High resolution phase to sine amplitude conversion

• US 4,905,177 A
• Filed: 01/19/1988
• Issued: 02/27/1990
• Est. Priority Date: 01/19/1988
• Status: Expired due to Term

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Qualcomm, Inc.
• Inventors: Weaver, Lindsay A. Jr., Kerr, Richard J.
• Primary Examiner(s): Harkcom, Gary V.
• Assistant Examiner(s): Nguyen, Long T.
• Application Number:
  US07/145,789
• Time in Patent Office: 770 Days
• Field of Search: 364/721, 364/724.01, 364/718, 364/724.06, 364/724.08, 332/1, 332/18
• US Class Current: 708/276
• CPC Class Codes: G06F 1/0356 by using two or more smaller tables, e.g. addressed by parts of the argument
• CPC Class Codes: G06F 2101/04 Trigonometric functions
• CPC Class Codes: G06F 2101/08 Powers or roots

Abstract

A method and apparatus for converting phase data into amplitude data. Input phase data is divided into upper and lower phase increments using an input splitter which can also fold in quadrant data. One or more storage elements, such as sine and cosine ROMs, are connected to the input splitter and generate an amplitude, which forms one of a plurality of series terms, and first and second derivatives of the amplitude in response to the upper phase increment. The first derivative and lower phase increment are multiplied together in a digital multiplier to produce a second series term. A third series term comprising a product of the amplitude and lower phase increment squared, is generated by a term generator connected to the splitter and to the storage element. The term generator can comprise elements such as a ROM addressable by the lower phase increment and the amplitude. The first, second and third series terms are subsequently added together by one or more digital adders connected to the multiplier, term generator, storage element and in series with each other to form an output amplitude. An output controller is used for transferring the output amplitude to other apparatus and for incorporating quadrant unfolding elements.

Claims

1. A method of converting phase data into amplitude data of a preselected periodic function, comprising the steps of:

splitting phase data, corresponding to an independent variable of a preselected periodic function, into an upper phase increment and a lower phase increment;

generating at least three Taylor series expansion terms relating to said periodic function from said upper and lower phase increments;

summing said series terms to form an output amplitude.

2. The method of claim 1 wherein said step of generating Taylor series terms from said upper and lower phase increments comprises:

generating at least three series terms with the Nth term of said series being generated according to the steps of;

generating an (N-1)th order derivative value corresponding to an (N-1)th order derivative of said periodic function evaluated at said upper phase increment;

raising said lower phase increment to an (N-1)th power;

forming an inverse of N-1 factorial; and

multiplying said (N-1)th order derivative value, raised lower phase increment and inverse of N-1 factorial together to form said Nth term.

3. The method of claim 2 wherein said step of generating three series terms from said upper and lower phase increments comprises generating at least three successive and immediately adjacent terms.

4. The method of claim 3 wherein said steps of generating a plurality of successive terms and summing said terms comprises:

generating an amplitude value corresponding to said periodic function evaluated at said upper phase increment, as the Nth series term, from upper phase increment;

generating a first derivative value corresponding to a first derivative of said periodic function evaluated at said upper phase increment from said upper phase increment;

generating to a second derivative of said periodic function evaluated at said upper phase increment from said upper phase increment;

multiplying said first derivative value and said lower phase increment together to produce (N+1)th series term; and

multiplying said second derivative value, said lower phase increment squared, and inverse of two factorial to generate an (N+2)th series term; and

summing said Nth, and (N+1)th, and (N+2)th series terms to form an output amplitude.

5. The method of claim 4 further comprising the steps of:

generating an (N+Y)th order derivative value corresponding to an (N+Y) order derivative of said periodic function evaluated at said upper phase increment from said upper phase increment, where Y is greater than 2;

raising said lower phase increment to an (N+Y)th power to form an (N+Y)th factor;

forming an (N+Y)th series term by multiplying said (N+Y)th factor, said (N+Y)th order derivative value, and an inverse of an (N+Y)th factorial; and

adding said (N+Y)th series term to said summer Nth, (N+1)th and (N+2)th series terms.

6. The method of claim 4 wherein said step of summing further comprises the steps of:

summing said (N+1)th and (N+2)th series terms to form a first addend; and

summing the first addend and said amplitude to form an output amplitude.

7. The method of claim 4 further comprising the steps of:

providing phase data in the range of 0 to 2π

;

detecting a quadrant position for said phase data; and

adjusting a sign of said output amplitude to reflect said detected quadrant.

