KR100901521B1 - Heuristic traffic load balanced HTLB channel sharing method and apparatus, medium accessing method using the channel sharing method, medium accessing network system using the channel sharing apparatus and recording medium storing program for performing the methods thereof - Google Patents

Abstract

The present invention discloses a channel sharing method and apparatus used for data packet transmission in a wavelength division multiple access network using a passive star coupler, a medium access method and a medium access network system. In particular, the channel sharing method in the wavelength division multiple access network of the present invention comprises the steps of: receiving traffic load information from each of the stations; Generating a traffic load information matrix according to each transmitting station and receiving station using the received traffic load information; And determining a station group using at least some column vectors or row vectors selected from column vectors and row vectors constituting the traffic load information matrix. According to the present invention, it is efficient for a pre-allocation based media access protocol and has a lower time complexity compared to a conventional sub-optimal traffic load balanced CSM and MULTIFIT CSM. It is possible to implement a medium access network system having a.

WDMA, channel sharing method, heuristic, media access method

Description

Heuristic traffic load balanced channel sharing method and apparatus, a medium access method using the channel sharing method, a medium access network system using the channel sharing apparatus, and a recording medium on which a program for performing the methods is recorded. channel sharing method and apparatus, medium accessing method using the channel sharing method, medium accessing network system using the channel sharing apparatus and recording medium storing program for performing the methods

The present invention relates to a traffic load balanced channel sharing method and apparatus, a medium access method using the channel sharing method, a medium access network system using the channel sharing apparatus, and a recording medium on which a program for performing the method is recorded. A channel sharing method for enabling efficient packet transmission in a wavelength division multiple access network using a star coupler.

A wavelength division multiple access network using a passive star coupler based on wavelength division multiplexing (WDMA) is a widely known optical network structure that can cover urban areas and efficiently solve the gap with a macro. Macro-wavelength division multiplexing (WDMA) networks with passive star coupler topologies can be applied to connect high-speed backbones (large transmission lines, aggregations), and to connect networks.

Among the existing media access control schemes for efficient packet transmission of the media access control layer, a protocol for controlling media access is mainly based on pre-allocation, reservation, and contention based media access control. It is classified by protocol. In particular, the pre-location based media access control protocol is based on a scheduling algorithm that periodically assigns a set of transmitting and receiving stations (SDP) and a specific wavelength based on the transmit-receiving station set diagram. This has the advantage of preventing various network collisions caused by transmission, reception, or contention of channels over the other two protocols. In addition, the pre-location-based media access control protocol has the advantage that the hardware of the system to be applied is not complicated, has a simple packet structure, and guarantees high throughput.

Wavelength division network systems using passive star couplers generally require stations to share channels (eg wavelength) for packet transmission due to the limited number of channels. Furthermore, given that massive amounts of traffic are pouring from various applications, such as VoIP, VoD, IPTV, and interactive conferencing at the same time, an efficient channel sharing method (CSM) is needed to balance the asymmetric multimedia traffic load. In addition, existing media access algorithms, particularly pre-allocation based MACs, need to be modified.

In the case of a pre-fixed CSM, the source group (group of stations sharing the same channel) is predetermined without attempting traffic load balancing. Unlike the above scheme, a conventional CSM scheme for attempting load balancing is a MULTIFIT CSM scheme proposed by E. G. Coffman et al.

MULTIFIT CSM preferentially determines the sum of minimum and maximum traffic loads by grouping stations first, and repeats grouping until the sum of minimum traffic loads is equal to the sum of maximum traffic loads. In addition, MULTIFIT CSM considers the traffic loads of all receiving stations while performing traffic load balancing, which is inefficient in terms of time complexity and traffic load balancing, especially for pre-allocation based MAC protocols. have.

In view of the above-described prior art, the present invention is suitable for pre-allocation based media access protocols by balancing traffic loads according to pre-allocated destination stations per time slot, and consistent traffic load balancing in various traffic load environments. An object of the present invention is to provide a novel channel sharing method and apparatus capable of doing the same.

In addition, an object of the present invention is to provide a medium access method using the channel sharing method, a medium access network system using the channel sharing apparatus, and a recording medium on which a program for performing the methods is recorded.

According to an aspect of the present invention, there is provided a channel sharing method in a wavelength division multiple access network, comprising: receiving traffic load information from each station; Generating a traffic load information matrix according to each transmitting station and receiving station using the received traffic load information; And determining a station group using at least some column vectors or row vectors selected from column vectors and row vectors constituting the traffic load information matrix.

According to another aspect of the present invention, there is provided a channel sharing apparatus in a wavelength division multiple access network, including: a traffic load information generation unit configured to generate traffic load information according to stations; And receiving traffic load information from each of the stations, using the received traffic load information, generating a traffic load information matrix according to each transmitting station and a receiving station, and forming the generated traffic load information matrix. And a group determiner for determining a station group using at least some column vectors or row vectors selected from among the column vectors and the row vectors.

