CA2408809C  A method of and a device for generating a signal having a predetermined quality criterion for use in or on a communication system  Google Patents
A method of and a device for generating a signal having a predetermined quality criterion for use in or on a communication system Download PDFInfo
 Publication number
 CA2408809C CA2408809C CA2408809A CA2408809A CA2408809C CA 2408809 C CA2408809 C CA 2408809C CA 2408809 A CA2408809 A CA 2408809A CA 2408809 A CA2408809 A CA 2408809A CA 2408809 C CA2408809 C CA 2408809C
 Authority
 CA
 Canada
 Prior art keywords
 signal
 amplitude
 frequency
 time domain
 noise
 Prior art date
 Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
 Expired  Lifetime
Links
 238000004891 communication Methods 0.000 title claims abstract description 49
 230000003595 spectral Effects 0.000 claims abstract description 89
 238000010998 test method Methods 0.000 claims abstract description 6
 230000000051 modifying Effects 0.000 claims description 36
 238000001228 spectrum Methods 0.000 claims description 34
 239000000969 carrier Substances 0.000 claims description 33
 230000001131 transforming Effects 0.000 claims description 21
 238000007493 shaping process Methods 0.000 claims description 20
 238000005315 distribution function Methods 0.000 claims description 10
 238000005259 measurement Methods 0.000 claims description 2
 238000004590 computer program Methods 0.000 claims 1
 239000000523 sample Substances 0.000 description 13
 206010063834 Oversensing Diseases 0.000 description 12
 238000000034 method Methods 0.000 description 10
 238000001914 filtration Methods 0.000 description 9
 230000001419 dependent Effects 0.000 description 7
 230000005540 biological transmission Effects 0.000 description 6
 230000004044 response Effects 0.000 description 6
 230000001186 cumulative Effects 0.000 description 5
 238000010586 diagram Methods 0.000 description 5
 230000035515 penetration Effects 0.000 description 3
 230000001360 synchronised Effects 0.000 description 3
 238000004364 calculation method Methods 0.000 description 2
 230000001364 causal effect Effects 0.000 description 2
 230000001808 coupling Effects 0.000 description 2
 238000010168 coupling process Methods 0.000 description 2
 238000005859 coupling reaction Methods 0.000 description 2
 238000001514 detection method Methods 0.000 description 2
 238000011144 upstream manufacturing Methods 0.000 description 2
 ZOCUOMKMBMEYQVGSLJADNHSAN 9αFluoro11β,17α,21trihydroxypregna1,4diene3,20dione 21acetate Chemical compound data:image/svg+xml;base64,<?xml version='1.0' encoding='iso-8859-1'?>
<svg version='1.1' baseProfile='full'
              xmlns='http://www.w3.org/2000/svg'
                      xmlns:rdkit='http://www.rdkit.org/xml'
                      xmlns:xlink='http://www.w3.org/1999/xlink'
                  xml:space='preserve'
width='300px' height='300px' viewBox='0 0 300 300'>
<!-- END OF HEADER -->
<rect style='opacity:1.0;fill:#FFFFFF;stroke:none' width='300.0' height='300.0' x='0.0' y='0.0'> </rect>
<path class='bond-0 atom-0 atom-1' d='M 192.7,189.1 L 216.9,185.8' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-29 atom-12 atom-0' d='M 177.8,169.9 L 192.7,189.1' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-1 atom-1 atom-2' d='M 216.9,185.8 L 226.0,163.3' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-2 atom-2 atom-3' d='M 226.0,163.3 L 250.2,159.9' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-2 atom-2 atom-3' d='M 229.0,157.9 L 245.9,155.6' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-31 atom-8 atom-2' d='M 211.1,144.0 L 226.0,163.3' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-3 atom-3 atom-4' d='M 250.2,159.9 L 259.3,137.4' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-4 atom-4 atom-5' d='M 259.7,139.8 L 267.5,138.7' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-4 atom-4 atom-5' d='M 267.5,138.7 L 275.4,137.6' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-4 atom-4 atom-5' d='M 259.0,135.0 L 266.9,133.9' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-4 atom-4 atom-5' d='M 266.9,133.9 L 274.8,132.8' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-5 atom-4 atom-6' d='M 259.3,137.4 L 244.4,118.2' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-6 atom-6 atom-7' d='M 244.4,118.2 L 220.3,121.5' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-6 atom-6 atom-7' d='M 241.4,123.5 L 224.6,125.8' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-7 atom-7 atom-8' d='M 220.3,121.5 L 211.1,144.0' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-8 atom-8 atom-9' d='M 208.5,139.9 L 207.7,140.5' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-8 atom-8 atom-9' d='M 205.9,135.8 L 204.4,136.9' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-8 atom-8 atom-9' d='M 203.3,131.6 L 201.0,133.4' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-8 atom-8 atom-9' d='M 200.7,127.5 L 197.6,129.9' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-8 atom-8 atom-9' d='M 198.1,123.3 L 194.2,126.3' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-9 atom-8 atom-10' d='M 211.1,144.0 L 187.0,147.4' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-10 atom-10 atom-11' d='M 187.0,147.4 L 190.6,154.0 L 192.5,152.5 Z' style='fill:#3B4143;fill-rule:evenodd;fill-opacity:1;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;' />
<path class='bond-10 atom-10 atom-11' d='M 190.6,154.0 L 198.1,157.7 L 194.3,160.7 Z' style='fill:#77D8ED;fill-rule:evenodd;fill-opacity:1;stroke:#77D8ED;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;' />
<path class='bond-10 atom-10 atom-11' d='M 190.6,154.0 L 192.5,152.5 L 198.1,157.7 Z' style='fill:#77D8ED;fill-rule:evenodd;fill-opacity:1;stroke:#77D8ED;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;' />
<path class='bond-11 atom-10 atom-12' d='M 187.0,147.4 L 177.8,169.9' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-32 atom-28 atom-10' d='M 172.0,128.1 L 187.0,147.4' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-12 atom-12 atom-13' d='M 177.8,169.9 L 153.7,173.2' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-33 atom-12 atom-30' d='M 175.1,174.4 L 176.6,175.1' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-33 atom-12 atom-30' d='M 172.4,179.0 L 175.4,180.2' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-33 atom-12 atom-30' d='M 169.6,183.5 L 174.2,185.3' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-13 atom-13 atom-14' d='M 153.7,173.2 L 140.0,193.4' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-30 atom-25 atom-13' d='M 138.8,154.0 L 153.7,173.2' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-34 atom-13 atom-31' d='M 153.7,173.2 L 138.5,168.0 L 137.6,172.8 Z' style='fill:#3B4143;fill-rule:evenodd;fill-opacity:1;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;' />
<path class='bond-14 atom-14 atom-15' d='M 140.0,193.4 L 116.6,186.6' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-15 atom-15 atom-16' d='M 116.6,186.6 L 115.9,162.3' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-16 atom-16 atom-17' d='M 115.9,162.3 L 91.7,165.6' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-23 atom-16 atom-24' d='M 115.9,162.3 L 114.6,155.5 L 112.3,156.4 Z' style='fill:#3B4143;fill-rule:evenodd;fill-opacity:1;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;' />
<path class='bond-23 atom-16 atom-24' d='M 114.6,155.5 L 108.7,150.4 L 113.3,148.7 Z' style='fill:#E84235;fill-rule:evenodd;fill-opacity:1;stroke:#E84235;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;' />
<path class='bond-23 atom-16 atom-24' d='M 114.6,155.5 L 112.3,156.4 L 108.7,150.4 Z' style='fill:#E84235;fill-rule:evenodd;fill-opacity:1;stroke:#E84235;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;' />
<path class='bond-24 atom-16 atom-25' d='M 115.9,162.3 L 138.8,154.0' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-17 atom-17 atom-18' d='M 89.5,164.7 L 86.5,171.9' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-17 atom-17 atom-18' d='M 86.5,171.9 L 83.6,179.2' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-17 atom-17 atom-18' d='M 94.0,166.5 L 91.0,173.8' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-17 atom-17 atom-18' d='M 91.0,173.8 L 88.1,181.0' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-18 atom-17 atom-19' d='M 91.7,165.6 L 76.8,146.4' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-19 atom-19 atom-20' d='M 76.8,146.4 L 68.9,147.5' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-19 atom-19 atom-20' d='M 68.9,147.5 L 61.0,148.6' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-20 atom-20 atom-21' d='M 47.0,142.3 L 42.4,136.4' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-20 atom-20 atom-21' d='M 42.4,136.4 L 37.8,130.5' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-21 atom-21 atom-22' d='M 40.0,131.4 L 42.6,125.1' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-21 atom-21 atom-22' d='M 42.6,125.1 L 45.1,118.8' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-21 atom-21 atom-22' d='M 35.5,129.6 L 38.1,123.3' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-21 atom-21 atom-22' d='M 38.1,123.3 L 40.6,117.0' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-22 atom-21 atom-23' d='M 37.8,130.5 L 13.6,133.8' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-25 atom-25 atom-26' d='M 136.2,149.8 L 135.4,150.4' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-25 atom-25 atom-26' d='M 133.7,145.6 L 132.1,146.8' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-25 atom-25 atom-26' d='M 131.2,141.5 L 128.8,143.2' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-25 atom-25 atom-26' d='M 128.6,137.3 L 125.5,139.6' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-25 atom-25 atom-26' d='M 126.1,133.1 L 122.2,136.0' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-26 atom-25 atom-27' d='M 138.8,154.0 L 147.9,131.5' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-27 atom-27 atom-28' d='M 147.9,131.5 L 172.0,128.1' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:2.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-28 atom-28 atom-29' d='M 174.5,124.3 L 173.0,123.6' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-28 atom-28 atom-29' d='M 177.0,120.4 L 174.0,119.2' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-28 atom-28 atom-29' d='M 179.4,116.5 L 174.9,114.7' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<text x='280.5' y='138.9' class='atom-5' style='font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >O</text>
<text x='199.0' y='171.5' class='atom-11' style='font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#77D8ED' >F</text>
<text x='79.6' y='193.0' class='atom-18' style='font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >O</text>
<text x='49.8' y='154.6' class='atom-20' style='font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >O</text>
<text x='44.0' y='112.8' class='atom-22' style='font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >O</text>
<text x='98.0' y='144.4' class='atom-24' style='font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >H</text>
<text x='104.3' y='144.4' class='atom-24' style='font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >O</text>
<text x='178.3' y='110.5' class='atom-29' style='font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >O</text>
<text x='185.0' y='110.5' class='atom-29' style='font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >H</text>
<text x='165.7' y='197.3' class='atom-30' style='font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#3B4143' >H</text>
<text x='126.8' y='173.8' class='atom-31' style='font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#3B4143' >H</text>
</svg>
 data:image/svg+xml;base64,<?xml version='1.0' encoding='iso-8859-1'?>
<svg version='1.1' baseProfile='full'
              xmlns='http://www.w3.org/2000/svg'
                      xmlns:rdkit='http://www.rdkit.org/xml'
                      xmlns:xlink='http://www.w3.org/1999/xlink'
                  xml:space='preserve'
width='85px' height='85px' viewBox='0 0 85 85'>
<!-- END OF HEADER -->
<rect style='opacity:1.0;fill:#FFFFFF;stroke:none' width='85.0' height='85.0' x='0.0' y='0.0'> </rect>
<path class='bond-0 atom-0 atom-1' d='M 53.6,52.7 L 60.3,51.8' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-29 atom-12 atom-0' d='M 49.4,47.3 L 53.6,52.7' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-1 atom-1 atom-2' d='M 60.3,51.8 L 62.9,45.4' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-2 atom-2 atom-3' d='M 62.9,45.4 L 69.7,44.5' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-2 atom-2 atom-3' d='M 63.7,44.0 L 68.5,43.3' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-31 atom-8 atom-2' d='M 58.7,40.1 L 62.9,45.4' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-3 atom-3 atom-4' d='M 69.7,44.5 L 72.2,38.2' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-4 atom-4 atom-5' d='M 72.3,38.9 L 74.7,38.5' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-4 atom-4 atom-5' d='M 74.7,38.5 L 77.1,38.2' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-4 atom-4 atom-5' d='M 72.1,37.5 L 74.5,37.2' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-4 atom-4 atom-5' d='M 74.5,37.2 L 76.9,36.9' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-5 atom-4 atom-6' d='M 72.2,38.2 L 68.1,32.8' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-6 atom-6 atom-7' d='M 68.1,32.8 L 61.3,33.7' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-6 atom-6 atom-7' d='M 67.2,34.3 L 62.5,34.9' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-7 atom-7 atom-8' d='M 61.3,33.7 L 58.7,40.1' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-8 atom-8 atom-9' d='M 57.5,38.1 L 57.1,38.4' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-8 atom-8 atom-9' d='M 56.3,36.2 L 55.6,36.7' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-8 atom-8 atom-9' d='M 55.1,34.2 L 54.0,35.1' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-9 atom-8 atom-10' d='M 58.7,40.1 L 52.0,41.0' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-10 atom-10 atom-11' d='M 52.0,41.0 L 53.0,42.9 L 53.6,42.5 Z' style='fill:#3B4143;fill-rule:evenodd;fill-opacity:1;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;' />
<path class='bond-10 atom-10 atom-11' d='M 53.0,42.9 L 55.2,44.0 L 54.1,44.8 Z' style='fill:#77D8ED;fill-rule:evenodd;fill-opacity:1;stroke:#77D8ED;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;' />
<path class='bond-10 atom-10 atom-11' d='M 53.0,42.9 L 53.6,42.5 L 55.2,44.0 Z' style='fill:#77D8ED;fill-rule:evenodd;fill-opacity:1;stroke:#77D8ED;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;' />
<path class='bond-11 atom-10 atom-12' d='M 52.0,41.0 L 49.4,47.3' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-32 atom-28 atom-10' d='M 47.8,35.6 L 52.0,41.0' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-12 atom-12 atom-13' d='M 49.4,47.3 L 42.6,48.2' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-33 atom-12 atom-30' d='M 48.6,48.7 L 49.0,48.8' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-33 atom-12 atom-30' d='M 47.8,50.0 L 48.6,50.4' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-33 atom-12 atom-30' d='M 47.0,51.4 L 48.2,51.9' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-13 atom-13 atom-14' d='M 42.6,48.2 L 38.8,53.9' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-30 atom-25 atom-13' d='M 38.4,42.9 L 42.6,48.2' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-34 atom-13 atom-31' d='M 42.6,48.2 L 38.0,46.7 L 37.8,48.1 Z' style='fill:#3B4143;fill-rule:evenodd;fill-opacity:1;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;' />
<path class='bond-14 atom-14 atom-15' d='M 38.8,53.9 L 32.2,52.0' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-15 atom-15 atom-16' d='M 32.2,52.0 L 32.0,45.2' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-16 atom-16 atom-17' d='M 32.0,45.2 L 25.3,46.1' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-23 atom-16 atom-24' d='M 32.0,45.2 L 31.7,43.4 L 31.1,43.7 Z' style='fill:#3B4143;fill-rule:evenodd;fill-opacity:1;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;' />
<path class='bond-23 atom-16 atom-24' d='M 31.7,43.4 L 30.2,42.2 L 31.4,41.7 Z' style='fill:#E84235;fill-rule:evenodd;fill-opacity:1;stroke:#E84235;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;' />
<path class='bond-23 atom-16 atom-24' d='M 31.7,43.4 L 31.1,43.7 L 30.2,42.2 Z' style='fill:#E84235;fill-rule:evenodd;fill-opacity:1;stroke:#E84235;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;' />
<path class='bond-24 atom-16 atom-25' d='M 32.0,45.2 L 38.4,42.9' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-17 atom-17 atom-18' d='M 24.6,45.8 L 23.7,48.0' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-17 atom-17 atom-18' d='M 23.7,48.0 L 22.9,50.2' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-17 atom-17 atom-18' d='M 25.9,46.4 L 25.0,48.5' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-17 atom-17 atom-18' d='M 25.0,48.5 L 24.1,50.7' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-18 atom-17 atom-19' d='M 25.3,46.1 L 21.1,40.7' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-19 atom-19 atom-20' d='M 21.1,40.7 L 18.7,41.0' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-19 atom-19 atom-20' d='M 18.7,41.0 L 16.3,41.4' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-20 atom-20 atom-21' d='M 12.8,39.7 L 11.5,38.0' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-20 atom-20 atom-21' d='M 11.5,38.0 L 10.1,36.3' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-21 atom-21 atom-22' d='M 10.8,36.5 L 11.4,34.9' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-21 atom-21 atom-22' d='M 11.4,34.9 L 12.0,33.4' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-21 atom-21 atom-22' d='M 9.5,36.0 L 10.1,34.4' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-21 atom-21 atom-22' d='M 10.1,34.4 L 10.8,32.9' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-22 atom-21 atom-23' d='M 10.1,36.3 L 3.4,37.2' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-25 atom-25 atom-26' d='M 37.3,40.9 L 36.9,41.2' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-25 atom-25 atom-26' d='M 36.1,38.9 L 35.3,39.5' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-25 atom-25 atom-26' d='M 34.9,37.0 L 33.8,37.8' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-26 atom-25 atom-27' d='M 38.4,42.9 L 41.0,36.5' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-27 atom-27 atom-28' d='M 41.0,36.5 L 47.8,35.6' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-28 atom-28 atom-29' d='M 48.4,34.6 L 48.0,34.5' style='fill:none;fill-rule:evenodd;stroke:#3B4143;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-28 atom-28 atom-29' d='M 49.0,33.7 L 48.2,33.3' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<path class='bond-28 atom-28 atom-29' d='M 49.7,32.7 L 48.4,32.2' style='fill:none;fill-rule:evenodd;stroke:#E84235;stroke-width:1.0px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1' />
<text x='77.2' y='40.3' class='atom-5' style='font-size:6px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >O</text>
<text x='54.3' y='49.4' class='atom-11' style='font-size:6px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#77D8ED' >F</text>
<text x='20.9' y='55.4' class='atom-18' style='font-size:6px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >O</text>
<text x='12.5' y='44.6' class='atom-20' style='font-size:6px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >O</text>
<text x='10.9' y='32.9' class='atom-22' style='font-size:6px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >O</text>
<text x='23.9' y='41.8' class='atom-24' style='font-size:6px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >H</text>
<text x='27.8' y='41.8' class='atom-24' style='font-size:6px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >O</text>
<text x='48.5' y='32.3' class='atom-29' style='font-size:6px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >O</text>
<text x='52.7' y='32.3' class='atom-29' style='font-size:6px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#E84235' >H</text>
<text x='45.0' y='56.6' class='atom-30' style='font-size:6px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#3B4143' >H</text>
<text x='34.1' y='50.0' class='atom-31' style='font-size:6px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#3B4143' >H</text>
</svg>
 C1CC2=CC(=O)C=C[C@]2(C)[C@]2(F)[C@@H]1[C@@H]1CC[C@@](C(=O)COC(=O)C)(O)[C@@]1(C)C[C@@H]2O ZOCUOMKMBMEYQVGSLJADNHSAN 0.000 description 1
 229940048207 Predef Drugs 0.000 description 1
 241000282485 Vulpes vulpes Species 0.000 description 1
 RYGMFSIKBFXOCRUHFFFAOYSAN copper Chemical compound data:image/svg+xml;base64,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 data:image/svg+xml;base64,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 [Cu] RYGMFSIKBFXOCRUHFFFAOYSAN 0.000 description 1
 229910052802 copper Inorganic materials 0.000 description 1
 239000010949 copper Substances 0.000 description 1
 230000001955 cumulated Effects 0.000 description 1
 230000000694 effects Effects 0.000 description 1
 239000003365 glass fiber Substances 0.000 description 1
 238000004519 manufacturing process Methods 0.000 description 1
 238000010606 normalization Methods 0.000 description 1
 238000011056 performance test Methods 0.000 description 1
 238000011160 research Methods 0.000 description 1
 230000002104 routine Effects 0.000 description 1
Classifications

