JP2004266524A - Frame synchronizing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a frame synchronizing method with which a frame starting code detection capability is more improved and a frame discard rate is reduced. <P>SOLUTION: A bit pattern of a preamble is set to make the number of bits which is different between a bit stream of an arbitrary continuous (n)-bit length extracted from a bit stream composed of the preamble and a frame starting code continued thereto and a bit stream of the frame starting code, equal to or greater than a humming distance (h). A synchronizing clock is extracted from the preamble in a received frame, and bit synchronism is established and transits into a hunt state. In the hunt state, the bit stream composed of the preamble and the frame starting code is shifted bit by bit successively from the top, the bit stream of the continuous (n)-bit length and the bit stream of the frame starting code are successively compared and when the different numbers of bits are within [(h-1)/2] (meaning an integer part of (h-1)/2) bits, it is judged that the frame starting code is detected, thereby starting a synchronizing state. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
[Field of the Invention]
The present invention relates to a frame synchronization method in an optical passive network (PON), and more particularly to a frame synchronization method in which the ability to detect a frame start code is further enhanced and the probability of frame discard is reduced.
[0002]
[Prior art]
An optical passive network (PON: transmission line branching multi-user connection network) is a technology for transmitting an ATM (Asynchronous Transfer Mode) cell between a subscriber line termination device and a subscriber accommodation device. 983.1 defines its function.
FIG. 1 shows a PON network configuration. The PON optically connects the subscriber accommodation device (OLT) 1 to a plurality of (n in the drawing) subscriber line terminating devices (ONUs) 2-1 to 2-n via an optical line and an optical branching circuit 3. . Between the OLT 1 and the ONU 2, an optical signal having a wavelength of 1.5 μm is used as a downstream signal from the OLT 1 to the ONU 2, and a 1.3 μm wavelength optical signal is used as an upstream signal from the ONU 2 to the OLT 1. In the meantime, transmission and reception of transmission frames are performed. When a single-core optical fiber is used, signal transmission is performed by wavelength multiplexing (WDM).
[0003]
The downstream frame selected from the OLT 1 is branched by the optical coupler (optical branching circuit) 3 and arrives at all the ONUs 2-1 to 2-n. The upstream frame transmitted from each ONU 2 is multiplexed by the optical coupler 3 and arrives at the OLT 1. At this time, the OLT 1 controls the transmission timing of the upstream frame from each ONU 2 so that the upstream frame from each ONU does not collide.
[0004]
FIG. 2 shows the configuration of the upstream frame of the PON. The upstream frame is composed of a guard time 10 for preventing collision, a preamble 11 for bit synchronization, a frame start code 12, a header 13 and a bay load 14 in order from the top, and a frame end code 15 is added as necessary. FIG. 2B shows an example in which the frame end code 15 is inserted at the last end of the upstream frame. Uplink synchronization performed in the OLT 1 is established by bit synchronization using a preamble 11 and synchronization of a code segment using a frame start code 12 (for example, see Non-Patent Document 1). That is, in the detection of the frame start code, a bit string having a length corresponding to the bit length of the continuous frame start code is shifted by m bits from the head of the bit string from the bit string composed of the preamble 11 and the frame start code 12 (m = 0, 1, 2,...) Are extracted, compared with the frame start code 12, and when they match with the start code, they are detected as frame start codes. In this case, assuming that the start code length is n bits and the transmission path error rate is ε, the probability of missing the start code is nC ₁ ε, which is proportional to the transmission path error rate.
[0005]
FIG. 3 shows an upstream frame defined by IEEE802.3ah, which is Non-Patent Document 2. In IEEE802.3ah, the 8B / 10B code / I / (= / K28.5 / D5.6 / or / K28 is preceded by the 8B / 10B code / S / (= / K27.7 /) which is the start code as the preamble. .5 / D16.2 /) is added as 2 × n (n is a natural number) symbols. n is indicated by the OLT 1 when the ONU 2 is started. The synchronization of the upstream frame from each ONU 2 is realized by exchanging a time stamp with a granularity of 16 ns between the OLT and the ONU, thereby synchronizing the counter of the OLT and the counter of each ONU. The transmission timing control of the upstream frame by the OLT is realized by the OLT using the counter value to instruct the ONU of the start position and the frame length.
