摘要:
A CORDIC unit for the iterative approximation of a vector rotation through a rotary angle θ by a number of elementary rotations through elementary angles αi, including elementary rotation stages for respectively affecting an elementary rotation through an elementary angle αi as an iteration step in the iterative approximation. After such an elementary rotation there remains a residual angle through which rotation is still to be affected. The elementary rotation stages of the CORDIC unit are adapted for rotation through elementary angles αi given by powers of two with a negative integral exponent. The CORDIC unit can also include a triggering device for triggering the elementary rotations, a triggering device which is adapted prior to each iteration step to compare the residual angle to at least one of the elementary angles and to omit those elementary rotation stages whose elementary angles are greater than the residual angle.
摘要:
A Field Programmable Gate Arrays (FPGA) design uses a Coordinate Rotation DIgital Computer (CORDIC) algorithm that can convert a Givens rotation of a vector to a set of shift-add operations. The CORDIC algorithm can be easily implemented in hardware architecture, therefore in FPGA. Since the computation of the inverse of the data correlation matrix involves a series of Givens rotations, the utility of the CORDIC algorithm allows a causal Constrained Energy Minimization (CEM) to perform real-time processing in FPGA. An FPGA implementation of the causal CEM is described and its detailed architecture is also described.
摘要:
The invention relates to continuous computing of an averaged value of a complex signal, in which values are produced by iterative processing, such as CORDIC processing, from digital complex input values of in-phase and quadrature components (si, sq) of the complex signal. The smoothed value is provided by processing the input values by two cascading CORDIC processing units with feedback, and a low-pass filtering contained implicitly therein.
摘要:
A dividing apparatus and method using coordinate rotation is disclosed. To this end, a plurality of rotation stages are sequentially performed until a divisor reaches a criterion and a rotation direction used in each of the plurality of stages is output. A division result acquired by performing rotation with respect to a dividend using the rotation direction for each of the plurality of stages is output.
摘要:
High quality epitaxial layers of monocrystalline materials can be grown overlying a monocrystalline substrate of a semiconductor structure by forming a compliant substrate for growing the monocrystalline layers. An accommodating buffer layer comprises a layer of monocrystalline oxide spaced apart from a silicon wafer by an amorphous interface layer of silicon oxide. The amorphous interface layer dissipates strain and permits the growth of a high quality monocrystalline oxide accommodating buffer layer. A semiconductor structure formed in accordance with this method includes a monocrystalline silicon substrate, a metal oxide semiconductor portion formed in the monocrystalline silicon substrate, and a compound semiconductor portion formed in the layer of monocrystalline compound semiconductor material. A circuit such as a microprocessor is formed in the complementary metal oxide semiconductor (CMOS) portion, and a coordinate rotation digital computer (CORDIC) functional unit formed in the compound semiconductor portion. The CORDIC algorithms are thus performed in a high speed compound semiconductor structure such as Gallium Arsenide (GaAs) which is integrated with a CMOS microprocessor on a common substrate.
摘要:
A fast CORDIC algorithm and the resulting VLSI architecture for the evaluation of trigonometric functions are disclosed. The new method employs signed digits to represent intermediate operands and requires a constant scaling or normalization factor which can be pre-computed and made available in read-only hardware for any desired target precision (i.e., word length). The speedup is achieved by performing CORDIC iterations in parallel in two separate modules. Each module executes a “double step” or two basic CORDIC rotations at every iteration cycle. Two angles arctan 2−2i and arctan 2−(2i+1) are used in each step i of the method. As a result, approximately {fraction (n/2)} steps (exactly ⌈ n + 3 2 ⌉ steps) are required to evaluate sine/cosine of n bit input argument up to n bits of accuracy. The VLSI architecture consists of two “zeroing” and two “rotator” modules, each consisting of signed digit adders, latches, shifters, etc. A novel decision block controls the iterations. The new ROM look-up table stores ⌈ n + 3 2 ⌉ values, each accurate to n+3 bits. For every pair of consecutive angles arctan 2−2i and arctan 2−(2i+1) for i=0, 1, 2, . . . , ⌈ n + 3 2 ⌉ only their sum and difference need be stored for the inventive double step branching CORDIC method. The result is a very fast VLSI architecture whose speedup to hardware-overhead ratio is close to 1. The double stepping method can be easily extended to the evaluation of inverse trigonometric, exponential as well as logarithm functions.