 H—ELECTRICITY
 H04—ELECTRIC COMMUNICATION TECHNIQUE
 H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
 H04L1/00—Arrangements for detecting or preventing errors in the information received
 H04L1/24—Testing correct operation
 H04L1/242—Testing correct operation by comparing a transmitted test signal with a locally generated replica
 H04L1/244—Testing correct operation by comparing a transmitted test signal with a locally generated replica test sequence generators

 H—ELECTRICITY
 H04—ELECTRIC COMMUNICATION TECHNIQUE
 H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
 H04L27/00—Modulatedcarrier systems
 H04L27/26—Systems using multifrequency codes
 H04L27/2601—Multicarrier modulation systems
Abstract
There is disclosed a signal having a predefined quality criterion for use with a communication systems, a method of and a system for generating such a signal, a method of testing the operation of a communication system using such a signal and a (tele)communication system arranged for operating such a method. The method for generating the signal having a predefined quality the steps of:  representing a first signal (10) comprising a plurality of frequency components each having spectral amplitude and phase properties, and processing the represented signal by arranging (11) its spectral amplitude properties, and  processing the represented signal by arranging (11) its spectral amplitude properties in accordance with the predefined quality criterion.
Description
Title A method of and a device for generating a signal having a predetermined quality criterion for use in or on a communication system.
Field of the Invention The present invention relates, generally, to communication systems and, more specifically, to a signal for use with a communication system, a method of and a system for generating such a signal, a method of testing the operation of a communication system using such a signal, a test system and a (tele)communication system arranged for operating such a method.
Background of the Invention Among others, for testing communication systems and communication equipment, such as xDSL transceivers and cables or networks, test signals are needed for stressing the communication system and the communication devices in a manner that is representative to actual deployment scenarios, with large numbers of systems or system devices per cable.
By measuring the transmission performance of the system or system device under realistic (noisy) test conditions, one can improve the design of the system or devices and/or prove that their performance is compliant with standards, such as issued by ETSI, ITU or ANSI or other (tele)communication bodies.
A method of executing such performance tests is to generate a signal which is known as impairment. More specifically, impairment can be subdivided into:
(r) crosstalk noise, having a noise profile characterized by a spectral envelope and spectral amplitude distribution e.g. from neighboring xDSL systems;
(ii) ingress noise, composed of discrete frequency components, also called rfitones, having a noise profile characterized by a number of discrete frequency CONFIRMATION C~PY
Field of the Invention The present invention relates, generally, to communication systems and, more specifically, to a signal for use with a communication system, a method of and a system for generating such a signal, a method of testing the operation of a communication system using such a signal, a test system and a (tele)communication system arranged for operating such a method.
Background of the Invention Among others, for testing communication systems and communication equipment, such as xDSL transceivers and cables or networks, test signals are needed for stressing the communication system and the communication devices in a manner that is representative to actual deployment scenarios, with large numbers of systems or system devices per cable.
By measuring the transmission performance of the system or system device under realistic (noisy) test conditions, one can improve the design of the system or devices and/or prove that their performance is compliant with standards, such as issued by ETSI, ITU or ANSI or other (tele)communication bodies.
A method of executing such performance tests is to generate a signal which is known as impairment. More specifically, impairment can be subdivided into:
(r) crosstalk noise, having a noise profile characterized by a spectral envelope and spectral amplitude distribution e.g. from neighboring xDSL systems;
(ii) ingress noise, composed of discrete frequency components, also called rfitones, having a noise profile characterized by a number of discrete frequency CONFIRMATION C~PY
components and spectral amplitude, modulation depth and modulation width parameters originating from radio and amateur broadcasting, for example, and (iii) impulse noise characterized by signal pulses caused by switching operations and components for example.
In the case of ingress noise, the frequency may vary (sweep) in time.
A device for generating impairment is known as an impairment generator and is arranged, in particular for use in or on communication systems, for generating at least one of said crosstalk noise and ingress noise.
In practice, for testing whether communication systems and communication devices are compliant with standards, various noise profiles have been defined which, among others, vary in accordance with system parameters such as the length and number of wire pairs in a communication cable and the transmission data rate, for example.
Further, each different type or length of a transmission medium such as a cable, a copper cable or an optical fiber or other cable type, request a different noise signal.
Methods and devices for generating noise profiles are known in the art. In particular, filtering techniques and filters are known for generating noise from an input signal providing an output signal having a particular spectral envelope and spectral amplitude distribution.
However, by using filtering techniques and filters, a causal relationship is established between the input signal and the output signal. Those skilled in the art will appreciate that such a type of signal is less suitable for a realistic imitation of real operational communication systems and communication devices.
WO 00/16181 discloses a method and a device for generating a random time domain signal approaching a predetermined histogram of amplitudes. In a first step, the signal is created by filtering a noise signal, such as a white noise signal, thereby producing a signal having a predetermined spectral envelope. In a next step, a nonlinear function is applied to the filtered noise signal, so as to produce the required time domain signal approaching the predetermined histogram of amplitudes. In a further step, pulse response filtering is applied to the time domain signal, to correct its spectral envelope and to obtain an output signal having a required spectral envelope. Both, the nonlinear function and the pulse response filtering function are special functions selected in accordance with the spectral envelope to be provided.
WO 00/16181 is limited in the sense that there is only provided a time domain signal only having a predetermined spectral envelope. WO 00/16181 is silent with respect to other quality criterion's to be imposed on the time domain signal to be provided, among others phase properties.
Summary of the Invention It is an object of the present invention to provide an improved signal for use with communication systems and communication devices, in particular for testing such systems and devices in accordance with predef ned (standardized) noise profiles.
In a first aspect of the present invention, there is disclosed a method of arranging a signal having a predefined quality criterion, preferably for use in or on a communication system, the method comprising the steps of  representing a first signal comprising a plurality of frequency components each having spectral amplitude and phase properties, and  processing the represented first signal by arranging the spectral amplitude properties in accordance with the or each predefined quality criterion, and arranging random phase properties.
The traditional way of modifying the envelope of a spectrum is the usage of a digital filter bank. This is far from ideal since, for the purpose of the present invention, no causal relationship between the represented first signal and the signal to be provided has to be established. This understanding of matters in accordance with the present invention simplifies the approach of frequency shaping significantly.
Starting from a first signal having random phase properties, frequency shaping of the represented first signal may be an adequate operation to provide a signal meeting the predefined quality criteria, for example. Frequency shaping in accordance with the present invention can be performed in several ways.
In an embodiment of the invention, the first signal is represented by a first set of numbers specifying a spectral amplitude and phase of each frequency component.
Scaling of the spectral amplitude of each frequency component suffices to effect frequency shaping of the represented signal in the frequency domain, while maintaining the random phase properties of the signal.
In a further embodiment of the invention, the first signal is represented by a second set of complex numbers having a real part and an imaginary part, which parts in combination specify a spectral amplitude and phase of each frequency component.
Frequency shaping is effected by adequately scaling the complex numbers, however such to maintain random phase properties after scaling of the represented first signal.
I0 In a yet further embodiment of the invention, the first signal is represented by a third set of numbers each specifying an amplitude of the first signal in the time domain.
By transforming this third set of numbers from the time domain into the frequency domain using, for example, a Fast Fourier Transform (FFT) algorithm, the first signal is represented by a fourth set of numbers specifying a spectral amplitude and phase of each 1 S frequency component. This fourth set of numbers can be further processed by a frequency shaping operation, as disclosed above in connection with the first set of numbers.
However, the third set of numbers may also be transformed, in accordance with the invention, from the time domain into the frequency domain for representing the first 20 signal by a fifth set of complex numbers having a real part and an imaginary part. As disclosed above, for the purpose of frequency shaping, the fifth set of complex numbers has to be adequately scaled.
In the case of a represented first signal having nonrandom phase properties, random phase properties can be approached by properly arranging the second, fourth 25 and fifth set of numbers.
Scaling in the frequency domain can be invoked by multiplication operations, using real or complex scaling factors. The scaling factor for multiplication of the spectral amplitude of a frequency component is found by dividing its desired value by its actual spectral amplitude.
30 In accordance with a further embodiment of the method according to the invention, in order to achieve a closer match to the or each predefined quality criterion, postprocessing of the processed represented first signal is provided.
For use in or on a communications system in accordance with the present invention, however, the represented first signal thus arranged in the frequency domain 5 has to be transformed into the time domain using, for example, an Inverse Fast Fourier Transform (IFFT) algorithm.
Further, the processing steps disclosed above may also a include operations such as convolution or deconvolution or multiplication or addon of signals.
In the time domain, the processed represented first signal meeting the or each predefined quality criterion may be represented among others by a sixth set of numbers in the time domain.
However, with the above approach the signal provided, meeting a quality criterion in the frequency domain, such as a predefined envelope of spectral amplitudes and random phase properties, may not yet meet a quality criterion in the time domain, such as a predefined time domain amplitude distribution.
In a yet further embodiment of the method according to the invention, the or each predefined quality criterion comprises any of a group including a predefined time domain amplitude distribution and a predefined envelope of spectral amplitudes.
Accordingly, in a further embodiment of the method according to the invention, the processed represented first signal is arranged in accordance with a predefined time domain amplitude distribution.
In a still further embodiment of the method according to the invention, the processed represented first signal is arranged in accordance with a predefined envelope of spectral amplitudes.
For providing a signal which accurately meets predefined quality criteria in both the frequency and time domain, according to the invention, at least one of the time domain amplitude distribution and the envelope of spectral amplitudes is approached by an iteration process. Amplitude and frequency shaping may be repeated as often as required until both shapes meet the requirements within reasonable accuracy.
In an embodiment of the invention, the iteration process comprises a comparison, after any iteration step, of any of the time domain amplitude distribution and envelope of spectral amplitudes of the processed represented first signal with a predefined time domain amplitude distribution and predefined envelope of spectral amplitudes.
It has been observed that there is no need to perform a full time domain characteristic check after frequency shaping to figure out whether the time domain characteristics are close enough to the requirements. A simple check of the crest factor requirement has proven to be adequate in practice to enable the decision whether to stop or to continue with the iteration. The crest factor of the signal is defined as the relation of the maximum or peak amplitude of the tones of the signal compared to the average ~or rms value of the tones of the signal.
The method according to the invention as disclosed above is, in particular, suitable for generating, among others, crosstalk noise.
If a signal having the characteristics of ingress noise is to be generated, in a second aspect of the method according to the invention, the or each predefined quality criterion comprises at least one modulated carrier, the or each modulated carrier including any of a group comprised of a carrier frequency, a carrier amplitude, a modulation depth, and a modulation width.
By shaping the represented first signal in accordance with a quality criterion ar quality criteria indicated above, a signal representing a particular type of ingress noise, having a particular time domain amplitude distribution, and a predefined envelope of spectral amplitudes can be easily and very efficiently provided.
In accordance with the method of the present invention, the signal meeting the or each predefined quality criterion can be provided by combining a plurality of signals processed as disclosed above.
For use of the signal in, for example, the testing of a communication network or a communication device, the processed represented signal has to be transformed from the frequency domain into the time domain using, among others, a FFT
algorithm, for example.
The invention further provides to combine the signals generated in accordance with the first and second aspect as disclosed above. However, also other signal components may be included.
In particular, in accordance with the method of the present invention, the signal having the or each predefined quality criterion is a noise signal.
In a third aspect of the invention, a method is disclosed of testing the operation of a communication system, which method comprises the steps of  generating a signal having a predetermined quality criterion in accordance with the method of the invention disclosed above, and  transferring the signal through the communication system under test.
The signals can be generated and stored using a set of instructions in a code format and executed in a predetermined order on a device. Such set of instructions can be software code compiled on a computer and stored in the computer or a network of computers or a floppy or CDROM or through the Internet. The software andlor signals produced can also be.stored on an Arbitrary Wave Form Generator (AWG) card and the AWG can be used to generate the signals or to reproduce stored signals from the memory. It is therefore possible to have a library of signals available stored on a data carrier that can be used in the execution or use of the method according to the invention.
The communication systems can be devices such as xDSL modems, or chips within or for such modems, or cables in the network, or networks for (tele)communication.
In a fourth aspect of the present invention a further method is disclosed of arranging a signal for use on or in a communication system. Preferably the signal is a noise signal. The signal may comprise crosstalk noise that is a random signal with predetermined properties in the frequency domain and in the time domain. The signal can fluthermore comprise rfitones that have a discrete frequency spectrum.
Also other signal components can be included in the signal.
The method comprising the steps of  representing a first signal in time domain having a time domain amplitude distribution, the signal having a spectral density in the frequency domain, thereby achieving a represented signal;
In the case of ingress noise, the frequency may vary (sweep) in time.
A device for generating impairment is known as an impairment generator and is arranged, in particular for use in or on communication systems, for generating at least one of said crosstalk noise and ingress noise.
In practice, for testing whether communication systems and communication devices are compliant with standards, various noise profiles have been defined which, among others, vary in accordance with system parameters such as the length and number of wire pairs in a communication cable and the transmission data rate, for example.
Further, each different type or length of a transmission medium such as a cable, a copper cable or an optical fiber or other cable type, request a different noise signal.
Methods and devices for generating noise profiles are known in the art. In particular, filtering techniques and filters are known for generating noise from an input signal providing an output signal having a particular spectral envelope and spectral amplitude distribution.
However, by using filtering techniques and filters, a causal relationship is established between the input signal and the output signal. Those skilled in the art will appreciate that such a type of signal is less suitable for a realistic imitation of real operational communication systems and communication devices.
WO 00/16181 discloses a method and a device for generating a random time domain signal approaching a predetermined histogram of amplitudes. In a first step, the signal is created by filtering a noise signal, such as a white noise signal, thereby producing a signal having a predetermined spectral envelope. In a next step, a nonlinear function is applied to the filtered noise signal, so as to produce the required time domain signal approaching the predetermined histogram of amplitudes. In a further step, pulse response filtering is applied to the time domain signal, to correct its spectral envelope and to obtain an output signal having a required spectral envelope. Both, the nonlinear function and the pulse response filtering function are special functions selected in accordance with the spectral envelope to be provided.
WO 00/16181 is limited in the sense that there is only provided a time domain signal only having a predetermined spectral envelope. WO 00/16181 is silent with respect to other quality criterion's to be imposed on the time domain signal to be provided, among others phase properties.
Summary of the Invention It is an object of the present invention to provide an improved signal for use with communication systems and communication devices, in particular for testing such systems and devices in accordance with predef ned (standardized) noise profiles.
In a first aspect of the present invention, there is disclosed a method of arranging a signal having a predefined quality criterion, preferably for use in or on a communication system, the method comprising the steps of  representing a first signal comprising a plurality of frequency components each having spectral amplitude and phase properties, and  processing the represented first signal by arranging the spectral amplitude properties in accordance with the or each predefined quality criterion, and arranging random phase properties.
The traditional way of modifying the envelope of a spectrum is the usage of a digital filter bank. This is far from ideal since, for the purpose of the present invention, no causal relationship between the represented first signal and the signal to be provided has to be established. This understanding of matters in accordance with the present invention simplifies the approach of frequency shaping significantly.
Starting from a first signal having random phase properties, frequency shaping of the represented first signal may be an adequate operation to provide a signal meeting the predefined quality criteria, for example. Frequency shaping in accordance with the present invention can be performed in several ways.
In an embodiment of the invention, the first signal is represented by a first set of numbers specifying a spectral amplitude and phase of each frequency component.
Scaling of the spectral amplitude of each frequency component suffices to effect frequency shaping of the represented signal in the frequency domain, while maintaining the random phase properties of the signal.
In a further embodiment of the invention, the first signal is represented by a second set of complex numbers having a real part and an imaginary part, which parts in combination specify a spectral amplitude and phase of each frequency component.
Frequency shaping is effected by adequately scaling the complex numbers, however such to maintain random phase properties after scaling of the represented first signal.
I0 In a yet further embodiment of the invention, the first signal is represented by a third set of numbers each specifying an amplitude of the first signal in the time domain.
By transforming this third set of numbers from the time domain into the frequency domain using, for example, a Fast Fourier Transform (FFT) algorithm, the first signal is represented by a fourth set of numbers specifying a spectral amplitude and phase of each 1 S frequency component. This fourth set of numbers can be further processed by a frequency shaping operation, as disclosed above in connection with the first set of numbers.
However, the third set of numbers may also be transformed, in accordance with the invention, from the time domain into the frequency domain for representing the first 20 signal by a fifth set of complex numbers having a real part and an imaginary part. As disclosed above, for the purpose of frequency shaping, the fifth set of complex numbers has to be adequately scaled.
In the case of a represented first signal having nonrandom phase properties, random phase properties can be approached by properly arranging the second, fourth 25 and fifth set of numbers.
Scaling in the frequency domain can be invoked by multiplication operations, using real or complex scaling factors. The scaling factor for multiplication of the spectral amplitude of a frequency component is found by dividing its desired value by its actual spectral amplitude.
30 In accordance with a further embodiment of the method according to the invention, in order to achieve a closer match to the or each predefined quality criterion, postprocessing of the processed represented first signal is provided.
For use in or on a communications system in accordance with the present invention, however, the represented first signal thus arranged in the frequency domain 5 has to be transformed into the time domain using, for example, an Inverse Fast Fourier Transform (IFFT) algorithm.
Further, the processing steps disclosed above may also a include operations such as convolution or deconvolution or multiplication or addon of signals.
In the time domain, the processed represented first signal meeting the or each predefined quality criterion may be represented among others by a sixth set of numbers in the time domain.
However, with the above approach the signal provided, meeting a quality criterion in the frequency domain, such as a predefined envelope of spectral amplitudes and random phase properties, may not yet meet a quality criterion in the time domain, such as a predefined time domain amplitude distribution.
In a yet further embodiment of the method according to the invention, the or each predefined quality criterion comprises any of a group including a predefined time domain amplitude distribution and a predefined envelope of spectral amplitudes.
Accordingly, in a further embodiment of the method according to the invention, the processed represented first signal is arranged in accordance with a predefined time domain amplitude distribution.
In a still further embodiment of the method according to the invention, the processed represented first signal is arranged in accordance with a predefined envelope of spectral amplitudes.
For providing a signal which accurately meets predefined quality criteria in both the frequency and time domain, according to the invention, at least one of the time domain amplitude distribution and the envelope of spectral amplitudes is approached by an iteration process. Amplitude and frequency shaping may be repeated as often as required until both shapes meet the requirements within reasonable accuracy.
In an embodiment of the invention, the iteration process comprises a comparison, after any iteration step, of any of the time domain amplitude distribution and envelope of spectral amplitudes of the processed represented first signal with a predefined time domain amplitude distribution and predefined envelope of spectral amplitudes.
It has been observed that there is no need to perform a full time domain characteristic check after frequency shaping to figure out whether the time domain characteristics are close enough to the requirements. A simple check of the crest factor requirement has proven to be adequate in practice to enable the decision whether to stop or to continue with the iteration. The crest factor of the signal is defined as the relation of the maximum or peak amplitude of the tones of the signal compared to the average ~or rms value of the tones of the signal.
The method according to the invention as disclosed above is, in particular, suitable for generating, among others, crosstalk noise.
If a signal having the characteristics of ingress noise is to be generated, in a second aspect of the method according to the invention, the or each predefined quality criterion comprises at least one modulated carrier, the or each modulated carrier including any of a group comprised of a carrier frequency, a carrier amplitude, a modulation depth, and a modulation width.