In accordance with another aspect of the present invention, there is provided a method of accessing a medium in a wavelength division multiple access network, comprising: transmitting, by each station, traffic load information to other stations at the same time; Generating a traffic load information matrix corresponding to each transmitting station and a receiving station using the traffic load information; Determining a station group by using at least some column vectors or row vectors selected from among column vectors and row vectors constituting the traffic load information matrix; And transmitting data to be transmitted by a transmitting station included in the determined station group to a receiving station included in the determined station group.

In accordance with another aspect of the present invention, there is provided a medium access network system in a wavelength division multiple access network, comprising: a transmission medium for transmitting data from stations; A coupler for coupling or branching a signal entering and exiting through the transmission medium; A traffic load information generator for generating traffic load information according to each station; And receiving traffic load information from each of the stations, using the received traffic load information, generating a traffic load information matrix according to each transmitting station and a receiving station, and forming the generated traffic load information matrix. And a group determiner which determines a station group by using at least some column vectors or row vectors selected from among the vectors and the row vectors.

The present invention also provides a computer-readable recording medium having recorded thereon a program for performing a channel sharing method and a medium access method in a wavelength division multiple access network on a computer.

According to the present invention, a heuristic traffic load balanced channel sharing method (HTLB CSM) is adopted as a channel sharing method for traffic load balancing, and the traffic load is balanced according to a pre-allocated destination station per time slot, thereby pre-allocation based Can improve the transmission efficiency of the medium access protocol. In particular, the HTLB CSM of the present invention has a lower time complexity than the existing sub-optimal traffic load balanced CSM and MULTIFIT CSM. In addition, the HTLB CSM has a higher Jane Index compared to MULTIFIT and pre-fixed CSMs, despite a variety of traffic environments, and therefore more consistent traffic load balancing than other channel sharing methods. Performance can be expressed.

Hereinafter, with reference to the accompanying drawings and embodiments, heuristic traffic load balanced channel sharing method and apparatus of the present invention, a medium access method using the channel sharing method, a station and a medium access network system using the channel sharing apparatus and the method The recording medium on which the program to be executed is recorded will be described in detail.

1 is a block diagram illustrating a medium access network system having a passive star coupled topology according to an embodiment of the present invention. The media access network system shown in FIG. 1 includes a passive star coupler 10, N stations 20, stations 1 -N and a transmission medium 30.

The passive star coupler 10 is a coupler that enables data communication between stations by combining or branching data packets generated at each station. The passive star coupler 10 is connected via a transmission medium 30 to enable data communication with the N stations. Unlike the present embodiment, it is also easy to implement a medium access network system using an active star coupler and, as described in the claims of the present invention, of course.

Each of the stations 1 to N 20 carries a data packet on a transmission medium and transmits the data packet to another station through a passive star coupler or receives a data packet from another station. Each station is connected to a passive star coupler 10 via a transmission medium 30. In general, since the number N of stations is larger than the number C of available wavelengths, some stations inevitably share the same channel, and it is necessary to group the same channel, that is, stations sharing the same wavelength. An algorithm according to a heuristic traffic load balanced (HTLB) channel sharing method (CSM), which is a method for grouping stations, is performed at each station, which will be described later.

The transmission medium 30 is a medium connecting the N stations and the coupler. There is no particular limitation on the type of transmission medium, but an optical fiber having a wide bandwidth and a high transmission speed is preferable.

2 is a block diagram illustrating a station 20 including a channel sharing apparatus according to an embodiment of the present invention. The station shown in FIG. 2 includes a processor 110, a first transmitter 120, a first receiver 130, a second transmitter 140, a second receiver 150, and a buffer 160. The station 20 is used as a transmitting station or a receiving station of a multiple access network system, in particular a wavelength division multiple access network.

The processor 110 includes a channel sharing device 112 and a medium access control (MAC) protocol processor 118, and the channel sharing device 112 includes a traffic load information generator 114 and a group determiner 116. Include.

The traffic load information generator 114 generates traffic load information for each station. The traffic load information is information indicating the degree of traffic load between a transmitting station transmitting data and a receiving station receiving the transmitted data. Traffic load information can be obtained by two parameters. One is traffic load patterns (TLPs) and one is traffic distribution patterns (TDPs). In particular, the traffic load information α _ij is expressed by Equation 1 below.

Equation 1

α _ij = σ _i × p _ij

Where σ _i is the normalized traffic load flowing into station i, p _ij is the probability that station i sends a packet to station j, and α _ij is normalized traffic when the transmitting station is i and the receiving station is j Indicates the load value. In other words, TLPs represent asymmetry and the traffic load flowing to all stations. TDP refers to the nonuniformity of traffic distribution from one station to another. The traffic load information may be normalized to be 0 ≦ α _ij ≦ 1, and the traffic load information may be represented by a decimal number of specific positions. For example, α ₄₅ = 0.45 indicates that the traffic load information from station 4 to station 5 is 0.45. When the traffic load information is represented to the second decimal place as described above, each number can be represented by 4 bits, so a total of 8 bits are required.

