Browse Prior Art Database

Method for an area-efficient, multichannel, pipelined data path for data mappers supporting flexible bandwidth allocation

IP.com Disclosure Number: IPCOM000029131D
Publication Date: 2004-Jun-16
Document File: 5 page(s) / 58K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method for an area-efficient, multichannel, pipelined data-path for data mappers supporting flexible bandwidth allocation. Benefits include improved functionality, improved performance and improved design cycle and robustness by reusing proven existing single channel data-path components.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 54% of the total text.

Method for an area-efficient, multichannel, pipelined data path for data mappers supporting flexible bandwidth allocation

Disclosed is a method for an area-efficient, multichannel, pipelined data-path for data mappers supporting flexible bandwidth allocation. Benefits include improved functionality, improved performance and improved design cycle and robustness by reusing proven existing single channel data-path components.

Background

              Single-channel implementations are the conventional processing solution (see Figure 1). At each stage there is some information that must be reused during the next clock cycle. These tasks typically include cyclic redundancy code (CRC) calculation, scrambling, packet/cell delineation and data realignment. In these tasks, data dependency occurs across cycles. The result of processing a data word at a particular stage depends on the processing of the previous data word at the same stage. Other tasks do not have this dependency. No information from the previous word is required to process the current one and no state-information must be stored.

General description

              The disclosed method is a multichannel, pipelined, data path implementation for applications that are typically used to perform full-packet processing tasks.

Advantages

              The disclosed method provides advantages, including:

•             Improved functionality due to enabling the mapper to accept data belonging to any channel and keeping the channel order at the output

•             Improved performance due to providing total flexibility regarding the order in which channels are processed

•             Improved robustness and speed-up design cycle by able to reuse proven existing single channel data-path components.

Detailed description

              The disclosed method includes multiple data flow. The input data may belong to a different channel each clock cycle. An additional input signal indicates the channel identifier (CID) for the data. The state information at each stage must be updated according to the channel being processed. It must be stored from one clock cycle to another and from one data-word to the next one belonging to the same channel. Current state information must be kept for all channels. To enable the modification of the bandwidth for any channel at any time, the mapper implementation must support any combination of data belonging to different channels.

              The registers conta...