By shaping the represented first signal in accordance with a quality criterion ar quality criteria indicated above, a signal representing a particular type of ingress noise, having a particular time domain amplitude distribution, and a predefined envelope of spectral amplitudes can be easily and very efficiently provided.
In accordance with the method of the present invention, the signal meeting the or each predefined quality criterion can be provided by combining a plurality of signals processed as disclosed above.
For use of the signal in, for example, the testing of a communication network or a communication device, the processed represented signal has to be transformed from the frequency domain into the time domain using, among others, a FFT
algorithm, for example.
The invention further provides to combine the signals generated in accordance with the first and second aspect as disclosed above. However, also other signal components may be included.
In particular, in accordance with the method of the present invention, the signal having the or each predefined quality criterion is a noise signal.
In a third aspect of the invention, a method is disclosed of testing the operation of a communication system, which method comprises the steps of  generating a signal having a predetermined quality criterion in accordance with the method of the invention disclosed above, and  transferring the signal through the communication system under test.
The signals can be generated and stored using a set of instructions in a code format and executed in a predetermined order on a device. Such set of instructions can be software code compiled on a computer and stored in the computer or a network of computers or a floppy or CDROM or through the Internet. The software andlor signals produced can also be.stored on an Arbitrary Wave Form Generator (AWG) card and the AWG can be used to generate the signals or to reproduce stored signals from the memory. It is therefore possible to have a library of signals available stored on a data carrier that can be used in the execution or use of the method according to the invention.
The communication systems can be devices such as xDSL modems, or chips within or for such modems, or cables in the network, or networks for (tele)communication.
In a fourth aspect of the present invention a further method is disclosed of arranging a signal for use on or in a communication system. Preferably the signal is a noise signal. The signal may comprise crosstalk noise that is a random signal with predetermined properties in the frequency domain and in the time domain. The signal can fluthermore comprise rfitones that have a discrete frequency spectrum.
Also other signal components can be included in the signal.
The method comprising the steps of  representing a first signal in time domain having a time domain amplitude distribution, the signal having a spectral density in the frequency domain, thereby achieving a represented signal;
 processing the represented signal in accordance with a nonlinear transformation, the nonlinear transformation achieving at least one predefined quality criterion,  the time domain amplitude distribution of the represented signal being processed at least with an inverse function of a predetermined time domain amplitude distribution.
The method may further comprise the step of comparing the time domain amplitude distribution of the represented signal with the predetermined time domain amplitude distribution, and thereafter arranging the nonlinear transformation in order to achieve a processed represented signal having a time domain amplitude distribution approaching the predetermined time domain amplitude distribution.
In a fifth aspect of the present invention a method is disclosed of further comprising the step of comparing the time domain amplitude distribution of the represented signal with the predetermined time domain amplitude distribution, and thereafter arranging the nonlinear transformation in order to achieve a processed represented signal having a time domain amplitude distribution approaching the predetermined time domain amplitude distribution.
According to the fifth aspect of the invention the method can also comprise the steps of representing a f rst signal in time domain and with an amplitude distribution and the signal having a spectral density in the frequency domain, thereby achieving a represented signal, and filtering the represented signal in the frequency domain including the steps of evaluating at least part of the signal representation in the frequency domain and thereafter processing the represented signal in the frequency domain.
The methods of the fourth and fifth aspect of the invention can be combined.
The methods of the fourth and fifth aspect of the invention allow to make a signal in different iterative steps that has a predetermined amplitude distribution and/or that has a predetermined spectral density or that has a amplitude distribution and/or that has a spectral density according to a predefined quality criterion. The predefined quality criterion can be the crest factor of the signal, that is the relation of the maximum or peak value of tones of the signal compared to the average value or rmsvalue of the tones of the signal. The processing steps as recited hereabove can comprise the steps of a Fast Fourier Transformation (FFT) or Inverse Fast Fourier Transformation (IFFT).
The processing steps can also a include operations such as a convolution or deconvolution or multiplication or addon of signals.
In the method of the fourth aspect, the amplitude distribution of the represented signal is processed including a function of the predetermined amplitude distribution, which can include an inverse function of the predetermined amplitude distribution.
The method as recited of the fourth and fifth aspect of the invention can further comprise the steps of transforming the first signal in the frequency domain;
multiplying the first signal in the frequency domain with a spectral envelope thereby achieving a multiplied signal; and thereafter representing the multiplied signal in time domain.
In the methods, the first signal in its representation in the frequency domain can be generated as a set of random numbers, preferably complex numbers the modulus of the complex number characterizing amplitude, the argument of the complex number characterizing phase and the real and/or the imaginary part of essentially each of the complex numbers can be chosen according to a Gaussian distribution. Each of the complex numbers can be substantially equal to the amplitude of the predetermined spectral density.
In a sixth aspect of the present invention, a signal is disclosed comprising at least a random noise signal, the random signal having an amplitude distribution in the time domain according to a predetermined quality criterion and a spectral density in the frequency domain according to a predetermined quality criterion, the random signal being composed of an array of random numbers. The signal can further comprise a discrete frequency spectrum. The noise signal can be generated using a set of instructions in a code format and being executed in a predetermined order.
Such set of instructions can be software code compiled on a computer and stored in the computer or a network of computers or a floppy or CDROM or through the Internet. The software can also be stored on an Arbitrary Wave Form Generator (AWG) card and the AWG
can be used to generate the signals or reproduce stored signals from the memory, It is therefore possible to have a library of signals available that can be used in the execution or use of the methods of the fourth and fifth aspect of the invention.
In a seventh aspect of the present invention, a method is disclosed of generating a signal comprising at least a random noise signal, the random signal having an 5 amplitude distribution in the time domain according to a predefined quality criterion and having a spectral density in the frequency domain according to a predefined quality criterion, the random signal being composed of an array of random numbers, the method comprising the step of generating a random set of numbers using a set of instructions in a code format and being executed in a predetermined order. The method can further 10 comprise the step of generating a discrete frequency spectrum, the discrete frequency spectrum using goniometry functions and modulating essentially each of the discrete frequencies with a noise characteristic. The random noise signal and the discrete frequency spectrum can be combined using a set of instructions in a code format and being executed in a predetermined order.
In an eight aspect of the present invention, a set of instructions is disclosed in a code format and executable in a predetermined order, the set of instructions being arranged for generating a random noise signal and a discrete frequency spectrum, the random signal having an amplitude distribution in the time domain according to a predetermined quality criterion and having a spectral density in the frequency domain according to a predetermined quality criterion. Such set of instructions can be softwaxe code compiled on a computer and stored in the computer or a network of computers or a floppy or CDROM or through the Internet. The software can also be stored on an Arbitrary Wave Form Generator (AWG) card and the AWG can be used to generate the signals or reproduce stored signals from the memory. It is therefore possible to have a library of signals available that can be used. The software can be Ccode or can be compiled in a MATLAB environment.
In a ninth aspect of the present invention, a system for testing the operation of a communication system is disclosed comprising a set of instructions in a code format and executable in a predetermined order and compiled on a device, the set of instructions being arranged for generating a noise signal comprising at least one of a random noise signal and a discrete frequency spectrum, the random signal having an amplitude distribution in the time domain according to a predetermined quality criterion and having a spectral density in the frequency domain according to a predetermined quality criterion. The test system according to this aspect of the invention can comprise an impairment generator for generating the noise signal.
The connection elements (transformers, active devices, attenuators, etc.) that connect the impairment generator to the communication system that is tested can have an unwanted frequency dependent response. The unwanted frequency dependent response can be measured for instance by generating specific test signals in the impairment generator. The unwanted frequency dependent response can be compensated by multiplying the desired spectral density of the signal divided by the unwanted frequency dependent response of the connection element.
In a tenth aspect of the present invention, a method of testing the operation of a communication system such as a xDSL modem is disclosed. The method comprises the step of superposing on a signal transceived by a the modem, a signal comprising at least one of a random noise signal and a discrete frequency spectrum, the random signal having an amplitude distribution in the time domain according to a predefined quality criterion and having a spectral density in the frequency domain according to a predetermined quality criterion., the noise signal furthermore being composed of an array of random numbers.
In an eleventh aspect of the present invention a method of testing the quality of operation of a communication system is disclosed. The method comprises the steps of superposing on a signal transceived by a the modem, a signal comprising at least one of a random noise signal and a discrete frequency spectrum, the random signal having an amplitude distribution in the time domain according to a predetermined quality criterion and having a spectral density in the frequency domain according to a predefined quality criterion, the noise signal fizrthermore being composed of an array of random numbers, and evaluating the transceived signal according to a predefined quality criterion.
Yet in a twelfth aspect of the present invention, a method of improving the design and/or production of a communication system is disclosed, the method comprising the steps of superposing on a signal transceived by a the modern, superposing on a signal transceived by the modem, a signal comprising at least one of a random noise signal and a discrete frequency spectrum, the random signal having an amplitude distribution in the time domain according to a predetermined quality criterion and having a spectral density in the frequency domain according to a predetermined quality criterion, the noise signal furthermore being composed of an array of random numbers; evaluating the transceived signal according to a predetermined quality criterion; and iteratively arranging the design of the modem in order to approach closer to the quality criterion for evaluating the transceived signal.
I0 In a thirteenth aspect of the present invention, a telecommunication network is disclosed including a signal comprising at least one of a random noise signal and a discrete frequency spectrum, the random signal having an amplitude distribution in the time domain according to a predetermined quality criterion and having a spectral density in the frequency domain according to a predetermined quality criterion, the noise signal furthermore being composed of an array of random numbers.
The features of the abovedescribed aspects and embodiments of the invention can be combined.
The signal, the methods and the set of instructions recited hereabove will allow to have a better quality of signal transmission over media such as telephone cables or wireless media. A better transmission of signals allows for a broader providing of more services for the users of communication systems.
Brief Description of the Figures Figure 1 shows, in a block diagram, a setup for a performance test in a communication system, using an impairment generator operating in accordance with the method of the present invention.
Figure 2 shows a flow diagram type of embodiments of the method according to the invention.
Figure 3' shows, in a graphic representation, an embodiment of the method according to the invention for generating ingress noise.
Figure 4 shows a flow diagram type of further embodiments of a method according to the invention.
Figure 5 shows an amplitude distortion (nonlinear transformation) function Q(x) S that amplifies the high amplitude peaks or tones in a signal.
Figure 6 shows in a flow diagram an example embodiment of the invention.
Figures 711 show results that are obtained according to an embodiment of the invention.
Detailed Description of the Embodiments For the purpose of teaching the invention, aspects and embodiments of the signal and method and systems of the invention are described below. It will be appreciated by those skilled in the art that other alternative and equivalent embodiments of the invention can be conceived and reduced to practice without departing form the true spirit of the invention. The scope of the invention being limited only by the appended claims.
In an embodiment of the invention, a system for testing the operation of a communication system such as a xDSL transceiver is disclosed. The set up of a test equipment for a high penetration of systems scenario in operational access networks is described.
A method is disclosed of arranging a signal for use on or in a communication system.
The purpose of transmission performance tests is to stress xDSL transceivers in a way that is representative to a high penetration of systems scenario in operational access networks. This high penetration approach enables:
(i) component and system designers to quantify the performance and to use it to improve their design and to prove compliance with standards; and (ii) operators to define deployment rules that apply to most operational situations.
The method may further comprise the step of comparing the time domain amplitude distribution of the represented signal with the predetermined time domain amplitude distribution, and thereafter arranging the nonlinear transformation in order to achieve a processed represented signal having a time domain amplitude distribution approaching the predetermined time domain amplitude distribution.
In a fifth aspect of the present invention a method is disclosed of further comprising the step of comparing the time domain amplitude distribution of the represented signal with the predetermined time domain amplitude distribution, and thereafter arranging the nonlinear transformation in order to achieve a processed represented signal having a time domain amplitude distribution approaching the predetermined time domain amplitude distribution.
According to the fifth aspect of the invention the method can also comprise the steps of representing a f rst signal in time domain and with an amplitude distribution and the signal having a spectral density in the frequency domain, thereby achieving a represented signal, and filtering the represented signal in the frequency domain including the steps of evaluating at least part of the signal representation in the frequency domain and thereafter processing the represented signal in the frequency domain.
The methods of the fourth and fifth aspect of the invention can be combined.
The methods of the fourth and fifth aspect of the invention allow to make a signal in different iterative steps that has a predetermined amplitude distribution and/or that has a predetermined spectral density or that has a amplitude distribution and/or that has a spectral density according to a predefined quality criterion. The predefined quality criterion can be the crest factor of the signal, that is the relation of the maximum or peak value of tones of the signal compared to the average value or rmsvalue of the tones of the signal. The processing steps as recited hereabove can comprise the steps of a Fast Fourier Transformation (FFT) or Inverse Fast Fourier Transformation (IFFT).
The processing steps can also a include operations such as a convolution or deconvolution or multiplication or addon of signals.
In the method of the fourth aspect, the amplitude distribution of the represented signal is processed including a function of the predetermined amplitude distribution, which can include an inverse function of the predetermined amplitude distribution.
The method as recited of the fourth and fifth aspect of the invention can further comprise the steps of transforming the first signal in the frequency domain;
multiplying the first signal in the frequency domain with a spectral envelope thereby achieving a multiplied signal; and thereafter representing the multiplied signal in time domain.
In the methods, the first signal in its representation in the frequency domain can be generated as a set of random numbers, preferably complex numbers the modulus of the complex number characterizing amplitude, the argument of the complex number characterizing phase and the real and/or the imaginary part of essentially each of the complex numbers can be chosen according to a Gaussian distribution. Each of the complex numbers can be substantially equal to the amplitude of the predetermined spectral density.
In a sixth aspect of the present invention, a signal is disclosed comprising at least a random noise signal, the random signal having an amplitude distribution in the time domain according to a predetermined quality criterion and a spectral density in the frequency domain according to a predetermined quality criterion, the random signal being composed of an array of random numbers. The signal can further comprise a discrete frequency spectrum. The noise signal can be generated using a set of instructions in a code format and being executed in a predetermined order.
Such set of instructions can be software code compiled on a computer and stored in the computer or a network of computers or a floppy or CDROM or through the Internet. The software can also be stored on an Arbitrary Wave Form Generator (AWG) card and the AWG
can be used to generate the signals or reproduce stored signals from the memory, It is therefore possible to have a library of signals available that can be used in the execution or use of the methods of the fourth and fifth aspect of the invention.
In a seventh aspect of the present invention, a method is disclosed of generating a signal comprising at least a random noise signal, the random signal having an 5 amplitude distribution in the time domain according to a predefined quality criterion and having a spectral density in the frequency domain according to a predefined quality criterion, the random signal being composed of an array of random numbers, the method comprising the step of generating a random set of numbers using a set of instructions in a code format and being executed in a predetermined order. The method can further 10 comprise the step of generating a discrete frequency spectrum, the discrete frequency spectrum using goniometry functions and modulating essentially each of the discrete frequencies with a noise characteristic. The random noise signal and the discrete frequency spectrum can be combined using a set of instructions in a code format and being executed in a predetermined order.
In an eight aspect of the present invention, a set of instructions is disclosed in a code format and executable in a predetermined order, the set of instructions being arranged for generating a random noise signal and a discrete frequency spectrum, the random signal having an amplitude distribution in the time domain according to a predetermined quality criterion and having a spectral density in the frequency domain according to a predetermined quality criterion. Such set of instructions can be softwaxe code compiled on a computer and stored in the computer or a network of computers or a floppy or CDROM or through the Internet. The software can also be stored on an Arbitrary Wave Form Generator (AWG) card and the AWG can be used to generate the signals or reproduce stored signals from the memory. It is therefore possible to have a library of signals available that can be used. The software can be Ccode or can be compiled in a MATLAB environment.
In a ninth aspect of the present invention, a system for testing the operation of a communication system is disclosed comprising a set of instructions in a code format and executable in a predetermined order and compiled on a device, the set of instructions being arranged for generating a noise signal comprising at least one of a random noise signal and a discrete frequency spectrum, the random signal having an amplitude distribution in the time domain according to a predetermined quality criterion and having a spectral density in the frequency domain according to a predetermined quality criterion. The test system according to this aspect of the invention can comprise an impairment generator for generating the noise signal.
The connection elements (transformers, active devices, attenuators, etc.) that connect the impairment generator to the communication system that is tested can have an unwanted frequency dependent response. The unwanted frequency dependent response can be measured for instance by generating specific test signals in the impairment generator. The unwanted frequency dependent response can be compensated by multiplying the desired spectral density of the signal divided by the unwanted frequency dependent response of the connection element.
In a tenth aspect of the present invention, a method of testing the operation of a communication system such as a xDSL modem is disclosed. The method comprises the step of superposing on a signal transceived by a the modem, a signal comprising at least one of a random noise signal and a discrete frequency spectrum, the random signal having an amplitude distribution in the time domain according to a predefined quality criterion and having a spectral density in the frequency domain according to a predetermined quality criterion., the noise signal furthermore being composed of an array of random numbers.
In an eleventh aspect of the present invention a method of testing the quality of operation of a communication system is disclosed. The method comprises the steps of superposing on a signal transceived by a the modem, a signal comprising at least one of a random noise signal and a discrete frequency spectrum, the random signal having an amplitude distribution in the time domain according to a predetermined quality criterion and having a spectral density in the frequency domain according to a predefined quality criterion, the noise signal fizrthermore being composed of an array of random numbers, and evaluating the transceived signal according to a predefined quality criterion.
Yet in a twelfth aspect of the present invention, a method of improving the design and/or production of a communication system is disclosed, the method comprising the steps of superposing on a signal transceived by a the modern, superposing on a signal transceived by the modem, a signal comprising at least one of a random noise signal and a discrete frequency spectrum, the random signal having an amplitude distribution in the time domain according to a predetermined quality criterion and having a spectral density in the frequency domain according to a predetermined quality criterion, the noise signal furthermore being composed of an array of random numbers; evaluating the transceived signal according to a predetermined quality criterion; and iteratively arranging the design of the modem in order to approach closer to the quality criterion for evaluating the transceived signal.
I0 In a thirteenth aspect of the present invention, a telecommunication network is disclosed including a signal comprising at least one of a random noise signal and a discrete frequency spectrum, the random signal having an amplitude distribution in the time domain according to a predetermined quality criterion and having a spectral density in the frequency domain according to a predetermined quality criterion, the noise signal furthermore being composed of an array of random numbers.
The features of the abovedescribed aspects and embodiments of the invention can be combined.
The signal, the methods and the set of instructions recited hereabove will allow to have a better quality of signal transmission over media such as telephone cables or wireless media. A better transmission of signals allows for a broader providing of more services for the users of communication systems.
Brief Description of the Figures Figure 1 shows, in a block diagram, a setup for a performance test in a communication system, using an impairment generator operating in accordance with the method of the present invention.
Figure 2 shows a flow diagram type of embodiments of the method according to the invention.
Figure 3' shows, in a graphic representation, an embodiment of the method according to the invention for generating ingress noise.
Figure 4 shows a flow diagram type of further embodiments of a method according to the invention.
Figure 5 shows an amplitude distortion (nonlinear transformation) function Q(x) S that amplifies the high amplitude peaks or tones in a signal.
Figure 6 shows in a flow diagram an example embodiment of the invention.
Figures 711 show results that are obtained according to an embodiment of the invention.
