Semiconductor memory device and bit error detection method thereof
A memory device detects and corrects bit errors. The memory device includes cyclic redundancy check (CRC) and error correction code (ECC) circuits. The CRC circuit generates a write CRC code corresponding to data to be stored in memory cells. The ECC circuit generates an ECC code corresponding to the data and detecting and correcting a bit error of the data by means of the ECC code during a read operation. The CRC circuit generates a read CRC code corresponding to data corrected by the ECC circuit during the read operation, and corrects a bit error of the data according to a comparison of the read CRC code and the write CRC code.
Latest Samsung Electronics Patents:
This U.S. non-provisional patent application claims priority under 35 U.S.C. §119 to Korean Patent Application 2005-100406 filed on Oct. 24, 2005, the entire contents of which are herein incorporated by reference.
BACKGROUND1. Technical Field
The present disclosure relates to semiconductor memory devices. In particular, the present disclosure relates to a semiconductor memory device and method for detecting bit errors therein.
2. Discussion of Related Art
Semiconductor memory devices are used for storing and retrieving data. Semiconductor memory devices are largely classified into random access memories (RAMs) and read-only memories (ROMs). RAMs are volatile memory devices that lose their stored data when a power supply is turned off. ROMs are nonvolatile memory devices that retain data even without power supply. RAMs include dynamic RAMs and static RAMs. ROMs include programmable ROMs (PROMs), erasable PROMs (EPROMs), electrically EPROMs (EEPROMs), and flash memories.
The flash memories are widely employed in mobile communication terminals, portable media players, digital cameras, mobile storage media, and so forth. In using the flash memories for storage media, data integrity needs to be assured. However, data stored on the flash memory devices typically includes bit errors. With their inherent property as memory devices, the flash memories need to have functions for detecting and correcting bit errors therein. Flash memories employ error correction code (ECC) circuits for detecting and correcting bit errors therein.
The ECC circuits used in the flash memory device are designed to correct a 1-bit error and to detect 2-bit errors. Any more then two error bits may not be detected in the flash memory. This limit on the number of detectable error bits degrades the reliability of the flash memory device.
SUMMARY OF THE INVENTIONThe memory device corrects bit errors in a predetermined number, detecting pluralities of erroneous bits. According to an embodiment of the present invention, a semiconductor memory device includes a CRC circuit generating a write CRC code corresponding to data to be stored in memory cells; and an ECC circuit generating an ECC code corresponding to the data and detecting and correcting a bit error of the data by means of the ECC code during a read operation. The CRC circuit generates a read CRC code corresponding to data corrected by the ECC circuit during the read operation, and corrects a bit error of the data according to a comparison of the read CRC code and the write CRC code.
The CRC circuit includes a CRC engine receiving the data corrected by the ECC circuit and generating the read CRC code, and a comparator generating the pass signal when the read CRC code matches the write CRC code, and generating the fail signal when the read CRC code does not match the write CRC code.
The semiconductor memory device is a NAND flash memory device including a cell array divided into main and spare regions, and the data is programmed into the main region and the ECC and CRC codes are programmed into the spare region. The ECC circuit detects two error bits and corrects one error bit, and the CRC circuit detects more than two error bits.
According to an embodiment of the present invention, a semiconductor memory device includes a CRC circuit generating a write CRC code corresponding to data to be stored in memory cells, a first ECC circuit generating a first ECC code corresponding to the data and detecting and correcting a bit error of the data by means of the first ECC code during a read operation, and a second ECC circuit generating a second ECC code corresponding to the write CRC data and detecting and correcting a bit error of the write CRC code by means of the second ECC code during the read operation. The CRC circuit generates a read CRC code corresponding to data corrected by the first ECC circuit during the read operation, and corrects a bit error of the data according to a comparison of the read CRC code and the write CRC code corrected by the second ECC circuit.
The CRC circuit includes a CRC engine receiving the data corrected by the first ECC circuit and generating the read CRC code, and a comparator generating the pass signal when the read CRC code matches the corrected write CRC code, and generating the fail signal when the read CRC code does not match the corrected write CRC code.
