Timing recovery in error recovery for iterative detection
AssigneeMaxtor Corporation
InventorsJingfeng Liu, Rose Shao, Jinghuan Chen
Life of this patent
Estimating the term…
Abstract
A decoding system includes a timing loop for use in error recovery mode operations, in which “fast decode” bit values that are used for timing recovery purposes in the normal modes of operation are replaced, at an appropriate iteration in the error recovery mode, with detected bit values from an iterative detector. The replacement occurs after a reread operation in which the timing loop maintains lock.
Claims20
A timing recovery system for use in data block error recovery operations, the system including:
- an analog-to-digital converter that produces digital values associated with samples of an analog signal that corresponds to the data block;
- a fast detector that produces first bit values that correspond to the digital values by decoding respective portions of the sampled signal;
- an iterative detector that produces second bits values that correspond to the digital values by iteratively decoding the entire data block;
- a timing sub-system for producing a clocking signal that sets a sampling rate for the analog-to-digital converter, the timing sub-system producing the clocking signal based in part on detected phase errors between a bit rate of the signal and the sampling rate, the timing sub-system operating during error recovery rereads to
- determine the phase error based on the digital values and the corresponding first bit values, or
- determine the phase error based on the digital values and the corresponding second bit values;
- a loss of lock detector that determines for each decoding of the data block if a loss of lock has occurred in the timing sub-system, and
- a controller for providing the first and second bit values to the timing sub-system, the controller providing the second bit values after an error recovery reread occurs without a loss of lock.
A method of performing error recovery operations to recover a data block, the method including the steps of:
- A. re-reading a sector of data to produce a read signal and decoding the read signal based on signal samples that are produced in accordance with timing signals that are based on fast decode bit values that are produced by decoding respective portions of the data block;
- B. determining if the data block decoded in step A contains a correctable number of errors and if so ending the error recovery operation;
- C. if a next re-read operation is required, determining if a loss of lock occurred in a timing loop and if so repeating steps A-C until the block is correctly decoded or a stop condition occurs,
- D. if a no loss of lock is determined to have occurred in step C rereading the sector and decoding the read signal based on signal samples that are produced in accordance with timing signals that are based on iteratively detected bit values that are produced by decoding the entire data block;
- E. determining if the data block decoded in step D contains a correctable number of errors and if so ending the error recovery operation;
- F. if a next reread operation is required and no loss of lock occurred in step D, repeating steps D-F until the data block is recovered or a stop condition occurs;
- G. if a next reread operation is required and a loss of lock occurred in step D, rereading the sector and decoding the read signal based on signal samples that are produced in accordance with timing signals that are based on the iteratively detected bit values used in the previous reread and decoding step;
- H. determining if the data block decoded in step G contains a correctable number of errors and if so ending the error recovery operation;
- I. if a next reread operation is required and no loss of lock occurred in step G, repeating steps D-H until the data block is recovered or a stop condition occurs.
A decoding system for decoding a block of data, the system including
- an analog-to-digital converter that samples an analog signal that corresponds to the data block and produces digital values that correspond to the respective samples;
- a fast detector that produces first bit values that correspond to the digital values by decoding respective portions of the sampled signal;
- an iterative detector that produces second bits values that correspond to the digital values by iteratively decoding the entire data block;
- an iterative decoder that receives the second bit values and produces corresponding hard decision bit values;
- a clock for generating a clock signal;
- a timing sub-system for producing an adjusted clock signal to drive the analog-to-digital converter, the timing sub-system adjusting the clock signal based on detected phase errors between a bit rate in the signal and the rate of sampling, the timing subsystem operating in a first mode by detecting phase errors between the first bit values and the corresponding digital values and in a second mode by detecting phase errors between the second bit values and the corresponding digital values;
- a loss of lock detector that detects loss of lock in the timing sub-system during a decoding of the data block, and
- a controller for providing the first bit values to the timing sub-system during block decoding operations and error recovery operations and the second bit values to the timing sub-system during an error recovery re-read operation that follows an error recovery re-read operation in which no loss of lock was detected.
A timing recovery system for use in error recovery operations of a data block decoder, the system including
- a fast detector that produces first bit values by decoding respective portions of the data block;
- an iterative detector that produces second bits values by iteratively decoding the entire data block;
- a loss of lock detector that detects a loss of lock in a timing loop;
- and
- a timing sub-system for producing a timing signal, the timing sub-system producing the timing signal based on the first bit values in a first mode and based on the second bit values in a second mode, the timing sub-system operating in the second mode after an error recovery re-read operation in which no loss of lock is detected.
Dependencies are inferred from the claim text.