A logarithmic number system (LNS) is an arithmetic system used for representing
real number
In mathematics, a real number is a number that can be used to measure a continuous one- dimensional quantity such as a duration or temperature. Here, ''continuous'' means that pairs of values can have arbitrarily small differences. Every re ...
s in computer and
digital hardware
Digital electronics is a field of electronics involving the study of digital signals and the engineering of devices that use or produce them. It deals with the relationship between binary inputs and outputs by passing electrical signals through ...
, especially for
digital signal processing
Digital signal processing (DSP) is the use of digital processing, such as by computers or more specialized digital signal processors, to perform a wide variety of signal processing operations. The digital signals processed in this manner are a ...
.
Overview
A number,
, is represented in an LNS by two components: the
logarithm
In mathematics, the logarithm of a number is the exponent by which another fixed value, the base, must be raised to produce that number. For example, the logarithm of to base is , because is to the rd power: . More generally, if , the ...
(
) of its
absolute value
In mathematics, the absolute value or modulus of a real number x, is the non-negative value without regard to its sign. Namely, , x, =x if x is a positive number, and , x, =-x if x is negative (in which case negating x makes -x positive), ...
(as a binary word usually in
two's complement
Two's complement is the most common method of representing signed (positive, negative, and zero) integers on computers, and more generally, fixed point binary values. Two's complement uses the binary digit with the ''greatest'' value as the ''s ...
), and its
sign bit
In computer science, the sign bit is a bit in a signed number representation that indicates the sign of a number. Although only signed numeric data types have a sign bit, it is invariably located in the most significant bit position, so the term ...
(
):
:
An LNS can be considered as a
floating-point
In computing, floating-point arithmetic (FP) is arithmetic on subsets of real numbers formed by a ''significand'' (a Sign (mathematics), signed sequence of a fixed number of digits in some Radix, base) multiplied by an integer power of that ba ...
number with the
significand
The significand (also coefficient, sometimes argument, or more ambiguously mantissa, fraction, or characteristic) is the first (left) part of a number in scientific notation or related concepts in floating-point representation, consisting of its s ...
being always equal to 1 and a non-integer
exponent
In mathematics, exponentiation, denoted , is an operation involving two numbers: the ''base'', , and the ''exponent'' or ''power'', . When is a positive integer, exponentiation corresponds to repeated multiplication of the base: that is, i ...
. This formulation simplifies the operations of multiplication, division, powers and roots, since they are reduced down to addition, subtraction, multiplication, and division, respectively.
On the other hand, the operations of addition and subtraction are more complicated and are calculated by the formulae
:
:
where the "sum" function is defined by
, and the "difference" function by
. These functions
and
are also known as
Gaussian logarithms.
The simplification of multiplication, division, roots, and powers is counterbalanced by the cost of evaluating these functions for addition and subtraction. This added cost of evaluation may not be critical when using an LNS primarily for increasing the precision of floating-point math operations.
History
Logarithmic number systems have been
independently invented and published at least three times as an alternative to
fixed-point and
floating-point
In computing, floating-point arithmetic (FP) is arithmetic on subsets of real numbers formed by a ''significand'' (a Sign (mathematics), signed sequence of a fixed number of digits in some Radix, base) multiplied by an integer power of that ba ...
number systems.
Nicholas Kingsbury and Peter Rayner introduced "logarithmic arithmetic" for
digital signal processing
Digital signal processing (DSP) is the use of digital processing, such as by computers or more specialized digital signal processors, to perform a wide variety of signal processing operations. The digital signals processed in this manner are a ...
(DSP) in 1971.
A similar LNS named "signed logarithmic number system" (SLNS) was described in 1975 by Earl Swartzlander and
Aristides Alexopoulos; rather than use two's complement notation for the logarithms, they
offset them (scale the numbers being represented) to avoid negative logs.
Samuel Lee and Albert Edgar described a similar system, which they called the "Focus" number system, in 1977.
The mathematical foundations for addition and subtraction in an LNS trace back to
Zecchini Leonelli and
Carl Friedrich Gauss
Johann Carl Friedrich Gauss (; ; ; 30 April 177723 February 1855) was a German mathematician, astronomer, geodesist, and physicist, who contributed to many fields in mathematics and science. He was director of the Göttingen Observatory and ...
in the early 1800s.
Applications
In the late 1800s, the Spanish engineer
Leonardo Torres Quevedo
Leonardo Torres Quevedo (; 28 December 1852 – 18 December 1936) was a Spanish civil engineer, mathematician and inventor, known for his numerous engineering innovations, including Aerial tramway, aerial trams, airships, catamarans, and remote ...
conceived a series of
analogue calculating mechanical machines and developed one that could solve algebraic equations with eight terms, finding the roots, including the complex ones. One part of this machine called an "endless spindle" allowed the mechanical expression of the relation
,
[F. Thomas. ]
A Short Account on Leonardo Torres' Endless Spindle
', Mechanism and Machine Theory, Vol. 43, No. 8, pp. 1055-1063, 2008. with the aim of extracting the logarithm of a sum as a sum of logarithms.