8. The method of claim 4 wherein said steps of generating said amplitude value and said first derivative value comprise directly introducing said upper phase increment split from said phase data into a sine Read Only Memory and a cosine Read Only Memory respectively as data address inputs, said sine Read Only Memory having memory locations in which sine values are stored with each memory location having a unique sine value and data address and said cosine Read Only Memory having memory locations, in which cosine values are stored with each memory location having a unique cosine value and data address, said sine Read Only Memory and cosine Read Only Memory respectively providing said amplitude value and said first derivative value in response to said data adddress inputs.

9. The method of claim 4 wherein the step of generating a (N+2)th series term comprises the steps of:

introducing said lower phase increment split from said phase data value directly into a term Read Only Memory as a portion of a term Read Only Memory data address input; and

introducing said amplitude value directly into said term Read Only Memory as a remaining portion of said term Read Only Memory data address, said term Read Only Memory having memory locations in which (N+2)th series term are stored with each memory location having a unique (N+2)th series term and data address, said term Read Only Memory providing said (N+2)th series in response to said term Read Only Memory data address input.

10. The method of claim 2 wherein said step of generating three series terms comprises generating three alternating terms spaced apart by one term.

11. An apparatus for converting phase data into amplitude data of a preselected periodic function, comprising:

input means for splitting phase data corresponding to an independent variable of a preselected periodic function into an upper phase increment and a lower phase increment;

expansion means connected to said input means for generating at least three Taylor series expansion terms about said periodic function from said upper and lower phase increments;

adder means for summing said series terms to form an output amplitude.

12. The apparatus of claim 11 wherein said expansion means comprises:

series generation means for generating at least three successive series terms with the Nth terms of said series being defined as the product of an (N-1)th order derivative of said periodic function evaluated at said upper phase increment, said lower phase increment raised to an (N-1)th power, and an inverse of N-1 factorial.

13. The apparatus of claim 12 wherein said series generation means comprises:

storage means, connected to said input means, for storing for each upper phase increment an amplitude value, a first derivative value and a second derivative value of said periodic function each evaluated at a corresponding upper phase increment, said storage means for receiving an input upper phase increment and responsive thereto for providing a corresponding amplitude value, first derivative value and second derivative value with said amplitude value being a first (N) series terms where N=1,

multiplication means connected to said input means and said storage means for respectively receiving said lower phase increment and said first derivative value as evaluated at said upper phase increment to which said lower phase increment corresponds, for multiplying said lower phase increment and said first derivative value so as to produce a second (N+1) series term;

term means connected to said input means and said storage means for respectively receiving said lower phase increment and said second derivative value for generating a third (N+1) series term corresponding to a product of a square of said lower phase increment, said second derivative value and an inverse of 2!.

14. The apparatus of claim 13 wherein said storage means comprises:

amplitude storage means for storing said amplitude values and for providing in response to an input of said upper phase increment a corresponding amplitude value; and

derivative storage means for storing said first derivative values and for providing in response to an input of said upper phase increment a corresponding first derivative value of said periodic function, evaluated at said upper phase increment.

15. The apparatus of claim 14 wherein said amplitude storage means comprises a sine Read Only Memory storing predetermined sine values for said upper phase increment and said derivative storage means comprises a cosine Read Only Memory storing predetermined cosine values for said upper increment phase.

16. The apparatus of claim 13 wherein said term means comprises a series term Read Only Memory storing predetermined values for the product of the square of said lower phase increment, the second derivative value and the inverse of 2!.

17. The apparatus of claim 13 further comprising:

second storage means connected to said input means for generating an (N+Y)th order derivative of said periodic function, evaluated at said upper phase increment, in response to an input of said upper phase increment wherein Y>

2;

at least a second term means connected to said input means and to said second storage means for receiving said (N+Y)th order derivative of said preselected function, evaluated at said upper phase increment and said lower phase increment, for generating an (N+Y+1)th series term, corresponding to a product of said (N+Y)th order derivative of said lower phase increment raised to the (N+Y)th power and further divided by N+Y factorial and for providing said (N+Y)th series term to said adder means for summing with said other series terms.

18. The apparatus of claim 11 wherein said input means further comprises a quadrant folder for determining which of four quadrants said phase data corresponds and setting a corresponding quadrant value;

and said apparatus further comprises quadrant indication means connected to said quadrant folder and said adder means for respectively receiving said quadrant value and said output amplitude and responsive to said quadrant value for adjusting a sign of said output amplitude corresponding to the quadrant to which said phase data corresponds.

19. The apparatus of claim 11 wherein said input means is configured to separate phase data, corresponding to said independent variable of said periodic function, in the form of N-bit phase data words into upper phase increment data words of A bits and lower phase increment data words of N-A bits where A is a predetermined number of most significant bits of said N-bit phase data words.

20. The apparatus of claim 11 further comprising output means for transferring output amplitude data indicative of said periodic function to other apparatus.