[0006]
Regarding the 8B / 10B code, there are two 8B / 10B codes corresponding to the original 8-bit data, and either value is taken based on the running disparity (RD) rule (for example, Non-Patent Document 3). reference). The 8B / 10B code is divided into two blocks of 6 bits in the first half and 4 bits in the second half, and the number of 0s and 1s is calculated in each block. If the number of 1s is large, it is positive, and if it is small, it is negative. The transmitter controls the code so that the RD values are transmitted alternately. Also, the RD value at the time of starting the transmitter is defined as negative.
[0007]
[Non-patent document 1]
ITU-TR Recommendation G983.1, published October 1998, Table 6, p27-28
[Non-patent document 2]
IEEE 802.3ah / D1.0, issued August 19, 2002, 57, Chapter 2, GMII data strcam, pp. 142-143
[Non-Patent Document 3]
IEEE 802.3, 2000 Edition (ISO / IEC8802-3, 2000 Edition), Chapter 36.2, Physical coding sublayer (PCS), pp. 279-280; 965-993
[0008]
[Problems to be solved by the invention]
In a PON, application of an error correction code is being studied for the purpose of preventing signal deterioration due to dispersion or loss of a transmission line. In applying the error correction code, it is necessary to make the detection miss rate of the frame start code and the frame end code sufficiently smaller than the bit error rate (less than or equal to the bit error rate after error correction). The frame is discarded before the correction, and a problem occurs that the error correction cannot be performed efficiently. In this case, in the conventional frame start code detection method, since the probability of missing the start code is proportional to the error rate of the transmission path, a situation occurs in which the error detection code cannot be used effectively using the conventional detection method. I will. Further, for an upstream frame, bit synchronization and frame synchronization must be performed for each upstream frame from each ONU. In consideration of transfer efficiency, a frame synchronization method that achieves faster frame synchronization and a lower frame discard rate is desired. .
[0009]
Accordingly, it is an object of the present invention to provide a frame synchronization method in a PON in which the detection capability of a frame start code is further improved.
Another object of the present invention is to provide a frame synchronization method capable of more quickly performing frame synchronization in a PON.
Another object of the present invention is to provide a frame synchronization method in which the frame discard rate is greatly reduced.
[0010]
Means and effects for solving the problem
A frame synchronization method according to the present invention comprises connecting a plurality of subscriber line terminating devices and a subscriber accommodation device via an optical line and an optical branching circuit, and transmitting a transmission frame between the subscriber line termination device and the subscriber accommodation device. A frame synchronization method for an upstream frame of a passive optical network for transmitting and receiving
The uplink frame includes, at least in order from the beginning, a guard time for collision prevention, a preamble for bit synchronization, and a frame start code for specifying a start position of the frame,
Defines the Hamming distance h,
When the bit pattern of the preamble is n bits in the frame start code, an arbitrary continuous n-bit bit string extracted from a bit string composed of the preamble and a continuous frame start code And the number of bits that differ between the bit string of the frame start code and the
Upon receiving the signal transmitted from the subscriber line terminating device, the subscriber accommodating device extracts a synchronization clock from a preamble in a received frame, establishes bit synchronization, and enters a hunt state after establishing bit synchronization. Transition,
In the hunt state, a bit string having a preamble and a frame start code of a received frame is sequentially shifted by one bit at a time from the head thereof, and a bit string having a continuous n-bit length is sequentially extracted. The bit sequence of the code is sequentially compared, and the number of different bits is [(h-1) / 2] (where the expression [(h-1) / 2] is (h-1) / 2 If the number of bits is less than or equal to the integer part, it is determined that the frame start code has been detected, and the state shifts to the synchronous state.
[0011]
In order to solve the above problems, the present invention improves the ability to detect a frame start code by using the characteristics of a PON. That is, in the PON, since the OLT instructs the ONU on the transmission range of the upstream frame and the preamble pattern, the detection range of the frame start code of the upstream frame and the preamble pattern can be limited. As a result, the length of the continuous frame start code extracted from the bit string composed of the frame start code and the preamble and the start code by sequentially shifting m bits (m = 0, 1, 2,...) From the head of the bit string It is possible to increase the hamming distance with the corresponding bit string. Assuming that these Hamming distances are equal to or longer than h (the minimum distance is h), it is possible to detect a start code up to [h-1] / 2] bit errors. Taking an IEEE802.3ah frame as an example, the granularity of the time stamp used to synchronize the timing between the OLT and each ONU is 16 ns, and the OLT determines the position of the upstream frame transmitted from the ONU by a maximum of 16 ns. It can be specified by the error of Therefore, the detection range of the start code requires 16 ns (two codes) after the start code. Considering that the timing is shifted, the detection range may be up to several codes behind. Even if the detection range is not limited, if an error is detected behind the start code, the OLT knows the positional relationship between the frame start code and the end position of the frame, as described in claim 2. By using this, it is possible to easily grasp that an erroneous detection has been made.