Detailed Description of the Embodiments For the purpose of teaching the invention, aspects and embodiments of the signal and method and systems of the invention are described below. It will be appreciated by those skilled in the art that other alternative and equivalent embodiments of the invention can be conceived and reduced to practice without departing form the true spirit of the invention. The scope of the invention being limited only by the appended claims.
In an embodiment of the invention, a system for testing the operation of a communication system such as a xDSL transceiver is disclosed. The set up of a test equipment for a high penetration of systems scenario in operational access networks is described.
A method is disclosed of arranging a signal for use on or in a communication system.
The purpose of transmission performance tests is to stress xDSL transceivers in a way that is representative to a high penetration of systems scenario in operational access networks. This high penetration approach enables:
(i) component and system designers to quantify the performance and to use it to improve their design and to prove compliance with standards; and (ii) operators to define deployment rules that apply to most operational situations.
Figure 1 illustrates the functional description of a possible test setup 1.
It includes:
~ a test loop 2, being a real cable or a cable simulator;
~ an adding element 3 to inject impairment noise into the test loop 2;
S ~ a high impedance, and well balanced differential voltage probe 4, connected with level detectors S such as a spectrum analyser or an rms volt meter, for example, (not shown), and ~ xDSL transceivers (modems) 6, 7 under test.
When external splatters are required for the xDSL system under test (for POTS
or ISDN
signals), these splatter can be included in the modems 6, 7 under test.
The signal flow through the test equipment setup 1 is from port Tx to port Rx, which means that measuring upstream and downstream performance requires an interchange of transceiver position and test "cable" ends. The received signal level at port Rx is the level, measured between node A2 and B2, when port Tx as well as port 1 S Rx are terminated with the xDSL transceivers (modems) 6, 7 under test. The impairment generator 8 is switched off during this measurement. The transmitted signal level at port Tx is the level, measured between node Al and B 1, under the same conditions.
The noise that the impairment generator 8 should inject into the test setup 1 is frequency dependent. The noise which the impairment generator 8 injects into the test setup 1 should be a realistic representation of a real (spectral polluted) access network, and is;
(a) dependent on the length of the test loop 2, and (b) different for downstream performance tests and upstream performance tests.
This impairment noise, measured between node A2 and B2, is usually a mix of random, 2S impulsive and harmonic noise (the rfitones). A set of characteristics is identified as a "noise profile".
The signal and noise levels are probed with a well balanced differential voltage probe 4.
In a fully automated test setup 1 the test loop 2, 3 the voltage probe 4 and level detector S, the modems under test 6, 7 and the impairment generator 8 may connect to a Central Processing Unit (CPU) 9, as schematically indicated with broken lines.
Those skilled in the art will appreciate that the connections with the CPU 9 may involve data links for remote testing by the CPU 9.
Definitions that are relevant for the use of the test equipment are the following 5 ~ Probing an rmsvoltage U,.",S [V] in this setup, over the full signal band, means a power level of P [dBm] that equals: P = 10 x logo ( U""SZ/ Rw x 1000) [dBm];
~ Probing an rmsvoltage U~",S [V] in this setup, within a small frequency band of 4f (in Hertz), means a power spectral density level of P [dBm/Hz] within that filtered band that equals: P =10 x logo ( U~,52 / R~ x 1000 / ~f) [dBm/Hz];
10 ~ The bandwidth ~f identifies the noise bandwidth of the filter, and not the 3dB
bandwidth.
Figure 2 shows schematically embodiments of the method for arranging a signal Ui(t) i=1,2,3,... for use on or in a communication system in accordance with the invention. The signal may comprise crosstalk noise, that is a random signal with 15 predetermined properties in the frequency domain and in the time domain.
As represented by flow [ 1 ] of Figure 2, the method comprises the steps of representing a first signal comprising a plurality of frequency components each having spectral amplitude and phase properties, block 10 "First signal", and processing the represented signal by arranging the spectral amplitude properties in accordance with at least one predefined quality criterion, as well as arranging random phase properties, block 11 "Arrange", thereby achieving a processed represented signal.
The first signal may be represented by a first set of numbers specifying a spectral amplitude and phase of each frequency component. Further, the first signal may be represented by a second set of complex numbers, having a real part and an imaginary part, which parts in combination specify a spectral amplitude and phase of each frequency component. That is, the modulus of a complex number characterises the spectral amplitude whereas the argument of the complex number characterises the phase of the frequency component.
In accordance with the present invention, the represented first signal 10 is processed to arrange random phase properties. However, starting from a represented first signal 10 having random phase properties, for shaping the frequency of the signal in accordance with the predefined quality criterion, it suffices to shape the spectral amplitude of the frequency components.
The method may also comprise the steps of representing the first signal in the time domain, in that the first signal is represented by a third set of numbers each specifying an amplitude of the first signal in the time domain. By transforming the third set of numbers from the time domain into the frequency domain, for example using an FFT algorithm, a fourth set of numbers is achieved specifying a spectral amplitude and phase of each frequency component. Likewise, the fourth set of numbers is to be processed by arranging the spectral amplitude properties in accordance with the at least one predefined quality criterion, as well as arranging its random phase properties.
By transforming the processed represented signal from the frequency domain into the time domain, for example using an Inverse FFT algorithm (IFFT), block "Transform", the signal U 1 (t) having the at least one predefined quality criterion is eventually generated.
While the signal U 1 (t) meets at least one predefined quality criterion in the frequency domain, such as a spectral envelope and/or preemphasise properties, it may yet be required to provide a signal having a predefined quality criterion in the time domain.
As disclosed in flows [2] and [3] of Figure 2, i.e. block 13 "Amplitude Shape"
and block 14 "Frequency Shape", the quality criterion in the time domain may comprise a predefined amplitude distribution and/or a predefined envelope of spectral amplitudes.
The method can also comprise the step of making a signal in different iterative steps, see Figure 2 [4]. Block 15 "Test Shape" and back coupling loop 16. Thus the signal can have a predetermined time dpmain amplitude distribution and/or a predetermined envelope of spectral amplitudes and/or a spectral density according to predetermined quality criterion's.
The at least one predetermined quality criterion can be the crest factor of the signal that is a relation of the maximum or peak value of the tones of the signal compared to the average or rms value of the tones of the signal.
It includes:
~ a test loop 2, being a real cable or a cable simulator;
~ an adding element 3 to inject impairment noise into the test loop 2;
S ~ a high impedance, and well balanced differential voltage probe 4, connected with level detectors S such as a spectrum analyser or an rms volt meter, for example, (not shown), and ~ xDSL transceivers (modems) 6, 7 under test.
When external splatters are required for the xDSL system under test (for POTS
or ISDN
signals), these splatter can be included in the modems 6, 7 under test.
The signal flow through the test equipment setup 1 is from port Tx to port Rx, which means that measuring upstream and downstream performance requires an interchange of transceiver position and test "cable" ends. The received signal level at port Rx is the level, measured between node A2 and B2, when port Tx as well as port 1 S Rx are terminated with the xDSL transceivers (modems) 6, 7 under test. The impairment generator 8 is switched off during this measurement. The transmitted signal level at port Tx is the level, measured between node Al and B 1, under the same conditions.
The noise that the impairment generator 8 should inject into the test setup 1 is frequency dependent. The noise which the impairment generator 8 injects into the test setup 1 should be a realistic representation of a real (spectral polluted) access network, and is;
(a) dependent on the length of the test loop 2, and (b) different for downstream performance tests and upstream performance tests.
This impairment noise, measured between node A2 and B2, is usually a mix of random, 2S impulsive and harmonic noise (the rfitones). A set of characteristics is identified as a "noise profile".
The signal and noise levels are probed with a well balanced differential voltage probe 4.
In a fully automated test setup 1 the test loop 2, 3 the voltage probe 4 and level detector S, the modems under test 6, 7 and the impairment generator 8 may connect to a Central Processing Unit (CPU) 9, as schematically indicated with broken lines.
Those skilled in the art will appreciate that the connections with the CPU 9 may involve data links for remote testing by the CPU 9.
Definitions that are relevant for the use of the test equipment are the following 5 ~ Probing an rmsvoltage U,.",S [V] in this setup, over the full signal band, means a power level of P [dBm] that equals: P = 10 x logo ( U""SZ/ Rw x 1000) [dBm];
~ Probing an rmsvoltage U~",S [V] in this setup, within a small frequency band of 4f (in Hertz), means a power spectral density level of P [dBm/Hz] within that filtered band that equals: P =10 x logo ( U~,52 / R~ x 1000 / ~f) [dBm/Hz];
10 ~ The bandwidth ~f identifies the noise bandwidth of the filter, and not the 3dB
bandwidth.
Figure 2 shows schematically embodiments of the method for arranging a signal Ui(t) i=1,2,3,... for use on or in a communication system in accordance with the invention. The signal may comprise crosstalk noise, that is a random signal with 15 predetermined properties in the frequency domain and in the time domain.
As represented by flow [ 1 ] of Figure 2, the method comprises the steps of representing a first signal comprising a plurality of frequency components each having spectral amplitude and phase properties, block 10 "First signal", and processing the represented signal by arranging the spectral amplitude properties in accordance with at least one predefined quality criterion, as well as arranging random phase properties, block 11 "Arrange", thereby achieving a processed represented signal.
The first signal may be represented by a first set of numbers specifying a spectral amplitude and phase of each frequency component. Further, the first signal may be represented by a second set of complex numbers, having a real part and an imaginary part, which parts in combination specify a spectral amplitude and phase of each frequency component. That is, the modulus of a complex number characterises the spectral amplitude whereas the argument of the complex number characterises the phase of the frequency component.
In accordance with the present invention, the represented first signal 10 is processed to arrange random phase properties. However, starting from a represented first signal 10 having random phase properties, for shaping the frequency of the signal in accordance with the predefined quality criterion, it suffices to shape the spectral amplitude of the frequency components.
The method may also comprise the steps of representing the first signal in the time domain, in that the first signal is represented by a third set of numbers each specifying an amplitude of the first signal in the time domain. By transforming the third set of numbers from the time domain into the frequency domain, for example using an FFT algorithm, a fourth set of numbers is achieved specifying a spectral amplitude and phase of each frequency component. Likewise, the fourth set of numbers is to be processed by arranging the spectral amplitude properties in accordance with the at least one predefined quality criterion, as well as arranging its random phase properties.
By transforming the processed represented signal from the frequency domain into the time domain, for example using an Inverse FFT algorithm (IFFT), block "Transform", the signal U 1 (t) having the at least one predefined quality criterion is eventually generated.
While the signal U 1 (t) meets at least one predefined quality criterion in the frequency domain, such as a spectral envelope and/or preemphasise properties, it may yet be required to provide a signal having a predefined quality criterion in the time domain.
As disclosed in flows [2] and [3] of Figure 2, i.e. block 13 "Amplitude Shape"
and block 14 "Frequency Shape", the quality criterion in the time domain may comprise a predefined amplitude distribution and/or a predefined envelope of spectral amplitudes.
The method can also comprise the step of making a signal in different iterative steps, see Figure 2 [4]. Block 15 "Test Shape" and back coupling loop 16. Thus the signal can have a predetermined time dpmain amplitude distribution and/or a predetermined envelope of spectral amplitudes and/or a spectral density according to predetermined quality criterion's.
The at least one predetermined quality criterion can be the crest factor of the signal that is a relation of the maximum or peak value of the tones of the signal compared to the average or rms value of the tones of the signal.
The signals can be generated and stored using a set of instructions in a code format and executable in a predetermined order and compiled on a device. Such set of instructions can be software code compiled on a computer and stored in the computer or a network of computers or a floppy or CDROM or through the Internet. The software can also be stored on an Arbitrary Wave Form Generator (AWG) card and the AWG
can be used to generate the signals or reproduce stored signals from the memory.
It is therefore possible to have a library of signals available that can be used in the execution or use of the methods of the first and second aspect of the invention. The communication systems can be devices such as xDSL modems 6, 7, or chips within or for such modems 6, 7, or networks for telecommunication.
The processing in block 13 "Amplitude Shape" in Figure 2 is done for achieving an impact or control on the time domain characteristics. An amplitude distortion (transformation) function Q(x) is chosen that amplifies the high amplitude peaks or tones in the signal. A nonlinear transformation function Q(x) can be reconstructed from the actual amplitude distribution function of the signal and the predetermined amplitude distribution function.
For a noise signal f(t) in the time period t in between 0 en T, the amplitude distribution F(a) of the signal is defined as a fraction of the time that the noise f in absolute value is larger than a. If G(a) is the predetermined amplitude distribution (such as an enhancedGaussian, see below), and G1 (a) is the inverse function thereof, the transformation function Q(x) to make an intermediate or final signal g(t) from the noise ' signal f(t) can be defined as:
Q(x) = sig~(x) ~ Gt ~F(~xI)) (1) g(t) = Q~.f (t)~ ~ (2) sign(x)=xl~x~ fox x~0; sign(x)=0 for x=0;
As a result g(t) will have the predetermined amplitude distribution G(a). Q(x) in a number of cases can be an analytical function but can also be numerically constructed.
can be used to generate the signals or reproduce stored signals from the memory.
It is therefore possible to have a library of signals available that can be used in the execution or use of the methods of the first and second aspect of the invention. The communication systems can be devices such as xDSL modems 6, 7, or chips within or for such modems 6, 7, or networks for telecommunication.
The processing in block 13 "Amplitude Shape" in Figure 2 is done for achieving an impact or control on the time domain characteristics. An amplitude distortion (transformation) function Q(x) is chosen that amplifies the high amplitude peaks or tones in the signal. A nonlinear transformation function Q(x) can be reconstructed from the actual amplitude distribution function of the signal and the predetermined amplitude distribution function.
For a noise signal f(t) in the time period t in between 0 en T, the amplitude distribution F(a) of the signal is defined as a fraction of the time that the noise f in absolute value is larger than a. If G(a) is the predetermined amplitude distribution (such as an enhancedGaussian, see below), and G1 (a) is the inverse function thereof, the transformation function Q(x) to make an intermediate or final signal g(t) from the noise ' signal f(t) can be defined as:
Q(x) = sig~(x) ~ Gt ~F(~xI)) (1) g(t) = Q~.f (t)~ ~ (2) sign(x)=xl~x~ fox x~0; sign(x)=0 for x=0;
As a result g(t) will have the predetermined amplitude distribution G(a). Q(x) in a number of cases can be an analytical function but can also be numerically constructed.
An example of an enhanced Gaussian function is as follows.
The amplitude distribution of Gaussian type noise is:
G(x) =1e~f x (3) with : erf (x) _ ~ f dt exp. tz ), (4) and with 6 being the RMS value of the signal.
The "enhanced" Gaussian distribution is defined as:
G x  1 a~~+erf x a+erf A 0Sx<A
( ) ~a. ~a~ ~ (5) 0 x>A
If VAS is the desired RMS value of the noise sample, and Cf being the desired crest factor, choose:
A = Cf x V,~s , and (6) 6=~((1+a) VZ,~s  AZ~a/3). ('7) Typical values for a that have proven useful are of a magnitude between 0.001 and 0.01, and this represents the deviation of enhanced Gaussian distributed from a true Gaussian distribution.
In block 14, Frequency Shape, of Figure 2, the frequency domain characteristics of the signal are improved, as a posedprocessing step to achieve a closer match to the or each quality criterion. The corrected frequency curve can be achieved, for example, by comparing (dividing) a predetermined spectral density through the measured spectral density of the (intermediate) signal U2(t). An example hereof is given in the best mode embodiment described in the sequel with a convolution of FFT functions.
In the part [4] of Figure 2, it is shown how an iterative procedure of the steps detailed here above may lead to a further improvement of the finish or final signal for use in or on a communication system. The iterative procedure, i.e. testing of the frequency shape by block I5, Test Shape, and back coupling loop 16, is executed until the predetermined quality criterions) are achieved.
Figure 3 illustrates in a schematic, graphic representation use of the method according to the invention for the provision of an ingress noise signal.
As disclosed in the preamble, ingress noise may be characterised by a plurality of frequency components at discrete carrier frequencies fci, i=1,2,3, .... The frequency components at the carrier frequency fci each having a carrier amplitude Aci, i=1,2,3, ..., and, if applicable, having a modulation width, i.e. a number of discrete frequencies at the left and right hand side of the associated carrier frequency fci, as well as having a modulation depth, that is the amplitude of the side frequencies associated with the respective carrier frequency fci.
Figure 3 shows, in a graphic representation having a horizontal or frequency axis f and a vertical or amplitude axis A, by way of example only, a signal comprised of two carrier frequencies fcl and fc2, having a carrier amplitude Acl and Ac2, respectively.
Around the carrier frequency fcl at each side thereof three side frequency components are arranged, each having an amplitude A1. For the frequency component at carrier frequency fc2 on each side thereof two side frequency components are arranged, each having an amplitude A2.
In accordance with the present invention, for providing a signal having at least one predefined quality criterion, the amplitude of the frequency components have to be shaped, such as disclosed by the dotted lines I and 2 in Figure 3.
Starting from a represented first signal having random phase properties, in accordance with the present invention, by the shaping of the amplitudes, the random phase properties are maintained in the signal to be provided having the predefined quality criterion.
Figure 4 shows schematically a further embodiment of a method for arranging a signal for use in or on a communication system, in particular for use if random phase properties are already provided for. The signal comprises crosstalk noise that is a random signal with predetermined properties in the frequency domain and in the time domain. The signal can furthermore comprise rfitones that have a discrete frequency spectrum. Also other signal components can be included in the signal.
The method further may comprise the steps of representing a first signal in time 5 domain and with an amplitude distribution and the signal having a spectral density in the frequency domain, thereby achieving a represented signal, and processing the represented signal according to a nonlinear transformation, the nonlinear transformation achieving a predetermined quality criterion. This is shown as amplitude shaping in Figure 4, flows [24].
10 The method further may comprise the step of representing a first signal in time domain and with an amplitude distribution and the signal having a spectral density in the frequency domain, thereby achieving a represented signal, and processing the represented signal until a signal is achieved having a spectral density according to a predetermined spectral density quality criterion. This is shown as frequency shaping in 15 Figure 4, flows [24]. The frequency shaping step can also comprise the step of filtering the represented signal in the frequency domain including the steps of evaluating at least part of the signal representation in the frequency domain and thereafter processing the signal representation in the frequency domain.
The method may also comprise the step of making a signal in different iterative 20 steps, see Figure 4 flow [4]. Thus the signal can have a predetermined time domain amplitude distribution and/or a predetermined spectral density or a time domain amplitude distribution and/or a spectral density according to a predefined quality criterion. The predefined quality criterion can be the crest factor of the signal, that is a relation of the maximum or peak value of the tones of the signal compared to the average value of the tones of the signal. The signals can be generated and stored using a set of instructions in a code format and executable in a predetermined order and compiled on a device.
Likewise, the set of instructions can be software code compiled on a computer and stored in the computer or a network of computers or a floppy or CDROM or through the Internet. The software can also be stored on ari Arbitrary Wave Form Generator (AWG) card and the AWG can be used to generate the signals or reproduce stored signals from the memory. It is therefore possible tv have a library of signals available that can be used in the execution or use of the methods of the invention. The communication systems can be devices such as xDSL modems, or chips within or for such modems, or networks for telecommunication. In detail the following embodiment is shown in Figure 4.
Using software, random numbers are generated, block 16, "Create Noise". In hardware white noise can be generated. The random numbers are filtered until a predetermined spectral density is achieved. The random numbers that are generated each represent a frequency component. The necessary processing to achieve a predetermined spectral density is executed by scaling the amplitude of the complex numbers and thereafter an IFFT processing is done in order to make the desired noise signal. Another way of executing the method is to generate random numbers that represent the phase of each frequency component and thereafter the amplitude of the complex numbers is arranged to approach or be equal to a predetermined spectral density The processing in block 13, "Amplitude Shape", is done for achieving an impact or control on the time domain characteristic. An amplitude distortion (transformation) function Q(x) is chosen that amplifies the high amplitude peaks or tones in the signal is shown in Figure 5. The nonlinear transformation function Q(x) can be reconstructed from the actual amplitude distribution function of the signal and the predetermined amplitude distribution function, as disclosed above in connection with the equations (17).
In block 14, "Frequency Shape" of Figure 4, the frequency domain characteristics of the signal are improved. The corrected frequency curve is achieved by comparing (dividing) a predetermined spectral density through the measured spectral density of the (intermediate) signal. An example hereof is given in the best mode embodiment described in the sequel with a convolution of FFT functions.
In the flow [4] of Figure 4, like in the flow [4] of Figure 2, again it is shown how an iterative procedure of the steps detailed hereabove may lead to a further improvement of the final for use in or on a communication system. The iterative procedure is executed until the predetermined quality criterions are achieved.
With the method according to the invention, as disclosed above, signals representing crosstalk noise and ingress noise can be generated with a device, such as an impairment generator 8, see Figure 1, which may be arranged for providing a signal comprised of both a crosstalk noise signal and an ingress noise signal, while other signal components may be added to the output signal to be provided, if required.
The signal to be provided, in an embodiment of the invention, can be advantageously provided as a sixth set of numbers in the time domain, for example an array of numbers.
Figure 6 shows, in a flow type diagram, an example embodiment of the invention, running on a Personal Computer 20. The impairment noise is generated by block 21, called SPOCS, comprising a block 22, "White noise signal", a block "Spectral shaping", provided by FFT, a block 24, "Desired noise signal", produced from the output of block 23 by IFFT, the resulting signal of which is stored on an AWG card 25. In the crosstalk scenario, i.e. block 26, a noise PSD is created, block 27, which is further processed by block 23.
A best mode embodiment of the set of instructions of the invention is disclosed here below. The code given here below is compiled in a MATLAB environment.
Comments related to the functionality of the code are given after the % signs.
For a person skilled in the art, the code provided is self explanatory.
Figures 79 show results obtained with the best mode embodiment. Figure 8 shows a plot of the spectrum of the generated noise sample plus the PSD of the noise profile. Figure 9 shows a plot of the generated noise sample in the time domain. Figure 10 shows a plot of the distribution function of the generated noise sample.