摘要:
An architecture in which a plurality of delay calculators are distributed throughout the beamformer. Each delay calculator provides beamforming delays and apodization values for a plurality of channels/elements. The delay calculators are initialized with element position coordinates. Then, while imaging, focal coordinates and steering angles are broadcast to all delay calculators. The delay calculator uses a cordic rotator to perform the aperture projections and calculate the hypotenuse. The cordic rotator performs successive coordinate transformations, in two dimensions, using coefficients that are simple powers of two. The resulting apodization values and time delays are then provided by a channel control bus to a plurality of channels.
摘要:
A 64-bit precision digital circuit for computing the exponential function and a related 64-bit precision digital circuit for computing sine and cosine, each circuit comprising a master circuit and a slave circuit. The master circuit computes the remainders x.sub.i for every "logical" iteration i using fast, low-precision circuit, thereby accumulating temporary errors. Only at the end of every 8 i's, which marks the end of a "physical" iteration, is a complete and fast correction to the accumulated errors performed. The slave circuit computes quantities called the y.sub.i 's, which will eventually converge to the desired output.
摘要翻译:用于计算指数函数的64位精密数字电路和用于计算正弦和余弦的相关64位精密数字电路,每个电路包括主电路和从电路。 主电路使用快速,低精度电路为每个“逻辑”迭代i计算余数+ E,otl x + EE i,从而累积临时错误。 只有在每8个我的结尾,这标志着“物理”迭代的结束,是对所执行的累积错误的完整和快速的校正。 从电路计算称为yi的数量,最终会收敛到期望的输出。
摘要:
A monolithic integrated circuit as may be used in combination with a plurality of sensors for generating respective sensor output signals, which monolithic integrated circuit includes means for converting each sensor output signal to bit-serial digital format, together with some initial processing circuitry comprising a bit-serial multiply-add processor. This processor includes a bit-serial digital multiplier for multiplying a first digital processor input signal in bit-serial form by a second digital processor input signal to generate a digital product signal, a digital adder for adding a third digital processor input signal to the digital product signal to generate a digital sum signal, and means for supplying a digital processor output signal with bits correspond-ing to those of said digital sum signal. A memory system provides memory for storing program instructions, memory for storing successive values of the second digital processor input signal, memory for storing successive values of the third digital processor input signal, and memory for storing successive values of the digital processor output signal as written into the memory system. The first digital processor input signal can be selected from among the sensor output signals as converted to bit-serial digital format. The second digital processor input signals applied to the bit-serial multiply-add processor are at least at selected times read from the memory system, as are the third digital processor input signals applied to the bit-serial multiply-add processor. A controller retrieves stored program instructions in a prescribed order from the memory for storing program instructions and generates control signals for controlling at least the reading and writing of the memory system, as well as the selecting of the first digital processor input signal.
摘要:
A trigonometric function arithmetic processor comprises a first arithmetic unit for executing, in m steps, a pseudo-division operation for obtaining from an initial value .theta. a sequence of numbers {ak} and a pseudoremainder .epsilon. which fulfill the following equation ##EQU1## where a.sub.k =+1 or -1, and the trigonometric function arithmetic processor also comprises a second arithmetic unit for executing the following pseudo-multiplication operation in m steps from initial values Xm=P and Ym=.epsilon..times.P (where P=constant) and the sequence of numbers {a.sub.k }, for k=m-1, m-2, . . . 1 and 0,X.sub.k-1 =X.sub.k -a.sub.k .times.2.sup.-2k .times.Y.sub.kY.sub.k-1 =(Y.sub.k +a.sub.k .times.X.sub.k)/2.sup.kso that X.sub.o =Q.times.cos .theta. nd Yo=Q.times.sin .theta. (Q=constant) are simultaneously obtained.