The group determiner 116 receives the traffic load information from each of the stations, and generates a traffic load information matrix (see FIG. 4) according to each transmitting station and the receiving station by using the received traffic load information. A station group is determined using at least some column vectors or row vectors selected from the column vectors and the row vectors constituting the generated traffic load information matrix. Here, the station group refers to a group using a specific wavelength for data transmission together.

In particular, the group determiner 116 searches for the maximum traffic load and the minimum traffic load in the column vector according to the first receiving station among the receiving stations, and transmits according to the found maximum traffic load and the minimum traffic load. Determine the stations as a first group, the maximum traffic load and the minimum traffic in the column vector according to the second receiving station of the received stations, excluding traffic load according to the transmitting stations determined as the first group It is preferable to configure the search for the load, determine the transmitting stations according to the maximum traffic load and the minimum traffic load found as the second group, and repeat the above process until the group has no pending transmission stations.

In addition, the group determining unit 116 searches for the maximum traffic load in each column vector according to the receiving stations of at least some of the receiving stations, and respectively transmits the stations according to the found maximum traffic load. Determining components of different station groups, searching for the maximum traffic load or the minimum traffic load in each of the column vectors with the station group excluding the determined transmission stations, and the searched maximum traffic load or minimum Determine a station group of transmitting stations according to the traffic load of the station, and determine the station group so that the sum of the traffic loads of the transmitting stations constituting each station group has a predetermined uniformity, and the group does not have any undetermined transmission stations. It is configured to repeat the above process until It is.

The medium access control (MAC) protocol processor 118 performs signal processing to control access to the transmission medium in order to share the transmission capability of the transmission medium. In particular, the MAC protocol processor 118 preferably controls media access based on pre-allocation.

In accordance with the periodic scheduling of source and destination station pairs, i.e., SDPs, for transmission, the pre-location based MAC protocol has a constant bit rate (CBR) and inconspicuous burstiness. Has the advantage of sending traffic that provides traffic flow with The pre-location based MAC protocol is implemented by pre-allocating SDPs and mapping channels on the basis of a source-destination map. The MAC protocol is meant as a hybrid of WDMA and TDMA. Unlike other MAC protocols, the pre-location based MAC protocol provides simultaneous packet transmission of all stations. The pre-allocation-based MAC protocol is advantageous in terms of “broadcast-and-select” in WDMA networks using passive star couplers.

Compared with other MAC protocols, packet transmissions of all stations are implemented simultaneously without control of signaling, simplifying hardware complexity since they have a simplified MAC algorithm and simple topological embedding. In particular, when compared to reservation based MAC protocol, more than one channel can be used for packet transmission. Moreover, pre-allocation of SDPs allows packet transmission without any contention or conflict. However, there are some problems with the pre-location based MAC protocol. For example, in an environment with light traffic load, the generation of idle time slots for a particular SDP results in low usage of the channel, and also results in continuously lower data rates. In addition, in order to be applied to various multimedia applications, it is necessary to be suitable for various traffic types, the present invention relates to a channel sharing method that can improve the transmission characteristics even under the input traffic conditions of various environments.

The first transmitter 120 transmits traffic load information according to the station of FIG. 2 to other stations. In addition, the first transmitter 120 transmits a control signal for preparing the receiving station so that the receiving station can receive data. That is, when a packet arrives at the station, the first transmitter transmits a control signal for pre-reservation to a type slot through a control channel. Since the first transmitter 120 transmits the traffic load information and the control signal, the first transmitter 120 is preferably implemented as a fixed-wavelength transmitter having a fixed transmission wavelength.

The first receiver 130 receives traffic load information from other stations. In addition, the first receiving unit 130 receives an acknowledgment signal, that is, an ACK signal from the receiving station. That is, whether or not the reservation is successful from each receiving station is received according to the reservation result for every time slot. Since the first receiver 130 receives the grant signal and does not need to convert the wavelength, the first receiver 130 may be implemented as a fixed-wavelength receiver in which the reception wavelength is fixed.

The second transmitter 140 transmits data through a transmission medium in a control cycle of the first transmitter 120 and the second receiver 130 and a data cycle after the ACK cycle. Like the first transmitter, the second transmitter 140 is preferably implemented as a fixed wavelength transmitter.

The second receiver 150 receives data from another station. Since the wavelengths used between stations are different from each other, a turnable-wavelength receiver capable of converting wavelengths is preferable.

The buffer unit 160 temporarily stores data received through a transmission medium and data to be transmitted to another station.

3 is a flowchart illustrating a method of sharing a heuristic traffic load balanced (HTLB) channel according to an embodiment of the present invention. The method shown in FIG. 3 includes the following steps performed in time series at the channel sharing device 112.

In step 210, the station i receives traffic load information from all stations. Traffic load information is information α _ij indicating the degree of traffic load between the transmitting station i transmitting data and the receiving station j receiving the transmitted data.

