summaryrefslogtreecommitdiff
path: root/validation/test/P2P_throughput/P2P_throughput.rst
blob: 61bcdab504e4d6e6e06df517ea7f6fc30e7dcf56 (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
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
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 installation
=======================

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

.. figure:: {bench_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

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

The PC of test and the hosts A and B must 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 must 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

The PC of test and the host A (resp. B) must 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

Software
--------

On the PC of test, the packages **python**, **dia** and **rst2pdf** must be installed .

On the hosts A and B, the package **proftpd-basic** must be installed, and an FTP server must be configured and running, so that the SPC300-like plugs can upload their files of traces.

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

* Use non-shielded Ethernet cables to connect the plugs to the PC's
* Connect the earth of the power strip to the T
* If there are AMN's:

 - 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

.. raw:: pdf

   PageBreak

.. raw:: pdf

   PageBreak

PPSD measurement
----------------

It is necessary to know the PPSD (Peak Power Spectral Density) in order to be able to:

 - Complete the `SNR calculation`_ and the `SJR calculation`_
 - Compare the results obtained with different test benches

Typical values are the following:

 - -51 dBm/Hz for a plug AV (SPR310 or MPR520), since the maximum authorized is -50 dBm/Hz
 - -48 dBm/Hz for a slave EoC SPC300 (SCR310)
 - -55 dBm/Hz for a master EoC MSE500 (MSK500 or MCR500)

The exact value highly depends on the board and must be carefully measured with a spectrum analyser, as detailed in the example below.

.. figure:: {power_fig_name}
 :scale: 250 %

In this example::

  Measured average output power (MP) = 113.54 dBµV
  Attenuation of the resistive coupler used for the measurement (ARC) = 6 dB
  Peak to Average Ratio (PAR) = {par_db} dB
  Signal Bandwith (SB) = 66 MHz

Therefore::

  Tx Signal Power (SP) = MP + ARC
                       = 113.54 dBµV + 6 dB
                       = 119.54 dBµV
		       = 12.55 dBm (on a 50 ohms load)
  PPSD = SP + PAR - 10 log10(SB)
       = 12.55 dBm + {par_db} dB - 10 log10(66 x 10^6) dB
       = -55.64 dBm/Hz


Test bench parameters
=====================

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

Relatively to the figure given in the `Description`_ section, the parameters used for this report are the following::

 Start order: The host A is started first
 Host A ({non_plc_ip_address_a}):
  Plug model: {plug_model_a}
  Plug version: {plug_version_a}
  Iperf version: {version_a}
 Host B ({non_plc_ip_address_b}):
  Plug model: {plug_model_b}
  Plug version: {plug_version_b}
  Iperf version: {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



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

The parameters used to compute the SNR for this report are the following::

 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) = PPSD - 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
                            = {snr_offset_db} - Noise Power (dBm)

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

The parameters used to compute the SJR for this report are the following::

 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)
                             = {sjr_offset_db} - Jammer Power (dBm)

.. raw:: pdf

   PageBreak

Results
=======

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