KR20050024938A - A method for determining a priority of packet transmission and a computer-readable medium having the program therof - Google Patents

Abstract

PURPOSE: A method for determining a priority level for packet transmission and a computer-readable recording medium storing the program are provided to give relative priority levels to packets by using delay time characteristics among different traffic packets and differentiate inter-packet priority levels among the same type of traffic. CONSTITUTION: A scheduler starts scheduling at a specific point(S100). The scheduler measures an SNR(Signal to Noise Ratio) of a user terminal(S210). An amount of data stored in a queue is measured(S310). A extent of a provided service during a certain slot is calculated in consideration of an average transmission data amount(S320). A packet interval is measured based on traffic characteristics of data stored in the queue(S410). A delay available time of a packet is calculated by using the packet interval(S420). A priority level of each packet is determined(S500). Data transmission is performed during a certain time with respect to a queue for a user with the highest priority level(S600).

Description

A method for determining a priority of packet transmission and a computer-readable medium having the program therof}

The present invention relates to packet type voice and data communication. More specifically, the present invention relates to a scheduling method for transmitting voice and data packets in consideration of channel status and delay time.

The channel environment of the mobile terminal may be relatively divided into a good channel environment and a poor channel environment according to the area where the terminal is located. When the channel environment of the terminal is in a good state by using such channel environment information, if packet data can be transmitted, it can be transmitted at a high data rate and a plurality of users can share a channel with each other.

 However, since users use different types of packet data services, a packet scheduler considering service characteristics as well as a wireless channel environment is required in order to fairly service different services. Data transmission services can be broadly classified into real-time data services and non-real-time data services. For example, voice data for a voice call belongs to a real time data service, and video content and file services such as http or ftp are non-real time data services.

In case of real-time service, delay time between packets is short for continuity, and service is available only when packet delay time is satisfied.

On the other hand, non-real-time data services such as video data, http or ftp have a relatively low demand for real-time services. Such a non-real-time data service is composed of a packet chunk consisting of several packets, but the delay time between the packet chunks is relatively long in consideration of the time of checking by the user requesting the service.

1 illustrates a state in which packets transmitted from a network are stored in a queue.

As shown in Figure 1, packets are sent, reclassified and stored in the scheduler, depending on the type of traffic, such as http or ftp. As shown, the scheduler stores data for each of the user identifiers UID 1 to 4 and then stores the data for each user.

In this case, according to the prior art, a packet is differentiated and transmitted from the non-real time packet service data and the real time packet service data using a probabilistic value that can be delayed. Equation 1 is the probability value ?? _i is defined.

Here, W _i represents a processable time while satisfying a minimum data rate, the data stored in the queue, T _i refers to the constant time it takes to satisfy the data service.

However, since the probability value should be obtained based on the experimental values that have undergone many experiments, many samples and labors are required. In addition, depending on the state of the queue, the duration of the packet in the queue, and the packet generation method, an additional scheduling algorithm is needed to relatively compare non-real-time services and real-time services such as voice.

On the other hand, in the EV-DO system for transmitting packet data, the forward data transmission method uses a method of transmitting all the power that can be transmitted from the base station to a terminal at regular time intervals.

Here, the conventionally used scheduler PFS (Proportionally Fair Scheduler) for the system in the prior art using the radio channel information measured by using a pilot burst (burst) is inserted into each slot in each terminal in the best wireless environment Set the transmission order to enable packet data transmission.

The amount of data that can be transmitted to each terminal determined by the scheduler, a modulation scheme, and a slot length (HDR transmission unit) vary according to channel conditions of each terminal. The slot is 1.67ms long and one packet consists of up to 16 slots. The UE measures signal to interference plus noise ratio (SINR) and indicates a data rate that can be transmitted in the measured SINR as a 4-bit data rate control (DRC) value. The DRC value considering the channel condition is transmitted to the base station through the reverse DRC channel. This information is reported continuously every 1.67 ms. The scheduler in the base station transmits the packet data amount to the terminal at the data rate requested by the terminal. The PFS gives a packet data transmission opportunity to a terminal having the highest DRC / R value. Here, R value means the average amount of data transmitted to each terminal for a certain time. The average data rate continues to change with feedback for 1000 slots. Changing average data rates provide relative fairness for all users.