In operation 220, the station i generates a traffic load information matrix M.

4 is a reference diagram illustrating an example of a traffic load information matrix M generated in this embodiment. In the matrix M shown in Fig. 4, a row indicates traffic load information according to any receiving station when the transmitting stations are the same, and a column shows traffic load information according to any transmitting station when the receiving stations are the same. ).

In operation 230, the station i determines a station group by using the traffic load information matrix. Station group c (1 ≦ c ≧ C) is determined based on traffic load information for all stations. All stations 1 to N forward their traffic load information to other stations. If k bits are available here to represent i for specifying the station, then the total number N of stations becomes 2 ^k . When the traffic load information is expressed to two decimal places, the number of bits required is eight, and since there are a total of N stations, the total number of bits required is k + 8N.

The traffic load information matrix M shown in FIG. 4 is updated every single time slot. In particular, the column vector of the traffic load information matrix is information about the normalized traffic load at which a particular receiving station j can receive packets from another station. The column vector corresponds to a pre-assigned destination (receiving station) in a time slot in a pre-allocation based MAC protocol. In this embodiment, the column vector is used to determine a group of stations while balancing traffic load.

FIG. 5 shows an example of an algorithm (LB-I CSM) for determining a station group in step 230 of FIG. 4. The LB-I CSM algorithm is a method of determining a station having a maximum traffic load and a minimum traffic load as one station group in order to balance traffic loads in a heat vector according to a pre-assigned destination station.

The LB-I CSM is based on the traffic load information matrix M, and load balancing is performed in ascending order of the station group c. First, for one column vector, the LB-I CSM algorithm selects stations corresponding to N / C (corresponding to s in FIG. 5) as stations forming one group. Half of the stations selected, s / 2, are the stations with the highest traffic load in the column vector, and the other half are the stations with the least traffic load. Next, the grouping of the station according to another column vector excludes the already grouped station, and selects the station in the same manner as the previous column vector.

FIG. 6 is a reference diagram for explaining a concept of the algorithm (LB-I CSM) of FIG. 5. The example shown in FIG. 6 is the case where N = 8 and C = 4. The row direction is the transmission stations 1-8, and the column direction shows the destination stations 1-8 in order. Here, it is assumed that the destinations, that is, the receiving stations 1, 2, 3, and 4 are preallocated in the current time slot. Therefore, the group determiner determines the group of stations in consideration of only the first, second, third and fourth column vectors in the traffic load information matrix M. Since N / C is 2, two stations are selected in each column vector, one station with the highest traffic load and one station with the least traffic load.

Stations 4 and 6 are organized into a first group because the station with the highest traffic load 0.35 in the first column vector 240 is station 4 241 and the station with the least traffic load 0.06 is station 6 242. . That is, S ₁ = {4, 6}, and R '= {1, 2, 3, 5, 7, 8}. Next, since the station with the highest traffic load in the second column vector 250 is station 8 252 and the station with the least traffic load is 2 251, stations 8 and 2 are organized into a second group. . That is, S ₂ = {2, 8} and R '= {1, 3, 5, 7}. Station 1 261 and station 5 262 can also be selected as the third group of stations for the third column vector 260, and station 3 271 and station 7 for the fourth column vector 270. 272 may be selected as the fourth group of stations.

FIG. 7 illustrates another example of an algorithm for determining a station group (LB-II CSM) in step 230 of FIG. 4. According to the LB-II CSM algorithm, the group determiner determines each station having the highest traffic load according to each column vector as a different group, and calculates the sum Lsum _c of traffic loads belonging to each group. In initial load balancing for the destination stations, the group determiner selects the station with the maximum traffic load in each column vector. The selected stations are organized in ascending order of c (station group number), and stations that are already grouped are excluded from subsequent groupings. After this grouping is performed, the group determining unit calculates the sum of the traffic loads. Next, the stations are selected in descending order of the sum of traffic loads. This process is repeated until all stations are grouped.

The LB-II CSM has a higher time complexity than the LB-I CSM because it calculates the sum of traffic loads. That is, although the time complexity of the HTLB CSM of the present invention is O (s × C) = O (N), the MULTIFIT CSM is O {N log (CN)}, which shows that the HTLB CSM is better in terms of time complexity. have. The station group of the present invention can confirm that the overall traffic load sum is well balanced.

FIG. 8 is a reference diagram for explaining a concept of the algorithm (LB-II CSM) of FIG. 7. In the case of FIG. 8, N = 8 and C = 4 similarly to FIG. The destination stations pre-allocated in the current time slot are assumed to be 1, 2, 3, 4. The group determiner may select one station having the maximum traffic load in each column vector as one group. Here, the group number (c) is determined in ascending order, except for already grouped stations. That is, the station with the maximum traffic load in the first column vector 310 is station 4 311, the station with the maximum traffic load in the second column vector 320 is station 8 321, and the third column. The vector 330 is a station 5 331, and the fourth column vector 340 is a station 1 341.

