Title of Invention

ADAPTIVE ESTIMATION METHOD OF MULTIMEDIA DATA TRANSMISSION RATE

Abstract

There is disclosed an adaptive estimation method of a multimedia data transmission rate. A transmission unit of a multimedia data transmission system transmits multimedia data to a reception unit through a RTP packet (S1 1), the reception unit transmits state information of a network to the transmission unit through a RTCP receiver report packet (S13), and the transmission unit detects a packet loss ratio from the RTCP receiver report packet (S15), and estimates an available transmission rate by a calculation method that varies according to a range within which the detected packet loss ratio falls into (S17). Accordingly, the estimation method can speedily resolve network congestion when the packet loss is large.

Full Text

1. Field of the Invention The present invention relates to a multimedia data transmission system based upon a real-time transport protocol (RTP)/real-time transport control protocol (RTCP) in a wired IP (internet protocol), and particularly, to a method for adaptively estimating a transmission rate of multimedia data by using a function of monitoring an RTCP network state. 2. Description of the Related Art In general, conditions such as a sufficient bandwidth, a small delay and a small packet loss and the like should be ensured to transmit multimedia data through a wired IP network such as the Internet. However, a network layer in a current wired IP network cannot provide a function suitable to meet quality of service (QoS) required for video transmission. Therefore, the QoS should be secured by a higher layer of the network layer. To this end, a real-time transport protocol (RTP) and a real-time transport control protocol (RTCP) operated on a transport layer have been proposed. The RTP is an Internet protocol for real-time multimedia data such as the audio and video generated in real-time. Although the RTP itself does not ensure real-time transmission of data, by using the RTP, application programs for 2l5transmission/reception in real-time multimedia data transmission system can support streaming data. The RTP is commonly executed on a UDP (user datagram protocol). The RTCP is a protocol used to maintain the QoS (Quality of Service) of the RTP. The RTP is related only to the data transmission whereas the RTCP relates to monitoring of the data transmission and the transmitting of session-related information. RTP nodes send to one another, RTCP packets in order to analyze a network state and to periodically report whether the network is congested. By the use of the RTP and the RTCP, characteristics according to time limit in transmitting multimedia data may be considered and it is possible to cope with a loss generated in the network. In a general multimedia data transmission system, a multimedia application program (application layer) detects the network state through the RTCP and controls an encoding rate of real-time multimedia data to be transmitted. The controlling of the encoding rate of the real-time multimedia data is made through transmission rate control, and a general method of estimating an ineffective transmission rate by using network state information by the RTCP uses the following equation 1. 1.22x5 (equation 1) R(t) = RTT(t)x p(t) The R(t) indicates an effective transmission rate, the p(t) indicates a packet loss ratio and is obtained by the RTCP transmitted from a receiving side. The RTT(t) indicates a round-trip delay time, and the s indicates a size of a packet. When the Round-Trip delay Time (RTT(t)) and the packet loss ratio (P(t)) are given, a general effective transmission rate is estimated by the equation 1. Namely, in the general method for estimating the transmission rate by using the equation 1, when the packet size (s) is fixed, the estimated transmission rate is varied according to the Round-Trip delay Time (RTT(t)) and the packet loss ratio (P(t))- Figure 1 shows a change in an available transmission rate estimated by the equation 1 when the packet loss ratio (p(t)) and the packet size (s) are fixed and the Round-Trip delay Time (RTT(t)) is linearly changed. When the packet loss ratio is fixed such that p(t)= 0.015 and s = 625 and the Round-Trip delay Time (RTT(t)) is linearly increased from 80ms to 380ms, the minimum transmission rate of a user is 50kbps and the maximum transmission rate is 500kbps. As shown in Figure 1, as the Round-Trip delay Time (RTT(t)) is increased, the estimated available transmission rate is appropriately decreased. Figure 2 shows a change in an available transmission rate estimated by the equation 1 when the Round-Trip delay Time (RTT(t)) and the packet size (s) are fixed and the packet loss ratio (p(t)) is linearly changed. When the Round-Trip l|5delay Time (RTT(t)) is fixed to 100ms and the packet size (s) is fixed to 625 and the packet loss ratio (p(t)) is changed from 0.1% to 20%, the minimum transmission rate of the user is 50kbps, and the maximum transmission rate is i 500kbps. As shown in Figure 2, in the