Digital signal processor having instruction set with one or more non-linear functions using reduced look-up table with exponentially varying step-size

Abstract

A digital signal processor and method are disclosed having an instruction set with one or more non-linear functions using a look-up table of reduced size and exponentially varying step-sizes. A digital signal processor evaluates a non-linear function for a value, x, by obtaining at least two values from at least one look-up table for the non-linear function that are near the value, x, wherein the at least one look-up table stores a subset of values for the non-linear function using exponentially-varying step sizes; and interpolating the at least two obtained values to obtain a result, y. A position of a leading zero in the value, x, can be used as an index into the at least one look-up table. The interpolation can comprise, for example, a linear interpolation or a polynomial interpolation. A modulo arithmetic operation can optionally be employed for a periodic non-linear function.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

The present application is related to U.S. patent application Ser. No. 12/324,926, entitled “Digital Signal Processor Having Instruction Set with One or More Non-Linear Complex Functions;” U.S. patent application Ser. No. 12/324,927, entitled “Digital Signal Processor Flaying Instruction Set with One or More Non-Linear Functions Using Reduced Look-Up Table;” and U.S. patent application Ser. No. 12/324.934, entitled “Digital Signal Processor with One or More Non-Linear Functions Using Factorized Polynomial Interpolation;” each filed contemporaneously herewith and incorporated by reference herein.

FIELD OF THE INVENTION

The present invention is related to digital signal processing techniques and, more particularly, to techniques for digital processing of non-linear functions.

BACKGROUND OF THE INVENTION

Digital signal processors (DSPs) are special-purpose processors utilized for digital processing. Signals are often converted from analog form to digital form, manipulated digitally, and then converted back to analog form for further processing. Digital signal processing algorithms typically require a large number of mathematical operations to be performed quickly and efficiently on a set of data.

DSPs thus often incorporate specialized hardware to perform software operations that are often required for math-intensive processing applications, such as addition, multiplication, multiply-accumulate (MAC) and shift-accumulate. A Multiply-Accumulate architecture, for example, recognizes that many common data processing operations involve multiplying two numbers together, adding the resulting value to another value and then accumulating the result. Such basic operations can be efficiently carried out utilizing specialized high-speed multipliers and accumulators.

DSPS, however, generally do not provide specialized instructions to support complex mathematical functions, such as exp, log, cos, 1/x and x^K. Increasingly however, there is a need for complex arithmetic operations in processors. A nonlinear function is any problem where the variable(s) to be solved for cannot be written as a linear sum of independent components. A complex number can be written in the form a+bi, where a and b are real numbers, referred to as the real part and the imaginary part of the complex number, respectively. Many applications require complex arithmetic operations, such as complex addition or complex multiplications.

If supported at all, a DSP supports a non-linear function by using a large table look-up (LUT). An exemplary LUT may store on the order of 2,000 16 bit values, and thus require 32 kilobits of random access memory (RAM). The LUT is typically implemented in a separate dedicated SRAM (so that data and the non-linear LUT can be accessed at the same time to achieve improved performance).

In cases where the DSP is based on VLIW (Very Long Instruction Word) or SIMD (Single Instruction Multiple Data) architectures with N issues slots, the memory size becomes even larger. The LUT must be replicated N times because each issue slot must be able to read different values in the look-up table simultaneously, as the values of the data in each issue slot may be different. This replication of memory results in an even greater silicon area. For example, assuming a LUT in a 4-way vector co-processor, a memory size of 128 Kb is required (32 Kb×4). In addition, if different non-linear functions are required for different parts of a program being executed, the various LUTs must be loaded into memory, thereby significantly increasing latency and potentially reducing performance.

A need therefore exists for a digital signal processor having an instruction set that supports one or more non-linear functions using a look-up table of reduced size.

SUMMARY OF THE INVENTION

Generally, a digital signal processor and method are disclosed having an instruction set with one or more non-linear functions using a look-up table of reduced size and exponentially varying step-sizes. According to one aspect of the invention, a digital signal processor evaluates a non-linear function for a value, x, by obtaining two or more values from at least one look-up table for the non-linear function that are near the value, x, wherein the at least one look-up table stores a subset of values for the non-linear function using exponentially-varying step sizes; and interpolating the two or more obtained values to obtain a result, y.

According to a further aspect of the invention, a position of a leading zero in the value, x, can be used as an index into the at least one look-up table. The interpolation can comprise, for example, a linear interpolation or a polynomial interpolation. A modulo arithmetic operation can optionally be employed for a periodic non-linear function.