Further, the group determining unit calculates the sum of traffic loads according to each group, that is, the sum of the station group 1 (S ₁ ) is Lsum ₁ , the sum of the station group 2 (S ₂ ) is Lsum ₂ , and the station group 3 The sum of (S ₃ ) is Lsum _3, and the sum of station group 4 (S ₄ ) is Lsum ₄ . That is, Lsum ₁ = 0.35, Lsum ₂ = 0.60, Lsum ₃ = 0.62, Lsum ₄ = 0.48. Using the result of the sum of the traffic loads, the group determining unit re-determines the grouping order. Ie S ₁ -> S ₄ -> S ₂ -> Determine the grouping order with S ₃ . Next, the group determination section selects the station 2 312 having the maximum traffic load in the first column vector corresponding to S ₁ , and then selects the above sequence, that is, S _4. -> S _2- > S ₃ to determine the group.

In the present invention, the heuristic traffic load balanced channel sharing method (HTLM CSM) is a concept including both the above-described LB-I CSM and LB-II CSM, and means LB-II CSM.

9 is a flowchart illustrating a medium access method of a wavelength division multiple access network according to an embodiment of the present invention. The media access method shown in FIG. 9 includes the following steps performed in time series at a station 20 constituting a wavelength division multiple access network.

In step 411, the station 20 simultaneously broadcasts traffic load information to all stations. The traffic load information is information according to Equation 1, and is a value obtained by normalizing the degree of traffic load according to a transmitting station and a receiving station. From the point of view of the data frame, this step is where the traffic cycle is performed.

In step 421, any station i determines a station group to which it belongs according to a heuristic traffic load balanced channel sharing algorithm. From the point of view of the data frame, this step is where the traffic cycle is performed. Since the station (i) has been described above to perform the station grouping, a description thereof will be omitted.

In step 431, the station i determines whether the station i has priority to transmit a data packet in the group c to which it belongs.

If it is determined that there is a priority, in step 432, the station i determines whether it has a data packet to be transmitted. If it is determined that there is no priority, in step 433, the stations except the station i among the stations belonging to the group c determine whether there is no data packet to be transmitted.

If it is determined in step 432 that there is a data packet to be transmitted, in step 434, the station i transmits a control packet to the station j. Although not shown in Figure 9, it is preferable to further include the step of checking the transmission traffic priority prior to this step.

If it is determined in step 432 that there is no data packet to be transmitted, in step 435, the station i determines whether there is a data packet to be transmitted to the station j. If it is determined in step 435 that there is a data packet to be transmitted, step 451 is performed.

If it is determined in step 432 that there is a data packet to be transmitted, the station i waits for an ACK cycle.

Steps 431 to 436 described above correspond to control cycles in terms of data frames.

In step 451, the station j determines whether the control packet has been received from the station i.

If only the control packet is received from the station i in step 451, the station j transmits an ACK packet to the station i in step 452. Accordingly, data packet transmission from the group c to the station j is performed (step 461).

If a control packet is received from a station other than the station i in step 451, the station j determines whether the station i has a priority to transmit the data packet in step 453.

If it is determined in step 453 that there is priority, step 452 is performed. If it is determined that there is no priority, in step 454, the station j transmits an ACK packet to the station i1 belonging to the group cc1. Accordingly, data packet transmission from the group cc1 to the station j is performed (step 462).

In step 436, the station j checks whether the station j is a priority station among idle groups.

In step 456, the station j selects the station ii2 having the minimum h. Here h denotes the number of time slots shifted forward through early transmission.

In step 457, the station j transmits an ACK packet to the station ii2 of the group cc2.

In operation 463, the station ii2 determines whether there is currently a data packet to be transmitted to the station j. If there is a data packet to be transmitted, the data packet transmission from the group cc2 to the station j is performed (step 464). If there is no data packet to transmit, no data packet is transmitted (465).

10 is a reference diagram illustrating a map of a transmitting station group and a receiving station according to time slots.

Groups 1 to C on the vertical axis represent station groups using different wavelengths, and time slots 1 to N on the horizontal axis represent N time slots belonging to one cycle time slots. In addition, the values shown in FIG. 10 indicate a destination station, that is, a receiving station, allocated according to a transmitting station group and a time slot.

11 is a frame structure diagram for media access using heuristic traffic load balanced channel sharing according to the present invention.

FIG. 11 is a structural diagram 510 illustrating the flow of data in the time domain for modified accelerative pre-allocation (MAP) WDMA, particularly extended MAP-WDMA. Extended MAP-WDMA combined with a fixed type of CSM has a control cycle, an ACK cycle, and a data cycle. However, according to the present invention, the CSM of the traffic load-balanced type further includes two cycles of dotted lines. That is, it further includes a traffic cycle and a CSM cycle.

