US 9,813,224 B2
Digital processor having instruction set with complex angle function
Kameran Azadet, San Ramon, CA (US)
Assigned to Intel Corporation, Santa Clara, CA (US)
Filed by Intel Corporation, Santa Clara, CA (US)
Filed on Apr. 17, 2014, as Appl. No. 14/255,491.
Claims priority of provisional application 61/812,858, filed on Apr. 17, 2013.
Prior Publication US 2014/0317376 A1, Oct. 23, 2014
Int. Cl. G06F 15/00 (2006.01); G06F 7/38 (2006.01); G06F 9/00 (2006.01); G06F 9/44 (2006.01); H04L 5/14 (2006.01); H04L 25/08 (2006.01); G06F 17/50 (2006.01); H04B 1/62 (2006.01); H04L 1/00 (2006.01); H04B 1/04 (2006.01); G06F 17/15 (2006.01); G06F 9/30 (2006.01); H04L 25/03 (2006.01); H04L 27/36 (2006.01); H04J 11/00 (2006.01); H04B 1/525 (2015.01)
CPC H04L 5/1461 (2013.01) [G06F 9/30036 (2013.01); G06F 17/15 (2013.01); G06F 17/50 (2013.01); G06F 17/5009 (2013.01); H04B 1/0475 (2013.01); H04B 1/525 (2013.01); H04B 1/62 (2013.01); H04J 11/004 (2013.01); H04L 1/0043 (2013.01); H04L 25/03012 (2013.01); H04L 25/03343 (2013.01); H04L 25/08 (2013.01); H04L 27/367 (2013.01); H04L 27/368 (2013.01); H04B 2001/0425 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method performed by a digital processor for evaluating a complex angle for an input value, x, comprising:
obtaining one or more complex angle software instructions having said input value, x, as an input;
in response to at least one of said complex angle software instructions, performing the following steps:
invoking at least one complex angle functional unit that implements said one or more complex angle software instructions to apply said complex angle function to said input value, x; and
generating an output corresponding to said complex angle of said input value, x, using one or more multipliers of a Multiply accumulate (MAC) unit of said digital processor,
wherein said complex angle software instruction is part of an instruction set of said digital signal processor, and
wherein said complex angle instruction computes the phase of an input complex number by successive approximation, computing a rotated version initialized to the input complex number at a first iteration, and at iteration k rotating the rotated signal from a previous iteration to produce a new rotated input.