function rtpAnalyze( input_file )
%RTP_ANALYZE Analyze RTP stream(s) from a txt file
% The function takes the output from the command line tool rtp_analyze
% and analyzes the stream(s) therein. First, process your rtpdump file
% through rtp_analyze (from command line):
% $ out/Debug/rtp_analyze my_file.rtp my_file.txt
% Then load it with this function (in Matlab):
% >> rtpAnalyze('my_file.txt')
% Copyright (c) 2015 The WebRTC project authors. All Rights Reserved.
%
% Use of this source code is governed by a BSD-style license
% that can be found in the LICENSE file in the root of the source
% tree. An additional intellectual property rights grant can be found
% in the file PATENTS. All contributing project authors may
% be found in the AUTHORS file in the root of the source tree.
[SeqNo,TimeStamp,ArrTime,Size,PT,M,SSRC] = importfile(input_file);
%% Filter out RTCP packets.
% These appear as RTP packets having payload types 72 through 76.
ix = not(ismember(PT, 72:76));
fprintf('Removing %i RTCP packets\n', length(SeqNo) - sum(ix));
SeqNo = SeqNo(ix);
TimeStamp = TimeStamp(ix);
ArrTime = ArrTime(ix);
Size = Size(ix);
PT = PT(ix);
M = M(ix);
SSRC = SSRC(ix);
%% Find streams.
[uSSRC, ~, uix] = unique(SSRC);
% If there are multiple streams, select one and purge the other
% streams from the data vectors. If there is only one stream, the
% vectors are good to use as they are.
if length(uSSRC) > 1
for i=1:length(uSSRC)
uPT = unique(PT(uix == i));
fprintf('%i: %s (%d packets, pt: %i', i, uSSRC{i}, ...
length(find(uix==i)), uPT(1));
if length(uPT) > 1
fprintf(', %i', uPT(2:end));
end
fprintf(')\n');
end
sel = input('Select stream number: ');
if sel < 1 || sel > length(uSSRC)
error('Out of range');
end
ix = find(uix == sel);
% This is where the data vectors are trimmed.
SeqNo = SeqNo(ix);
TimeStamp = TimeStamp(ix);
ArrTime = ArrTime(ix);
Size = Size(ix);
PT = PT(ix);
M = M(ix);
SSRC = SSRC(ix);
end
%% Unwrap SeqNo and TimeStamp.
SeqNoUW = maxUnwrap(SeqNo, 65535);
TimeStampUW = maxUnwrap(TimeStamp, 4294967295);
%% Generate some stats for the stream.
fprintf('Statistics:\n');
fprintf('SSRC: %s\n', SSRC{1});
uPT = unique(PT);
if length(uPT) > 1
warning('This tool cannot yet handle changes in codec sample rate');
end
fprintf('Payload type(s): %i', uPT(1));
if length(uPT) > 1
fprintf(', %i', uPT(2:end));
end
fprintf('\n');
fprintf('Packets: %i\n', length(SeqNo));
SortSeqNo = sort(SeqNoUW);
fprintf('Missing sequence numbers: %i\n', ...
length(find(diff(SortSeqNo) > 1)));
fprintf('Duplicated packets: %i\n', length(find(diff(SortSeqNo) == 0)));
reorderIx = findReorderedPackets(SeqNoUW);
fprintf('Reordered packets: %i\n', length(reorderIx));
tsdiff = diff(TimeStampUW);
tsdiff = tsdiff(diff(SeqNoUW) == 1);
[utsdiff, ~, ixtsdiff] = unique(tsdiff);
fprintf('Common packet sizes:\n');
for i = 1:length(utsdiff)
fprintf(' %i samples (%i%%)\n', ...
utsdiff(i), ...
round(100 * length(find(ixtsdiff == i))/length(ixtsdiff)));
end
%% Trying to figure out sample rate.
fs_est = (TimeStampUW(end) - TimeStampUW(1)) / (ArrTime(end) - ArrTime(1));
fs_vec = [8, 16, 32, 48];
fs = 0;
for f = fs_vec
if abs((fs_est-f)/f) < 0.05 % 5% margin
fs = f;
break;
end
end
if fs == 0
fprintf('Cannot determine sample rate. I get it to %.2f kHz\n', ...
fs_est);
fs = input('Please, input a sample rate (in kHz): ');
else
fprintf('Sample rate estimated to %i kHz\n', fs);
end
SendTimeMs = (TimeStampUW - TimeStampUW(1)) / fs;
fprintf('Stream duration at sender: %.1f seconds\n', ...
(SendTimeMs(end) - SendTimeMs(1)) / 1000);
fprintf('Stream duration at receiver: %.1f seconds\n', ...
