Electrical computers and digital processing systems: multicomput – Computer-to-computer data routing – Least weight routing
Reexamination Certificate
1998-05-26
2003-11-04
Courtenay, III, St. John (Department: 2126)
Electrical computers and digital processing systems: multicomput
Computer-to-computer data routing
Least weight routing
Reexamination Certificate
active
06643680
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to an information processing system including a central processing unit (CPU) and a signal processor and, in particular, to a signal processing system including a signal processor which simultaneously processes a plurality of mutually different application programs (hereinafter referred to as “applications”) under the control of a CPU.
Signal processing systems of this type have been available, wherein, in general, a signal processor (DSP, RISC) for executing applications of communications, image processing, data collection, etc. is combined with a CPU (microcomputer etc.). Following the increasing processing capabilities of signal processors such as DSPs, such signal processing systems in which a plurality of applications are combined and simultaneously executed have been proposed and actually put to practical use. For example, the signal processing system includes a signal processor which combines an application for modem, an application for compression/decompression of audio signals, an application for audio equalizer and an application for image code, and processes them simultaneously. In such a signal processing system, a variety of user's requests for sounds, images and communications can be satisfied through the simultaneous processing of them.
In the foregoing signal processing systems, for ensuring real-time signal processing, it is necessary to manage the maximum signal processing capability (MIPS) of the signal processor, memory relocation therein and the like by the CPU or the signal processor having the high processing speed.
For explaining it in more detail, operations in a conventional signal processing system including a CPU and a signal processor will be explained. In this case, stored programs in the CPU are divided into an application program interface (API) section defining user's programs, that is, applications, and a combination of them, and a driver section for driving the signal processor.
The driver section executes applications of the signal processor and derives memory locations etc. so as to control programs on the side of the signal processor. More specifically, the driver section sends resource requests of the respective applications to be executed in the signal processor to the signal processor and, when a plurality of applications are executed in a combined fashion, the driver section refers to execution order scheduling information and a table representing the current state of the signal processor so as to carry out processes for the optimum scheduling and memory allocation, and so on. After these processes, the driver section controls a memory and ports of the signal processor, thereby to control the signal processor so as to satisfy requests from the API section.
On the other hand, the signal processor carries out, based on commands from the driver section, scheduling of a plurality of applications and memory allocation therefor so as to achieve real-time signal processing.
However, in the foregoing signal processing system wherein the plurality of applications are simultaneously processed, every time there raises an entry request or a de-entry request from an application, a combination of the applications is changed. Therefore, the CPU or the signal processor should perform processes, such as calculation of resources, scheduling adjustment and memory allocation, depending on the combination of the applications. Since these processes are carried out every time a combination of the applications changes, burden is large for the CPU or the signal processor. Further, this also prolongs time required for designing the whole signal processing system, thereby to cause an increase in cost.
More specifically, applications to be processed in the signal processing system of this type include, in general, applications for modem, applications for audio signal compression/decompression, applications for audio equalizer, applications for image code, and so on. Accordingly, combinations of the applications should be diversified and, as a result, the foregoing processes, such as the calculation of resources, the scheduling adjustment and the memory allocation, to be executed in the CPU or the signal processor become complicated.
SUMMARY OF THE INVENTION
It is therefore an object of the present invention to provide a signal processing system which can reduce burden applied to a CPU or a signal processor.
It is another object of the present invention to provide a signal processing system which can largely shorten time required for designing it and further contribute to reduction in cost.
It is still another object of the present invention to provide a signal processing system which can deal with a variety of combinations of applications.
Other objects of the present invention will become clear as the description proceeds.
According to one aspect of the present invention, there is provided a signal processing, system which comprises A table storing a plurality of pattern data corresponding to a plurality of combinations, respectively, each of which is of applications different from one another and is determined in accordance with a given condition, obtaining means connected to said table for obtaining, as an obtained datum, one of said pattern data with reference to said table in response to a request relating to the use of said applications, and executing means connected to said table and said obtaining means for simultaneously executing said applications in accordance with said obtained datum.
According to another aspect of the present invention, there is provided a signal processing system in which a plurality of mutually different applications are simultaneously executed. The signal processing system comprises a table storing a plurality of pattern data each representing a processing operation corresponding to each of combinations of the applications. The combinations are predetermined in accordance with a given condition. The signal processing system further comprises obtaining means connected to said table and responsive to a request of a change of the applications for obtaining the pattern datum corresponding to the combination of changed applications from said table and means connected to said obtaining means for carrying out the processing operation in accordance with the obtained pattern datum.
According to another aspect of the present invention, there is provided a signal processing method which comprises the steps of storing, on a table, a plurality of pattern data corresponding to a plurality of combinations, respectively, each of which is of applications different from one another and is determined in accordance with a given condition, obtaining, as an obtained datum, one of said pattern data with reference to said table in response to a request relating to a choice of said combinations, and simultaneously executing said applications in accordance with said obtained datum.
REFERENCES:
patent: 5479606 (1995-12-01), Gray
patent: 5758129 (1998-05-01), Gray
patent: 5805827 (1998-09-01), Chau et al.
patent: 5815686 (1998-09-01), Earl et al.
patent: 5893166 (1999-04-01), Frank et al.
patent: 6006314 (1999-12-01), Suzuki
patent: 6021408 (2000-02-01), Ledain et al.
patent: 06-282446 (1994-10-01), None
Kakisada Emi
Kato Miki
Courtenay III St. John
Foley & Lardner
NEC Electronics Corporation
Nguyen Van H.
LandOfFree
Signal processing system which can reduce burden applied to... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Signal processing system which can reduce burden applied to..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Signal processing system which can reduce burden applied to... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3182141