<?xml version="1.0" encoding="UTF-8" ?>
<?xml-stylesheet type="text/xsl" href="https://community.arm.com/utility/feedstylesheets/rss.xsl" media="screen"?><rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:slash="http://purl.org/rss/1.0/modules/slash/" xmlns:wfw="http://wellformedweb.org/CommentAPI/"><channel><title>Convolution problem</title><link>https://community.arm.com/developer/tools-software/tools/f/armds-forum/2386/convolution-problem</link><description> Note: This was originally posted on 20th August 2013 at http://forums.arm.com When i obtain high pass filter using arm_conv_partial_f32 function instead of arm_fir_f32 function; i get same output array with opposite signs. This situation is not seen</description><dc:language>en-US</dc:language><generator>Telligent Community 10</generator><item><title>RE: Convolution problem</title><link>https://community.arm.com/thread/8253?ContentTypeID=1</link><pubDate>Wed, 11 Sep 2013 11:34:33 GMT</pubDate><guid isPermaLink="false">dd9e70c8-6d3c-4c71-b136-2456382a7b5c:431e016b-d7ae-4b92-a629-f9a867224b86</guid><dc:creator>deniz gurhan</dc:creator><description>&lt;div&gt;&lt;i&gt;Note: This was originally posted on 22nd August 2013 at &lt;a href="http://forums.arm.com"&gt;http://forums.arm.com&lt;/a&gt;&lt;/i&gt;&lt;br /&gt;&lt;br /&gt;&lt;span&gt;Of course, stupid of me. It makes sense. Thanks a lot Paul.&lt;/span&gt;&lt;br /&gt;&lt;br /&gt;&lt;blockquote&gt;&lt;br /&gt;I have a hunch what is happening.&amp;#160; The FIR filter function arm_fir_f32() requires that the filter coefficients be in time reversed order.&amp;#160; The arm_conv_partial_f32() has the coefficients in normal order.&amp;#160; In a high pass filter, the coefficients typically alternate in sign.&amp;#160; If the length of the filter is even then time flipping the filter is equivalent to scaling by -1.&lt;br /&gt;&lt;br /&gt;&lt;br /&gt;&lt;/blockquote&gt;&lt;/div&gt;&lt;div style="clear:both;"&gt;&lt;/div&gt;</description></item><item><title>RE: Convolution problem</title><link>https://community.arm.com/thread/8249?ContentTypeID=1</link><pubDate>Wed, 11 Sep 2013 11:34:33 GMT</pubDate><guid isPermaLink="false">dd9e70c8-6d3c-4c71-b136-2456382a7b5c:53331967-3b3c-4c35-b9a6-c56d22a29c63</guid><dc:creator>Dr. Paul Beckmann</dc:creator><description>&lt;div&gt;&lt;i&gt;Note: This was originally posted on 22nd August 2013 at &lt;a href="http://forums.arm.com"&gt;http://forums.arm.com&lt;/a&gt;&lt;/i&gt;&lt;br /&gt;&lt;br /&gt;&lt;span&gt;I have a hunch what is happening.&amp;#160; The FIR filter function arm_fir_f32() requires that the filter coefficients be in time reversed order.&amp;#160; The arm_conv_partial_f32() has the coefficients in normal order.&amp;#160; In a high pass filter, the coefficients typically alternate in sign.&amp;#160; If the length of the filter is even then time flipping the filter is equivalent to scaling by -1.&lt;/span&gt;&lt;br /&gt;&lt;br /&gt;&lt;blockquote&gt;&lt;br /&gt;When i obtain high pass filter using arm_conv_partial_f32 function instead of arm_fir_f32 function; i get same output array with opposite signs. This situation is not seen when i use low pass or mid pass filter. Anyone help?&lt;br /&gt;&lt;/blockquote&gt;&lt;/div&gt;&lt;div style="clear:both;"&gt;&lt;/div&gt;</description></item></channel></rss>