Mercurial > hg > audiostuff
comparison spandsp-0.0.3/spandsp-0.0.3/DueDiligence @ 5:f762bf195c4b
import spandsp-0.0.3
| author | Peter Meerwald <pmeerw@cosy.sbg.ac.at> |
|---|---|
| date | Fri, 25 Jun 2010 16:00:21 +0200 |
| parents | |
| children |
comparison
equal
deleted
inserted
replaced
| 4:26cd8f1ef0b1 | 5:f762bf195c4b |
|---|---|
| 1 Intellectual Property Due Diligence | |
| 2 ----------------------------------- | |
| 3 | |
| 4 Modems and voice coding are heavily patented areas. Implementing these without | |
| 5 serious consideration of IP issues would be foolish. This document describes | |
| 6 the basis on which the software has been implemented. | |
| 7 | |
| 8 | |
| 9 A check of the intellectual property information at the ITU web site shows a | |
| 10 number of patent claims against the current standards implemented by spandsp. | |
| 11 It is important to realise, however, that some of these patents have long | |
| 12 since expired (group III fax dates back to the 1970s). Also, many are | |
| 13 related to recent additions to the FAX standard, such as colour FAX handling, | |
| 14 which few people ever use. | |
| 15 | |
| 16 The V.14 rate adaption standard seems free of patent encumberance. | |
| 17 | |
| 18 One patent is listed as relevant to the V.17 standard. It is a patent from | |
| 19 IBM, but the ITU database does not specify its nature. I believe it is | |
| 20 related to the trellis coding used, and I think it has expired. I do not | |
| 21 know for sure. The techniques used in the implementation should be free of | |
| 22 patent encumberance. Most of the implementation is similar to the V.29 | |
| 23 modem. The key addition the trellis code processing. The trellis encoding | |
| 24 is trivial. The decoding uses Viterbi techniques, which are quite old. | |
| 25 | |
| 26 The V.21 standard dates from the 1950s. The V.23 standard is also very old. | |
| 27 There is no possibility that any patents related to it are still in force. | |
| 28 However, the implementation also needs to be free of patented techniques. | |
| 29 The implementation only uses very mature numerical oscillator and quadrature | |
| 30 correlation techniques, so there should be no patent issues. | |
| 31 | |
| 32 Only one patent is listed as relevant to the V.29 standard. This dates from | |
| 33 the 1970s, and must have expired. The modem has been implemented using only | |
| 34 very mature techniques, none of which can be less than 20 years old. There | |
| 35 seem no possibility, therefore, that any patents are still in force related | |
| 36 to the techniques used. | |
| 37 | |
| 38 Some aspect of the V.8 standard seems to have patents associated with it, | |
| 39 according to the ITU patent database. I am unclear what these are. V.8 is a | |
| 40 very simple standard. There seems to be nothing innovative about it. | |
| 41 | |
| 42 Many patents are listed as relevant to the T.30 standard. However, they all | |
| 43 appear to relate to newer features, such as colour FAX, added in recent years. | |
| 44 The current implementation only covers the original features from the late | |
| 45 1970s, where there appear to be patent issues. | |
| 46 | |
| 47 The T.4 standard defines the image compression and decompression techniques | |
| 48 used for group 3 FAXes. The spandsp implementation is based on code derived | |
| 49 from freely available implementations of T.4. These have existed for a number | |
| 50 of years without IP issues. The standard is old enough for any patents to have | |
| 51 expired, anyway. | |
| 52 | |
| 53 V.42bis compression uses the LZW algorithm. This is the same algorithm used in | |
| 54 GIF files. Unisys patented this algorithm. However, the Unisys patent has now | |
| 55 expired. |