Method and device for filtering transmission

• US 5,959,976 A
• Filed: 12/09/1996
• Issued: 09/28/1999
• Est. Priority Date: 12/09/1996
• Status: Expired due to Term

• Current Assignee: Winbond Electronics Corporation
• Original Assignee: Winbond Electronics Corporation
• Inventors: Kuo, Yung-tien
• Primary Examiner(s): MARCELO, MELVIN C
• Application Number:
  US08/763,693
• Time in Patent Office: 1,023 Days
• Field of Search: 370/254-258, 370/401-405
• US Class Current: 370/257
• CPC Class Codes: H04L 12/4625 Single bridge functionality, e.g. connection of two networks over a single bridge
• CPC Class Codes: H04L 45/742 Route cache; Operation thereof

Abstract

The present invention relates to a method and device for filtering transmission addresses in Ethernet, more particularly, to a filtering method and device provided to reduce the network loads of different groups in Ethernet. The present invention solves the problems that a large amount of memory space is required and a packet can not be filtered effectively since the the memory contents are lost when power fails in the filtering method of the prior art. The filtering method of the present invention is only to store the flag of the corresponding source address rather than the entire address data. The work for data comparing only reads out the state of the flag of the corresponding address, then refers to the differences between the state of the "0" and "1" of the flag to determine whether the packet should be blocked or allowed to is pass. The memory space will be very small due to only storing the state of the flag. The present invention is further equipped with a programmable timer such that the address data can be just stored into a memory rather than proceeding a comparison and filter in a specific time or every other time interval after power is abruptly disconnected or fails.

Claims

1. A method for filtering transmission addresses in Ethernet for determining whether a packet is allowed to pass or is blocked between two network groups, comprising:

a step for extracting source address data and destination address data from said packet formed in every network group through an extractor;

a step for compressing said source address data and said destination address data extracted from every network group to source address data and destination address data with a small quantity of bits, namely, compressed source address data and compressed destination address data respectively;

a step for setting compressed source address data corresponding to said source address data with a small quantity of bits to be a setup signal for writing said compressed source address data into a memory address and setting a corresponding memory address to be an effective flag;

a step for setting compressed destination address data corresponding to said destination address data with a small quantity of bits to be a read signal for reading out said compressed destination address data from a memory address and reading out a state of said effective flag; and

a step for determining whether said packet is allowed to pass or is blocked according to said state of said effective flag.

2. The method for filtering transmission addresses in Ethernet as claimed in claim 1, wherein said effective flag is represented by one bit.

3. The method for filtering transmission addresses in Ethernet as claimed in claim 1, wherein said extractor is formed by a counter.

4. The method for filtering transmission addresses in Ethernet as claimed in claim 1, wherein said source address data extracted from one of said two network groups becomes a basis of said destination address data of the other one of said two network groups to be searched for.

5. The method for filtering transmission addresses in Ethernet as claimed in claim 1, wherein said compressed source address data and said compressed destination address data are implemented with an operation method of Cyclic Redundancy Check.

6. The method for filtering transmission addresses in Ethernet as claimed in claim 1, wherein said compressed source address data and said compressed destination address data are implemented with an operation method of Exclusive OR.

7. The method for filtering transmission addresses in Ethernet as claimed in claim 1, wherein said compressed source address data and said compressed destination address data each have a length of 12 bits.

8. The method for filtering transmission addresses in Ethernet as claimed in claim 1, further comprising a broadcast mode set to be not controlled by said effective flag in a specific time interval, and only for learning rather than for filtering, to reduce the effect of losing memory contents.

9. The method for filtering transmission addresses in Ethernet as claimed in claim 1, further comprising a mode set to be not controlled by said effective flag and proceeding to allow said packet to pass directly instead of filtering said packet during a certain time interval after power has failed.

10. The method for filtering transmission addresses in Ethernet as claimed in claim 1, further comprising a step proceeding to broadcast rather than proceeding to filter said packet when two address data collide.

11. A device for filtering transmission addresses is in Ethernet for determining whether a packet is allowed to pass or is blocked between two network groups, comprising:

extractors extracting source address data and destination address data from said packet formed in every network group;

compressed address generators compressing said source address data and said destination address data extracted from every network group to address data with a small quantity of bits;

memories having the provision of sending said address data with a small quantity of bits described above and representing said address data corresponding to a memory address by a flag, and proceeding to set said flag or reading out the state of said flag by said memory address corresponding to source address data or destination address data respectively; and

filters being in between two different networks and determining their operations of being enabled or disabled according to said state of said flag.

12. The device for filtering transmission addresses in Ethernet as claimed in claim 11, wherein said flag is represented by one bit.

13. The device for filtering transmission addresses in Ethernet as claimed in claim 11, wherein said extractors are formed by counters.

14. The device for filtering transmission addresses in Ethernet as claimed in claim 11, wherein said source address data extracted from one of said two network groups become an input of said compressed address generators of the other one of said two network groups to proceed the comparison between said source address data and said destination address data of said two network groups.

15. The device for filtering transmission addresses in Ethernet as claimed in claim 11, wherein said compressed source address data and said compressed destination address data are implemented with an operation tool of Cyclic Redundancy Check.

16. The device for filtering transmission addresses in Ethernet as claimed in claim 11, wherein said compressed source address data and said compressed destination address data are implemented with an operation tool of Exclusive OR.

17. The device for filtering transmission addresses in Ethernet as claimed in claim 11, wherein said compressed source address data and said compressed destination address data each have a length of 12 bits.

18. The device for filtering transmission addresses in Ethernet as claimed in claim 11, further comprising a programmable timer proceeding to learn and to filter said packet through a filter in a broadcast mode set to be not controlled by said flag in a specific time interval.

19. The device for filtering transmission addresses in Ethernet as claimed in claim 11, further comprising a programmable timer set to be not controlled by said flag and said filter proceeds not to block said packet but to allow said packet to pass directly during a certain time interval after power has failed.