However, the prior art considers only non-real time data in which data of all users who want to transmit data can be stored in an infinite memory size. As such, there is a problem in that scheduling considering only a wireless channel environment cannot satisfy a requirement of a real-time data service such as a voice call.

Another prior art Modified Largest Weighted Delay First (M-LWDF) scheduling algorithm prioritizes probabilistic values that can cause packets to be delayed and exceed delay thresholds, and at least data processing requirements, as well as user data requirements according to the wireless channel environment. As an important factor in determining the ranking. For this purpose, the queue status is prioritized by dividing the data amount of each user stored in the queue by the arrival data rate.

However, the M-LWDF requires a probabilistic value regarding packet delay exceedance and does not provide a reasonable criterion for at least data processing requirements for each user.

The present invention provides a packet prioritization method for efficient and fair radio resource management in consideration of channel conditions while satisfying service requirements for traffic characteristics such as real-time services.

In addition, the present invention provides a method for assigning a relative priority between packets using delay time characteristics between different traffic packets, and differentiating the priority between packets even among the same traffic types.

In addition, the present invention proposes a method for satisfying the quality of service (QoS) without using a probabilistic value regarding packet delay excess.

In order to achieve the above object of the present invention, a packet transmission method according to a feature of the present invention,

Calculating a transmission state of the channel; A second step of calculating the amount of data in the queue over time; Calculating a packet delay possible time according to the traffic characteristic of the packet; A fourth step of calculating a priority decision value proportional to the calculated goodness of the channel state and the amount of data in the queue and inversely proportional to the packet delay time, and transmitting a packet to a user terminal having a higher priority decision value. do.

In addition, the first step may include measuring a signal-to-noise ratio of the wireless channel.

The second step may further include measuring the amount of data existing in the existing queue and the amount of newly arrived data; The method may include measuring an average value of the transmitted data.

Further, the third step may include calculating a packet interval according to the traffic characteristic, when the traffic characteristic is a voice packet, calculating a constant packet interval, and when the traffic characteristic is a block of packets, Repartitioning the packets at the minimum data rate may further comprise calculating a modified packet interval over time.

DETAILED DESCRIPTION Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention. As those skilled in the art would realize, the described embodiments may be modified in various different ways, all without departing from the spirit or scope of the present invention.

In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention. Like parts are designated by like reference numerals throughout the specification.

Now according to an embodiment of the present invention will be described in detail with reference to the drawings.

2 shows the types of data stored in the queues.

As shown in Fig. 2, the amount of data stored in the queues 11 and 12 of the scheduler is different. Here, Q _i (t) means the amount of data already stored, and N _i means the amount of newly arrived data. r (t) means the amount of data transmitted to the user terminals 31 and 32.

Most wireless schedulers use the radio channel state, that is, the data rate required by the terminal, as an important factor in determining the priority between packets in order to transmit a lot of data. In this case, however, it is assumed that the size of the queue is infinite as described above, but the size of the queue is actually finite.

Data stored in a finite sized queue arrives at a queue with a different time difference depending on the data arrival process. The amount of data for each user in the queue is different for each user. Therefore, even if the channel status is assigned the priority of data transmission, there may be no data to be transmitted in the queue.

Therefore, even though a channel state is relatively bad because a queue size is relatively small, and a small amount of data can be transmitted than a user having a good channel state, there are cases where a user having a large amount of data stored in a queue should be selected first.

Therefore, in order to maintain a finite queue state, the present invention actually transmits an average value of data, , Are used in the algorithm and different queue arrival parallaxes are used to determine the priority between packets.

In Equation 2 below, QA _i (t) is a formula for all data stored in a queue for user i.

Since the present invention considers all the data stored in the queue and the average amount of transmitted data, it prevents the queue from overflowing with a finite size and measures the extent to which the data is in the queue, thereby measuring the degree of service for a certain slot. This can be seen through.

