Redundant binary representation
 | This article needs additional citations for verification. (July 2008) |
The redundant binary representation (RBR) is a numeral system that uses more bits than needed to represent a single binary digit so that some numbers have several representations. RBR is unlike usual binary numeral systems, including two's complement, which use a single bit for each digit. RBR has many different properties when compared to regular binary representation systems. Most importantly, RBR allows carry-free addition[1], but makes bitwise logical operation slower. Usually, every bit has a sign that is not necessarily the same as the sign of the number represented. When digits have signs, the RBR is also a signed-digit representation.
Conversion from RBR
| Digit | Interpreted value |
|---|---|
| 00 | −1 |
| 01 | 0 |
| 10 | 0 |
| 11 | 1 |
RBR is a place-value notation system. In RBR, digits are pairs of bits, that is, for every place, RBR uses a pair of bits. The number represented by an RBR digit can be given using a translation table. This table indicates the mathematical value of each possible pair of bits.
As in conventional binary representation, the integer value of a given representation is a weighted sum of the values of the digits. The weight starts at 1 for the rightmost position and goes up by a factor of 2 for each next position. Usually, RBR allows negative values. There is no single sign bit that tells if a RBR represented number is positive or negative. Some integers have many possible representations in a RBR.
Not all RBR have the same properties. For example, using the translation table on the right, the number 1 can be represented in RBR in many ways: "01·01·01·11", "01·01·10·11", "01·01·11·00", "11·00·00·00". Also, for this translation table, flipping all bits corresponds to finding the additive inverse (multiplication by −1) of the integer represented.
An integer value can be converted back from RBR using the following formula, where n is the number of digit and dk is the interpreted value of the k-th digit, where k starts at 0 at the rightmost position:
Arithmetic operations
Inherently, the RBR needs different rule of arithmetic.
Addition
The addition operation in RBR is carry-free which means that the carry does not have to propagate through all the width of the addition unit. In effect, the addition in RBR is a constant-time operation. The addition will always take the same amount of time independent of the bit width of the operands. This does not imply that the addition is always faster in RBR than is two's complement representation, but that the addition will eventually be faster in RBR with increasing bit width because the two's complement addition unit's delay is proportional to log(n) (where n is the bit width)[2]. The addition in RBR take a constant time because each digit of the result can be calculated independently of one another, implying that each digit of the result can be calculated in parallel.
Subtraction
Subtraction is the same as the addition except that you need to find the additive inverse of one of the operand. The additive inverse is usually found using a translation table.
Logical operations
Implementing logical operations in RBR using digital logic is more complicated compared to the usual representations. The expected result of the bitwise AND operation on a pair of representations of 1 is expected to have value 1. Since there are many ways to represent 1 in RBR, it is not possible to simply use the basic logic gate AND between every digit. (While it is possible to do bitwise operations directly in RBR, it is not clear that this is a meaningful operation.) Assuming one wants the result to represent the same integer value as if the operation had been carried out using a standard binary representation, it is necessary to convert the two operands first to non-redundant representations. Consequently, logical operations are slower in RBR.