 In addition, extended MAP-WDMA is performed under a slotted time named type slot 511, and all stations perform the cycles during one time slot. One time slot for a CSM of pre-fixed type has a control cycle, an ACK cycle, a data cycle, a tuning delay, and a propagation delay. However, as shown in FIG. 11, one time slot in a traffic load balanced type CSM is a traffic cycle 521, a CSM cycle 522 that depends on a given load balanced type CSM, a control cycle 523, an ACK. Cycle 524, and data cycle 525. Although not shown in FIG. 11, there is a tuning delay between the control cycle and the ACK cycle, and a tuning delay and propagation delay between the ACK cycle and the data cycle. In FIG. 11, 526 is a traffic cycle for station 1, 527 is a control cycle for station 1, and 528 is an ACK cycle for station N. In FIG.

12A and 12B illustrate a Jane Index according to a medium access method using an LB-II CSM among a heuristic traffic load balanced (HTLB) channel sharing method (CSM) according to the present invention. ) Is a reference diagram.

The Jane Index is used as a criterion to determine whether the traffic loads of all station groups are well balanced. In this experimental example, the zein index of HTLB CSM was calculated and the zein index was compared with other CSM. The zein index is defined by Equation 2 below.

Equation 2

이다. 본 실험 예에서 대조예로 사용되는 CSM은 3개의 사전 할당된 CSM을 포함한다. 사전 할당된 CSM은 N-CSM(neighboring-CSM), I-CSM(interleaved-CSM) 및 EON-CSM(even-odd neighboring CSM)이다. N-CSM의 스테이션 그룹(c)는 이론적으로

로 정의되어 사용되었고, I-CSM의 경우는

로 정의되어 사용되었다. EON-CSM의 경우 기수의 경우

이고, 우수의 경우

로 정의되어 사용되었다.From here,

to be. The CSM used as a control in this experimental example includes three pre-assigned CSMs. Pre-assigned CSMs are neighboring-CSM (N-CSM), interleaved-CSM (I-CSM) and even-odd neighboring CSM (EON-CSM). Station group (c) of N-CSM is theoretically

Defined and used for I-CSM

Defined and used. For EON-CSM For Radix

, For rain

Defined and used.

In this experiment, various TLPs were considered to investigate the effect of traffic load balancing. First, three TLPs (U ₁ , U ₂ , U _3I ) were applied to investigate the effects of TLP imbalance on traffic load balancing. The traffic load flowing into all three TLPs is typically U _1. = U ₂ = U _3I = {a, (1-a), a, (1-a), a, (1-a) ..., a, (1-a)}. Here, a is 0≤a≤1. Three TLPs have the same TLP configuration and are modeled to have an average value of 0.5. But δU ₁ = 0.05, δU ₂ = 0.1, δU ₁ = 0.2 so that the standard deviation value is different from each other. The value of a is determined by the variance and the N value. In the following, the present experimental example, and the product was used for the traffic load, defined as in Equation 3 _3I U, U _{3II, 3III} U to confirm the results of using the different types of TLP.

Equation 3

_{Here, U 3I, U 3II, U} 3III The TLPs have the same average value of 0.5 and the same variance of 0.2, but with different combinations of traffic loads. In addition, this experimental example examines the effect of TLP on traffic load balancing when p _ij is 1 / N-1 (i ≠ j) or 0 (i = j).

12A and 12B are graphs showing zein indices according to different CSMs and zein indices of HTLB CSMs. The Janet Index of HTLB CSM was higher than the pre-allocated CSM, even higher than the MULTIFIT CSM. From this, it can be seen that the HTLB CSM can guarantee even traffic load balancing even when the TLP imbalance is severe. In FIG. 12A, the N-CSM is observed to have a higher zein index than the pre-assigned CSM. However, the N-CSM can be seen that the zein index of the N-CSM in FIG. 12B fluctuates depending on the input traffic conditions. In addition, although the TLP is changed in FIG. 12B, the HTLB CSM still appears to have a better Jane index than the pre-allocated CSM, and it can also be seen that the traffic load balancing can be performed equally with the MULTIFIT CSM. 12A and 12B, it was confirmed that the HTLM CSM performs steady traffic load balancing regardless of the input traffic condition.

Meanwhile, the channel sharing method and the medium access method of the present invention can be embodied in computer readable codes on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored.

Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disks, optical data storage devices, and the like, which may be implemented in the form of a carrier wave (for example, transmission over the Internet). Include. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion. And functional programs, codes and code segments for implementing the present invention can be easily inferred by programmers in the art to which the present invention belongs.

So far I looked at the center of the preferred embodiment for the present invention. Those skilled in the art will understand that the present invention can be implemented in a modified form without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments should be considered in descriptive sense only and not for purposes of limitation. The scope of the present invention is shown not in the above description but in the claims, and all differences within the scope should be construed as being included in the present invention.

The heuristic traffic load balanced channel sharing method of the present invention and a medium access method using the same are suitable for use in a wavelength division multiple access network using a passive star coupler. In particular, the present invention is low in time complexity, and can be consistently loaded in various traffic environments, and thus can be applied to a variety of multimedia applications such as VoIP, IPTV, VoD, and video conferencing. It is suitable to be.