A more complete understanding of the present invention, as well as further features and advantages of the present invention, will be obtained by reference to the following detailed description and drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic block diagram of an exemplary digital signal processor that incorporates features of the present invention;

FIG. 2 is a schematic block diagram of an exemplary vector-based digital signal processor that processes one or more numbers simultaneously in accordance with an embodiment of the present invention;

FIG. 3 illustrates a plot of a non-linear function 300 that may be processed in accordance with the present invention;

FIG. 4 illustrates a plot of a non-linear function 400 that may be processed in accordance with the present invention; and

FIG. 5 is a flow chart illustrating an exemplary interpolation process for a non-linear function that employs exponential step-sizes in accordance with the present invention.

DETAILED DESCRIPTION

The present invention provides a digital signal processor having an instruction set that supports one or more non-linear functions using one or more look-up tables of reduced size. Generally, one or more look-up tables having exponentially varying step size store a subset of values for a non-linear function and an interpolator interpolates two or more values obtained from the look-up table(s) to obtain the result. As used herein, the term “digital signal processor” shall be a processor that executes instructions in program code. Further, a hard-wired logic implementation of digital signal processing functions is not considered herein.

In this manner, the present invention supports non-linear functions by using a smaller look-up table than required by conventional techniques. As previously indicated, an exemplary look-up table may store on the order of 2,000 16 bit values, and thus require 32 kilobits of random access memory (RAM). With the present invention, a smaller look-up table can be employed to store a subset of the 2,000 values, such as 128 values. An interpolator will interpolate two or more values obtained from the smaller look-up table(s) to obtain the result. As discussed further below in conjunction with FIG. 3, the interpolation can be a linear interpolation or a higher order interpolation such as quadratic or cubic interpolation.

FIG. 1 is a schematic block diagram of an exemplary digital signal processor 100 that incorporates features of the present invention. As shown in FIG. 1, the exemplary digital signal processor 100 includes one or more functional units 110 for non-linear functions. In addition, the digital signal processor 100 comprises one or more look-up tables 120 that store a subset of values for a non-linear function. The present invention recognizes that performance of the digital signal processor 100 may be improved by employing a technique for mapping the input data to non-uniformly spaced sub-samples stored in the look-up table 120, such as an exponentially varying step-size, as discussed further below in conjunction with FIGS. 4 and 5. Generally, the input data is mapped to nonuniformly spaced sub-samples, such as an exponential spacing, and then stored in the look-up table 120. The exemplary exponential spacing scheme performs efficiently for common exponent-based functions, such as 1/x, 1/sqrt(x), and ln(x).

An interpolator 130 interpolates two or more values obtained from a look-up table 120 to obtain the result. The interpolator 130 may also include a modulo arithmetic unit (not shown) to support periodic functions, such as sine and cosine (e.g., by wrapping the input angle of the trigonometric sine and cosine functions).

Thus, in implementing the non-linear function, the non-linear functional unit(s) 110 obtain two or more values from the subset of values stored in the look-up table(s) 120 and then initiate the interpolator 130 to interpolate the two or more obtained sub-sample values to obtain the desired result.

In various embodiments, the digital signal processor 100 may use hardware or a look-up table (or a combination thereof) to compute the non-linear values of the data. Generally, if the digital signal processor 100 is processing software code that includes a predefined instruction keyword corresponding to a non-linear math function and any appropriate operands for the function, the instruction decoder must trigger the appropriate non-linear functional units 110 that are required to process the instruction (e.g., connect two registers, apply one or more nonlinear mathematical operations on the values contained therein, and store the result in a third register). It is noted that a non-linear functional unit 110 can be shared by more than one instruction.

Generally, the present invention extends conventional digital signal processors to provide an enhanced instruction set that supports one or more non-linear functions using one or more look-up tables and an interpolation function. As shown in FIG. 1, a digital signal processor 100 in accordance with the present invention receives at least one number as an input, applies one or more non-linear functions to the input and generates an output value.

The disclosed digital signal processors may have a scalar architecture that processes a single number at a time, or a vector architecture, as discussed hereinafter in conjunction with FIG. 2, that processes one or more numbers simultaneously. In the case of a vector-based digital signal processor implementation, the input number is a vector comprised of a plurality of scalar numbers that are processed in parallel.

FIG. 2 is a schematic block diagram of an exemplary vector-based digital signal processor 200 that processes one or more numbers simultaneously in accordance with an embodiment of the present invention. Generally, the vector-based implementation of FIG. 2 increases the number of MIPs (instructions per second), relative to the scalar implementation of FIG. 1, by performing different processes concurrently. Thus, the vector-based digital signal processor 200 contains plural non-linear functional units 210-1 through 210-N. For example, a dual digital signal processor 200 contains two non-linear functional units 210-1 and 210-1 that are capable of performing two independent non-linear function operations concurrently.