The semiconductor memory device is a NAND flash memory device including a cell array divided into main and spare regions, wherein the data is programmed into the main region and the first and second ECC codes and the CRC code are programmed into the spare region. The first and second ECC circuits each detect two error bits and correct one error bit, and the CRC circuit detects more than two error bits.
According to an embodiment of the present invention, a method of detecting a bit error includes generating ECC and CRC codes corresponding to data to be stored in memory cells, storing the data in the memory cells, correcting a bit error for the data stored in the memory cells by means of the ECC code, generating a read CRC code corresponding to the data corrected with the bit error, and detecting a bit error of the data according to a comparison of the read CRC code and the write CRC code. Correcting the bit error includes treating the data as being failed when the number of error bits is over a correctable number of error bits. The method includes determining a failure of the data when the read CRC code does not match the write CRC code.
The semiconductor memory device is a NAND flash memory device including a cell array divided into main and spare regions. The method further includes programming the data into the main region and programming the ECC and CRC codes into the spare region. Correcting the bit error includes detecting two error bits and correcting one error bit, and detecting the bit error of the data includes determining a bit error in a corrected bit error.
According to an embodiment of the present invention, a method for detecting a bit error in a semiconductor memory device includes generating a write CRC code corresponding to data to be stored in memory cells, and generating a first ECC code corresponding to the data to be stored in the memory cells, and a second ECC code corresponding to the write CRC code. The method includes programming the data, the write CRC code, and the first and second ECC codes into the memory cells, correcting a bit error of the data by means of the first ECC code, and a bit error of the write CRC code by means of the second ECC code, generating a read CRC code corresponding to the data corrected with the bit error, and detecting a bit error of the data according to a comparison of the read CRC code and the write CRC code corrected with the bit error. Correcting the bit error includes treating the data as being failed when the number of error bits is over a correctable number of error bits.
The semiconductor memory device is a NAND flash memory, device including a cell array divided into main and spare regions, and the data is programmed into the main region and the first and second ECC codes and the CRC code are programmed into the spare region. Correcting the bit error includes detecting two error bits and correcting one error bit, and detecting the bit error of the data further includes determining a bit error in a corrected the bit error.
BRIEF DESCRIPTION OF THE FIGURESNon-limiting and non-exhaustive embodiments of the present invention will be described with reference to the following figures, wherein like reference numerals refer to like parts throughout the various figures unless otherwise specified. In the figures:
Preferred embodiments of the present invention will be described below in more detail with reference to the accompanying drawings. The present invention may, however, be embodied in different forms and should not be constructed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the present invention to those skilled in the art.
Referring to
The cell array 110 stores data input by way of the data buffer 120. In the operational structure of the semiconductor memory device, physical and architectural characteristics may cause bit errors on data stored in the cell array 110. For example, a bit error occurs when data is changed from ‘1’ to ‘0’ or from ‘0’ to ‘1’. The ECC circuit 140 that is able to detect and correct bit errors (or error bits). The ECC circuit 140 can detect and correct single error bits and detect two error bits. The semiconductor memory device 100 further comprises the CRC circuit 130 in addition to the ECC circuit 140 so as to increase the number of detectable error bits.
The CRC circuit 130 receives write data Data_W from the data buffer 120 during a write operation and receives read data Data_R from the ECC circuit 140. The CRC circuit 130 generates a write CRC code CRC_W from the write data Data_W. The CRC circuit 130 internally generates a read CRC code CRC_R (refer to
The ECC circuit 140 receives the write data Data_W from the data buffer 120 during a write operation and receives the read data Read_R from the cell array 110 during the-read operation. The ECC circuit 140 generates an ECC code ECC from the write data Data_W. The ECC circuit 140 provides the write data Data_W and the ECC code ECC to the cell array 110. During the read operation the ECC circuit 140 detects and corrects an error bit of the read data Data_R, which has been input from the cell array 110, by means of the ECC code ECC stored in the cell array.
