Data storage device encoding and interleaving codewords to improve trellis sequence detection

This application is a continuation of U.S. patent application Ser. No. 15/154,165, filed on May 13, 2016, entitled “DATA STORAGE DEVICE ENCODING AND INTERLEAVING CODEWORDS TO IMPROVE TRELLIS SEQUENCE DETECTION,” which is a continuation-in-part of U.S. patent application Ser. No. 14/930,898, filed on Nov. 3, 2015, entitled “DATA STORAGE DEVICE ENCODING AND INTERLEAVING CODEWORDS TO IMPROVE TRELLIS SEQUENCE DETECTION”; the disclosures of which are hereby incorporated by reference in their entirety.

Data storage devices such as disk drives comprise a disk and a head connected to a distal end of an actuator arm which is rotated about a pivot by a voice coil motor (VCM) to position the head radially over the disk. The disk comprises a plurality of radially spaced, concentric tracks for recording user data sectors and servo sectors. The servo sectors comprise head positioning information (e.g., a track address) which is read by the head and processed by a servo control system to control the actuator arm as it seeks from track to track.

FIG. 1 shows a prior art disk format 2 as comprising a number of servo tracks 4 defined by servo sectors 6 0 - 6 N recorded around the circumference of each servo track. Each servo sector 6 i comprises a preamble 8 for storing a periodic pattern, which allows proper gain adjustment and timing synchronization of the read signal, and a sync mark 10 for storing a special pattern used to symbol synchronize to a servo data field 12 . The servo data field 12 stores coarse head positioning information, such as a servo track address, used to position the head over a target data track during a seek operation. Each servo sector 6 i further comprises groups of servo bursts 14 (e.g., N and Q servo bursts), which are recorded with a predetermined phase relative to one another and relative to the servo track centerlines. The phase based servo bursts 14 provide fine head position information used for centerline tracking while accessing a data track during write/read operations. A position error signal (PES) is generated by reading the servo bursts 14 , wherein the PES represents a measured position of the head relative to a centerline of a target servo track. A servo controller processes the PES to generate a control signal applied to a head actuator (e.g., a voice coil motor) in order to actuate the head radially over the disk in a direction that reduces the PES.

FIG. 1 shows a prior art disk format comprising a plurality of servo tracks defined by servo sectors.

FIG. 2A shows a data storage device in the form of a disk drive according to an embodiment comprising a head actuated over a disk.

FIG. 2B is a flow diagram according to an embodiment wherein first and second mixed rate codewords are interleaved to generate an interleaved codeword written to the disk.

FIG. 2C shows an embodiment wherein a first codeword having a code rate of N is interleaved with a larger, second codeword having a code rate greater than N.

FIG. 3 shows control circuitry according to an embodiment wherein the lower rate of the first codeword increases the accuracy of a trellis detector during a second and subsequent global iterations.

FIGS. 4A and 4B illustrate that when the first codeword is successfully decoded, the corresponding reliability metrics effectively prune branches from the trellis of the trellis detector when performing subsequent iterations to recover the second codeword.

FIG. 5 is a flow diagram according to an embodiment wherein the encoded codewords are first interleaved in a non-uniform (e.g., random) manner, and then the non-uniform interleaved codewords are interleaved in a uniform manner to generate an interleaved codeword written to the disk.

FIG. 6A is a flow diagram according to an embodiment wherein first data is encoded into a first codeword comprising a plurality of i-bit symbols and second data is encoded into a second codeword comprising a plurality of j-bit symbols, wherein i is different than j.

FIG. 6B shows an embodiment wherein the first codeword comprises 1-bit symbols and the second codeword comprises 2-bit symbols, and the first and second codewords are symbol interleaved to generate the interleaved codeword.

FIG. 7 shows control circuitry according to an embodiment wherein the trellis detector processes the 1-bit reliability metrics corresponding to the i-bit symbols of the first codeword and the j-bit symbols of the second codeword.