[0012]
The preamble pattern is an idle code / I / of an 8B / 10B code, and has the property of making the RD value at the end of the code negative. The idle code /I1/=/K28.5/D5.6/ is used to invert the RD value, and /I2/=/K28.5/D16.2/ is used to hold the RD value. The preamble length is indicated by the OLT as the idle sequence length. Also, since the initial value of the RD value is negative, the preamble pattern arriving at the OLT is limited by the RD rule and is the following pattern.
/K28.5(-)/D16.2(+)/K28.5(-)/D16.2(+)/. . . /D28.5(-)/D16.2(+)/start code Here, / I /, / I1 /, / I2 // is an idle order set of 8B / 10B code, and /K28.5/ is 8B / 10B special code, comma (0011111xxx or 11000000xxx, where x includes 1 or 0. /D5.6/, /D16.2/ are 8B / 10B codes corresponding to 8-bit data. /K28.5(- ) / The code in parentheses indicates the leading RD value of the 8B / 10B code, which is positive for + and negative for-.
[0013]
The frame start code is /K27.7/D21.6/ having a length of 20 bits, and a bit string composed of the preamble of the received frame and the frame start code is extracted while shifting one bit at a time from the head of the bit string. If the Hamming distance h between the n-bit long bit string and the bit string of the frame start code is all h = 5 or more (the minimum distance is 5). This enables detection of the frame start code up to a two-bit error. In this case, the probability of missing the frame start code is about ₂₀ C ₃ epsilon ^3, transmission path error rate epsilon <10 ^- it can be seen that sufficiently smaller than the epsilon in the range of ^3. Also, the probability of erroneous synchronization part of the preamble is proportional to 3Ipushiron ^3, sufficiently smaller than the still epsilon. The probability of erroneous synchronization after the frame start code is approximately (probability of overlooking the frame start code) × (probability that a code string that is one bit different from the frame start code is erroneously detected as the frame start code due to a one-bit error). Yes, and will not be greater than the probability of missing a frame start code.
[0014]
BEST MODE FOR CARRYING OUT THE INVENTION
Embodiment 1
FIG. 4 shows an embodiment of the present invention. The upstream frame transmitted from the subscriber line terminating device and received by the subscriber accommodating device includes a guard time for preventing collision, a preamble for bit synchronization, and a frame start code for specifying a start position of the upstream frame in order from the beginning. , And a payload with a header. If necessary, a frame end code for specifying the end position of the frame is added to this.
[0015]
First, detection of a frame start code will be described. The bit length of the frame start code is n bits. Further, the Hamming distance is defined as h. In this example, the bit pattern of the preamble differs between an arbitrary continuous n-bit-length bit sequence extracted from a bit sequence composed of a preamble and a continuous frame start code and a bit sequence of a frame start code. The bit pattern of the preamble is set in advance so that the number of bits (Hamming distance) is h or more. Note that the detection range of the frame start code is up to A symbols after the frame start code. Here, the number of the A symbols is set to be approximately equal to the timing error between the OLT and the ONU.
[0016]
The subscriber accommodating apparatus starts receiving an upstream frame transmitted from the subscriber line terminal (20). When an uplink frame is received, a synchronization clock is extracted from the preamble in the received frame to establish bit synchronization (21). Next, after establishing bit synchronization, the state transits to the hunt state. In the hunt state, successive n-bit length bit strings are sequentially extracted from the bit string formed by the preamble of the received frame and the frame start code while shifting one bit at a time from the beginning. Then, the extracted bit string and the bit string of the frame start code are sequentially compared, and if the number of different bits is within [(h-1) / 2] bits of the frame start code, it is determined that the frame start code has been detected. Then, the state transits to the frame synchronization state (23). On the other hand, if a bit string within [(h-1) / 2] bits of the frame start code is not detected in the detection range, synchronization fails (24). Here, the expression [(h-1) / 2] means taking an integer part of (h-1) / 2.