1 is a block diagram illustrating a medium access network system having a passive star coupled topology according to an embodiment of the present invention.

2 is a block diagram illustrating a station including a channel sharing apparatus according to an embodiment of the present invention.

3 is a flowchart illustrating a method of sharing a heuristic traffic load balanced (HTLB) channel according to an embodiment of the present invention.

4 is a reference diagram illustrating an example of a traffic load information matrix M generated in this embodiment.

FIG. 5 shows an example of an algorithm (LB-I CSM) for determining a station group in step 230 of FIG. 4.

FIG. 6 is a reference diagram for explaining a concept of the algorithm (LB-I CSM) of FIG. 5.

FIG. 7 illustrates another example of an algorithm for determining a station group (LB-II CSM) in step 230 of FIG. 4.

FIG. 8 is a reference diagram for explaining a concept of the algorithm (LB-II CSM) of FIG. 7.

9 is a flowchart illustrating a medium access method of a wavelength division multiple access network according to an embodiment of the present invention.

10 is a reference diagram illustrating a map of a transmitting station group and a receiving station according to time slots.

11 is a frame structure diagram for media access using heuristic traffic load balanced channel sharing according to the present invention.

12A and 12B are reference diagrams illustrating a Jane index according to a media access method using a heuristic traffic load balanced (HTLB) channel sharing method according to the present invention.

Claims (26)