FIG. 2A shows a data storage device in the form of a disk drive according to an embodiment comprising a head 16 actuated over a disk 18 . The disk drive further comprises control circuitry 20 configured to execute the flow diagram of FIG. 2B , wherein first data is encoded into a first codeword (block 22 ), and second data is encoded into a second codeword (block 24 ). The first codeword and the second codeword are interleaved to generate an interleaved codeword (block 26 ) that is written to the disk (block 28 ). In one embodiment, a first code rate of the first codeword is less than a second code rate of the second codeword as illustrated in the example shown in FIG. 2C . Also in the embodiment of FIG. 2C , the first codeword is smaller in size than the second codeword such that the code rate of the concatenated (interleaved) codeword achieves a desired format efficiency when written to the disk. As described in greater detail below, in one embodiment the lower code rate of the first codeword improves the probability of successfully decoding the first codeword, and when the first codeword is successfully decoded (or nearly decoded), the corresponding reliability metrics may improve the accuracy of a trellis detector when performing subsequent iterations to recover the second codeword.

In the embodiment of FIG. 2A , the disk 18 comprises a plurality of servo sectors 30 0 - 30 N that define a plurality of servo tracks 32 , wherein data tracks are defined relative to the servo tracks at the same or different radial density. The control circuitry 20 processes a read signal 34 emanating from the head 16 to demodulate the servo sectors 30 0 - 30 N and generate a position error signal (PES) representing an error between the actual position of the head and a target position relative to a target track. A servo control system in the control circuitry 20 filters the PES using a suitable compensation filter to generate a control signal 36 applied to a voice coil motor (VCM) 38 which rotates an actuator arm 40 about a pivot in order to actuate the head 16 radially over the disk 18 in a direction that reduces the PES. The servo sectors 30 0 - 30 N may comprise any suitable head position information, such as a track address for coarse positioning and servo bursts for fine positioning. The servo bursts may comprise any suitable pattern, such as an amplitude based servo pattern or a phase based servo pattern ( FIG. 1 ).

FIG. 3 shows control circuitry according to an embodiment wherein first data is encoded by a first rate encoder 42 A to generate a first codeword 44 A, and second data is encoded by a second rate encoder 42 B to generate a second codeword 44 B. The encoders may operate according to any suitable encoding scheme, such as a low density parity check (LDPC) code, and the encoders may generate the codewords having any suitable code rate and size. In one embodiment, the first codeword 44 A may be 1K in size and encoded to have a code rate of 0.65, and the second codeword 44 B may be 3K in size and encoded to have a code rate of 0.85, such that the concatenated, interleaved codeword may be 4K in size and have an overall code rate of 0.8 which is proximate to the code rate of the larger, second codeword 44 B. A first non-uniform interleaver 46 A interleaves the first codeword 44 A to generate a first non-uniform interleaved codeword 48 A, and a second non-uniform interleaver 46 B interleaves the second codeword 44 B to generate a second non-uniform interleaved codeword 48 B. Any suitable non-uniform interleaver may be employed, wherein in one embodiment the non-uniform interleavers 46 A and 46 B operate to randomly interleave the symbols of the respective codewords. In one embodiment, randomly interleaving the symbols of each codeword improves the performance of an iterative detection scheme, such as with an outer LDPC code and an inner trellis detector. In the embodiment of FIG. 3 , a uniform interleaver 50 interleaves the first and second non-uniform interleaved codewords 48 A and 48 B to generate the interleaved codeword 52 written to the disk 18 . In the embodiment, the uniform interleaver 50 interleaves the codewords 48 A and 48 B based on the ratio of the codeword sizes. For example in an embodiment where the first non-uniform interleaved codeword 48 A is 1K in size and the second non-uniform interleaved codeword 48 B is 3K in size, the uniform interleaver 50 may perform a 1:3 interleaving of the symbols such that there is a repeating sequence of one symbol from the first codeword 48 A followed by three symbols from the second codeword 48 B (e.g., as illustrated in FIG. 2C ).