[0017]
Embodiment 2
FIG. 5 shows another embodiment of the present invention. The difference from FIG. 4 is that a transition from the frame synchronization state (23) to the frame discard state (25) is added. That is, in the frame synchronization state, after the end of the frame is detected, the positional relationship between the frame end position assumed by the OLT and the frame start position does not match the positional relationship calculated from the frame start code detected by the OLT and the frame end position. Transitions the upstream frame to discard. According to this example, even if erroneous synchronization occurs after the frame start code, the erroneous synchronization can be determined, so that the detection range of the frame start code can be set as the upstream frame transmission range.
[0018]
Embodiment 3
FIG. 6 shows another embodiment of the present invention. The difference from the embodiment shown in FIG. 4 is that the transmission frame is an Ethernet (registered trademark) frame with a MAC preamble in which the 8B / 10B encoding is performed, and the upstream frames are arranged in order from the top, a guard time for collision prevention and a bit synchronization. And a preamble for detecting a code position delimiter, a frame start code for specifying a start position of an upstream frame, an Ethernet (registered trademark) frame with a MAC preamble, and a frame end code for specifying an end position of the frame. . If this frame is a frame defined by IEEE802.3ah, and the start code is /S/D21.6=/K27.7/D21.6 of the 8B / 10B code, the minimum value (minimum distance) of the Hamming distance h Is h = 5. Also, since the time stamp in 1EE802.3ah is 16 ns (2 symbol length), the start code detection range may be about 2 symbols after the start code or about 3 symbols with a margin. However, if there is an allowable value of the fluctuation of the arrival of the upstream frame, the allowable value is added to this value. In the hunt state, the same processing as in the first embodiment is performed.
[0019]
Embodiment 4
FIG. 7 shows another embodiment of the present invention. The difference from FIG. 5 is that, similar to the third embodiment, the transmission frame is an Ethernet (registered trademark) frame with a MAC preamble in which 8B / 10B encoding is performed, and the upstream frames are sequentially arranged from the top in order to prevent guard time for collision prevention. , A preamble for detecting bit synchronization and code position delimitation, a frame start code for specifying the start position of an upstream frame, an Ethernet (registered trademark) frame with a MAC preamble, and a frame end code for specifying the end position of the frame. is there.
This frame is a frame defined by IEEE802.3ah. Assuming that the start code is /S/D21.6/=/K27.7/D21.6 of the 8B / 10B code, the hamming distance h is h = 5. is there. The start code detection range may be about two symbols after the start symbol or three symbols with a margin since the time stamp in IEEE802.3ah is 16 ns (two symbol length). However, if there is an allowable value of the fluctuation of the arrival of the upstream frame, the allowable value is added to this value. Since the end position of the upstream frame does not have six or more consecutive bits 0 in the 8B / 10B code, it can be determined that the upstream frame has ended when six or more bits 0 continue. If the relationship between the frame start code and the end position does not match the predetermined positional relationship, the frame is discarded. Note that the processing in the hunt state and the frame discard processing in the case where the end positions do not match are the same as in the second embodiment.
[0020]
Embodiment 5
FIG. 8 shows still another embodiment of the present invention. In this example, similarly to the fourth embodiment, frame synchronization of an Ethernet (registered trademark) frame with a MAC preamble in which a transmission frame is 8B / 10B encoded will be described. The upstream frame includes, in order from the beginning, a guard time for collision prevention, a preamble for detecting bit synchronization and code position delimiter, a frame start code for specifying the start position of the upstream frame, an Ethernet frame with a MAC preamble, and a frame. Has a frame end code that specifies the end position of. The upstream frame includes a plurality of commas in the preamble. In a normal Ethernet (registered trademark) frame, code boundaries can be synchronized by detecting a comma. This embodiment is an example in which a normal synchronization procedure of an Ethernet (registered trademark) frame is also performed at the same time.
[0021]
Synchronization is established and the state transits to the hunt state. In the hunt state, as in the above-described embodiment, a bit string composed of the preamble of the received frame and the frame start code is shifted by one bit from the beginning, and has a continuous n-bit length (bit of the frame start code). ) Is sequentially extracted. Then, the extracted bit sequence and the bit sequence of the frame start code are sequentially compared. In parallel, a comma detection is also performed in pre-synchronization 1 (26). This comma detection is performed by sequentially shifting the bit sequence composed of the preamble of the received frame and the frame start code one bit at a time from the beginning, and setting the bit length corresponding to the bit length of the comma (for example, 7 bits). Bit strings are sequentially extracted. Then, the extracted bit string and the comma bit string are sequentially compared. In this comma detection, if the extracted bit string matches the comma bit string, it is determined that a comma has been detected, and a transition is made to pre-synchronization 1.