As noted above, the input to the vector-based digital signal processor 200 is a vector, X, comprised of a plurality of scalar numbers, x_n, that are processed in parallel. For example, assume a vector-based digital signal processor 200 supports a non-linear function for a vector, X, where X is comprised of scalar numbers x₁ through x₄. The exemplary function may be expressed as follows:NL_vec4(x₁,x₂,x₃,x₄,type,mod),where type may be, for example, ‘sin’, ‘cos’, ‘log 10’, ‘exp’ or ‘sqrt’. Mod=‘2*π’ for a period of 2π, and ‘0’ for non-periodic functions.

As previously indicated, an interpolator 130 interpolates two or more values obtained from the look-up table 120 to obtain the result.

Interpolation

In one embodiment, the interpolator 130 employs a linear interpolation, such as a linear spline, to interpolate two or more sub-sample values obtained from the look-up table 120 to obtain the result.

FIG. 3 illustrates a plot of a non-linear function 300 that may be processed in accordance with the present invention. As shown in FIG. 3, the non-linear function 300 is sub-sampled and the sub-sampled values, such as x₁ and x₂, can be stored in the look-up table 120. If the non-linear function 300 must be evaluated for a value, x, that is not stored in the look-up table 120, the digital signal processor 100 retrieves the two or more stored sub-sampled values, such as y₁ and y₂ that are stored for x₁ and x₂, and interpolates to find the value, y, for x, as follows:y=m(x−x₁)+y₁ where m equals (y₂−y₁)/(x₂−x₁).

It is noted that performance may optionally be further improved by employing one or more polynomial interpolation algorithms, such as a Taylor series, as described in U.S. patent application Ser. No. 12/324.934, filed contemporaneously herewith and entitled “Digital Signal Processor with One or More Non-Linear Functions Using Factorized Polynomial Interpolation,” incorporated by reference herein.

Exponentially Spaced Look-Up Table

As previously indicated, one or more look-up tables 120 can have exponentially varying step sizes between stored values for the non-linear function. FIG. 4 illustrates a plot of a non-linear function 400 that may be processed in accordance with the present invention. According to one aspect of the present invention, the step size between each successive stored value increases exponentially. Generally, for the function 400 shown in FIG. 4, with an exponential step size, additional granularity is obtained in the curved region closest to the y axis and less granularity is provided in the linear region as the x axis values increase.

If the non-linear function 400 must be evaluated for a value, x, that is not stored in the look-up table 120, the digital signal processor 100 retrieves two or more stored sub-sampled values, based on the leading zero in the binary representation of the value, x. As used herein, a leading zero is any zero that leads a number string with a non-zero value. For example, the 8 bit binary number 00100011 has two leading zeros (the two left zeros). This definition can be extended to negative twos complementary values, by counting the number of consecutive most To significant ones. The present invention recognizes that the number of leading zeros provides an indication of the magnitude of the corresponding number (i.e., how large or small in a logarithmic scale).

The leading zero can be used as an index into the look-up table 120 to obtain the two or more stored sub-sampled values.

FIG. 5 is a flow chart illustrating an exemplary interpolation process 500 for a non-linear function that employs exponential step-sizes in accordance with the present invention. The interpolation process 500 may be implemented, for example, by one or more functional units 210 from the exemplary vector-based digital signal processor 200 of FIG. 2.

As shown in FIG. 5, the exemplary interpolation process 500, initially employs the leading zero detection techniques during step 510, as described above, to compute the look-up table address. Thereafter, the appropriate parameters are read from the look-up table during step 520 (using non-uniform powers of 2 spacing). Finally, additional operations, such as interpolation, are optionally performed during step 530 in hardware.

CONCLUSION

While exemplary embodiments of the present invention have been described with respect to digital logic blocks and memory tables within a digital signal processor, as would be apparent to one skilled in the art, various functions may be implemented in the digital domain as processing steps in a software program, in hardware by circuit elements or state machines, or in combination of both software and hardware. Such software may be employed in, for example, a digital signal processor, application specific integrated circuit or micro-controller. Such hardware and software may be embodied within circuits implemented within an integrated circuit.

Thus, the functions of the present invention can be embodied in the form of methods and apparatuses for practicing those methods. One or more aspects of the present invention can be embodied in the form of program code, for example, whether stored in a storage medium, loaded into and/or executed by a machine, wherein, when the program code is loaded into and executed by a machine, such as a processor, the machine becomes an apparatus for practicing the invention. When implemented on a general-purpose processor, the program code segments combine with the processor to provide a device that operates analogously to specific logic circuits. The invention can also be implemented in one or more of an integrated circuit, a digital signal processor, a microprocessor, and a micro-controller.