The ECC circuit 140 may be implemented in various coding schemes such as Hamming code, BCH (Bose, Chaudhuri, Hocquenghem) code, or Reed-Solomon code. For example, an ECC circuit using Hamming codes with predetermined bits is able to detect a 2-bit error (i.e., two error bits) and to correct 11-bit error (i.e., one error bit). Namely, the ECC circuit with Hamming codes functions to detect and correct one error bit and to detect two error bits.
The semiconductor memory device 100 uses the ECC circuit 140 to correct a 1-bit error (i.e., one error bit) and uses the CRC circuit 130 to detect a 3-bit error or greater (i.e., three or more error bits).
The selection circuit 131 alternatively outputs one of the write data Data W and the read data Data_R in response to a command CMD. The command may be a write command or a read command. The selection circuit 131 outputs the write data Data_W in response to the write command and outputs the read data Data_R in response to the read command. The read data Data_R is data corrected by the ECC circuit 140.
The CRC engine 132 receives the write data Data_W and outputs the write CRC code CRC_W to the cell array. The CRC engine 132 receives the read data Data_R and outputs the read CRC code CRC_R. The CRC engine may be utilized for verifying data reliability in the fields of data communication, data compression, data storage media (e.g., magnetic tapes, magnetic disc, and so forth), and the like. For example, with a 32-bit CRC engine, the probability that a bit error will be missed is than 2−32. The probability that the CRC engine won't detect a bit error is close to zero.
The comparator 133 compares the read CRC code CRC_R up with the write CRC code CRC_W. The comparator 133 generates the pass signal Pass when the read CRC code CRC_R matches the write CRC code CRC_W, or generates the fail signal Fail when the read CRC code CRC_R mismatches the write CRC code CRC_W.
Returning to
The second ECC circuit 250 detects error bits that are generated from the write CRC code CRC_W in the semiconductor memory device 200. The semiconductor memory device 200 achieves higher reliability for bit error as compared to the semiconductor memory device 100 shown in
The second ECC circuit 250 receives the write CRC code CRC_W from the CRC circuit 230 during the write operation, and receives the read CRC code CRC_R from the cell array 110 during the read operation. The second ECC circuit 250 generates a second ECC code ECC_2 from the write CRC code CRC_W. The ECC code generated from the first ECC circuit 240 is referred to as the first ECC code ECC_1. The second ECC circuit 250 provides the cell array 210 with the write CRC code CRC_W and the second ECC code ECC_2.
The second ECC circuit 250 detects and corrects an error bit of the read CRC code CRC_R, which is input from the cell array 210, by means of the second ECC code ECC_2 stored in the cell array 210. The second ECC circuit 250 corrects the error bit of the read CRC code CRC_R and thereafter provides a corrected read CRC code CRC_R′ to the CRC circuit 230.
The CRC circuit 230 internally generates a read CRC code CRC_R″ from the read data Data_R. The CRC circuit 230 compares the internally generated read CRC code CRC_R″ with the corrected read CRC code CRC_R′, and generates the pass or fail signal.
Referring to
Referring to
The cell array of the NAND flash memory device 300 is divided into the main and spare regions 310 and 311. The page buffers are also divided into the main page buffer 312 corresponding to the main region 310 and the spare page buffer 313 corresponding to the spare region 311. Configurations and operations of the cell array and page buffers in the NAND flash memory device 300 are well known by those skilled in this art, and further descriptions are omitted..
Referring to
Referring to
Referring to
At block S120, the ECC codes ECC_1 and ECC_2 are generated corresponding to the write data Data_W and the write CRC code CRC_W, respectively. Referring to
At block S130, the main region 310 is programmed with the write data Data_W, and the spare region 311 is programmed with the first ECC code ECC_1, the second ECC code ECC 2, and the write CRC code CRC_W.
At block S220 checks are performed for bit errors on the read data Data_R of the main page buffer 312 and the read CRC code CRC_R of the spare page buffer 313. Referring to
At block S230, a bit error (or an error bit) is detected from the read data Data_R and the read CRC code CRC_R. If there is a bit error, the procedure goes to block S240. Unless there is a bit error, the procedure moves to block S232.