[0022]
When detecting the frame start code, if the number of different bits is less than [(h-1) / 2] bits of the frame start code in the comparison process between the bit sequence of the frame start code and the extracted bit sequence Is determined to have been detected, and a transition is made to the frame synchronization state (23). In this case, the state transits to the frame synchronization regardless of the result of the comma detection executed in parallel. On the other hand, if no bit string within [(h-1) / 2] bits of the frame start code is detected in the detection range, the result of the comma detection performed in pre-synchronization 1 is referred to. In this case, if an event in which a comma is detected in the last two symbols of the comma detected immediately before occurs consecutively N-1 times (for example, N = 3), a transition is made to pre-synchronization 2. At this time, if a comma is not detected, the process shifts to hunt execution. In the pre-synchronization 1, a frame start code is detected by shifting one symbol at a time. Even if a comma is not detected in two symbols after the comma detected immediately before, if a frame start code is detected, the state transitions to the frame synchronization state without transition to the hunt state. In pre-synchronization 2, a bit string is extracted by shifting one symbol at a time, and only detection of a frame start code is executed. This detection of the frame start code again can be, for example, the same processing as the above-described frame detection.
[0023]
In pre-synchronization 1, if the detection range ends before the comma is detected N-1 times during the period of two symbols after the comma detected immediately before, a transition to synchronization failure occurs (24). Further, even if the frame start code is detected in the pre-synchronization 2, if the frame start code is not detected within the detection range, the synchronization similarly fails. On the other hand, when the frame start code is detected in pre-synchronization 2, the state transits to frame synchronization (23). In this way, by detecting the comma in parallel with the detection of the frame start code and establishing the symbol synchronization by detecting the frame start code even if no comma is detected, the preamble required for symbol synchronization can be greatly reduced. And the transfer efficiency can be greatly improved.
[0024]
The frame described in this example is a frame defined by IEEE802.3ah. If the start code is /S/D21.6/=/K27.7/D21.6/ of an 8B / 10B code, h = 5. It is. Further, if the preamble is composed only of /K28.5/D16.2/, h = 6. The start code detection range is. Since the time stamp in IEEE802.3ah is 16 ns (2 symbol length), it may be about 2 symbols or 3 symbols after the start code. However, if there is an allowable value for the fluctuation of the arrival of the upstream frame, the allowable value is added to this provision.
[0025]
Embodiment 6
FIG. 9 shows another embodiment of the present invention. In this embodiment, a frame discarding step is added to the embodiment shown in FIG. In the 8B / 10B code, since six or more bits 0 are not consecutive, it is possible to determine the end position of the upstream frame from the fact that six or more bits 0 are consecutive. Therefore, in this example, in the frame synchronization state, after the end of the frame, the positional relationship between the assumed frame end position of the OLT and the frame start code, and the positional relationship calculated from the frame start code detected by the OLT and the end position of the upstream frame. Do not match, the upstream frame is shifted to discard, and the upstream frame is discarded. In this example, even if erroneous synchronization occurs after the frame start code, the erroneous synchronization can be determined, so that the detection range of the frame start code can be set as the transmission range of the upstream frame.
[0026]
【The invention's effect】
As described above, according to the present invention, the ability to detect the start code of an upstream frame in a PON can be improved, and frame discarding due to a bit error can be prevented.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating a configuration of an example of a PON.
FIG. 2 is a diagram illustrating a configuration of an example of an upstream frame.
FIG. 3 is a diagram showing an Ethernet (registered trademark) (registered trademark) frame with a MAC preamble that has been 8B / 10B encoded.
FIG. 4 is a diagram showing a frame synchronization processing step according to the first embodiment of the present invention.
FIG. 5 is a diagram showing a frame synchronization processing step according to a second embodiment of the present invention.
FIG. 6 is a diagram showing a frame synchronization processing step according to a third embodiment of the present invention.
FIG. 7 is a diagram showing a frame synchronization processing step according to a fourth embodiment of the present invention.
FIG. 8 is a diagram showing a frame synchronization processing step according to a fifth embodiment of the present invention.
FIG. 9 is a diagram showing a frame synchronization processing step according to a sixth embodiment of the present invention.
