {plug_model_b} - |date| - page ###Page### / ###Total###

.. raw:: pdf

   PageBreak

Abstract
========
This document describes the maximum throughput obtained between 2 PCs, connected in PLC, with the tool Iperf, for different types of traffic and different channel conditions.

Test bench
==========

Description
-----------

.. figure:: {fig_name}
 :scale: 200 %

On the figure, the host A controls the attenuator and the wave form generator but it is not technically necessary and it could be any other PC.

The attenuation of P dB on the wave form generator is useful when the generator is off, in order to avoid any reflection.

The purpose of the attenuation of N dB on the output of each plug is to avoid any dazzling between the plugs.

The AMN's are necessary only if the plugs are not EoC, i.e have no native output in coax.

.. raw:: pdf

   PageBreak

Configuration
-------------

 * Start order: The host A is started first
 * Iperf version on the host A ({non_plc_ip_address_a}): {version_a}
 * Iperf version on the host B ({non_plc_ip_address_b}): {version_b}
 * AMN attenuation (M): {amn_attenuation_db} dB
 * Signal attenuation (N): {signal_attenuation_db} dB
 * T attenuation (T): {t_attenuation_db} dB
 * Wave form attenuation (P): {waveform_attenuation_db} dB

.. raw:: pdf

   PageBreak

Earthing guidelines
-------------------
The earthing guidelines are the following:

* Separate the ground and the earth on the AMN's so that the signal does not radiate through the earth
* Connect the grounds of the AMN's with a coax without core
* Use non-shielded Ethernet cables to connect the plugs to the PC's
* Connect the earth of the power strip to the T

Network settings
----------------

The PC of test and the hosts A and B should be configured so that the PC of test is able to connect to the host A and B in SSH without having to enter a password for each SSH session.

The PC of test should be configured to be able to ping the power strip. To that aim, on the PC of test, execute the following command (where ethX is the interface between the PC of test and the power strip)::

 sudo route add -net $POWER_STRIP_ADDRESS netmask 255.255.255.255 $ethX

For MSTAR plugs, the PC of test and the host A (resp. B) should be configured so that the PC of test is able to ping and telnet the plug A (resp. B) via the host A (resp. B). To that aim:

* On the PC of test, execute the following commands::

   sudo route add -net $PLUG_A_ADDRESS netmask 255.255.255.255 gw $HOST_A
   sudo route add -net $PLUG_B_ADDRESS netmask 255.255.255.255 gw $HOST_B

* On the hosts A and B:

 - Uncomment the following line in the file /etc/sysctl.conf::

    #net.ipv4.ip_forward=1

 - Execute the following command::

    sudo /etc/init.d/procps restart

 - Execute the following command, where ethX is the interface between the host and the plug::

    sudo iptables -t nat -A POSTROUTING -s $TEST_PC_NETWORK -o ethX -j MASQUERADE

.. raw:: pdf

   PageBreak

Notes
=====

SNR calculation
---------------

Hypothesis:

 * Attenuation of the attenuator = 0 dB
 * Peak to Average Ratio (PAR) = {par_db} dB
 * Peak Power Spectral Density (PPSD) = {ppsd_dbm_per_hz} dBm/Hz
 * Signal Bandwith (SB - {signal_freq_min_mhz} MHz to {signal_freq_max_mhz} MHz) = {signal_bandwith_mhz} MHz
 * Noise Bandwith (NB - {noise_freq_min_mhz} MHz to {noise_freq_max_mhz} MHz) = {noise_bandwith_mhz} MHz

Therefore:

 * Tx Signal Power (SP) = PSD - PAR + 10 log10(SB) = {tx_signal_power_dbm} dBm
 * Bandwiths Ratio (BR) = 10 log10 (SB/NB) = {bandwiths_ratio_db} dB
 * Signal to Noise Ratio (dB) = (SP - {amn_attenuation_db} dB - {signal_attenuation_db} dB) -  (Noise Power (dBm) - {waveform_attenuation_db} dB) - BR
 * Signal to Noise Ratio (dB) = {snr_offset_db} - Noise Power (dBm)

SJR calculation
---------------

Hypothesis:

 * Attenuation of the attenuator = {sjr_attenuator_attenuation_db} dB
 * Peak to Average Ratio (PAR) = {par_db} dB
 * Peak Power Spectral Density (PPSD) = {ppsd_dbm_per_hz} dBm/Hz
 * Signal Bandwith (SB - {signal_freq_min_mhz} MHz to {signal_freq_max_mhz} MHz) = {signal_bandwith_mhz} MHz

Therefore:

 * Tx Signal Power (SP) = PSD - PAR + 10 log10(SB) = {tx_signal_power_dbm} dBm
 * Signal to Jammer Ratio (dB) = (SP - {amn_attenuation_db} dB - {signal_attenuation_db} dB - {sjr_attenuator_attenuation_db} dB) - (Jammer Power (dBm) - {waveform_attenuation_db} dB)
 * Signal to Jammer Ratio (dB) = {sjr_offset_db} - Jammer Power (dBm)

Channel Estimation (CE)
-----------------------

When the channel changes, the tone map must be recalculated, which is called the Channel Estimation (CE) and which takes some time.

For that reason, in some tests related to the adaptation to the channel (attenuation, noise, etc.), a traffic is first generated during a certain time, in order to force the tone map adaptation, and then the measurement is done.

.. raw:: pdf

   PageBreak