Figure 11 shows a plot of the cumulative distribution functiopn of the generated noise sample.
Figure 7 shows a graphical User Interface (UI) and settings of the AWG
control.
%______________________________________________________________________________ ______________________ function DemoImpair2;
%______________________________________________________________________________ ______________________ DemoImpair2 S % Code, programmed in the Matlab programming language, that demonstrates the basic algorithms of an Impairment Generator.
The demonstrated algorithms have full control over the predefined quality criteria, such as:
 frequency and time domain characteristics (spectrum; probability % distribution) when generating noise with continuous spectra  carrier amplitude, carrier frequency, modulation depth and modulation width, when generating noise with discrete spectra Both types of noise axe calculated independently, and represented in the time domain as arrays with numbers.
% Both types of noise can be made available simultaneously by adding these arrays element wise.
(c) 20002001 KPN Research;
DEMO FUNCTIONS
% DemoXtalkNoise  shows the process of creating continuous noise DemoIngressNoise  shows the process of creating discrete noise MAIN FUNCTIONS: Noise is represented as an array with random numbers DefineShape  initialize all userdefinable parameters % CreateNoiseCont  generates continuous noise CreateNoiseDiscr Fast  generate discrete noise, fast algorithm CreateNoiseDiscr Slow  generate discrete noise, slow algorithm FrequencyShape  modify spectral density of continuous noise AmplitudeShape  modify amplitude distribution of continuous % noise SUPPORTING FUNCTIONS:
CalcSpec  calculates the spectral density of noise CaIcNBSV  calculates the narrow band signal voltage of % noise CalcCrest  calculates the crest factor of noise CalcDistrib  calculates the probability distribution of noise CalcCumDistrib  calculates the cumulated distribution of noise CalcSmooth  smoothes a spectrum, like in a real spectrum % analyzer CalcEnhancedGaussDistribution  a sample of a neargaussian distribution CalcDemodulation  calculate the noise modulated on a carrier %______________________________________________________________________________ ______________________ Shape=DefineShape;
DemoXtalkNoise(Shape);
DemoIngressNoise(Shape);
%_____________________________________________________ ,.
____________________________________________ function [U,t]=DemoXtalkNoise(Shape);
%______________________________________________________________________________ ______________________ demonstrates the generation of noise with continuous spectrum e.g. for crosstalk testing R = Shape.R;
CF min = Shape.Xtalk.CF min;
[U,t]=CreateNoiseCont(Shape); plot(t,U); title('Xtalk method 1'); shg; pause [X,f]=CalcSpec(U,t); plot(f,X); title('Xtalk method 1'); shg; pause [X,f]=CalcSpec(U,t); plot(fdBm(X,R)); title('Xtalk method 1'); shg; pause [P,u]=CalcCumDistrib(U); plot(u,P); title('Xtalk method 1'); shg; pause U=AmplitudeShape(U,Shape); plot(t,U); title('Xtalk method 2'); shg; pause [X,fJ=CalcSpec(U,t); plot(f,X) title('Xtalk method 2'); shg; pause [X,fJ=CalcSpec(U,t); plot(fdBm(X,R)); title('Xtalk method 2'); shg; pause [P,u]=CalcCumDistrib(U); plot(u,P); title('Xtalk method 2'); shg; pause 5 U=FrequencyShape(U,Shape); plot(t,U); title('Xtalk method 3'); shg; pause [X,fJ=CalcSpec(U,t); plot(f,X); title('Xtalk method 3'); shg; pause [X,f]=CalcSpec(U,t); plot(fdBm(X,R)); title('Xtalk method 3'); shg; pause [P,u]=CalcCumDistrib(U); plot(u,P); title('Xtalk method 3'); shg; pause 10 for i=2:10 i U=AmplitudeShape(U,Shape);
[X,f]=CalcSpec(LJ,t);
15 U=FrequencyShape(U,Shape);
[X,f]=CalcSpec(U,t);
if CalcCrest(I7)>CF min, break; end;
end;
20 [P,u]=CalcCumDistrib(U);
plot(t,U); title('Xtalk method 4'); shg; %pause plot(f,dBm(X,R)); title('Xtalk method 4'); shg; %pause plot(u,P); title('Xtalk method 4'); shg; %pause 25 %__________________________________________________________________ function [U,t]=DemoIngressNoise(Shape);
%____________________________,_____,___________________________________________ ______________________ demonstrates the generation of noise with discrete spectrum e.g. for ingress testing R = Shape.R;
[U,t]=CreateNoiseDiscr Fast(Shape);
The amplitude distribution of Gaussian type noise is:
G(x) =1e~f x (3) with : erf (x) _ ~ f dt exp. tz ), (4) and with 6 being the RMS value of the signal.
The "enhanced" Gaussian distribution is defined as:
G x  1 a~~+erf x a+erf A 0Sx<A
( ) ~a. ~a~ ~ (5) 0 x>A
If VAS is the desired RMS value of the noise sample, and Cf being the desired crest factor, choose:
A = Cf x V,~s , and (6) 6=~((1+a) VZ,~s  AZ~a/3). ('7) Typical values for a that have proven useful are of a magnitude between 0.001 and 0.01, and this represents the deviation of enhanced Gaussian distributed from a true Gaussian distribution.
In block 14, Frequency Shape, of Figure 2, the frequency domain characteristics of the signal are improved, as a posedprocessing step to achieve a closer match to the or each quality criterion. The corrected frequency curve can be achieved, for example, by comparing (dividing) a predetermined spectral density through the measured spectral density of the (intermediate) signal U2(t). An example hereof is given in the best mode embodiment described in the sequel with a convolution of FFT functions.
In the part [4] of Figure 2, it is shown how an iterative procedure of the steps detailed here above may lead to a further improvement of the finish or final signal for use in or on a communication system. The iterative procedure, i.e. testing of the frequency shape by block I5, Test Shape, and back coupling loop 16, is executed until the predetermined quality criterions) are achieved.
Figure 3 illustrates in a schematic, graphic representation use of the method according to the invention for the provision of an ingress noise signal.
As disclosed in the preamble, ingress noise may be characterised by a plurality of frequency components at discrete carrier frequencies fci, i=1,2,3, .... The frequency components at the carrier frequency fci each having a carrier amplitude Aci, i=1,2,3, ..., and, if applicable, having a modulation width, i.e. a number of discrete frequencies at the left and right hand side of the associated carrier frequency fci, as well as having a modulation depth, that is the amplitude of the side frequencies associated with the respective carrier frequency fci.
Figure 3 shows, in a graphic representation having a horizontal or frequency axis f and a vertical or amplitude axis A, by way of example only, a signal comprised of two carrier frequencies fcl and fc2, having a carrier amplitude Acl and Ac2, respectively.
Around the carrier frequency fcl at each side thereof three side frequency components are arranged, each having an amplitude A1. For the frequency component at carrier frequency fc2 on each side thereof two side frequency components are arranged, each having an amplitude A2.
In accordance with the present invention, for providing a signal having at least one predefined quality criterion, the amplitude of the frequency components have to be shaped, such as disclosed by the dotted lines I and 2 in Figure 3.
Starting from a represented first signal having random phase properties, in accordance with the present invention, by the shaping of the amplitudes, the random phase properties are maintained in the signal to be provided having the predefined quality criterion.
Figure 4 shows schematically a further embodiment of a method for arranging a signal for use in or on a communication system, in particular for use if random phase properties are already provided for. The signal comprises crosstalk noise that is a random signal with predetermined properties in the frequency domain and in the time domain. The signal can furthermore comprise rfitones that have a discrete frequency spectrum. Also other signal components can be included in the signal.
The method further may comprise the steps of representing a first signal in time 5 domain and with an amplitude distribution and the signal having a spectral density in the frequency domain, thereby achieving a represented signal, and processing the represented signal according to a nonlinear transformation, the nonlinear transformation achieving a predetermined quality criterion. This is shown as amplitude shaping in Figure 4, flows [24].
10 The method further may comprise the step of representing a first signal in time domain and with an amplitude distribution and the signal having a spectral density in the frequency domain, thereby achieving a represented signal, and processing the represented signal until a signal is achieved having a spectral density according to a predetermined spectral density quality criterion. This is shown as frequency shaping in 15 Figure 4, flows [24]. The frequency shaping step can also comprise the step of filtering the represented signal in the frequency domain including the steps of evaluating at least part of the signal representation in the frequency domain and thereafter processing the signal representation in the frequency domain.
The method may also comprise the step of making a signal in different iterative 20 steps, see Figure 4 flow [4]. Thus the signal can have a predetermined time domain amplitude distribution and/or a predetermined spectral density or a time domain amplitude distribution and/or a spectral density according to a predefined quality criterion. The predefined quality criterion can be the crest factor of the signal, that is a relation of the maximum or peak value of the tones of the signal compared to the average value of the tones of the signal. The signals can be generated and stored using a set of instructions in a code format and executable in a predetermined order and compiled on a device.
Likewise, the set of instructions can be software code compiled on a computer and stored in the computer or a network of computers or a floppy or CDROM or through the Internet. The software can also be stored on ari Arbitrary Wave Form Generator (AWG) card and the AWG can be used to generate the signals or reproduce stored signals from the memory. It is therefore possible tv have a library of signals available that can be used in the execution or use of the methods of the invention. The communication systems can be devices such as xDSL modems, or chips within or for such modems, or networks for telecommunication. In detail the following embodiment is shown in Figure 4.
Using software, random numbers are generated, block 16, "Create Noise". In hardware white noise can be generated. The random numbers are filtered until a predetermined spectral density is achieved. The random numbers that are generated each represent a frequency component. The necessary processing to achieve a predetermined spectral density is executed by scaling the amplitude of the complex numbers and thereafter an IFFT processing is done in order to make the desired noise signal. Another way of executing the method is to generate random numbers that represent the phase of each frequency component and thereafter the amplitude of the complex numbers is arranged to approach or be equal to a predetermined spectral density The processing in block 13, "Amplitude Shape", is done for achieving an impact or control on the time domain characteristic. An amplitude distortion (transformation) function Q(x) is chosen that amplifies the high amplitude peaks or tones in the signal is shown in Figure 5. The nonlinear transformation function Q(x) can be reconstructed from the actual amplitude distribution function of the signal and the predetermined amplitude distribution function, as disclosed above in connection with the equations (17).
In block 14, "Frequency Shape" of Figure 4, the frequency domain characteristics of the signal are improved. The corrected frequency curve is achieved by comparing (dividing) a predetermined spectral density through the measured spectral density of the (intermediate) signal. An example hereof is given in the best mode embodiment described in the sequel with a convolution of FFT functions.
In the flow [4] of Figure 4, like in the flow [4] of Figure 2, again it is shown how an iterative procedure of the steps detailed hereabove may lead to a further improvement of the final for use in or on a communication system. The iterative procedure is executed until the predetermined quality criterions are achieved.
With the method according to the invention, as disclosed above, signals representing crosstalk noise and ingress noise can be generated with a device, such as an impairment generator 8, see Figure 1, which may be arranged for providing a signal comprised of both a crosstalk noise signal and an ingress noise signal, while other signal components may be added to the output signal to be provided, if required.
The signal to be provided, in an embodiment of the invention, can be advantageously provided as a sixth set of numbers in the time domain, for example an array of numbers.
Figure 6 shows, in a flow type diagram, an example embodiment of the invention, running on a Personal Computer 20. The impairment noise is generated by block 21, called SPOCS, comprising a block 22, "White noise signal", a block "Spectral shaping", provided by FFT, a block 24, "Desired noise signal", produced from the output of block 23 by IFFT, the resulting signal of which is stored on an AWG card 25. In the crosstalk scenario, i.e. block 26, a noise PSD is created, block 27, which is further processed by block 23.
A best mode embodiment of the set of instructions of the invention is disclosed here below. The code given here below is compiled in a MATLAB environment.
Comments related to the functionality of the code are given after the % signs.
For a person skilled in the art, the code provided is self explanatory.
Figures 79 show results obtained with the best mode embodiment. Figure 8 shows a plot of the spectrum of the generated noise sample plus the PSD of the noise profile. Figure 9 shows a plot of the generated noise sample in the time domain. Figure 10 shows a plot of the distribution function of the generated noise sample.
Figure 11 shows a plot of the cumulative distribution functiopn of the generated noise sample.
Figure 7 shows a graphical User Interface (UI) and settings of the AWG
control.
%______________________________________________________________________________ ______________________ function DemoImpair2;
%______________________________________________________________________________ ______________________ DemoImpair2 S % Code, programmed in the Matlab programming language, that demonstrates the basic algorithms of an Impairment Generator.
The demonstrated algorithms have full control over the predefined quality criteria, such as:
 frequency and time domain characteristics (spectrum; probability % distribution) when generating noise with continuous spectra  carrier amplitude, carrier frequency, modulation depth and modulation width, when generating noise with discrete spectra Both types of noise axe calculated independently, and represented in the time domain as arrays with numbers.
% Both types of noise can be made available simultaneously by adding these arrays element wise.
(c) 20002001 KPN Research;
DEMO FUNCTIONS
% DemoXtalkNoise  shows the process of creating continuous noise DemoIngressNoise  shows the process of creating discrete noise MAIN FUNCTIONS: Noise is represented as an array with random numbers DefineShape  initialize all userdefinable parameters % CreateNoiseCont  generates continuous noise CreateNoiseDiscr Fast  generate discrete noise, fast algorithm CreateNoiseDiscr Slow  generate discrete noise, slow algorithm FrequencyShape  modify spectral density of continuous noise AmplitudeShape  modify amplitude distribution of continuous % noise SUPPORTING FUNCTIONS:
CalcSpec  calculates the spectral density of noise CaIcNBSV  calculates the narrow band signal voltage of % noise CalcCrest  calculates the crest factor of noise CalcDistrib  calculates the probability distribution of noise CalcCumDistrib  calculates the cumulated distribution of noise CalcSmooth  smoothes a spectrum, like in a real spectrum % analyzer CalcEnhancedGaussDistribution  a sample of a neargaussian distribution CalcDemodulation  calculate the noise modulated on a carrier %______________________________________________________________________________ ______________________ Shape=DefineShape;
DemoXtalkNoise(Shape);
DemoIngressNoise(Shape);
%_____________________________________________________ ,.
____________________________________________ function [U,t]=DemoXtalkNoise(Shape);
%______________________________________________________________________________ ______________________ demonstrates the generation of noise with continuous spectrum e.g. for crosstalk testing R = Shape.R;
CF min = Shape.Xtalk.CF min;
[U,t]=CreateNoiseCont(Shape); plot(t,U); title('Xtalk method 1'); shg; pause [X,f]=CalcSpec(U,t); plot(f,X); title('Xtalk method 1'); shg; pause [X,f]=CalcSpec(U,t); plot(fdBm(X,R)); title('Xtalk method 1'); shg; pause [P,u]=CalcCumDistrib(U); plot(u,P); title('Xtalk method 1'); shg; pause U=AmplitudeShape(U,Shape); plot(t,U); title('Xtalk method 2'); shg; pause [X,fJ=CalcSpec(U,t); plot(f,X) title('Xtalk method 2'); shg; pause [X,fJ=CalcSpec(U,t); plot(fdBm(X,R)); title('Xtalk method 2'); shg; pause [P,u]=CalcCumDistrib(U); plot(u,P); title('Xtalk method 2'); shg; pause 5 U=FrequencyShape(U,Shape); plot(t,U); title('Xtalk method 3'); shg; pause [X,fJ=CalcSpec(U,t); plot(f,X); title('Xtalk method 3'); shg; pause [X,f]=CalcSpec(U,t); plot(fdBm(X,R)); title('Xtalk method 3'); shg; pause [P,u]=CalcCumDistrib(U); plot(u,P); title('Xtalk method 3'); shg; pause 10 for i=2:10 i U=AmplitudeShape(U,Shape);
[X,f]=CalcSpec(LJ,t);
15 U=FrequencyShape(U,Shape);
[X,f]=CalcSpec(U,t);
if CalcCrest(I7)>CF min, break; end;
end;
20 [P,u]=CalcCumDistrib(U);
plot(t,U); title('Xtalk method 4'); shg; %pause plot(f,dBm(X,R)); title('Xtalk method 4'); shg; %pause plot(u,P); title('Xtalk method 4'); shg; %pause 25 %__________________________________________________________________ function [U,t]=DemoIngressNoise(Shape);
%____________________________,_____,___________________________________________ ______________________ demonstrates the generation of noise with discrete spectrum e.g. for ingress testing R = Shape.R;
[U,t]=CreateNoiseDiscr Fast(Shape);
[U,t]=CreateNoiseDiscr Slow(Shape); %gives same result [X,f]=CalcNBSV(U,t); plot(fdBm(X,R)); title('Ingress method'); shg; pause for ToneNr=[ 1:2]
[Uac,Uac rms]=CalcDemodulation(U,t,Shape,ToneNr);
plot(t,Uac); title('demodulated ingress noise of one carrier'); shg; pause [P,u]=CalcDistrib(Uac/Uac rms);
plot(u,P); title('distribution of demod noise'); shg; pause end;
%_______________________________________________________________________________ ____________________ function [Shape] = DefineShape;
%______________________________________________________________________________ ______________________ %Create the noise profiles for the noise that should be generated, in terms of  spectral density (in this example rectangular in nature) I 5 %  probability distribution (in this example near Gaussian)  tones and modulation Spectra in Volt per sqrt(FIz) %______________________________________________________________________________ ______________________ Fmax=4E6; Fl=300E3; Fh=700E3; N=2~18; R=I35;
c~5.5; % desired crest factor) cf min=5.1; % desired crest factor) m N/2;
Shape.N=N; % number of time samples Shape.m=m; % number of freq samples Shape.dF= Fmax./(m1); % frequency spacing Shape.dT= 1/(N~Shape.dF); % time spacing Shape.R=R; % impedance of desired noise source;
define crosstalk noise target (Spectral density & Amplitude Distribution) Shape.Xtalk.freq=[O:m1]' * Shape.dF;
[Uac,Uac rms]=CalcDemodulation(U,t,Shape,ToneNr);
plot(t,Uac); title('demodulated ingress noise of one carrier'); shg; pause [P,u]=CalcDistrib(Uac/Uac rms);
plot(u,P); title('distribution of demod noise'); shg; pause end;
%_______________________________________________________________________________ ____________________ function [Shape] = DefineShape;
%______________________________________________________________________________ ______________________ %Create the noise profiles for the noise that should be generated, in terms of  spectral density (in this example rectangular in nature) I 5 %  probability distribution (in this example near Gaussian)  tones and modulation Spectra in Volt per sqrt(FIz) %______________________________________________________________________________ ______________________ Fmax=4E6; Fl=300E3; Fh=700E3; N=2~18; R=I35;
c~5.5; % desired crest factor) cf min=5.1; % desired crest factor) m N/2;
Shape.N=N; % number of time samples Shape.m=m; % number of freq samples Shape.dF= Fmax./(m1); % frequency spacing Shape.dT= 1/(N~Shape.dF); % time spacing Shape.R=R; % impedance of desired noise source;
define crosstalk noise target (Spectral density & Amplitude Distribution) Shape.Xtalk.freq=[O:m1]' * Shape.dF;
Shape.Xtalk.spec =(Shape.Xtalk.freq>=Fl).*(Shape.Xtalk.freq<=Fh)*(1/300);
Shape.Xtalk.DistU = O:cf/1000:cf;
Shape.Xtalk.DistP = CalcEnhancedGaussDistribution(Shape.Xtalk.DistU, cfJ; %P
Shape.Xtalk.CF min=cf min;
% define ingress noise target (RFITones) P dBm =[70;50;60;60;40;60;60;40;70;40]; % dBm @ 135 ohm P=(10).~(P dBm/10)*lE3;
Shape.Ingress.ToneU =sqrt(P*135); % U=sqrt(P*R); effective value Shape.Ingress.ToneF =[99;207;333;387;531;603;711;801;909;981]*1E3;
Shape.Ingress.ModDepth = 0.32*ones(10,1);%=mod index > 0.8, at CF>2.5 Shape.Ingress.ModWidth = 2*4.SE3*ones(10,1); %_ 10 kHz .. +10 kHz) %______________________________________________________________________________ ______________________ function [U,t] = CreateNoiseCont(Shape);
1 5 %___________________________________________________________________________________________________ create a noise voltage U(t), with predefined frequency domain characteristics (spectrum), but with uncontrolled time domain characteristics (distribution) N = Shape.N; % number of samples, to be generated %U = rand(N,1 ); % Uniform distributed white noise U = randn(N,1 ); % Gaussian distributed white noise U = FrequencyShape(U,Shape); % shaped noise t=[O:N1]' * Shape.dT; % associated time axis %______________________________________________________________________________ _____________________ function [U,t] = CreateNoiseDiscr Fast(Shape);
%__________________________________________________________________________________________________ Create a voltage U(t), with AM modulated carriers (RFI Tones); each having an individual predefined frequency, amplitude, modulation width and modulation depth.
The random phase of the lower side band of each Garner tone is mirrored to % convert arbitrary QAM modulation into (no mirroring) into the more restricted AM modulation (full mirroring) %' Mark that X refers in this algorithm to the components of the Fourier series of the (near harmonic) ingress noise signal , while it refers to the spectral density in case of the (pseudo random) crosstalk noise signal Calculation time increases about linear with the number of samples About 80% of all calculation time is caused by the inverse Fourier transform %______________________________________________________________________________ ______________________ N = Shape.N; % number of samples m = Shape.m; % half this number Nc=TOUnd(Shape.Ingress.ToneF/Shape.dF)+1; % index of carrier freq (pos only) Nm=round(Shape.Ingress.ModWidth/Shape.dF/2); % number of modulation components Xc=0.5*Shape.Ingress.ToneU; % amplitude of Garner amplitude Xm=Shape.Ingress.ModDepth.*Xc.lsqrt(2*Nm); % amplitude of modulation band X=zeros(N,1); % initialization Xc=Xc.*exp(j* 1000*rand(size(Xc))); % random carrier phase Xcc=(Xc.*Xc)./abs(Xc.*Xc);
for k=l :length(Nc) % for all modulated carriers, do:
Nmp=Nc(k)+[1:Nm(k)]'; . % locate upper side band frequencies Nmn Nc(k)[ 1:Nm(k)]'; % locate upper side band frequencies Xmp=Xm(k).*exp(j* 1000*rand(size(Nmp))); % create upper side band Xmn=conj(Xmp)*Xcc(k); % mirror lower side band X(Nmp)=Xmp; % insert upper side band X(Nmn)=Xmn; % insert lower side band end;
X(Nc)=Xc; % insert all carriers X(N:l:m+2)=conj(X(2:ceil(m))); % Append spectrum (negative freq.) %U= real(ifft(X))*N; % Transform to timedomain U = real(fft(X)); % Transform to timedomain 10% faster) t = Shape.dT*[O:N1]'; % associated time axis %__,_______,___________________________________________________________________ ______________________ function [U,t] = CreateNoiseDiscr Slow(Shape);
%______________________________________________________________________________ ______________________ Create a voltage U(t), with RFI Tones at predefined frequency, amplitude and modulation bandwidth and modulation depth This algorithm is straightforward, very inefficient, and for demo purposes only It can prove that CreateNoiseDiscr Fast returns the same results %______________________________________________________________________________ ______________________ N = Shape.N; % number of samples, to be generated m = Shape.m;
f=[O:N1]' * Shape.dF;
t=[O:N1]' ~ Shape.dT;
Fc=Shape.Ingress.ToneF; % list of carrier frequencies Fc=Shape.dF * round(Fc/Shape.dF); % force an integer number of periods U=0;
for k=l :length(Shape.Ingress.ToneF);
%  create noisy modulate, having U avg=0 and U rms=1.
Nm =round(Shape.Ingress.ModWidth(k)/Shape.dFl2);
Xn0=([ 1:N]<=(Nm+1 )) ; % shape modulation noise amplitude Xn =XnO.*exp(j*1000*rand(N,1)); % shape modulation noise phase Xn(1)= 0; % Eliminate DC component.
Xn(N:I:m+2)=conj(Xn(2:ceil(m))); % Append spectrum (negative freq.) Noise = real(ifft(Xn)); % Transform to timedomain Noise=Noise/sqrt(sum(Noise.*Noise)/N); % force rms=1  perform modulation Carner = Shape.Ingress.ToneU(k) * cos(2~pi*Fc(k)*t+1000*rand);
5 Modulate = Shape.Ingress.ModDepth(k) * Noise;
U = U + Carrier .* (1 + Modulate);
end;
%______________________________________________________________________________ ______,_______________ function [U] = FrequencyShape(IJ,Shape) %______________________________________________________________________________ ______________________ Reshape the spectrum of the sample U, as specified by the target shape INPUT:
15 % U: the consecutive values of the sample fs: the sample frequency spectrum: the desired PSD (in V/sqrt(Hz)) %______________________________________________________________________________ ______________________ N = length(U);
20 m = length(Shape.Xtalk.spec); % m=N/2 t= [O:N1]'*Shape.dT;
perform the frequency scaling Scaling = Shape.Xtalk.spec ./ CalcSpec(U,t);
X = fft(U); % Transform to frequency domain 25 X(1) = 0; % Eliminate DC component.
X(2:m+1) =X(2:m+1) .* Scaling; % Scale spectrum (positive freq.) X(N:l:m+2)=conj(X(2:cei1(m))); % Append spectrum (negative freq.) U = real(ifft(X)); % Transform to time_domain %______________________________________________________________________________ ______________________ function [U] = AmplitudeShape(U,Shape) %______________________________________________________________________________ ______________________ This function shapes the amplitude distribution of the function U
% by an amplitude dependent (nonlinear) distortion function Q(x).
The result is U(t) = Q{U(t)) Let FF be the actual cumulative distribution function of the sample, and °10 let GG be the desired cumulative distribution function, % then the distortion function is given by:
