Abstract: A method and system utilized by a multi-processor computer for storing exponent codes is provided. The multi-processor computer includes a plurality of processing elements. Each processing element executes a task having an instruction set. The processing elements are assigned a position in an execution sequence that defines the order in which the processing elements generate their respective outputs. Instruction sets are stored in a computer memory attached to the processing elements. The instruction sets are stored so that the processing elements produce outputs according to the execution schedule. The computer memory can be accessed using a two-dimensional addressing scheme where one dimension designates processing elements and the other designates instructions. Each instruction set includes one or more exponent codes that cause a processing element to raise an input value to a specified power.

Abstract: A computer processor that performs operations in a logarithmic number system (LNS) domain includes a log converter (20) which generates log signals, a data pipeline (22), a plurality of processing elements (231a-f) coupled to respective stages (24a-d) of the data pipeline, an inverse-log converter (28), and a programmable accumulator (232) that performs various summing operations to produce an output signal. An instruction, selected from a set of instructions, is decoded by a control unit (234) to configure the computer processor to perform operations on one or more data streams. Mathematical operations that can be performed by the processor include matrix multiplication, matrix-inversion, fast Fourier transforms (FFT), auto-correlation, cross-correlation, discrete cosine transforms (DCT), polynomial equations, and difference equations in general, such as those used to approximate infinite impulse response (IIR) and finite impulse response (FIR) filters.

Abstract: A computer processor that performs operations in a logarithmic number system (LNS) domain includes an input log converter (20), a feedback log converter (303), a first data pipeline (304), a second data pipeline (306), a plurality of processing elements (26a-f) coupled to respective stages of the data pipelines, an inverse-log converter (28), and a programmable accumulator (232) which produces output signals. An instruction, selected from a set of instructions, is decoded by a control unit (235) to configure the computer processor to perform operations on one or more data streams. Mathematical operations that can be performed by the processor include matrix multiplication, matrix-inversion, fast Fourier transforms (FFT), auto-correlation, cross-correlation, discrete cosine transforms (DCT), polynomial equations, and difference equations in general, such as those used to approximate infinite impulse response (IIR) and finite impulse response (FIR) filters.

Abstract: A circuit and method for computing an exponential signal x.sup.g is provided. The circuit includes a logarithm converter which converts an input signal to binary word that represents the logarithm of an input signal x. A first shift register shifts the binary word in a bit-wise fashion to produce a first intermediate value; while a second shift register shifts the binary word in a bit-wise fashion to produce a second intermediate value. The shift registers may be implemented using multiplexers. The shifting operations are equivalent to multiplying the intermediate values by a factor which is a power of two. The first intermediate value is either added to or subtracted from the second intermediate value to produce a combined value. An inverse-logarithm converter converts the combined value to the exponential signal.

Abstract: A computer processor for performing mathematical operations includes a logarithm converter which generates log values on a data bus, a plurality of processing elements connected to the data bus, an inverse-logarithm converter which receives values from the processing elements, an accumulator which sums converted values from the inverse-logarithm converter, and a control unit for configuring the accumulator to perform various summing operations. The computer processor also includes a switch for providing processor outputs as feedback. An instruction, selected from a set of instructions, is decoded by the control unit to configure the computer processor to perform operations on a data stream.

Abstract: A computer processor for performing mathematical operations includes a logarithm converter which generates log values on a data bus, a plurality of processing elements connected to the data bus, an inverse-logarithm converter which receives values from the processing elements, an accumulator which sums converted values from the inverse-logarithm converter, and a control unit for configuring the accumulator to perform various summing operations. The computer processor also includes a switch for providing processor outputs as feedback. An instruction, selected from a set of instructions, is decoded by the control unit to configure the computer processor to perform operations on a data stream.

Abstract: A method and system is described which allows execution overlap in a computer having a plurality of processing elements. The method and system provide an accumulation schedule based on the expected completion times of the processing elements. Outputs from the processing elements are accumulated according to the accumulation schedule. The accumulation schedule includes a plurality of accumulation flags which indicate when the outputs are to be accumulated.

Abstract: A computer processor that performs operations in a logarithmic number system (LNS) domain includes an input log converter (20), a feedback log converter (303), a first data pipeline (304), a second data pipeline (306), a plurality of processing elements (26a-f) coupled to respective stages of the data pipelines, an inverse-log converter (28), and a programmable accumulator (232) which produces output signals. An instruction, selected from a set of instructions, is decoded by a control unit (235) to configure the computer processor to perform operations on one or more data streams. Mathematical operations that can be performed by the processor include matrix multiplication, matrix-inversion, fast Fourier transforms (FFT), auto-correlation, cross-correlation, discrete cosine transforms (DCT), polynomial equations, and difference equations in general, such as those used to approximate infinite impulse response (IIR) and finite impulse response (FIR) filters.