Abstract Hospital global budget payment of national health insurance in Taiwan starts from July 1, 2002. The payment of emergency department has fixed point-value from July 1 , 2002 to December 31,2003 ( one point of hospital global budget payment equals one NT dollar) . But the payment of emergency department becomes floating point-value, shared with out patient department from January 1,2004. Hospital global budget payment of national health insurance in Taiwan suppressed excessive medical supply through the mechanism of floating point-value. The emergency department of hospital in Taiwan can¡¦t refused patient¡¦s visiting by the regulation of law. We collect the figure of people¡¦s medical utility in emergency department of one regional hospital with 653 beds in Kaohsiung city from July 1 , 2002 to December 31,2003.( one point of hospital global budget payment equals one NT dollar)then the same data also collected from January 1,2004.to July 31,2005.( floating point-value),analyzed by T test. The research result indicates that : the numbers of patient¡¦s transfer to emergency department increased (p value: ¡Õ¡Õ0.05), the numbers of triage1,2 at emergency department increased (p value : ¡Õ¡Õ0.05), the numbers of emergency visiting each month increased (p value:0.036), the growth of admission don¡¦t have statistic significance (p value:0.509). Key word: global budget payment¡Bfloating point-value¡Btransfer¡Btriage

Floating-point multipliers are typically power hungry which is undesirable in many embedded applications. This paper proposes a variable-latency floating-point multiplier architecture, which is suitable for low-power, high-performance, and high-accuracy applications. The architecture splits the significand multiplier into upper and lower parts, and predicts the required significand product and sticky bit from upper part. In the case of correct prediction, the computation of lower part is disabled and the rounding operation is significantly simplified so that floating-point multiplication can be completed early. Finally, detailed design and simulation of the floating-point multiplier is presented, together with its evaluation by comparing power consumption with the fast and conventional floating-point multipliers. Experimental results demonstrate that the proposed double-precision multiplier consumes up to 26.41% and 24.97% less power and energy than the fast floating-point multiplier respectively at the expense of only small area and delay overhead. In addition, the results also show that the performance of proposed floating-point multiplier is very approximate to that of fast floating-point multipliers.

Floating-point arithmetic is attractive for the implementation for a variety of Digital Signal Processing (DSP) applications because it allows the designer and user to concentrate on the algorithms and architecture without worrying about numerical issues. In the past, many DSP applications used fixed point arithmetic due to the high cost (in delay, silicon area, and power consumption) of floating-point arithmetic units. In the realization of modern general purpose processors, fused floating-point multiply add units have become attractive since their delay and silicon area is often less than that of a discrete floating-point multiplier followed by a floating point adder. Further the accuracy is improved by the fused implementation since rounding is performed only once (after the multiplication and addition). This work extends the consideration of fused floating-point arithmetic to operations that are frequently encountered in DSP. The Fast Fourier Transform is a case in point since it uses a complex butterfly operation. For a radix-2 implementation, the butterfly consists of a complex multiply and the complex addition and subtraction of the same pair of data. For a radix-4 implementation, the butterfly consists of three complex multiplications and eight complex additions and subtractions. Both of these butterfly operations can be implemented with two fused primitives, a fused two-term dot-product unit and a fused add-subtract unit. The fused two-term dot-product multiplies two sets of operands and adds the products as a single operation. The two products do not need to be rounded (only the sum is normalized and rounded) which reduces the delay by about 15% while reducing the silicon area by about 33%. For the add-subtract unit, much of the complexity of a discrete implementation comes from the need to compare the operand exponents and align the significands prior to the add and the subtract operations. For the fused implementation, sharing the comparison and alignment greatly reduces the complexity. The delay and the arithmetic results are the same as if the operations are performed in the conventional manner with a floating-point adder and a separate floating-point subtracter. In this case, the fused implementation is about 20% smaller than the discrete equivalent. / text

Floating-point computer arithmetic units are used for modern-day computers for 2D/3D graphic and scientific applications due to their wider dynamic range than a fixed-point number system with the same word-length. However, the floating-point arithmetic unit has larger area, power consumption, and latency than a fixed-point arithmetic unit. It has become a big issue in modern low-power processors due to their limited power and performance margins. Therefore, fused architectures have been developed to improve floating-point operations. This dissertation introduces new improved fused architectures for add-subtract, sum-of-squares, and magnitude operations for graphics, scientific, and signal processing. A low-power dual-path fused floating-point add-subtract unit is introduced and compared with previous fused add-subtract units such as the single path and the high-speed dual-path fused add-subtract unit. The high-speed dual-path fused add-subtract unit has less latency compared with the single-path unit at a cost of large power consumption. To reduce the power consumption, an alternative dual-path architecture is applied to the fused add-subtract unit. The significand addition, subtraction and round units are performed after the far/close path. The power consumption of the proposed design is lower than the high-speed dual-path fused add-subtract unit at a cost in latency; however, the proposed fused unit is faster than the single-path fused unit. High-performance and low-power floating-point fused architectures for a two-term sum-of-squares computation are introduced and compared with discrete units. The fused architectures include pre/post-alignment, partial carry-sum width, and enhanced rounding. The fused floating-point sum-of-squares units with the post-alignment, 26 bit partial carry-sum width, and enhanced rounding system have less power-consumption, area, and latency compared with discrete parallel dot-product and sum-of-squares units. Hardware tradeoffs are presented between the fused designs in terms of power consumption, area, and latency. For example, the enhanced rounding processing reduces latency with a moderate cost of increased power consumption and area. A new type of fused architecture for magnitude computation with less power consumption, area, and latency than conventional discrete floating-point units is proposed. Compared with the discrete parallel magnitude unit realized with conventional floating-point squarers, an adder, and a square-root unit, the fused floating-point magnitude unit has less area, latency, and power consumption. The new design includes new designs for enhanced exponent, compound add/round, and normalization units. In addition, a pipelined structure for the fused magnitude unit is shown. / text