It is to be understood that the embodiments and variations shown and described herein are merely illustrative of the principles of this invention and that various modifications may be implemented by those skilled in the art without departing from the scope and spirit of the invention.

Claims

1. A method performed by a vector-based digital signal processor for evaluating a non-linear function for a vector, x, comprising: receiving the vector, x, the vector including scalar numbers, an identifier and a parameter for the non-linear function;in response to a predefined software instruction keyword for said non-linear function, the vector-based digital signal processor processes the scalar numbers simultaneously by:invoking at least one hardware non-linear functional unit that implements said non-linear function, wherein said at least one hardware non-linear functional unit is configured to perform the following steps:obtaining at least two sets of values from at least one look-up table for said non-linear function that are near each scalar number of said vector, x, wherein said at least one look-up table stores a subset of values for said non-linear function using exponentially-varying step sizes; andinterpolating said at least two obtained values to obtain a vector result, y, wherein said predefined software instruction keyword for said non-linear function is part of an instruction set of said vector-based digital signal processor.

2. The method of claim 1, further comprising the step of using a position of one or more leading zeros in said scalar numbers to access said at least one look-up table.

3. The method of claim 1, wherein said interpolating step comprises a linear interpolation.

4. The method of claim 1, wherein said interpolating step comprises a polynomial interpolation.

5. The method of claim 1, further comprising a modulo arithmetic operation for a periodic non-linear function.

6. The method of claim 1, wherein said vector-based digital signal processor executes said predefined software instruction keyword from program code.

7. A vector-based digital signal processor, comprising: a circuitry to receive a vector including a set of scalar numbers, an identifier and a parameter for a non-linear function;at least one hardware non-linear functional unit that implements a non-linear function, wherein said hardware non-linear functional unit is invoked by said vector-based digital signal processor in response to a predefined software instruction keyword for said non-linear function, wherein said at least one hardware non-linear functional unit is configured to:obtain at least two values from at least one look-up table, the at least two values near each scalar number of said vector, x, wherein said at least one look-up table stores a subset of values for said non-linear function using exponentially-varying step sizes; andinterpolate said at least two values obtained from said at least one look-up table to obtain a vector result, y, wherein said predefined software instruction keyword for said non-linear function is part of an instruction set of said vector-based digital signal processor, wherein the set of scalar numbers are simultaneously processed by said at least one hardware non-linear functional unit.

8. The vector-based digital signal processor of claim 7, wherein a position of one or more leading zeros in said scalar numbers are used to access said at least one look-up table.

9. The vector-based digital signal processor of claim 7, wherein said hardware non-linear functional unit performs a linear interpolation.

10. The vector-based digital signal processor of claim 7, wherein said hardware non-linear functional unit performs a polynomial interpolation.

11. The vector-based digital signal processor of claim 7, further comprising a modulo arithmetic operation for a periodic non-linear function.

12. The vector-based digital signal processor of claim 7, wherein said vector-based digital signal processor executes said predefined software instruction keyword from program code.

13. An integrated circuit, comprising: a vector-based digital signal processor for evaluating a non-linear function for a vector, x, the vector including scalar numbers, an identifier and a parameter for the non-linear function, the vector-based digital signal processor, comprising:at least one hardware non-linear functional unit that implements a non-linear function, wherein said at least one hardware non-linear functional unit is invoked by said vector-based digital signal processor in response to a predefined software instruction keyword for said non-linear function, wherein said at least one hardware non-linear functional unit is configured to simultaneously:obtain at least two values from at least one look-up table near each scalar number of said vector, x, wherein said at least look-up table stores a subset of values for a non-linear function using exponentially-varying step sizes; andan interpolator that interpolates at least two values obtained from said at least one look-up table to obtain a result.

14. The integrated circuit of claim 13, wherein a position of one or more leading zeros in said scalar numbers are used to access said at least one look-up table.

15. The integrated circuit of claim 13, wherein said interpolator performs a linear interpolation.

16. The integrated circuit of claim 13, wherein said interpolator performs a polynomial interpolation.

17. The integrated circuit of claim 13, further comprising a modulo arithmetic operation for a periodic non-linear function.

18. The integrated circuit of claim 13, wherein said digital signal processor executes software instructions from program code.

19. The method of claim 1, wherein the parameter defines a period for periodic non-linear functions.

20. The method of claim 1, wherein the parameter includes a value of zero that defines a non-periodic and non-linear function.