general method for estimating an effective 2jotransmission rate by using the equation 1, it is not considered that a phenomenon such as time out occurs due to the large packet loss. Accordingly, when the packet loss ratio is small, the available transmission rate is estimated at an appropriate value. However, if the packet loss ratio is increased, the available transmission rate is overestimated. Namely, although the packet loss ratio is 10% which is considerably large, the available transmission rate is undesirably overestimated at about 200kbps by the general estimation method of the transmission rate. Also, the general estimation method of the transmission rate is disadvantageous in that the network congestion cannot be quickly resolved and reception quality of multimedia data is degraded because the available transmission rate is overestimated when the packet loss ratio is large. US 2002/004841 A1 discloses a method for adjusting the transmission rate of a multimedia data transmitting apparatus based on a data loss rate indicated in RTCP receiver report messages fed back from a receiver of the multimedia data. The transmission rate is Initially set to a given initialisation value. Depending on the reported data loss rate, the transmission rate is increased, decreased or maintained. More specifically, if the reported data loss rate is lower than a first threshold value, the transmission rate is increased. If the data loss rate is larger than a second threshold value which is greater than the first threshold value, the transmission rate is decreased. Also, if the data loss rate is reported to lie between the first and second threshold values, the transmission rate remains unchanged. D. Slsatem et al., In "The Loss-Delay Based Adjustment Algorithm: A TCA friendly Adaptation Scheme", Proceedings of NoSSDAV, Cambridge, UK 1998presenta congestion control mechanism for RTP-based multimedia applications wherein a sender calculates, on each received RTCP packet from a receiver, a transmission rate with respect to the receiver in a recursive manner, distinguishing between a no-loss case and a case of losses being reported. In the no-loss case, the sender Increases the transmission rate by an increment referred to as an additive increase rate (AIR) in this article. This additive increment is calculated based on the current transmission rate, a so-called bottleneck bandwidth indicative of the inter-packet spacing of two co-travelling packets due to a bottleneck along the transmission route, and a previous value of the increment In the tossy case, the transmission rate is reduced proportional to the reported loss factor. The mentioned article also refers to a TCP-throughput model estimating the throughput of a TCP connection from the maximum packet length, the round trip time of the connection and the average loss measured during the lifetime of the connection. US 2003/103243 Al discloses a method of adjusting a packet sending interval of a RTP-based communications apparatus so that the transmission rate approximates a target value. Principles of RTCP reporting can be found, e.g., in WO 2004/040928 Al. SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide an adaptive estimation method of transmission rate of multimedia data capable of improving accuracy of an estimated value of an available transmission rate by detecting a packet loss ratio from a RTCP receiver report packet transmitted from a reception side and varying an estimation method of an available transmission rate according to the packet loss ratio. To achieve these and other advantages and in accordance with the purpose of the present invention, as embodied and broadly described herein, there is provided an adaptive estimation method of transmission rate of multimedia data comprising: receiving a real-time transport control protocol i(RTCP) receiver report packet from a multimedia data reception unit; detecting a packet loss ratio from the RTCP receiver report packet; and adaptively estimating an available transmission rate according to a range within which the packet loss ratio falls into. The foregoing and other objects, features, aspects and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS E The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a unit of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention. lt> In the drawings: Figure 1 is a view which illustrates a change in an available transmission rate estimated by a general estimation method when a packet loss ratio is fixed and a round-trip delay time is linearly changed; Figure 2 is a view which illustrates a change in an available transmission rate estimated by a general estimation method when the round-trip delay time is fixed and the packet loss ratio is linearly changed; Figure 3 is a view which illustrates configuration of a general real-time multimedia data transmission system using a RTP and a RTCP; Figure 4 is a view which illustrates an adaptive estimation method of transmission rate of multimedia data in accordance with the present invention; and Figure 5 is a view showing transmission rate values estimated by a calculation method that varies according to the magnitude of the packet loss ratio. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Reference will now be made in detail to the preferred embodiments of the present invention, examples of which are illustrated in the accompanying ^drawings. Figure 3 shows configuration of a general real-time multimedia data transmission system using a RTP and a RTCP. As shown, a transmission unit and a reception unit of the general real time multimedia data transmission system are divided into transport domains and lpcompression domains (included in an application layer), respectively. The compression domain of the transmission unit includes a video encoder which encodes multimedia data according to an estimated transmission rate; and a video compression unit which compresses the encoded multimedia data. The transport domain of the transmission unit includes a real-time transport l|5protocol layer (RTP layer) controlling transport of multimedia data transported from the video compression unit, a UDP (user datagram protocol) layer and an IP (internet protocol) layer for transmitting to the reception unit, the multimedia data transported from the RTP layer, and a transmission rate control unit connected to the RTP layer, detecting a packet loss ratio from a RTCP receiver report (RR) 2|opacket transmitted from the reception unit, measuring a Round-Trip delay Time (RTT) by using the RTCP RR packet, estimating a transmission rate by using the packet loss ratio and the round-trip delay time, and providing the estimated transmission rate to the video encoder. The transport domain of the reception unit includes an IP layer and a UDP layer, a RTP layer controlling transport of multimedia data transmitted from the UDP layer, a service quality monitoring unit monitoring QoS (Quality of Service) of multimedia data through information included in a header of a RTP packet of the RTP layer, and a channel report unit for reporting to the transmission unit on channel information based on the monitored QoS. The channel report unit {transmits the channel information to the transmission unit through a RTCP RR packet. The RTCP RR packet includes a reception report block for feeding back to the transmission unit, the statistical information of RTP packets transmitted from the transmission unit. The reception report block further includes a packet loss ratio and a LSR (last SR timestamp). The LSR indicates a transmission time of the last RTCP sender report (SR) packet. The compression layer of the reception unit includes a video decoder which decodes multimedia data transmitted from the RTP layer. The multimedia data packet transmitted by the transmission unit of the l^general multimedia data transmission system having such configuration may be lost on the Internet or may be abandoned by the reception unit due to excessive time delay. Other packets arriving in the reception unit on time are transmitted to the vide decoder through the IP/UDP/RTP layers and then decoded. The reception unit receives the multimedia data packet and then 2fc)calculates information on a network state such as a packet loss ratio, a delay or the like through various information existing in a RTP header of the multimedia data packet, thereby monitoring QoS (Quality of Service). The monitored information is fed back to the transmission unit through a real-time transport control protocol (RTCP) of the channel report unit. The transmission unit predicts 2j5an available bandwidth of a current channel by using the information which has been fed back, and the predicted available bandwidth, namely, the available transmission rate, is transmitted to the video encoder. The video encoder having received the predicted available transmission rate controls an output encoding rate of a multimedia data bit stream corresponding to the predicted transmission rate. The present invention proposes a new estimation method of a transmission rate in which the transmission rate is estimated through information included in a RR packet when a transmission unit receives a RTCP RR packet from a reception unit in the multimedia data transmission system, while the method is not against a basic concept of a TCP (transmission control protocol)- friendly transmission-rate estimation. In the multimedia data transmission system to which the present invention is applied, one RTCP packet is transmitted every time a predetermined number of multimedia data packets of a RTP packet format are transmitted, and the predetermined number may be 20 as one example. Figure 4 shows an adaptive estimation method of a transmission rate of multimedia data. In the multimedia data transmission system, the transmission unit transmits multimedia data in RTP packet format to the reception unit (S11). Here, one RTCP packet is transmitted every time a predetermined number of RTP packets, for example, 20 RTP packets, are transmitted. Thus, the transmission unit transmits to the reception unit, one