As described above, the data stored in the queues have different traffic characteristics according to characteristics of each data service, and packet delay time required for providing each service.

3 illustrates a packet generation form of a voice service.

As shown in FIG. 3, all packets of a voice service are divided into a predetermined size and are sequentially generated at regular intervals. Here, the constant size and spacing of the packets means the spacing and the size at which the service can be performed without disconnection of voice. The Voice Packet Interval (Vpi) between voice packets is constant, but the time that packets can be stored in the queue without voice interruption is very short and finite.

However, the time that a voice packet to be stayed in the queue may change over time without always maintaining a constant value according to arrival time, scheduler period, or scheduled time point. Therefore, in the present invention, the time that a packet stays in the queue is newly defined as a possible packet delay time d _ij (t) as shown in Equation (3).

Here, by subtracting the arrival time (ATOPj (t); Arrival Time of Packet j) of the j th packet from the current scheduling time (CST _i (t)), it is possible to calculate the time that the packet j to be transmitted currently stays in the queue. The packet delay possible time d _ij (t) indicates a packet delay possible time for j packet of user i who has requested a voice service.

If, here, If Vpi is greater than Vpi, voice disconnection occurs.

4 illustrates a generation form of a packet chunk such as HTTP.

Unlike voice services, services such as HTTP are made up of chunks of packets, and the arrival intervals between packets are not constant. As shown in Fig. 4, the arrival times of the packets are not constant, but the arrival intervals of the packet chunks are considerably longer compared to services such as voice because they include time to read data and the like.

Therefore, considering only the delay time of the packet chunks, the packet chunk delay time can be defined as the average time of the packet chunk delay time that the user who requested the http service can wait for the packet chunk without giving up the http service. .

Such a definition of packet delay time is possible if all the packet chunk data stored in the queue can be transmitted at the same time. However, it is not possible because several users share the radio channel situation and radio resources that change from time to time.

Therefore, after assuming that a minimum radio resource is allocated to each channel, the packet chunks must be transmitted in multiple times at least by the amount of data that can be transmitted so that the packet chunks requested by the user can be delivered within a time that the user does not give up.

Therefore, embodiments of the present invention repartition packet chunks in consideration of minimum data rate and available time information.

5 illustrates packet chunks that have been subdivided to have a constant transmission interval according to an embodiment of the present invention.

As shown in Fig. 5, the packet chunk in the present invention is subdivided back into packets of minimum data rate with constant transmission interval. In order to subdivide the packet chunks, parameters such as the minimum data rate, available time information, and the expected number of transmissions must be defined and required.

The packet chunk of user i is broken up into 1,2,, j ,, j + n packets. As shown in equation (4), the minimum data rate (Minimum Data Rate) is the number of data bits that can be transmitted at the minimum bandwidth (Bw) allocable to the user, You can get it by multiplying by.

The amount of data stored in the queue QA _i (t) is the minimum data rate By dividing by, it is possible to obtain the expected number of transfers required when data is transmitted at the minimum data rate.

On the other hand, when sending a packet chunk, the available time information, , Is a constant time that the user can endure without giving up the service is called Et, and packet call request time Pct can be obtained as shown in Equation 5.

As the scheduling time continues to increase, the available time continues to decrease, and when the available time becomes less than zero, the scheduler may not meet the service requirements.

Equation 6 is a constant interval between packets that have been subdivided at the minimum data rate. It is defined.

here , Means the time that the resegmented packet should be processed without staying in the queue longer than this value.

However, as can be seen from Equation 6, a constant interval between the subdivided packets Does not always have a constant value, such as the interval between voice packets, but changes with time depending on the data rate stored in the queue or the radio channel information that determines the minimum data rate. This feature is well illustrated in FIG. 5.

Thus, the packet delay possible time for packet j of packet chunk user i Can be obtained from equation (7).

here, Denotes a j th packet request time.

According to the present invention, the packet delay possible time having characteristics of the packet clump form is made in a form similar to that of Equation 3 which defines the voice packet delay time through Equations 4 to 7.