During a read operation, the read signal 34 is processed by an analog-front-end (AFE) and equalizer circuitry 54 (which may comprise analog circuitry or digital circuitry or a combination of both). For example, in one embodiment block 54 may comprise an analog equalizer, followed by a sampler for sampling the read signal, followed by a digital equalizer that equalizes the signal samples based on a suitable response, such as a suitable partial response (e.g., PR4). The equalized signal samples 56 are input into a trellis detector 60 which processes the signal samples to generate a sequence of reliability metrics 62 representing the interleaved codeword that was written to the disk. Any suitable trellis detector 60 may be employed, such as a soft-input soft-output Viterbi detector or a Bahl, Cocke, Jelinek and Raviv (BCJR) detector. In addition, the trellis detector 60 may generate any suitable reliability metric for each symbol in the interleaved codeword, such as a log-likelihood ratio (LLR) representing a likelihood that each symbol was detected correctly.

A uniform de-interleaver 64 and respective non-uniform de-interleavers 66 A and 66 B de-interleave the sequence of reliability metrics 62 output by the trellis detector 60 into a first set of reliability metrics 68 A representing the first codeword 44 A and a second set of reliability metrics 68 B representing the second codeword 44 B. A first rate decoder 70 A decodes the first set of reliability metrics 68 A to generate a third set of reliability metrics 72 A representing the first codeword 44 A, and a second rate decoder 70 B decodes the second set of reliability metrics 68 B to generate a fourth set of reliability metrics 72 B representing the second codeword 44 B. The third set of reliability metrics 72 A and the fourth set of reliability metrics 72 B are re-interleaved to generate interleaved reliability metrics 74 representing the interleaved codeword. The interleaved reliability metrics 74 are input into the trellis detector 60 which performs another iteration to generate another sequence of reliability metrics 62 representing the interleaved codeword that was written to the disk.

In one embodiment, the decoders 70 A and 70 B execute local iterations on the respective sets of reliability metrics 68 A and 68 B. If the decoders reach their maximum iteration limit, the codeword is considered unrecoverable during the current global iteration, and therefore the current reliability metrics 72 A and 72 B generated by the decoders are re-interleaved and fed back to the trellis detector 60 . In one embodiment, the increased redundancy that causes the lower code rate of the first codeword 44 A helps improve the accuracy of the first rate decoder 70 A such that the probability of successfully decoding the first codeword increases. If the first codeword is successfully decoded, but the second codeword is not successfully decoded, the reliability metrics 72 A for the first codeword will be 100% which, as described below, effectively prunes branches from the trellis of the trellis detector 60 . Even if the first codeword is not successfully decoded, the interim reliability metrics 72 A generated by the first rate decoder 70 A may be more accurate (i.e., closer to 100%) due to the increase in redundancy of the first codeword 44 A. In either case, the higher reliability metrics 72 A of the first codeword 44 A bias the trellis detector 60 during the next global iteration so as to improve the accuracy of the trellis detector (for either or both of the codewords). That is, interleaving the higher reliability metrics 72 A of the first codeword 44 A with the lower reliability metrics 72 B of the second codeword 44 B may improve the accuracy of the trellis detector by pruning or biasing the branches corresponding to the symbols of the first codeword 44 A.

FIGS. 4A and 4B illustrate how in one embodiment the lower rate of the first codeword 44 A may prune the corresponding branches of the trellis during a second or subsequent global iteration. FIG. 4A shows a full trellis as implemented during the first global iteration wherein the symbols of both the first and second codeword are unknown. During a subsequent global iteration, the first codeword 44 A may be successfully decoded by the first rate decoder 70 A due to the lower code rate of the first codeword 44 A. Accordingly as shown in FIG. 4B , the branches that correspond to the known symbols of the first codeword 44 A are pruned from the trellis which increases the accuracy of the trellis detector 60 when detecting the symbols of the second codeword 44 B. In one embodiment, the branches of the trellis may be literally pruned from the trellis based on the known symbols of the first codeword 44 A, and in another embodiment, the reliability metrics being 100% for each symbol of the first codeword 44 A biases the trellis so as to effectively prune the corresponding branches from the trellis.