Q(x) = GG~ f 1 } FF (x) %______________________________________________________________________________ ______________________ UO=sqrt(sum(U.*U)/length(U)); %scaling farct (for normalization) Calculate the distortion function Q
[DistPl,DistU1] = CalcCumDistrib(U/UO); % the actual distribution Q = interpl(Shape.Xtalk.DistP, Shape.Xtalk.DistU, DistPl); % the distortion function U = UO *interpl{DistUl,Q,abs(U/UO)) .~ sign(U); % Perform the distortion plot(DistUl,Q); shg; %pause %______________________________________________________________________________ ______________________ function [X,f] = CalcSpec(U,t);
_______________________________________________________________________________ _____________________ calculate the spectral density of a signal, when it would be'measured' at specified resolution bandwidth RBW=1E3; %RBW: the desired resolution for the spectrum of U
N = length(U); m N/2;
dT = t(2)t( 1 ); % time spacing dF = 1 /dT/N; % frequency spacing f [O:m1]'*dF; % all positive frequencies X = fft(U)*dT; % to frequency domain X = abs(X(2:m+1)); % No DC and no negative frequencies.
X = sqrt(CalcSmooth(X.*X, f, RBW)); % average it over bandwidth RBW
%______________________________________________________________________________ ______________________ function [X,fJ = CaIcNBSV(U,t);
%______________________________________________________________________________ ______________________ calculate the narrow band signal voltage of a signal, when it would be'measured' at specified resolution bandwidth RBW=1E3; %RBW: the desired resolution for the spectrum of U
N = length(U); m=N/2;
dT = t(2)t( 1 ); % time spacing dF = 1/dT/N; % frequency spacing f= [O:m1]'*dF; % all positive frequencies %X = fft(U)*dT * sqrt(dF); % to frequency domain %X = fft(IT)/N; % to frequency domain X = fft(LT)/N*2; % to frequency domain X = abs(X(2:m+1)); % No DC and no negative frequencies.
%X = sqrt(CalcSmooth(X.*X, f, RBW)); % average it over bandwidth RBW
%______________________________________________________________________________ ______________________ function [CF] = CalcCrest(U) _______________________________________________________________________________ _____________________ Calculate the Crest Factor of a signal (U(t), which is the peak value divided by the rmsvalue Urms = sqrt(sum(U.~2)/length(U));
Upeak = max(abs(U));
CF = Upeak/Urms;
%____________________________________________,_________________________________ _______________________ function [F]=CalcEnhancedGaussDistribution(x,Cf);
%______________________________________________________________________________ _______________________ Generate a Cumulative distribution function F(x) that is identified as % "enhanced gaussian distribution"
Cf = crest factor Alpha = 1 e3;
Sigma = sqrt( (1+Alpha)  Cf~2 * Alpha/3);
x = x .* (x>0) .* ( x<Cf) + Cf * (x>=Cfj;
denominator = Alpha + erf(Cf/(sqrt(2)*Sigma));
F = 1  (Alpha * x/Cf + erf(x/(sqrt(2)*Sigma)))/denominator;
%___________________________________________________________________________________________________ function [DistP, DistU,P] = CalcDistrib(U) %______________________________________________________________________________ ______________________ calculate the amplitude distribution of signal U
N = length(U);
Nbins=100;
[cumbin,xx] = hist(U,Nbins);
dX=xx(3)xx(2);
DistP = cumbin(:)/N/dX; % force sum(DistP)*dX
DistU = xx(:);
Urms=sqrt(sum(U. *U)/N);
P=exp(0.5*(DistU/Urms).~2); P=P/sum(P)/dX;
DistU=[DistU,DistU];
DistP=[DistP,P];
_______________________________________________________________________________ _______________,_____ function [DistP, DistU] = CalcCumDistrib(U) %______________________________________________________________________________ ______________________ calculate the (backward) cumulative amplitude distribution of signal U
len = length(U);
Ueff = sqrt(sum(U .* U)/length(U));
U = abs(U/Ueff);
 evaluate distribution function Nbins = min([SO,floor(len/10)]);
[cumbin,xx] = hist(U,Nbins);
BinWidth=xx(2)  xx(1);
DistU = xx  BinWidth/2; % shift for n = [Nbins1:1:1 ]; cumbin(n) = cumbin(n) + curnbin(n+1 ); end DistP = cumbin/len;
%  improve numerical stability for other routines, when they use this result DistU = [ 0 , DistU(2:end)]; % start at x = 0 DistP = [DistP, 1/len];
DistU = [DistU,xx(Nbins) + 0.999 * BinWidth/2]; % add final (single) point DistP = [DistP, 1e100];
DistU = [DistU,xx(Nbins) + (1.001) * BinWidth/2]; % factor 1.001 for stability ___________________________________________________________________________________________________ function [PSD,freq]=CalcSmooth(PSD,freq,RBW) %_________________________________________________,____________________________ _______________________ % Imitate a real Spectrum Analyzer, with finite resolution bandwidth, and Gaussian shaped band filters PSD = "power spectral density" which is de square of the "spectral density";
in Volts per square Hertz.
%______________________________________________________________________________ _______________________ N = length(PSD);
df = freq(2)freq( I );
br = 3 * floor(RBW/dfj;
factor = 2*br + 1;
if (factor > 1) 5 ..ff= df * (br:br); % smooth interval ..mask = exp(f~*ff/(2*RBW~2));
..mask = mask/sum(mask); % Gaussian mask of resolution band filter ..xhelp = [PSD;zeros(2*br,l)];
1p = filter(mask,l,xhulp); % smart convolution 10 PSD = yhelp(br+1:endbr);
end;
%______________________________________________________________________________ ______________________ function [Uac,Uac rms]=CalcDemodulation(U,t,Shape,ToneNr);
%______________________________________________________________________________ _____________________ Demodulate the noise that has been modulated on the carriers of the discrete noise, it is for demonstration purposes only to prove that the discrete noise meets the predefined parameters.
20 % The demodulator uses synchronous detection, that is not locked in phase The consequence is an unknown attenuation over the full demodulation band.
This is corrected afterward by measuring the DC level, and amplify the demodulated signal until this DC level has been normalized to 1 Volt %
25 % PROOF: (psi is unknown) let Ur~cos(w*t+psi)*(1+Uac); % = carrier modulated with "1+Uac"
Uc =cos(w*t); % = carrier Ud =Urf~Uc; % = synchronous detected signal 30 % then Ud=1/2*(cos(psi+2*w*t)+cos(psi))*(1+Uac);
Ul~cos(psi)/2*(1+Uac); % after lowpas filtering % Udc=cos(psi)/2; %= by averaging Ulf % Uac=(Ulf/Udc)1;
_______________________________________________________________________________ _____________________ N=Shape.N;
Fc=Shape.Ingress.ToneF(ToneNr); %select carrier frequency Fc=Shape.dF * round(Fc/Shape.dF); %force an integer number of periods ModWidth=Shape.Ingress.ModWidth(ToneNr);
ModDepth=Shape.Ingress.ModDepth(ToneNr);
Ud=U.*cos(2*pi*Fc*t); % synchronous detection of modulated carrier Nm=round(l .l *ModWidth/Shape.dF/2); % calculate filter frequency mask=zeros(N,1); mask([l:Nm, NNm:N])=1; % create filter Ulf=real(ifft(fft(Ud).*mask)); % perform lowpass filtering Udc=sum(Ulf)!N; % find not normalized DC level Uac=Ulf/Udc1; % normalize overall level, and remove DC.
Uac rms=sqrt(sum(LTac.*Uac/N)); % must be equal to ModDepth, since Udc=1 Scale=Uac rms/ModDepth; % must be "one"
Shape.Xtalk.DistU = O:cf/1000:cf;
Shape.Xtalk.DistP = CalcEnhancedGaussDistribution(Shape.Xtalk.DistU, cfJ; %P
Shape.Xtalk.CF min=cf min;
% define ingress noise target (RFITones) P dBm =[70;50;60;60;40;60;60;40;70;40]; % dBm @ 135 ohm P=(10).~(P dBm/10)*lE3;
Shape.Ingress.ToneU =sqrt(P*135); % U=sqrt(P*R); effective value Shape.Ingress.ToneF =[99;207;333;387;531;603;711;801;909;981]*1E3;
Shape.Ingress.ModDepth = 0.32*ones(10,1);%=mod index > 0.8, at CF>2.5 Shape.Ingress.ModWidth = 2*4.SE3*ones(10,1); %_ 10 kHz .. +10 kHz) %______________________________________________________________________________ ______________________ function [U,t] = CreateNoiseCont(Shape);
1 5 %___________________________________________________________________________________________________ create a noise voltage U(t), with predefined frequency domain characteristics (spectrum), but with uncontrolled time domain characteristics (distribution) N = Shape.N; % number of samples, to be generated %U = rand(N,1 ); % Uniform distributed white noise U = randn(N,1 ); % Gaussian distributed white noise U = FrequencyShape(U,Shape); % shaped noise t=[O:N1]' * Shape.dT; % associated time axis %______________________________________________________________________________ _____________________ function [U,t] = CreateNoiseDiscr Fast(Shape);
%__________________________________________________________________________________________________ Create a voltage U(t), with AM modulated carriers (RFI Tones); each having an individual predefined frequency, amplitude, modulation width and modulation depth.
The random phase of the lower side band of each Garner tone is mirrored to % convert arbitrary QAM modulation into (no mirroring) into the more restricted AM modulation (full mirroring) %' Mark that X refers in this algorithm to the components of the Fourier series of the (near harmonic) ingress noise signal , while it refers to the spectral density in case of the (pseudo random) crosstalk noise signal Calculation time increases about linear with the number of samples About 80% of all calculation time is caused by the inverse Fourier transform %______________________________________________________________________________ ______________________ N = Shape.N; % number of samples m = Shape.m; % half this number Nc=TOUnd(Shape.Ingress.ToneF/Shape.dF)+1; % index of carrier freq (pos only) Nm=round(Shape.Ingress.ModWidth/Shape.dF/2); % number of modulation components Xc=0.5*Shape.Ingress.ToneU; % amplitude of Garner amplitude Xm=Shape.Ingress.ModDepth.*Xc.lsqrt(2*Nm); % amplitude of modulation band X=zeros(N,1); % initialization Xc=Xc.*exp(j* 1000*rand(size(Xc))); % random carrier phase Xcc=(Xc.*Xc)./abs(Xc.*Xc);
for k=l :length(Nc) % for all modulated carriers, do:
Nmp=Nc(k)+[1:Nm(k)]'; . % locate upper side band frequencies Nmn Nc(k)[ 1:Nm(k)]'; % locate upper side band frequencies Xmp=Xm(k).*exp(j* 1000*rand(size(Nmp))); % create upper side band Xmn=conj(Xmp)*Xcc(k); % mirror lower side band X(Nmp)=Xmp; % insert upper side band X(Nmn)=Xmn; % insert lower side band end;
X(Nc)=Xc; % insert all carriers X(N:l:m+2)=conj(X(2:ceil(m))); % Append spectrum (negative freq.) %U= real(ifft(X))*N; % Transform to timedomain U = real(fft(X)); % Transform to timedomain 10% faster) t = Shape.dT*[O:N1]'; % associated time axis %__,_______,___________________________________________________________________ ______________________ function [U,t] = CreateNoiseDiscr Slow(Shape);
%______________________________________________________________________________ ______________________ Create a voltage U(t), with RFI Tones at predefined frequency, amplitude and modulation bandwidth and modulation depth This algorithm is straightforward, very inefficient, and for demo purposes only It can prove that CreateNoiseDiscr Fast returns the same results %______________________________________________________________________________ ______________________ N = Shape.N; % number of samples, to be generated m = Shape.m;
f=[O:N1]' * Shape.dF;
t=[O:N1]' ~ Shape.dT;
Fc=Shape.Ingress.ToneF; % list of carrier frequencies Fc=Shape.dF * round(Fc/Shape.dF); % force an integer number of periods U=0;
for k=l :length(Shape.Ingress.ToneF);
%  create noisy modulate, having U avg=0 and U rms=1.
Nm =round(Shape.Ingress.ModWidth(k)/Shape.dFl2);
Xn0=([ 1:N]<=(Nm+1 )) ; % shape modulation noise amplitude Xn =XnO.*exp(j*1000*rand(N,1)); % shape modulation noise phase Xn(1)= 0; % Eliminate DC component.
Xn(N:I:m+2)=conj(Xn(2:ceil(m))); % Append spectrum (negative freq.) Noise = real(ifft(Xn)); % Transform to timedomain Noise=Noise/sqrt(sum(Noise.*Noise)/N); % force rms=1  perform modulation Carner = Shape.Ingress.ToneU(k) * cos(2~pi*Fc(k)*t+1000*rand);
5 Modulate = Shape.Ingress.ModDepth(k) * Noise;
U = U + Carrier .* (1 + Modulate);
end;
%______________________________________________________________________________ ______,_______________ function [U] = FrequencyShape(IJ,Shape) %______________________________________________________________________________ ______________________ Reshape the spectrum of the sample U, as specified by the target shape INPUT:
15 % U: the consecutive values of the sample fs: the sample frequency spectrum: the desired PSD (in V/sqrt(Hz)) %______________________________________________________________________________ ______________________ N = length(U);
20 m = length(Shape.Xtalk.spec); % m=N/2 t= [O:N1]'*Shape.dT;
perform the frequency scaling Scaling = Shape.Xtalk.spec ./ CalcSpec(U,t);
X = fft(U); % Transform to frequency domain 25 X(1) = 0; % Eliminate DC component.
X(2:m+1) =X(2:m+1) .* Scaling; % Scale spectrum (positive freq.) X(N:l:m+2)=conj(X(2:cei1(m))); % Append spectrum (negative freq.) U = real(ifft(X)); % Transform to time_domain %______________________________________________________________________________ ______________________ function [U] = AmplitudeShape(U,Shape) %______________________________________________________________________________ ______________________ This function shapes the amplitude distribution of the function U
% by an amplitude dependent (nonlinear) distortion function Q(x).
The result is U(t) = Q{U(t)) Let FF be the actual cumulative distribution function of the sample, and °10 let GG be the desired cumulative distribution function, % then the distortion function is given by:
Q(x) = GG~ f 1 } FF (x) %______________________________________________________________________________ ______________________ UO=sqrt(sum(U.*U)/length(U)); %scaling farct (for normalization) Calculate the distortion function Q
[DistPl,DistU1] = CalcCumDistrib(U/UO); % the actual distribution Q = interpl(Shape.Xtalk.DistP, Shape.Xtalk.DistU, DistPl); % the distortion function U = UO *interpl{DistUl,Q,abs(U/UO)) .~ sign(U); % Perform the distortion plot(DistUl,Q); shg; %pause %______________________________________________________________________________ ______________________ function [X,f] = CalcSpec(U,t);
_______________________________________________________________________________ _____________________ calculate the spectral density of a signal, when it would be'measured' at specified resolution bandwidth RBW=1E3; %RBW: the desired resolution for the spectrum of U
N = length(U); m N/2;
dT = t(2)t( 1 ); % time spacing dF = 1 /dT/N; % frequency spacing f [O:m1]'*dF; % all positive frequencies X = fft(U)*dT; % to frequency domain X = abs(X(2:m+1)); % No DC and no negative frequencies.
X = sqrt(CalcSmooth(X.*X, f, RBW)); % average it over bandwidth RBW
%______________________________________________________________________________ ______________________ function [X,fJ = CaIcNBSV(U,t);
%______________________________________________________________________________ ______________________ calculate the narrow band signal voltage of a signal, when it would be'measured' at specified resolution bandwidth RBW=1E3; %RBW: the desired resolution for the spectrum of U
N = length(U); m=N/2;
dT = t(2)t( 1 ); % time spacing dF = 1/dT/N; % frequency spacing f= [O:m1]'*dF; % all positive frequencies %X = fft(U)*dT * sqrt(dF); % to frequency domain %X = fft(IT)/N; % to frequency domain X = fft(LT)/N*2; % to frequency domain X = abs(X(2:m+1)); % No DC and no negative frequencies.
%X = sqrt(CalcSmooth(X.*X, f, RBW)); % average it over bandwidth RBW
%______________________________________________________________________________ ______________________ function [CF] = CalcCrest(U) _______________________________________________________________________________ _____________________ Calculate the Crest Factor of a signal (U(t), which is the peak value divided by the rmsvalue Urms = sqrt(sum(U.~2)/length(U));
Upeak = max(abs(U));
CF = Upeak/Urms;
%____________________________________________,_________________________________ _______________________ function [F]=CalcEnhancedGaussDistribution(x,Cf);
%______________________________________________________________________________ _______________________ Generate a Cumulative distribution function F(x) that is identified as % "enhanced gaussian distribution"
Cf = crest factor Alpha = 1 e3;
Sigma = sqrt( (1+Alpha)  Cf~2 * Alpha/3);
x = x .* (x>0) .* ( x<Cf) + Cf * (x>=Cfj;
denominator = Alpha + erf(Cf/(sqrt(2)*Sigma));
F = 1  (Alpha * x/Cf + erf(x/(sqrt(2)*Sigma)))/denominator;
%___________________________________________________________________________________________________ function [DistP, DistU,P] = CalcDistrib(U) %______________________________________________________________________________ ______________________ calculate the amplitude distribution of signal U
N = length(U);
Nbins=100;
[cumbin,xx] = hist(U,Nbins);
dX=xx(3)xx(2);
DistP = cumbin(:)/N/dX; % force sum(DistP)*dX
DistU = xx(:);
Urms=sqrt(sum(U. *U)/N);
P=exp(0.5*(DistU/Urms).~2); P=P/sum(P)/dX;
DistU=[DistU,DistU];
DistP=[DistP,P];
_______________________________________________________________________________ _______________,_____ function [DistP, DistU] = CalcCumDistrib(U) %______________________________________________________________________________ ______________________ calculate the (backward) cumulative amplitude distribution of signal U
len = length(U);
Ueff = sqrt(sum(U .* U)/length(U));
U = abs(U/Ueff);
 evaluate distribution function Nbins = min([SO,floor(len/10)]);
[cumbin,xx] = hist(U,Nbins);
BinWidth=xx(2)  xx(1);
DistU = xx  BinWidth/2; % shift for n = [Nbins1:1:1 ]; cumbin(n) = cumbin(n) + curnbin(n+1 ); end DistP = cumbin/len;
%  improve numerical stability for other routines, when they use this result DistU = [ 0 , DistU(2:end)]; % start at x = 0 DistP = [DistP, 1/len];
DistU = [DistU,xx(Nbins) + 0.999 * BinWidth/2]; % add final (single) point DistP = [DistP, 1e100];
DistU = [DistU,xx(Nbins) + (1.001) * BinWidth/2]; % factor 1.001 for stability ___________________________________________________________________________________________________ function [PSD,freq]=CalcSmooth(PSD,freq,RBW) %_________________________________________________,____________________________ _______________________ % Imitate a real Spectrum Analyzer, with finite resolution bandwidth, and Gaussian shaped band filters PSD = "power spectral density" which is de square of the "spectral density";
in Volts per square Hertz.
%______________________________________________________________________________ _______________________ N = length(PSD);
df = freq(2)freq( I );
br = 3 * floor(RBW/dfj;
factor = 2*br + 1;
if (factor > 1) 5 ..ff= df * (br:br); % smooth interval ..mask = exp(f~*ff/(2*RBW~2));
..mask = mask/sum(mask); % Gaussian mask of resolution band filter ..xhelp = [PSD;zeros(2*br,l)];
1p = filter(mask,l,xhulp); % smart convolution 10 PSD = yhelp(br+1:endbr);
end;
%______________________________________________________________________________ ______________________ function [Uac,Uac rms]=CalcDemodulation(U,t,Shape,ToneNr);
%______________________________________________________________________________ _____________________ Demodulate the noise that has been modulated on the carriers of the discrete noise, it is for demonstration purposes only to prove that the discrete noise meets the predefined parameters.
20 % The demodulator uses synchronous detection, that is not locked in phase The consequence is an unknown attenuation over the full demodulation band.
This is corrected afterward by measuring the DC level, and amplify the demodulated signal until this DC level has been normalized to 1 Volt %
25 % PROOF: (psi is unknown) let Ur~cos(w*t+psi)*(1+Uac); % = carrier modulated with "1+Uac"
Uc =cos(w*t); % = carrier Ud =Urf~Uc; % = synchronous detected signal 30 % then Ud=1/2*(cos(psi+2*w*t)+cos(psi))*(1+Uac);
Ul~cos(psi)/2*(1+Uac); % after lowpas filtering % Udc=cos(psi)/2; %= by averaging Ulf % Uac=(Ulf/Udc)1;
_______________________________________________________________________________ _____________________ N=Shape.N;
Fc=Shape.Ingress.ToneF(ToneNr); %select carrier frequency Fc=Shape.dF * round(Fc/Shape.dF); %force an integer number of periods ModWidth=Shape.Ingress.ModWidth(ToneNr);
ModDepth=Shape.Ingress.ModDepth(ToneNr);
Ud=U.*cos(2*pi*Fc*t); % synchronous detection of modulated carrier Nm=round(l .l *ModWidth/Shape.dF/2); % calculate filter frequency mask=zeros(N,1); mask([l:Nm, NNm:N])=1; % create filter Ulf=real(ifft(fft(Ud).*mask)); % perform lowpass filtering Udc=sum(Ulf)!N; % find not normalized DC level Uac=Ulf/Udc1; % normalize overall level, and remove DC.
Uac rms=sqrt(sum(LTac.*Uac/N)); % must be equal to ModDepth, since Udc=1 Scale=Uac rms/ModDepth; % must be "one"
Claims (17)
1. A method of generating a signal having a defined envelope of spectral amplitudes, said method comprising the steps of:
 representing a first signal comprising a plurality of frequency components each having spectral amplitude and phase properties, and  processing said represented first signal by setting said spectral amplitude properties in accordance with the defined envelope of spectral amplitudes, and setting random phase properties;
 applying an amplitude distortion function to the first signal;
 frequency shaping the distorted first signal.
 representing a first signal comprising a plurality of frequency components each having spectral amplitude and phase properties, and  processing said represented first signal by setting said spectral amplitude properties in accordance with the defined envelope of spectral amplitudes, and setting random phase properties;
 applying an amplitude distortion function to the first signal;
 frequency shaping the distorted first signal.
2. A method according to claim 1, comprising iterating said steps of applying the amplitude distortion function, frequency shaping until a test of a shape of the amplitude distribution of the first signal indicates that the first signal satisfies an amplitude distribution requirement.
3. A method according to claim 2, comprising checking a crest factor requirement for the first signal after said frequency shaping and iterating said steps of applying the amplitude distortion function, frequency shaping until the first signal satisfies the crest factor requirement.
4. A method according to claim 1, wherein said first signal is represented by a set of numbers specifying a spectral amplitude and phase of each frequency component during said processing step wherein the spectral amplitude properties and the random phase properties are set.
5. A method according to claim 1, wherein said first signal is represented by a set of complex numbers having a real part and an imaginary part during said processing step wherein the spectral amplitude properties and the random phase properties are set, said real part and an imaginary part in combination specifying a spectral amplitude and phase of each frequency component.
6. A method according to any one of claims 1 to 5, further comprising the step of transforming said processed represented signal from the frequency domain into the time domain before applying an amplitude distortion function.
7. A method according to any one of claims 1 to 6, wherein the defined envelope of spectral amplitudes defines a signal with at least one modulated carrier, with a defined carrier frequency, carrier amplitude, modulation depth, and modulation width.
8. A computer program product having a memory with computer readable code embodied therein, comprising a set of instructions in code format and executable in a predetermined order on a processing device, wherein the instructions when executed by the processing device cause the processing device to execute the method of any one of claims 1 to 7.
9. A device comprising processing means, memory means and arbitrary wave generator means, arranged to generate the first signal according to the method of any one of claims 1 to 8.
10. A method of testing the operation of a communication system, said method comprising the steps of:
 generating a signal in accordance with the method of any one of claims 1 to 9, and  transferring said signal through said communication system.
 generating a signal in accordance with the method of any one of claims 1 to 9, and  transferring said signal through said communication system.
11. A system comprising  means for generating the first signal according to the method of any one of claims 1 to 10,  modem means,  cable means and  processor means, wherein said processor means are arranged for controlling said generating means, modem means and cable means for automated measurement and/or monitoring purposes.
12. A system according to claim 11, wherein the system is a telecommunication system.
13. A method according to claim 1, comprising  defining a predetermined time domain amplitude distribution;
 comparing a time domain amplitude distribution of said first signal with said predetermined time domain amplitude distribution prior to said applying of the amplitude distortion function, and  selecting a nonlinear transformation based on said comparing so that the application of the selected nonlinear transformation changes the first signal in a way that makes its time domain amplitude distribution approach said predetermined time domain amplitude distribution.
 comparing a time domain amplitude distribution of said first signal with said predetermined time domain amplitude distribution prior to said applying of the amplitude distortion function, and  selecting a nonlinear transformation based on said comparing so that the application of the selected nonlinear transformation changes the first signal in a way that makes its time domain amplitude distribution approach said predetermined time domain amplitude distribution.
14. A method according to claim 13, wherein applying the amplitude distortion function results in a signal g(t), computed as a function Q{f(t)) of said represented signal f(t) and wherein said function Q is defined as:
Q (x) = sign (x) .cndot. G1 (F ( ¦ x ¦ ) ) with: sign(x)=x/ ¦ x ¦ for x ~ 0; sign(x)=0 for x=0;
F being said time domain amplitude distribution of said represented signal; and G being said predetermined time domain amplitude distribution function.
Q (x) = sign (x) .cndot. G1 (F ( ¦ x ¦ ) ) with: sign(x)=x/ ¦ x ¦ for x ~ 0; sign(x)=0 for x=0;
F being said time domain amplitude distribution of said represented signal; and G being said predetermined time domain amplitude distribution function.
15. A method of testing the operation of a communication system having a modem, said method comprising the step of superposing on a signal transceived by said modem, a signal comprising at least one of a random noise signal and a discrete frequency spectrum, said random signal having an amplitude distribution in the time domain according to a predetermined quality criterion and having a spectral density in the frequency domain according to a predetermined quality criterion, said noise signal furthermore being composed of an array of random numbers.
16. A method according to claim 1, wherein the quality of operation of a communication system having a modem is tested, said method comprising the steps of:
 superposing the first signal on a signal transceived by said modem;
 evaluating said transceived signal according to a predetermined quality criterion.
 superposing the first signal on a signal transceived by said modem;
 evaluating said transceived signal according to a predetermined quality criterion.
17. A method according to claim 16, comprising iteratively arranging the design of said modem in order to approach closer to said quality criterion for evaluating said transceived signal.
Priority Applications (5)
Application Number  Priority Date  Filing Date  Title 