Block S232 is carried out when there is no bit error; the first ECC circuit 340 directly outputs the read data Data_R, wherein, Data_R=Data_R′. The second ECC circuit 350 directly outputs the read CRC code CRC_R wherein, CRC_R=CRC_R′.
At block S240 the bit error is corrected. It is assumed that the first and second ECC circuits 340 and 350 are designed to correct a 1-bit error. If more than two error bits are detected at block S230, the first and second ECC circuits 340 and 350 may not correct the error bits. Where three or more error bits are detected, the procedure goes to block S280. At block S280, the read data is treated as being failed. Otherwise, if one error bit is detected at block S230, the error bit (i.e., the 1-bit error) is corrected therein. The procedure moves to block S242.
At block S242, the detected error bit is corrected. Referring to
At block S250, the read CRC code CRC_R″ is generated in correspondence with the corrected read data Data_R′. Referring to
At block S260, the internal read CRC code CRC_R″ is compared to the corrected read CRC code CRC_R′. Referring to
Referring to
The NAND flash memory device 300 programs the CRC code along with data, without programming an additional confirm mark or flag. The write operation need only be performed once, wherein the CRC code functions as the confirm mark.
As described above, the semiconductor memory device and bit error correction method according to an embodiment of the present invention detects a bit error, which has not been detected by the ECC circuit, by means of the CRC circuit, and corrects a predetermined bit error (e.g., a 1-bit error) by means of the ECC circuit.
The above-disclosed subject matter is to be considered illustrative, and not restrictive, and the appended claims are intended to cover all such modifications, enhancements, and other embodiments, which fall within the true spirit and scope of the present invention. Thus, to the maximum extent allowed by law, the scope of the present invention is to be determined by the broadest permissible interpretation of the following claims and their equivalents, and shall not be restricted or limited by the foregoing detailed description.
Claims
1. A semiconductor memory device comprising:
- a cyclic redundancy check (CRC) circuit generating a write CRC code corresponding to data to be stored in memory cells; and
- an error correction code (ECC) circuit generating an ECC code corresponding to the data and detecting and correcting a bit error of the data by means of the ECC code during a read operation,
- wherein the CRC circuit generates a read CRC code corresponding to data corrected by the ECC circuit during the read operation, and corrects a bit error of the data according to a comparison of the read CRC code and the write CRC code.
2. The semiconductor memory device as set forth in claim 1, wherein the CRC circuit generates a pass signal when the read CRC code matches the write CRC code and generates a fail signal when the read CRC code does not match the write CRC code.
3. The semiconductor memory device as set forth in claim 1, wherein the CRC circuit comprises:
- a CRC engine receiving the data corrected by the ECC circuit and generating the read CRC code; and
- a comparator generating the pass signal when the read CRC code matches with the write CRC code, and generating the fail signal when the read CRC code does not match the write CRC code.
4. The semiconductor memory device as set forth in claim 1, wherein the semiconductor memory device is a NAND flash memory device comprising a cell array divided into main and spare regions, and
- wherein the data is programmed into the main region and the ECC and CRC codes are programmed into the spare region.
5. The semiconductor memory device as set forth in claim 4, wherein the ECC circuit detects two error bits and corrects one error bit, and
- wherein the CRC circuit detects more than two error bits.
6. A semiconductor memory device comprising:
- a cyclic redundancy check (CRC) circuit generating a write CRC code corresponding to data to be stored in memory cells;
- a first error correction code (ECC) circuit generating a first ECC code corresponding to the data and detecting and correcting a bit error of the data by means of the first ECC code during a read operation; and
- a second ECC circuit generating a second ECC code corresponding to the write CRC data and detecting and correcting a bit error of the write CRC code by means of the second ECC code during the read operation,
- wherein the CRC circuit generates a read CRC code corresponding to data corrected by the first ECC circuit during the read operation, and corrects a bit error of the data according to a comparison of the read CRC code and the write CRC code corrected by the second ECC circuit.