The scheduling algorithm proposed in the present invention determines packet priorities in consideration of channel status, queue status, and packet delay possible time that varies with time. Each user calculates and transmits a noise ratio SNR value for a signal transmitted from a base station to obtain a transmittable channel state. Based on the SNR value, the base station transmits a data rate proportional to the channel state of each user, Determine the value. The determination of the priority may be determined in the order in which the value of Equation 8 is maximized.

As described above, from the value of Equation 8, the channel state, the data state stored in the queue, and the packet delay possible time are considered, and the packets are transmitted in the order of maximizing them.

6 is a flowchart illustrating a flow of priority determination between packets according to an embodiment of the present invention.

In step S100, the scheduler starts scheduling at a specific time. At this time, the traffic characteristics, channel status, data amount, and delayed transmission time of the packets stored in the queue for each user are all different.

In step 210, the scheduler measures the SNR of the user terminal. The measurement of such SNR can be measured by a known technique using a pilot channel or the like. If the channel state is determined from the SNR, the transmittable data rate is determined in step S220. To calculate.

Simultaneously with the step S210, in step S310, the amount of data stored in the queue is measured. The amount of data stored in the queue is measured in consideration of the amount of previously stored data and the amount of newly arrived data. Thereafter, in step S310, the degree of service for a predetermined slot is calculated by further considering the average amount of transmitted data. When the amount of data stored in the queue and the degree of service during a predetermined slot are calculated, the data state of the current queue can be calculated.

Simultaneously with the steps S210 and S310, in step S410, the packet interval is measured based on the traffic characteristics of the data stored in the queue. If the traffic is a voice packet, the packet interval is constant, but if it is in the form of a packet chunk, the packet interval modified by repartitioning To calculate. Thereafter, in step S420, the delayable time of a packet is calculated using the packet interval.

When all of the steps 220, 320, and 420 are completed, in step S500, priority between packets is determined. In the prioritization method, as shown in Equation 8, a numerical value for priority that is proportional to the amount of data in the queue according to the transmittable data rate and time and inversely proportional to the delayable time may be calculated.

Step S600 performs data transmission for a predetermined time for the queue for the user having the highest priority value. When the predetermined time elapses, the process returns to step S100 to perform scheduling on the changed queue state.

Hereinafter, a specific example of the embodiment according to the present invention will be described according to each case.

7A and 7B illustrate a method of transmitting voice data to a plurality of users according to an embodiment of the present invention.

In the embodiment of the present invention, it is assumed that User 1, User 2 and User 3 are present, User 1 and User 2 are located at the same distance from the base station, and User 3 is located at a much closer distance from the base station than User 1 and 2. .

  Thus, although the channel states of users 1 and 2 are similar, the channel states of user 3 are very good compared to the other two users.

It can be seen from FIG. 7A that the average data rate of each user and the amount of data stored in the queue are similar at the scheduling time point t1, and the delayable time of user 2 is shorter than that of users 1 and 3.

However, if the channel transmission state of user 3 is very good, at time t1, user 3 is preferentially selected and data transmission is performed.

Fig. 7B shows the queue state and the possible delay time at time t2 after data transmission for user 3 is performed. At time t2, the scheduler must select one of the two users with similar conditions.To guarantee the voice service, the scheduler compares the value of the delay time dij (t) and selects user 2 with a relatively short delay time. It will transmit data.

Accordingly, the voice data service according to the embodiment of the present invention performs data transmission efficiently and fairly without interruption of service.

8 illustrates a method of transmitting a packet chunk and a voice packet to a plurality of users according to an embodiment of the present invention.

As shown in Fig. 8, user 1 has to transmit a voice packet, and user 2 has to transmit a packet chunk which is non-real time data.

In this case, if two users have similar radio channel conditions, priority should be given to voice users having a short packet duration first. This differentiation is shown in FIG. end It is possible to follow the embodiment of the present invention as it is much larger.

9A and 9B illustrate a method of transmitting a packet chunk to a plurality of users, according to an embodiment of the present invention.