EP00202378.6  20000707  
EP20000202378 EP1189381B1 (en)  20000707  20000707  Random test signal for communication systems 
EP01200038.6  20010108  
EP01200038  20010108  
PCT/EP2001/007833 WO2002005473A1 (en)  20000707  20010706  A method of and a device for generating a signal having a predetermined quality criterion for use in or on a communication system 
Publications (2)
Publication Number  Publication Date 

CA2408809A1 CA2408809A1 (en)  20020117 
CA2408809C true CA2408809C (en)  20120911 
Family
ID=26072458
Family Applications (1)
Application Number  Title  Priority Date  Filing Date 

CA2408809A Expired  Lifetime CA2408809C (en)  20000707  20010706  A method of and a device for generating a signal having a predetermined quality criterion for use in or on a communication system 
Country Status (8)
Country  Link 

US (2)  US7729417B2 (en) 
EP (1)  EP1302015B1 (en) 
JP (2)  JP2004503177A (en) 
CN (2)  CN101674172B (en) 
AU (1)  AU2001289631A1 (en) 
CA (1)  CA2408809C (en) 
HK (1)  HK1058734A1 (en) 
WO (1)  WO2002005473A1 (en) 
Families Citing this family (10)
Publication number  Priority date  Publication date  Assignee  Title 