FIG. 5 is a flow diagram according to an embodiment wherein first data is encoded into a first codeword (block 76 ), and second data is encoded into a second codeword (block 78 ). A non-uniform interleaving of the first codeword is executed to generate a first non-uniform interleaved codeword (block 80 ), and a non-uniform interleaving of the second codeword is executed to generate a second non-uniform interleaved codeword (block 82 ). A uniform interleaving of the first non-uniform interleaved codeword and the second non-uniform interleaved codeword is executed to generate an interleaved codeword (block 84 ), and the interleaved codeword is written to the disk (block 86 ). In one embodiment, the first codeword has a rate and/or size that is different from a rate and/or size of the second codeword as described above. In another embodiment, the first and second codewords may have the same rate and/or the same size. As described above, interleaving the codewords using a uniform interleaver may improve the accuracy of a trellis detector, particularly when the decoding of one of the codewords provides higher reliability metrics for the second and subsequent global iterations by the trellis detector.

FIG. 6A is a flow diagram according to an embodiment wherein first data is encoded into a first codeword comprising a plurality of i-bit symbols (block 88 ), and second data is encoded into a second codeword comprising a plurality of j-bit symbols (block 90 ), wherein i is different than j and a first code rate of the first codeword is less than a second code rate of the second codeword. The first codeword and the second codeword are symbol interleaved to generate an interleaved codeword (block 92 ), and the interleaved codeword is written to the storage medium (block 94 ). An example of this embodiment is shown in FIG. 6B wherein the first codeword comprises 1-bit symbols and the second codeword comprises 2-bit symbols. The codewords are symbol interleaved such that each 1-bit symbol from the first codeword is followed by a 2-bit symbol from the second codeword as shown in FIG. 6B . In one embodiment, i is less than j which enables the lower rate codeword to have a shorter length than the higher rate codeword. Accordingly in this embodiment, the i-bit symbol codeword is more likely to be decoded first (due to the increased redundancy), thereby increasing the likelihood of successfully decoding the j-bit symbol codeword during subsequent global iterations. Although i is one in the embodiment of FIG. 6B , in other embodiments i may be greater than one such that both codewords may implement a non-binary code.

FIG. 7 shows control circuitry according to an embodiment wherein at least one of the first and second encoders 96 A and 96 B implement a non-binary encoding of the input data. In the example of FIG. 6B , the first encoder 96 A implements a binary encoding of first data into a first codeword comprising 1-bit symbols, and the second encoder 96 B implements a non-binary encoding of second data into a second codeword comprising 2-bit symbols. In this embodiment, the non-uniform interleavers 46 A and 46 B and the uniform interleaver 50 perform the interleaving of the codewords on a symbol basis. For example, in one embodiment the uniform interleaver 50 interleaves the 1-bit symbols of the 1-bit codewords with the 2-bit symbols of the 2-bit codewords as shown in FIG. 6B . The uniform de-interleaver 64 and the non-uniform de-interleavers 66 A and 66 B of FIG. 7 similarly operate on a symbol basis to de-interleave the interleaved codeword prior to decoding by a i-bit symbol decoder 98 A and a j-bit symbol decoder 98 B.

In one embodiment, at least one of the decoders 98 A and 98 B in FIG. 7 implement a non-binary decoding of a non-binary codeword (e.g., a 2-bit symbol codeword such as shown in FIG. 6B ), while the trellis detector 60 operates as a binary detector (1-bit symbol detector). In this embodiment, the non-binary decoder generates the 1-bit reliability metrics (e.g., RM_ 2 ) for each bit of each non-binary symbol. Consider the example embodiment of FIG. 6B where the second codeword comprises 2-bit symbols. The j-bit decoder 98 B of FIG. 7 may generate a 4-ary reliability metric (e.g., LLR) which, after taking the exponential of each term, are probabilities:

P ( b 1=0, b 2=0), P ( b 1=0, b 2=1), P ( b 1=1, b 2=0), P ( b 1=1, b 2=1).