A LNS has been used in the
Gravity Pipe (
GRAPE-5) special-purpose supercomputer
that won the
Gordon Bell Prize in 1999.
A substantial effort to explore the applicability of LNSs as a viable alternative to floating point for general-purpose processing of single-precision real numbers is described in the context of the ''European Logarithmic Microprocessor'' (ELM).
A fabricated prototype of the processor, which has a 32-bit cotransformation-based LNS
arithmetic logic unit
In computing, an arithmetic logic unit (ALU) is a Combinational logic, combinational digital circuit that performs arithmetic and bitwise operations on integer binary numbers. This is in contrast to a floating-point unit (FPU), which operates on ...
(ALU), demonstrated LNSs as a "more accurate alternative to floating-point", with improved speed. Further improvement of the LNS design based on the ELM architecture has shown its capability to offer significantly higher speed and accuracy than floating-point as well.
LNSs are sometimes used in
FPGA
A field-programmable gate array (FPGA) is a type of configurable integrated circuit that can be repeatedly programmed after manufacturing. FPGAs are a subset of logic devices referred to as programmable logic devices (PLDs). They consist of a ...
-based applications where most arithmetic operations are multiplication or division.
See also
*
Decibel
The decibel (symbol: dB) is a relative unit of measurement equal to one tenth of a bel (B). It expresses the ratio of two values of a Power, root-power, and field quantities, power or root-power quantity on a logarithmic scale. Two signals whos ...
*
Subnormal number
In computer science, subnormal numbers are the subset of denormalized numbers (sometimes called denormals) that fill the arithmetic underflow, underflow gap around zero in floating-point arithmetic. Any non-zero number with magnitude smaller than ...
*
Tapered floating point
In computing, tapered floating point (TFP) is a format similar to floating point, but with variable-sized entries for the significand and exponent instead of the fixed-length entries found in normal floating-point formats. In addition to this, ta ...
(TFP)
*
Level-index arithmetic (LI) and
symmetric level-index arithmetic (SLI)
*
Gaussian logarithm
*
Zech's logarithm
Zech logarithms are used to implement addition in finite fields when elements are represented as powers of a generator \alpha.
Zech logarithms are named after Julius Zech, and are also called Jacobi logarithms, after Carl G. J. Jacobi who used ...
*
ITU-T G.711
*
A-law algorithm
An A-law algorithm is a standard companding algorithm, used in European 8-bit PCM digital communications systems to optimize, i.e. modify, the dynamic range of an analog signal for digitizing. It is one of the two companding algorithms in th ...
*
μ-law algorithm
The μ-law algorithm (sometimes written Mu (letter), mu-law, often abbreviated as u-law) is a companding algorithm, primarily used in 8-bit PCM Digital data, digital telecommunications systems in North America and Japan. It is one of the two c ...
*
Slide rule
A slide rule is a hand-operated mechanical calculator consisting of slidable rulers for conducting mathematical operations such as multiplication, division, exponents, roots, logarithms, and trigonometry. It is one of the simplest analog ...
References
Further reading
* Previously published in:
* (NB. Describes a 13-bit LNS used in
Yamaha music synthesizers during the 1980s.)
*
*
*
Also reprinted in:
* {{cite book , editor-last1=Amir Sabbagh , editor-first1=Molahosseini , editor-first2=Leonel Seabra , editor-last2=de Sousa , editor3=Chip-Hong Chang , title=Embedded Systems Design with Special Arithmetic and Number Systems , publisher=
Springer International Publishing AG
Springer Nature or the Springer Nature Group is a German-British academic publishing company created by the May 2015 merger of Springer Science+Business Media and Holtzbrinck Publishing Group's Nature Publishing Group, Palgrave Macmillan, and Mac ...
, date=2017-03-21 , edition=1 , isbn=978-3-319-49741-9 , doi=10.1007/978-3-319-49742-6 , lccn=2017934074 , url=https://link.springer.com/book/10.1007/978-3-319-49742-6 (389 pages)
External links
A site that lists LNS papersesprit – European Logarithmic Microprocessor (formerly the 'High Speed Logarithmic Arithmetic' (HSLA) project)https://web.archive.org/web/20180707181644/http://www.ens-lyon.fr/LIP/Arenaire/Ware/FPLibrary/ --> A VHDL library for LNS hardware generationA Short Account on Leonardo Torres’ Endless Spindle
Computer arithmetic
Digital signal processing
Logarithms