RTCP SR packet every time it transmits 20 RTP packets. The RTCP SR packet includes information related to a transmission time of the SR packet, namely, a value of a NTP (network time protocol) time stamp at the time of SR packet transmission. When receiving the RTCP SR packet, the reception unit obtains a transmission time of the SR packet based on the NTP time stamp value of the RTCP SR packet, records the obtained transmission time of the SR packet on a LSR field to generate a RTCP RR packet and transmits the generated RTCP RR packet to the transmission unit (S13). The RTCP RR packet includes a packet loss ratio. The transmission unit having received the RTCP RR packet detects a packet loss ratio from the RTCP RR packet (S15) and estimates an available transmission rate (R(tn)) by a calculation method that varies according to the detected packet ioss ratio as shown in the following equation A. Namely, the transmission unit estimates an available transmission rate (Ri (tn)) when the detected packet loss ratio does not exceed 5%(t,) transmission unit estimates an available transmission rate (R2tn)) when the packet loss ratio is greater than 5% and smaller than 10% (0.05 transmission unit estimates an available transmission rate (R3(tn)) when the packet loss ratio is not smaller than 10% (p(tn) >.1). In other words, when the detected packet loss ratio does not exceed 5% (p(ti) (tn)) by the general estimation method using the equation 1. When the detected packet loss ratio is greater than 5% and smaller than 10% (0.05 normally arriving in the reception unit, of 20 RTP packets by a time required to transmit the 20 packets, thereby estimating the available transmission rate (R2(tn)). The time required to transmit the 20 RTP packets is calculated by using a LSR value included in the RTCP RR packet. Namely, the transmission unit subtracts a LSR value (LSR(tn-1)of a previously-received RTCP RR packet from a LSR value (LSR(tn)) of the currently-received RTCP RR packet, thereby obtaining the time required to transmit the 20 RTP packets. Namely, the transmission unit obtains a transmission time of the nth-transmitted SR packet (SR(tn)) by the LSR value (LSR(tn)) of the nm-fed back RR packet (RR(tn)) and subtracts a transmission time of the n-1lh SR packet (SR(tn-1)) from the transmission time of the nth SR packet (SR(tn)), thereby obtaining a time required to transmit the 20 RTP packets (LSR(tn)-LSR(tn-l)). Also, the size of total data having normally arrived in the reception unit is calculated as follows. The transmission unit checks the number of RTP packets having been normally transmitted between the transmission time of the n-1th SR packet (SR(ti-i)) and the transmission time of the nth SR packet (SR(tn)), and calculates the size of the normally-arriving data by using the checked number of the normally-transmitted RTP packets. The number of RTP packets having normally transmitted during an interval between the n-l"1 SR packet (SR(tn-i)) and the n"1 SR packet (SR(tn)) is s-20-(1-p(t,)). The p(tn) indicates a packet loss ratio detected from the nm RTCP RR packet, and the s indicates the packet size. The a and 0 are weighting factors given to allow transmission rates ! estimated by different equations to have continuous values without great differences therebetween when the packet loss ratios are 5% and 10%. The a and P may be calculated by using the three expressions in equation A when the packet loss ratio of the RTCP RR packet transmitted from the reception unit is 5% and the packet loss ratio of the RTCP RR packet is 10%. Here, if a time required to send the data is constant, the equation for estimating the available transmission rate (R2(tn)) is defined as a linear expression with respect to the packet loss ratio p. If the packet loss ratio detected in the step S15 is not smaller than 10% (p (tn) > 0.1), the transmission unit estimates the available transmission rate (R3(tn)) at a minimum transmission rate set by a user. Figure 5 shows transmission rates estimated by a calculation method that varies corresponding to the magnitude of the packet loss ratio in accordance with the present invention. Particularly, in Figure 5, shown are available transmission rates estimated corresponding to an increase in packet loss ratio p when s=625, RTT=100ms, and LSR(tn)- LSR(tn-i) = RTT/2 * 20. Here, a=44, (3=395600. As shown in Figure 5, the available transmission rate estimated by the equation A is the same as a resulting value of the equation 1 (a resulting value obtained by the general estimation method of a transmission rate) when the packet loss ratio does not exceed 5%. When the packet loss ratio is greater than 5% and smaller than 10%, the available transmission rate estimated by the equation A is lower than the resulting value of the equation 1, and when the packet loss ratio is not smaller than 10%, the estimated available transmission rate is the minimum transmission rate input by the user. Accordingly, in the estimation method of the transmission rate in accordance with the present