[Explanation of symbols]
REFERENCE SIGNS LIST 1 subscriber accommodation device 2 subscriber line termination device 3 coupler 10 guard time 11 preamble 12 frame start code 13 header 14 payload 15 frame end code

Claims (5)

Upstream of a passive optical network that connects a plurality of subscriber line terminating devices and a subscriber accommodation device via an optical line and an optical branching circuit, and transmits and receives a transmission frame between the subscriber line termination device and the subscriber accommodation device. A frame synchronization method for frames,
The uplink frame includes, at least in order from the beginning, a guard time for collision prevention, a preamble for bit synchronization, and a frame start code for specifying a start position of the frame,
Defines the Hamming distance h,
When the bit pattern of the preamble is n bits in the frame start code, an arbitrary continuous n-bit bit string extracted from a bit string composed of the preamble and a continuous frame start code And the number of bits that are different between the bit string of the frame start code and the bit string of the frame start code are set in advance so as to be equal to or longer than the Hamming distance h,
Upon receiving the signal transmitted from the subscriber line terminating device, the subscriber accommodating device extracts a synchronization clock from a preamble in a received frame, establishes bit synchronization, and enters a hunt state after establishing bit synchronization. Transition,
In the hunt state, a bit string having a preamble and a frame start code of a received frame is sequentially shifted by one bit at a time from the head thereof, and a bit string having a continuous n-bit length is sequentially extracted. The bit sequence of the code is sequentially compared, and the number of different bits is within [(h-1) / 2] bits of the frame start code (where the expression [(h-1) / 2] is (h-1) / 2 means an integer part), it is determined that a frame start code has been detected, and a transition is made to a synchronous state. 2. The frame synchronization method according to claim 1, wherein the upstream frame has a payload including data to be transmitted continuously after a frame start code and a frame end code. 3. When the code is detected, the positional relationship between the detected frame start code and the frame end code is compared with the positional relationship between the predetermined frame start code and the frame end code. A frame synchronization method comprising discarding an upstream frame. 3. The frame synchronization method according to claim 1, wherein the transmission frame is an Ethernet (registered trademark) frame with a MAC preamble encoded by 8B / 10B, and an upstream frame includes a guard time for preventing collision in order from the top. , A preamble for bit synchronization composed of an 8B / 10B idle code, a frame start signal for specifying a start position of an upstream frame, an Ethernet (registered trademark) frame with a MAC preamble, and a frame end position. A frame end signal. 4. The frame synchronization method according to claim 3, wherein after the bit synchronization is established, a transition is made to a hunt state.
With respect to the bit sequence composed of the preamble of the received frame and the frame start code, successive n-bit length bit sequences are sequentially extracted while sequentially shifting one bit at a time from the head thereof, and the extracted bit sequence and the bit sequence of the frame start code are extracted. Are sequentially compared with each other, and a bit string composed of the preamble and the frame start code of the received frame is shifted in parallel one bit at a time from the beginning of the bit sequence formed in parallel with the received frame. A comma detection step of sequentially extracting a continuous bit string having a bit length equal to the bit length of the comma included in the idle code and sequentially comparing the extracted bit string with the comma bit string,
In the frame start code detecting step, when a bit string whose bit difference number is within [(h-1) / 2] of the frame start code is detected, it is determined that the frame start code has been detected, and a transition is made to a synchronous state.
If it is determined that the frame start code has not been detected in the frame start code detection step, and if the bit string matches the comma bit string in the comma detection step, it is extracted at the boundary of the 8B / 10B code in the next detection step. When the beginning of the bit string matches, a frame start code detecting step is performed, and a comma detecting step is performed on the bit string two codes after the leading 8B / 10B code from the head of the comma detected immediately before. And if neither the frame start code is detected, the head of the bit string to be extracted is again shifted one bit at a time, and the detection step is executed.
In the comma detection step performed every two codes, it is determined whether or not the bit string matches the comma's bit continuation consecutively N times in a predetermined manner. A frame synchronization method comprising: performing a frame start code detection step by aligning a head of a bit string to be extracted with a boundary. 5. The frame synchronization method according to claim 4, wherein when the subscriber accommodation device detects a frame end code in a synchronized state, the positional relationship between the detected frame start code and the detected frame end position and a predetermined frame. A frame synchronization method comprising comparing a positional relationship between a start code and a frame end position, and discarding the upstream frame if they do not match.

Images