The ePUMA architecture is a novel master-multi-SIMD DSP platform aimed at low-power computing, like for embedded or hand-held devices for example. It is both a configurable and scalable platform, designed for multimedia and communications. Numbers with both integer and fractional parts are often used in computers because many important algorithms make use of them, like signal and image processing for example. A good way of representing these types of numbers is with a floating-point representation. The ePUMA platform currently supports a fixed-point representation, so the goal of this thesis will be to implement twelve basic floating-point arithmetic operations and two conversion operations onto an already existing datapath, conforming as much as possible to the IEEE 754-2008 standard for floating-point representation. The implementation should be done at a low hardware and power consumption cost. The target frequency will be 500MHz. The implementation will be compared with dedicated DesignWare components and the implementation will also be compared with floating-point done in software in ePUMA. This thesis presents a solution that on average increases the VPE datapath hardware cost by 15% and the power consumption increases by 15% on average. Highest clock frequency with the solution is 473MHz. The target clock frequency of 500MHz is thus not achieved but considering the lack of register retiming in the synthesis step, 500MHz can most likely be reached with this design.

Processors used in lower-end scientific applications like graphic cards and video game consoles have IEEE single precision floating-point hardware [23]. Double precision offers higher precision at higher implementation cost and lower performance. The need for high precision computations in these applications is not enough to justify the use double precision hardware and the extra hardware complexity needed [23]. Native-pair arithmetic offers an interesting and feasible solution to this problem. This technique invented by T. J. Dekker uses single-length floating-point numbers to represent higher precision floating-point numbers [3]. Native-pair arithmetic has been proposed by Dr. William R. Dieter and Dr. Henry G. Dietz to achieve better accuracy using standard IEEE single precision floating point hardware [1]. Native-pair arithmetic results in better accuracy however it decreases the performance by 11x and 17x for addition and multiplication respectively [2]. The proposed implementation uses a residual register to store the error residual term [2]. This addition is not only cost efficient but also results in acceptable accuracy with 10 times the performance of 64-bit hardware. This thesis demonstrates the implementation of a 32-bit floating-point unit with residual register and estimates the hardware cost and performance.

Thesis (M.S.)--Ohio University, June, 1994. / Title from PDF t.p.

Thesis (M.S.)--State University of New York at Binghamton, Department of Computer Science, 2005. / Includes bibliographical references.

The effects of using limited precision floating point for intermediate storage in an embedded MP3 decoder are investigated in this thesis. The advantages of using limited precision is that the values need shorter word lengths and thus a smaller memory for storage. The official reference decoder was modified so that the effects of different word lengths and algorithms could be examined. Finally, a software and hardware prototype was implemented that uses 16-bit wide memory for intermediate storage. The prototype is classified as a limited accuracy MP3 decoder. Only layer III is supported. The decoder could easily be extended to a full precision MP3 decoder if a corresponding increase in memory usage was accepted.

This master thesis consists of implementation and evaluation of an AEC, Acoustic Echo Canceller, algorithm in a floating-point architecture. The most important question this thesis will try to answer is to determine benefits or drawbacks of using a floating-point architecture, relative a fixed-point architecture, to do AEC. In a telephony system there is two common forms of echo, line echo and acoustic echo. Acoustic echo is introduced by sound emanating from a loudspeaker, e.g. in a handsfree or speakerphone, being picked up by a microphone and then sent back to the source. The problem with this feedback is that the far-end speaker will hear one, or multiple, time-delayed version(s) of her own speech. This time-delayed version of speech is usually perceived as both confusing and annoying unless removed by the use of AEC. In this master thesis the performance of a floating-point version of a normalized least-mean-square AEC algorithm was evaluated in an environment designed and implemented to approximate live telephony calls. An instruction-set simulator and assembler available at the initiation of this master thesis were extended to enable; zero-overhead loops, modular addressing, post-increment of registers and register-write forwarding. With these improvements a bit-true assembly version was implemented capable of real-time AEC requiring 15 million instructions per second. A solution using as few as eight mantissa bits, in an external format used when storing data in memory, was found to have an insignificant effect on the selected AEC implementation’s performance. Due to the relatively low memory requirement of the selected AEC algorithm, the use of a small external format has a minor effect on the required memory size. In total this indicates that the possible reduction of the memory requirement and related energy consumption, does not justify the added complexity and energy consumption of using a floating-point architecture for the selected algorithm. Use of a floating-point format can still be advantageous in speech-related signal processing when the introduced time delay by a subband, or a similar frequency domain, solution is unacceptable. Speech algorithms that have high memory use and small introduced delay requirements are a good candidate for a floating-point digital signal processor architecture.