(ArrTime(end) - ArrTime(1)) / 1000);
fprintf('Clock drift: %.2f%%\n', ...
100 * ((ArrTime(end) - ArrTime(1)) / ...
(SendTimeMs(end) - SendTimeMs(1)) - 1));
fprintf('Sent average bitrate: %i kbps\n', ...
round(sum(Size) * 8 / (SendTimeMs(end)-SendTimeMs(1))));
fprintf('Received average bitrate: %i kbps\n', ...
round(sum(Size) * 8 / (ArrTime(end)-ArrTime(1))));
%% Plots.
delay = ArrTime - SendTimeMs;
delay = delay - min(delay);
delayOrdered = delay;
delayOrdered(reorderIx) = nan; % Set reordered packets to NaN.
delayReordered = delay(reorderIx); % Pick the reordered packets.
sendTimeMsReordered = SendTimeMs(reorderIx);
% Sort time arrays in packet send order.
[~, sortix] = sort(SeqNoUW);
SendTimeMs = SendTimeMs(sortix);
Size = Size(sortix);
delayOrdered = delayOrdered(sortix);
figure
plot(SendTimeMs / 1000, delayOrdered, ...
sendTimeMsReordered / 1000, delayReordered, 'r.');
xlabel('Send time [s]');
ylabel('Relative transport delay [ms]');
title(sprintf('SSRC: %s', SSRC{1}));
SendBitrateKbps = 8 * Size(1:end-1) ./ diff(SendTimeMs);
figure
plot(SendTimeMs(1:end-1)/1000, SendBitrateKbps);
xlabel('Send time [s]');
ylabel('Send bitrate [kbps]');
end
%% Subfunctions.
% findReorderedPackets returns the index to all packets that are considered
% old compared with the largest seen sequence number. The input seqNo must
% be unwrapped for this to work.
function reorderIx = findReorderedPackets(seqNo)
largestSeqNo = seqNo(1);
reorderIx = [];
for i = 2:length(seqNo)
if seqNo(i) < largestSeqNo
reorderIx = [reorderIx; i]; %#ok<AGROW>
else
largestSeqNo = seqNo(i);
end
end
end
%% Auto-generated subfunction.
function [SeqNo,TimeStamp,SendTime,Size,PT,M,SSRC] = ...
importfile(filename, startRow, endRow)
%IMPORTFILE Import numeric data from a text file as column vectors.
% [SEQNO,TIMESTAMP,SENDTIME,SIZE,PT,M,SSRC] = IMPORTFILE(FILENAME) Reads
% data from text file FILENAME for the default selection.
%
% [SEQNO,TIMESTAMP,SENDTIME,SIZE,PT,M,SSRC] = IMPORTFILE(FILENAME,
% STARTROW, ENDROW) Reads data from rows STARTROW through ENDROW of text
% file FILENAME.
%
% Example:
% [SeqNo,TimeStamp,SendTime,Size,PT,M,SSRC] =
% importfile('rtpdump_recv.txt',2, 123);
%
% See also TEXTSCAN.
% Auto-generated by MATLAB on 2015/05/28 09:55:50
%% Initialize variables.
if nargin<=2
startRow = 2;
endRow = inf;
end
%% Format string for each line of text:
% column1: double (%f)
% column2: double (%f)
% column3: double (%f)
% column4: double (%f)
% column5: double (%f)
% column6: double (%f)
% column7: text (%s)
% For more information, see the TEXTSCAN documentation.
formatSpec = '%5f%11f%11f%6f%6f%3f%s%[^\n\r]';
%% Open the text file.
fileID = fopen(filename,'r');
%% Read columns of data according to format string.
% This call is based on the structure of the file used to generate this
% code. If an error occurs for a different file, try regenerating the code
% from the Import Tool.
dataArray = textscan(fileID, formatSpec, endRow(1)-startRow(1)+1, ...
'Delimiter', '', 'WhiteSpace', '', 'HeaderLines', startRow(1)-1, ...
'ReturnOnError', false);
for block=2:length(startRow)
frewind(fileID);
dataArrayBlock = textscan(fileID, formatSpec, ...
endRow(block)-startRow(block)+1, 'Delimiter', '', 'WhiteSpace', ...
'', 'HeaderLines', startRow(block)-1, 'ReturnOnError', false);
for col=1:length(dataArray)
dataArray{col} = [dataArray{col};dataArrayBlock{col}];
end
end
%% Close the text file.
fclose(fileID);
%% Post processing for unimportable data.
% No unimportable data rules were applied during the import, so no post
% processing code is included. To generate code which works for
% unimportable data, select unimportable cells in a file and regenerate the
% script.
%% Allocate imported array to column variable names
SeqNo = dataArray{:, 1};
TimeStamp = dataArr