7. The semiconductor memory device as set forth in claim 6, wherein the CRC circuit generates a pass signal when the read CRC code matches the corrected write CRC code, and generates a fail signal when the read CRC code does not match the corrected write CRC code.
8. The semiconductor memory device as set forth in claim 6, wherein the CRC circuit comprises:
- a CRC engine receiving the data corrected by the first ECC circuit and generating the read CRC code; and
- a comparator generating the pass signal when the read CRC code matches the corrected write CRC code, and generating the fail signal when the read CRC code does not match the corrected write CRC code.
9. The semiconductor memory device as set forth in claim 6, wherein the semiconductor memory device is a NAND flash memory device that has a cell array divided into main and spare regions, and
- wherein the data is programmed into the main region and the first and second ECC codes and the CRC code are programmed into the spare region.:
10. The semiconductor memory device as set forth in claim 9, wherein the first and second ECC circuits each detect two error bits and correct one error bit, and
- wherein the CRC circuit detects more than two error bits.
11. The semiconductor memory device as set forth in claim 9, wherein the data, the first and second ECC codes, and the CRC code are programmed substantially simultaneously.
12. A method of detecting a bit error in a semiconductor memory device, comprising:
- generating error correction code (ECC) and cyclic redundancy check (CRC) codes corresponding to data to be stored in memory cells;
- storing the data in the memory cells;
- correcting a bit error for the data stored in the memory cells by means of the ECC code;
- generating a read CRC code corresponding to the data corrected with the bit error; and
- detecting a bit error of the data according to a comparison of the read CRC code to the write CRC code.
13. The method as set forth in claim 12, wherein correcting the bit error comprises treating the data as being failed when the number of error bits is over a correctable number of error bits.
14. The method as set forth in claim 12, further comprising determining a failure of the data when the read CRC code does not match the write CRC code.
15. The method as set forth in claim 12, wherein the semiconductor memory device is a NAND flash memory device comprising a cell array divided into main and spare regions,
- the method further comprising programming the data into the main region and programming the ECC and CRC codes into the spare region.
16. The method as set forth in claim 15, wherein correcting the bit error comprises detecting two error bits and correcting one error bit, and
- wherein detecting the bit error of the data comprises determining a bit error in a corrected bit error.
17. A method of detecting a bit error in a semiconductor memory device, comprising:
- generating a write cyclic redundancy check (CRC) code corresponding to data to be stored in memory cells;
- generating a first error correction code (ECC) code corresponding to the data to be stored in the memory cells, and a second ECC code corresponding to the write CRC code;
- programming the data, the write CRC code, and the first and second ECC codes into the memory cells;
- correcting a bit error of the data by means of the first ECC code, and a bit error of the write CRC code by means of the second ECC code;
- generating a read CRC code corresponding to the data corrected with the bit error; and
- detecting a bit error of the data according to a comparison of the read CRC code and the write CRC code corrected with the bit error.
18. The method as set forth in claim 17, wherein correcting the bit error further comprises treating the data as being failed when the number of error bits is over a correctable number of error bits.
19. The method as set forth in claim 17, wherein the semiconductor memory device is a NAND flash memory device comprising a cell array divided into main and spare regions, and
- the method further comprising programmed the data into the main region, and the first ECC codes, the second ECC codes, and the CRC code are programmed into the spare region.
20. The method as set forth in claim 19, wherein correcting the bit error comprises detecting two error bits and correcting one error bit, and
- wherein detecting the bit error of the data further comprises determining a bit error in a corrected bit error.
Type: Application
Filed: Oct 17, 2006
Publication Date: Jan 17, 2008
Applicant: Samsung Electronics Co., Ltd. (Suwon-si)
Inventors: Shea-Yun Lee (Seoul), Dong-Hyun Song (Yongsin-si), Jang-Hwan Kim (Suwon-si), Sang-Lyul Min (Seoul)
Application Number: 11/582,106
International Classification: G11C 29/00 (20060101); H03M 13/00 (20060101);