FIG. 9A illustrates a case where both user 1 and user 2 use a service generated by a packet chunk. As shown in FIG. 9A, at the time t1, the amount of data stored in the queue for both users is similar, and the delayable time values of the two users are similar to each other. However, if user 2 has a relatively good channel condition, user 2 transmits data at time t1.

As described above, if user 2 transmits data, as shown in FIG. 9B, at the time t2, the amount of data stored in the queues of the two users and the delayable time are different from each other. Therefore, at time t2, the data is transmitted to user 1 having a large amount of data in the queue and a short delay.

Therefore, according to the embodiment of the present invention, even if user 2 has a good channel state, the scheduler selects user 1 at time t2 and transmits a packet due to a value that changes with time.

Therefore, according to the configuration of the present invention, it is possible to fairly and efficiently transmit data services having various traffic characteristics.

The above-mentioned method of prioritizing packets is recorded in a computer-readable recording medium, and can efficiently and efficiently perform data transmission scheduling on a computer.

Although the preferred embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements of those skilled in the art using the basic concepts of the present invention defined in the following claims are also provided. It belongs to the scope of rights.

As described above, according to the configuration of the present invention, a significant effect of performing efficient and fair transmission in consideration of channel status, queue status, and delayed transmission time for packets having different traffic characteristics without using probabilistic values. You can expect.

1 illustrates a state in which packets transmitted from a network are stored in a queue.

2 shows the types of data stored in the queues.

3 illustrates a packet generation form of a voice service.

4 illustrates a generation form of a packet chunk.

5 illustrates packet chunks that have been subdivided to have a constant transmission interval according to an embodiment of the present invention.

6 is a flowchart illustrating a flow of priority determination between packets according to an embodiment of the present invention.

7A and 7B illustrate a method of transmitting voice data to a plurality of users according to an embodiment of the present invention.

8 illustrates a method of transmitting a packet chunk and a voice packet to a plurality of users according to an embodiment of the present invention.

9A and 9B illustrate a method of transmitting a packet chunk to a plurality of users, according to an embodiment of the present invention.

Claims (6)

In a method for transmitting a data packet to a plurality of user terminals: Calculating a transmission state of the channel; A second step of calculating the amount of data in the queue over time; Calculating a packet delay possible time according to the traffic characteristic of the packet; A fourth step of calculating a priority determination value proportional to the calculated goodness of the channel state and the amount of data in the queue and inversely proportional to the packet delay time, and transmitting a packet to a user terminal having a high priority determination value.  Packet transmission method comprising. The method of claim 1, The first step is to calculate the transmission state of the channel by measuring the signal-to-noise ratio of the wireless channel. The method of claim 1, The second step is A packet transmission method for calculating the state of data in a queue over time by measuring an amount of data existing in an existing queue, a newly arrived data amount, and an average value of transmitted data. The method of claim 1, The third step is When the traffic characteristic is a voice packet, by measuring a constant packet interval, and when the traffic characteristic is a block of packets, by re-segmenting the packet at the minimum data rate by measuring the modified packet interval over time, the possible delay time of the packet Packet transmission method for calculating the. The method according to any one of claims 1 to 4, If the data transmission to the user terminal selected by the fourth step for a predetermined time, repeating the steps 1, 2, 3 again. A computer-readable recording medium having recorded thereon a program having a function of transmitting a data packet to a plurality of user terminals, the method comprising: A first function of measuring a signal-to-noise ratio of a wireless channel to calculate a transmission state of the channel; A second function of calculating the amount of data in the queue over time by measuring the amount of data existing in the existing queue, the amount of newly arrived data, and the average value of the transmitted data; When the traffic characteristic is a voice packet, by measuring a constant packet interval, and when the traffic characteristic is a block of packets, by subdividing the packet at the minimum data rate by measuring the modified packet interval over time, the packet according to the traffic characteristic A third function of calculating a delayable time; A fourth function of calculating a priority determination value proportional to the calculated goodness of the channel state and the amount of data in the queue, and inversely proportional to the packet delay time, and transmitting the packet to a user terminal having a high priority determination value. The computer-readable recording medium having the program recorded thereon.