The probabilities for b 1 may be represented as:

P ( b 1=0)= P ( b 1=0, b 2=0)+ P ( b 1=0, b 2=1)

P ( b 1=1)= P ( b 1=1, b 2=0)+ P ( b 1=1, b 2=1)

such that the 1-bit reliability metric for b 1 may be represented as:

b 1=log( P ( b 1=0)/ P ( b 1=1))

with a similar computation used to generate the 1-bit reliability metric for b 2 . The output of the trellis detector 60 in this embodiment is a binary (1-bit) reliability metric (e.g., LLR) that is converted into a non-binary reliability metric for each symbol of the non-binary codeword(s). In the above example where j=2, taking the exponential of each of the 1-bit reliability metrics generated for two adjacent bits representing a symbol of the second codeword generates the probabilities for each bit:

P ( b 1=0), P ( b 1=1), P ( b 2=0), P ( b 2=1).

The joint probabilities for the two adjacent bits may then be generated as:

P ( b 1) P ( b 2)

wherein taking the log of the four joint probabilities generates the 4-ary reliability metric for each 2-bit symbol processed by the j-bit decoder 98 B.

In the embodiments described above, the symbols of two codewords are interleaved to generate the interleaved codeword written to the disk. However, in other embodiments more than two codewords may be interleaved, wherein the code rate and/or size of each codeword as well as the symbol size for each codeword may vary (or be the same). For example, in one embodiment each codeword may have a progressively higher code rate and larger size (and optionally a larger symbol size), wherein the lower code rate codewords may be more readily decoded due to the increase in redundancy. In another embodiment when interleaving three or more codewords having the same or different code rate and/or size, the likelihood of successfully decoding at least one of the codewords may improve which may then cascade into the ability to successfully recover the remaining codewords.

Any suitable control circuitry may be employed to implement the flow diagrams in the above embodiments, such as any suitable integrated circuit or circuits. For example, the control circuitry may be implemented within a read channel integrated circuit, or in a component separate from the read channel, such as a disk controller, or certain operations described above may be performed by a read channel and others by a disk controller. In one embodiment, the read channel and disk controller are implemented as separate integrated circuits, and in an alternative embodiment they are fabricated into a single integrated circuit or system on a chip (SOC). In addition, the control circuitry may include a suitable preamp circuit implemented as a separate integrated circuit, integrated into the read channel or disk controller circuit, or integrated into a SOC.

In one embodiment, the control circuitry comprises a microprocessor executing instructions, the instructions being operable to cause the microprocessor to perform the flow diagrams described herein. The instructions may be stored in any computer-readable medium. In one embodiment, they may be stored on a non-volatile semiconductor memory external to the microprocessor, or integrated with the microprocessor in a SOC. In another embodiment, the instructions are stored on the disk and read into a volatile semiconductor memory when the disk drive is powered on. In yet another embodiment, the control circuitry comprises suitable logic circuitry, such as state machine circuitry.

In various embodiments, a disk drive may include a magnetic disk drive, an optical disk drive, etc. In addition, while the above examples concern a disk drive, the various embodiments are not limited to a disk drive and can be applied to other data storage devices and systems, such as magnetic tape drives, solid state drives, hybrid drives, etc. In addition, some embodiments may include electronic devices such as computing devices, data server devices, media content storage devices, etc. that comprise the storage media and/or control circuitry as described above.

The various features and processes described above may be used independently of one another, or may be combined in various ways. All possible combinations and subcombinations are intended to fall within the scope of this disclosure. In addition, certain method, event or process blocks may be omitted in some implementations. The methods and processes described herein are also not limited to any particular sequence, and the blocks or states relating thereto can be performed in other sequences that are appropriate. For example, described tasks or events may be performed in an order other than that specifically disclosed, or multiple may be combined in a single block or state. The example tasks or events may be performed in serial, in parallel, or in some other manner. Tasks or events may be added to or removed from the disclosed example embodiments. The example systems and components described herein may be configured differently than described. For example, elements may be added to, removed from, or rearranged compared to the disclosed example embodiments.

While certain example embodiments have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions disclosed herein. Thus, nothing in the foregoing description is intended to imply that any particular feature, characteristic, step, module, or block is necessary or indispensable. Indeed, the novel methods and systems described herein may be embodied in a variety of other forms; furthermore, various omissions, substitutions and changes in the form of the methods and systems described herein may be made without departing from the spirit of the embodiments disclosed herein.