IP Library Granted Patent US 7,733,764
Granted Patent B2
US 7,733,764 · App. 10/731,730 · Granted Jun 8, 2010

Method and system for dynamically reducing length of a delay chain

View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 7,733,764
App. No.
10/731,730
Granted
Jun 8, 2010
Kind
B2
Abstract

Upon a triggering event, a delay chain shifts data out at a higher rate than incoming packets and a processor controls bypassing circuitry to reduce the latency of hardware implementations of, for example, 802.11a OFDM receivers, with long delay chains. The signal processing algorithms used to recover symbol timing need a large number of samples stored in a delay chain, often consisting of pipelined registers. Such a delay chain introduces a large lag between the time samples have been acquired by the data converters and the time they are processed. This delay makes it difficult for higher level network layer implementations to meet the deadlines of 802.11a WLAN protocol. The proposed scheme implements dynamic reduction in the depth of the delay chain once timing recovery has been performed. A multi-step scheme achieves exponential reduction in the number of elements in the delay chain in every step.

Claims (17)

1. A method of reducing data path latency in digitally processing a sequence of data samples, the data path latency being associated with a transient processing operation on the data samples, comprising:

at the transition into the transient processing operation on the data samples reading the sequence of data samples into a tapped clocked delay chain, wherein each data storing element of the chain has an associated enabling signal for controlling, on any given clock cycle, whether to update or not its respective stored data sample;

during the transient processing operation on the data samples, processing data samples from taps on the clocked delay chain;

in response to receiving a signal of completion of the transient processing operation on the data samples, controlling the enabling signal of each data storing element in the chain to allow shifting data samples through an end portion of the clocked delay chain on a selected first set of clock cycles determining a first shifting rate and shifting data samples in an initial portion of the delay chain on a selected second set of clock cycles determining a second shifting rate, wherein the initial portion is complementary to the end portion and wherein the first shifting rate is higher than the second shifting rate; and

dynamically reducing the length of the clocked delay chain by moving the output of the delay chain to the last data storing element of the initial portion, after a number of clock cycles.

2. The method of claim 1 wherein the data samples are from a data packet.

3. The method of claim 2 wherein the data packet conforms to a transmission system selected from the group of 802.11a, 802.11g and HIPERLAN/2 transmission systems.

4. The method of claim 3 wherein the transient processing operation includes a synchronization of the data packet.

5. The method of claim 4 wherein the clocked delay chain comprises a plurality of pipelined registers.

6. The method of claim 5 wherein the reducing the length of the clocked delay chain is performed until a desired length of the clocked delay chain is achieved.

7. The method of claim 5 wherein reducing the length of the clocked delay chain further includes bypassing empty registers.

8. A method of reducing data path latency in digitally processing a sequence of data samples of a data packet, the data path latency being associated with synchronization of the data packet, comprising:

upon reception of the data packet, reading the sequence of data samples from the data packet into a tapped clocked delay chain comprising a plurality of pipelined registers, wherein each register has an associated enabling signal for controlling, on any given clock cycle, whether to update or not its respective stored data sample;

processing data samples from taps on the clocked delay chain to synchronize the data packet;

in response to receiving a signal of completion of synchronization of the data packet, controlling the enabling signal of each data storing element in the chain to allow shifting samples through an end portion of the clocked delay chain on a selected first set of clock cycles determining a first shifting rate and shifting data samples in an initial portion of the delay chain on a selected second set of clock cycles determining a second shifting rate, wherein the initial portion is complementary to the end portion and wherein the first shifting rate is higher than the second shifting rate;

reducing the length of the clocked delay chain by moving the output of the delay chain to the last register of the initial portion, after a number of clock cycles; and

repeating the steps of shifting data samples through an end portion of the reduced delay chain and through the initial portion and reducing the length of the clocked delay chain.

Assignments (4)
CORRECTIVE ASSIGNMENT TO CORRECT THE ASSIGNEE'S NAME FROM EDGEWATER WIRELESS TO EDGEWATER WIRELESS SYSTEMS INC. PREVIOUSLY RECORDED ON REEL 026049 FRAME 0176. ASSIGNOR(S) HEREBY CONFIRMS THE CORRECT SPELLING OF ASSIGNEE'S NAME TO BE EDGEWATER WIRELESS SYSTEMS INC. Recorded Oct 25, 2013
From: EDGEWATER COMPUTER SYSTEMS INC.
To: EDGEWATER WIRELESS SYSTEMS INC.
Reel/Frame 031495/0573 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 30, 2011
From: EDGEWATER COMPUTER SYSTEMS INC.
To: EDGEWATER WIRELESS
Reel/Frame 026049/0176 →
MERGER Recorded Feb 7, 2006
From: ENGIM, INC.
To: EDGEWATER COMPUTER SYSTEMS, INC.
Reel/Frame 017240/0071 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 21, 2004
From: SONI, MANEESH; CHADHA, KANU; BHARDWAJ, MANISH
To: ENGIM, INC.
Reel/Frame 014657/0398 →