AU2001289631A1 (en) *  20000707  20020121  Koninklijke Kpn N.V.  A method of and a device for generating a signal having a predetermined quality criterion for use in or on a communication system 
WO2004071004A1 (en) *  20030204  20040819  Tracespan Communications Ltd.  Nonintrusive modem performance analysis 
EP1703763A1 (en) *  20050318  20060920  Alcatel  Communication interface and testing method therefore 
DE102007038337A1 (en) *  20070814  20090219  Rohde & Schwarz Gmbh & Co. Kg  Method for testing devices for a mobile radio system, signal generator, device for a mobile radio system and measuring system 
US8238450B2 (en) *  20090130  20120807  Futurewei Technologies Inc.  Dynamic transmitter noise level adjustment for digital subscriber line systems 
TWI419475B (en) *  20100402  20131211  Faraday Tech Corp  Test system and method for analogtodigital converter 
US8749248B2 (en)  20110406  20140610  ConSonics, Inc.  Shielding flaw detection and measurement in quadrature amplitude modulated cable telecommunications environment 
US9491027B2 (en)  20110406  20161108  Comsonics, Inc.  Miniature mobile marker system and method 
CN104901915B (en) *  20150507  20180515  北京邮电大学  A kind of communication means and device for supporting multiuser 
EP3174210B1 (en) *  20151124  20220518  Nxp B.V.  A data processor 
Family Cites Families (7)
Publication number  Priority date  Publication date  Assignee  Title 

GB2031197B (en) *  19781006  19821215  Marconi Instruments Ltd  Pseudorandom noise generator 
US4317214A (en) *  19800714  19820223  Attinello John S  Apparatus for simulating interference transmissions 
US6445733B1 (en) *  19971003  20020903  Conexant Systems, Inc.  Method of and apparatus for performing line characterization in a nonidle mode in a subscriber line communication system 
FR2783374B1 (en) *  19980911  20001208  Thomson Csf  METHOD AND DEVICE FOR GENERATING A RANDOM SIGNAL AND DIGITALANALOG CONVERSION SYSTEMS USING SUCH A RANDOM SIGNAL 
CN1157954C (en) *  19980929  20040714  皇家菲利浦电子有限公司  Conversion of coded video data 
EP1303962B1 (en) *  20000616  20080227  Inari, Inc.  Slidingwindow processing for the reception of multicarrier signals 
AU2001289631A1 (en) *  20000707  20020121  Koninklijke Kpn N.V.  A method of and a device for generating a signal having a predetermined quality criterion for use in or on a communication system 

2001
 20010706 AU AU2001289631A patent/AU2001289631A1/en not_active Abandoned
 20010706 WO PCT/EP2001/007833 patent/WO2002005473A1/en active Application Filing
 20010706 CN CN200910145437.3A patent/CN101674172B/en not_active Expired  Lifetime
 20010706 CA CA2408809A patent/CA2408809C/en not_active Expired  Lifetime
 20010706 CN CN018124178A patent/CN1440603B/en not_active Expired  Lifetime
 20010706 EP EP01969351.4A patent/EP1302015B1/en not_active Expired  Lifetime
 20010706 JP JP2002509216A patent/JP2004503177A/en active Pending
 20010706 US US10/311,161 patent/US7729417B2/en not_active Expired  Fee Related

2004
 20040301 HK HK04101494.5A patent/HK1058734A1/en not_active IP Right Cessation

2007
 20071228 JP JP2007338507A patent/JP4335279B2/en not_active Expired  Lifetime

2010
 20100414 US US12/759,894 patent/US7957460B2/en active Active
Also Published As
Publication number  Publication date 

JP2008219864A (en)  20080918 
JP2004503177A (en)  20040129 
US20030174765A1 (en)  20030918 
US7729417B2 (en)  20100601 
AU2001289631A1 (en)  20020121 
JP4335279B2 (en)  20090930 
HK1058734A1 (en)  20040528 
US20100197240A1 (en)  20100805 
WO2002005473A1 (en)  20020117 
EP1302015B1 (en)  20170830 
CN101674172B (en)  20160803 
CN1440603B (en)  20110713 
CN101674172A (en)  20100317 
US7957460B2 (en)  20110607 
CN1440603A (en)  20030903 
WO2002005473A9 (en)  20020919 
CA2408809A1 (en)  20020117 
EP1302015A1 (en)  20030416 
Similar Documents
Publication  Publication Date  Title 

US7957460B2 (en)  Method of and a device for generating a signal having a predetermined quality criterion for use in or on a communication system  
US6697768B2 (en)  Adaptive method and apparatus for transmission line analysis  
US5748001A (en)  Method and apparatus for fast response and distortion measurement  
LagunaSanchez et al.  On the use of alphastable distributions in noise modeling for PLC  
WO2006110403A2 (en)  Doubleended line probing (delp) for dsl systems  
Verspecht et al.  Characterizing amplifier modulation distortion using a vector network analyzer  
Boets et al.  Preprocessing of signals for singleended subscriber line testing  
Amini et al.  Filterbank multicarrier reflectometry for cognitive live wire testing  
De Vito et al.  A Compressive SamplingBased Channel Estimation Method for Network Visibility Instrumentation  
EP1189381B1 (en)  Random test signal for communication systems  
Lindqvist et al.  Estimation of nonhomogeneous and multisection twistedpair transmissionline parameters  
US7133497B2 (en)  Use of lower bandwidth technology to determine channel capability for supporting higher bandwidth technology  
Cañete et al.  Cyclic signals and systems in power line communications  
Han et al.  Noise characterization and emulation for lowvoltage power line channels between 150 kHz and 10 MHz  
Pagani et al.  Path identification in a powerline network based on channel transfer function measurements  
CN105763226A (en)  Optimal wavelet basis medium voltage carrier denoising method based on signaltonoise ratio  
Willink et al.  Validation of HF channel simulators  
Van Laere et al.  Development, validation and utilization of an ITUT G. 9903 PHY simulator for communication performance evaluation  
Lafata et al.  Analysis of simulation methods for farend crosstalk cancellation  
Perrett et al.  Verification of FPGA Generated SEFDM Signals  
Drakulić et al.  Physical Communication  
Geens et al.  NPR and cochannel distortion ratio: a happy marriage?  
Gori et al.  Measurement of the increase in delay distortion for ADSL splitters  
Hosseini et al.  Design and implementation of a kalman filterbased timevarying harmonics analyzer  
Arrabal et al.  System Parameters Effect on DMTBased Broadband Indoor Power Line Communications 
Legal Events
Date  Code  Title  Description 

EEER  Examination request  
MKEX  Expiry 
Effective date: 20210706 