Electrical computers and digital processing systems: support – Computer power control
Reexamination Certificate
1999-02-16
2001-08-28
Ray, Gopal C. (Department: 2181)
Electrical computers and digital processing systems: support
Computer power control
C713S322000, C708S322000, C714S047300
Reexamination Certificate
active
06282661
ABSTRACT:
TECHNICAL FIELD
This invention relates to dynamically reducing power consumption in integrated circuits and, more particularly, in digital signal processors or the like.
BACKGROUND OF THE INVENTION
General-purpose digital signal processors (DSPs) are used in low cost/low power applications, for example, in modems. Such modems implement more than a single protocol and as such employ software processes to implement these protocols on a programmable DSP. There is a trend in industry to move more of the processing in modem applications into the digital domain from the analog domain. This trend involves doing much of the signal processing in software on a general-purpose DSP core. The advantages of this approach are reduced size and cost resulting from increased levels of integration. Such integration minimizes the number of passive components and increases the number of protocols that can be implemented in a single implementation. However, the power consumption of the programmable DSP core dominates the power consumption of the modem and, therefore, becomes a critical design issue, especially for battery operated portable applications.
DSPs typically spend much of their operating cycles performing multiply-accumulate (MAC) operations to implement filters and these operations dominate the power consumption. The protocols being implemented are always defined in terms of a worst-case operating environment. This includes a worst-case model of the communication channel including worst-case interference, temperature, echo, etc. A model of this environment is used to determine the amount of digital signal processing that is needed to realize a required bit-error-rate for the modem being implemented. Any modem implementation must provide at least this level of performance.
It is well known that much of the DSP complexity in a modem, i.e., the operations that require the most machine cycles to implement, is in implementing the filters in the modem receiver for channel equalization and echo cancellation. The specifications of these filters, for example, A/D precision, number of taps, precision of taps, update algorithm, adaptation rate, etc., are set by the characteristics of the worst-case model of the channel.
Some studies have suggested that the specifications of the filters in hardware implementations can be relaxed in non-worst-case operating environments. For example, U.S. Pat. No. 5,777,914 to C. J. Nicol et al. discloses a hardware arrangement that monitors the signal to noise performance of an adaptive filter over a period of time to yield an average error value. This average error value is used to “scale-back” the precision of the filter tap coefficients in an adaptive manner that, in turn, reduces the power consumption in the filter because the filter response is represented with fewer bits. The signal to noise performance of the filter is reflected in the error used to update the filter coefficients. If the error is very small, the update rate can be reduced without impacting the receiver performance. See an article by C. J. Nicol et al. entitled “A low power 128-tap digital adaptive equalizer for broadband modems”,
IEEE Journal of Solid State Circuits,
Vol. 32, No. 11, November 1997, pp. 1777-1789. Furthermore, the number of taps in the filter can be reduced to reduce the number of multiplications required for equalization. See an article by J. T. Ludwig et al. entitled “Low power filtering using approximate processing for DSP applications”,
IEEE
1995
Custom Integrated Circuits Conference,
pp. 185-188. Although these techniques have been used for filter implementations in hardware, it has historically made little or no sense to use them in software modem implementations because the programmable DSP operates at a fixed frequency and provides adequate performance to implement the worst-case protocol. Indeed, in hardware implementations, the objective of these adaptive techniques is to minimize switching capacitance to reduce power.
SUMMARY OF THE INVENTION
Power consumption in program implemented circuits and the like is dynamically controlled in accordance with the circuit performance over time, not by reducing switched capacitance as was done in prior hardware circuit implementations but, in accordance with the invention, by dynamically reducing the number of machine cycles required to implement the desired circuit at an acceptable performance level.
Power consumption is dynamically reduced in program implemented circuits, for example, circuits implemented on a DSP, that include filters employed for channel equalization and for echo cancellation. This reduction in power usage results from monitoring the circuit performance over time and dynamically scaling back filter parameters during intervals that less than worst case performance is required. Scaling back of the DSP implemented filter parameters results in fewer machine cycles being required to effect the filters while maintaining adequate performance that, in turn, allows reduction in the DSP clock frequency, resulting in lower power consumption. The reduced DSP clock frequency also allows the DSP to be operated at a lower supply voltage that yields significant additional power savings.
REFERENCES:
patent: 5307003 (1994-04-01), Fairbanks et al.
patent: 5734598 (1998-03-01), Abbott et al.
patent: 5777914 (1998-07-01), Larsson et al.
patent: 5926640 (1999-07-01), Mason et al.
patent: 6078209 (2000-06-01), Linoff
patent: 0505884A (1992-09-01), None
patent: 0700171A2 (1996-03-01), None
J. Goodman et al., “An Energy/Security Scalable Encryption Processor Using an Embedded Variable Voltage DC/DC Coverter”, ISSCC98, 1998.
Z. Kostic, “Digital Signal Ptocessors in Cellular Radio Communications”, IEEE Communications Magazine, Dec. 1997, pp. 22-35.
K. Suzuki et al., “A 300MIPS/W RISC Core Processor with Variable Supply-Voltage Scheme in Variable Threshold-Voltage CMOS”, IEEE Custom Integrated Circuits Conference, 1997, pp. 587-590.
C. J. Nicol et al., “A Low-Power 128-Tap Digital Adaptive Equalizer for Broadband Modems”, IEEE Journal of Solid State Circuits, vol. 32, No. 11, Nov. 1997, pp. 1777-1789.
J. T. Ludwig et al., “Low Power Filtering Using Approximate Processing For DSP Applications”, IEEE 1995 Custom Integrated Circuits Conference, pp. 185-188.
G. Wei, “A Low Power Switching Power Supply for Self-Clocked Systems”, ISLPED96 Conference Proceedings, Aug. 1996, pp. 313-317.
W. Namgoong, “A High-Efficiency Variable Voltage CMOS Dynamic dc-dc Switching Regulator”, ISSCC97, Feb. 1997, pp. 380-381.
P. Larsson et al., “Self-Adjusting Bit-Precision for Low-Power Digital Filters”, 1997 Symposium on VLSI Circuits Digest of Technical Papers, 1997, pp. 123-124.
P. Lapsley et al., “DSP Processor Fundamentals”, IEEE Press, New York, N. Y., 1997, pp. 1-7.
Agere Systems Guardian Corp.
Ray Gopal C.
LandOfFree
Apparatus and method for adaptive reduction of power... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Apparatus and method for adaptive reduction of power..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and method for adaptive reduction of power... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2495755