A channel sharing method in a wavelength division multiple access network having a plurality of stations, a) receiving traffic load information from each of the stations; b) generating a traffic load information matrix according to each transmitting station and receiving station using the received traffic load information; And c) determining a station group by using at least some column vectors or row vectors selected from among column vectors and row vectors constituting the traffic load information matrix. The method of claim 1, C) determining a group of stations using row vectors or column vectors according to predetermined receiving stations in the traffic load information matrix. The method of claim 1, Step c) is to determine a group of stations using row vectors or column vectors according to predetermined receiving stations in the traffic load information matrix, Determining the station group is to determine the station group using the traffic load information according to each row vector or column vector in a predetermined order, except that the station that has already been determined is excluded from the later station group determination. Channel sharing method characterized by. The method of claim 1, And in step a), the traffic load information is information representing a degree of traffic load between a transmitting station transmitting data and a receiving station receiving the transmitted data, and is expressed by the following equation. Equation α _ij = σ _i × p _ij Where σ _i is the normalized traffic load flowing into station i, p _ij is the probability that station i sends a packet to station j, and α _ij is normalized traffic when the transmitting station is i and the receiving station is j Indicates the load value. The method of claim 1, The row of the traffic load information matrix is for transmitting stations, the column is for receiving stations, and step c) c1) searching for a maximum traffic load and a minimum traffic load in a column vector according to a first receiving station of the receiving stations; c2) determining transmitting stations according to the maximum traffic load and the minimum traffic load found in step c1) as a first group; c3) searching for the maximum traffic load and the minimum traffic load in a column vector according to a second receiving station of the received stations, with the exception of the traffic load according to the transmitting stations determined as the first group in step c2). Making; And c4) determining transmission stations according to the maximum traffic load and the minimum traffic load found in step c3) as a second group. The method of claim 1, The row of the traffic load information matrix is for transmitting stations, the column is for receiving stations, In step c), in accordance with the reverse order of the sum of the traffic loads of the station group, each of the transmitting stations having the maximum traffic load found in each of the predetermined some of the column vectors is determined as different station groups. And determining a group of stations in a state of excluding the already grouped stations. The method of claim 6, wherein step c) c1) determining, according to a predetermined order of grouping, transmitting stations having the maximum traffic loads found in each of the column vectors of some of the column vectors as different station groups, except for the already grouped stations. Determining a station group; c2) calculating the sum of the traffic loads of each of the station groups, and determining the order of grouping in reverse order of the sum of the traffic loads; c3) in accordance with the ordering of the grouping determined in step c2), determine a transmitting station having the maximum traffic load found in each of the column vectors as a different station group, except for the station group that is already grouped. And determining the channel. The method of claim 6, c4) calculating a sum of traffic loads of each station group by further considering stations grouped in step c3), and determining the order of grouping in the reverse order of the sum of traffic loads; And c5) In accordance with the ordering of the grouping determined in step c4), the transmission station having the maximum traffic load found in each of the column vectors is determined as a different station group, except for the station group that is already grouped. And determining the channel. The method according to claim 5 or 6, The network is a network having a passive star coupling topology, and the stations perform data communication according to a pre-allocation based MAC protocol. A computer-readable recording medium having recorded thereon a program for performing on a computer a method of sharing a channel in a wavelength division multiple access network having a plurality of stations according to any one of claims 1 to 8. In the channel sharing apparatus of the wavelength division multiple access method having a plurality of stations, A traffic load information generator configured to generate traffic load information according to each of the stations; And Receives traffic load information from each of the stations, generates a traffic load information matrix according to each transmitting station and receiving station using the received traffic load information, and comprises a column vector constituting the generated traffic load information matrix. And a group determiner configured to determine a station group by using at least some column vectors or row vectors selected from among the plurality of rows and row vectors. The method of claim 11, And the group determiner determines a station group by using row vectors or column vectors according to a predetermined receiving station in the traffic load information matrix. The method of claim 11, The group determiner determines a station group by using row vectors or column vectors according to predetermined reception stations in the traffic load information matrix. The determining of the station group may include determining a station group by using traffic load information according to each row vector or column vector in a predetermined order, but excluding a station that has already been determined by the station group from a later station group determination. Channel sharing device. The method of claim 12, The row of the traffic load information matrix is for transmitting stations, the column is for receiving stations, The group determining unit searches for a maximum traffic load and a minimum traffic load in a column vector according to a first receiving station among the receiving stations, and transmits the transmitting stations according to the found maximum traffic load and minimum traffic load. Determine the group, and search for the maximum traffic load and the minimum traffic load in the column vector according to the second receiving station among the received stations, excluding the traffic load according to the transmitting stations determined as the first group, And determine transmission stations according to the found maximum traffic load and minimum traffic load as a second group. The method of claim 12, The row of the traffic load information matrix is for transmitting stations, the column is for receiving stations, The group deciding unit determines, according to the reverse order of the sum of the traffic loads of the station group, each transmitting station having the maximum traffic load found in each of the column vectors of some of the column vectors as different station groups. And determining the station group with the excluded station excluded. A medium access method in a wavelength division multiple access network having a plurality of stations, a) each of the stations transmitting traffic load information to other stations simultaneously; b) generating a traffic load information matrix according to each transmitting station and receiving station using the traffic load information; c) determining a station group using at least some column vectors or row vectors selected from among column vectors and row vectors constituting the traffic load information matrix; And d) transmitting the data to be transmitted by the transmitting station included in the determined station group to the receiving station included in the determined station group. The method of claim 16, C) determining a station group using row vectors or column vectors according to a predetermined receiving station in the traffic load information matrix. The method of claim 16, The row of the traffic load information matrix is for transmitting stations, the column is for receiving stations, and step c) c1) searching for a maximum traffic load and a minimum traffic load in a column vector according to a first receiving station of the receiving stations; c2) determining transmitting stations according to the maximum traffic load and the minimum traffic load found in step c1) as a first group; c3) excluding the maximum traffic load and the minimum traffic load in the column vector according to the second receiving station among the received stations, excluding the traffic load according to the transmitting stations determined as the first group in step c2). Searching; And c4) determining transmission stations according to the maximum traffic load and the minimum traffic load found in step c3) as a second group. The method of claim 16, The row of the traffic load information matrix is for transmitting stations, the column is for receiving stations, The step c) determines, according to the reverse order of the sum of the traffic loads of the station group, the transmitting stations having the maximum traffic loads found in each of the column vectors of some of the column vectors as different station groups, respectively. And determining the group of stations in the state of excluding the grouped stations. 20. The method of claim 19, wherein step c) c1) determining, according to a predetermined order of grouping, transmitting stations having the maximum traffic loads found in each of the column vectors of some of the column vectors as different station groups, except for the already grouped stations. Determining a station group; c2) calculating the sum of the traffic loads of each of the station groups, and determining the order of grouping in reverse order of the sum of the traffic loads; And c3) in accordance with the ordering of the grouping determined in step c2), determine a transmitting station having the maximum traffic load found in each of the column vectors as a different station group, except for the station group that is already grouped. Determining the media access method. The method of claim 16, wherein step d) d1) determining whether the transmitting station has priority in the station group to which the transmitting station belongs; d2) the transmitting station transmitting a control packet to a receiving station in consideration of the priority in step c1); And d3) the transmitting station receiving an ACK packet from the receiving station and transmitting the data to be transmitted to the receiving station. The method of claim 21, wherein step d2) If it is determined in step d1) that the transmission station has a priority, the transmitting station transmits a control packet to the receiving station. And transmitting the control packet to the receiving station in consideration of whether there is a desired packet. The method of claim 16, And said network is a network having a passive star combined topology and is a pre-location based medium access method. 24. A computer-readable recording medium having recorded thereon a program for performing a medium access method on a computer in a wavelength division multiple access network having a plurality of stations according to any one of claims 16 to 23. A medium access network system in a wavelength division multiple access network having a plurality of stations, A transmission medium for signal transmission between the stations; A coupler coupled to the stations via the transmission medium, for coupling or branching a signal entering and exiting through the transmission medium; A traffic load information generator for generating traffic load information according to each station; And A column for receiving traffic load information from each of the stations, generating a traffic load information matrix according to each transmitting station and a receiving station using the received traffic load information, and forming the generated traffic load information matrix And a group determination unit for determining a station group by using at least some column vectors or row vectors selected from among vectors and row vectors. The method of claim 25, And the coupler is a passive star coupled device and the media access system is a pre-location based media access system.

Images