invention, when the packet loss ratio is large, the available transmission rate is estimated at a value smaller than an estimated value obtained by the general estimation method of the transmission rate. As described so far, in the present invention, a packet loss ratio is sdetected from a RTCP RR packet transmitted from a reception unit and an estimation method of an available transmission rate is adaptively selected according to a range within which the detected packet rate falls into, so that the available transmission rate is estimated at a smaller value when the packet loss is large. Accordingly, the network congestion can be quickly resolved and reception l Equality of multimedia data due to the packet loss can be improved. As the present invention may be embodied in several forms without departing from the spirit or essential characteristics thereof, it should also be understood that the above-described embodiments are not limited by any of the details of the foregoing description, unless otherwise specifjed^bjutjather...sJiouJd be construed broadly within its spirit and scope as defined in the appended claims, and therefore all changes and modifications that fall within the metes and bounds of the claims, or equivalence of such metes and bounds are therefore intended to be embraced by the appended claims. WE CLAIM : 1. An adaptive estimation method of multimedia data transmission rate comprising the steps of: receiving (SI3) a real-time transport control protocol (RTCP) receiver report packet from a multimedia data reception unit; detecting (SI5) a packet loss ratio from the RTCP receiver report packet; and adaptively estimating (SI7) an available transmission rate according to a range within which the packet loss ratio falls ; wherein the step of estimating (SI7) comprises estimating the available transmission rate by a first equation when the packet loss ratio falls into a first range; estimating the available transmission rate by a second equation when the packet loss ratio falls into a second range ; and estimating the available transmission rate by a third equation when the packet loss ratio falls into a third range ; the first equation being as under : wherein : s : packet size RTT(tn): Round-Trip delay Time p(tn): packet loss ratio detected from nth RTCP receiver report (RR) packet Ri(.): available transmission rate in the first range ; the second equation being as under : wherein : s : packet size p(tn): packet loss ratio detected from n* RTCP receiver report (RR) packet LSR(tn): transmission time of n* RTCT sender report (SR) packet a,p : weight factor R2(.): available transmission rate in the second range ; and the third equation comprising a minimum transmission rate set by a user. 2. The method as claimed in claim 1, wherein the first range is a range of 5% or less, the second range is a range between 5% and 10%, and the third range is a range of 10% or more. 3. The method as claimed in claim 1 or 2, wherein one RTCP sender report packet is transmitted and one RTCP receiver report packet is received every time a predetermined number of RTP packets are transmitted. 4. The method as claimed in claim 3, wherein the predetermined number of RTP packets is 20. 5. The method as claimed in claim 3 or 4, wherein the RTCP receiver report packet comprises a reception report block for feeding back to a multimedia data transmission unit, statistical information of RTP packets transmitted from the multimedia data transmission unit, wherein the reception report block contains a packet loss ratio and a transmission time of the last RTCP sender report packet. There is disclosed an adaptive estimation method of a multimedia data transmission rate. A transmission unit of a multimedia data transmission system transmits multimedia data to a reception unit through a RTP packet (S1 1), the reception unit transmits state information of a network to the transmission unit through a RTCP receiver report packet (S13), and the transmission unit detects a packet loss ratio from the RTCP receiver report packet (S15), and estimates an available transmission rate by a calculation method that varies according to a range within which the detected packet loss ratio falls into (S17). Accordingly, the estimation method can speedily resolve network congestion when the packet loss is large.

Documents:

611-KOL-2005-CORRESPONDENCE.1.2.pdf

611-KOL-2005-CORRESPONDENCE.pdf

611-KOL-2005-FORM 27.pdf

611-KOL-2005-FORM-27.pdf

611-kol-2005-granted-abstract.pdf

611-kol-2005-granted-assignment.pdf

611-kol-2005-granted-claims.pdf

611-kol-2005-granted-correspondence.pdf

611-kol-2005-granted-description (complete).pdf

611-kol-2005-granted-drawings.pdf

611-kol-2005-granted-examination report.pdf

611-kol-2005-granted-form 1.pdf

611-kol-2005-granted-form 13.pdf

611-kol-2005-granted-form 18.pdf

611-kol-2005-granted-form 2.pdf

611-kol-2005-granted-form 3.pdf

611-kol-2005-granted-form 5.pdf

611-kol-2005-granted-gpa.pdf

611-kol-2005-granted-priority document.pdf

611-kol-2005-granted-reply to examination report.pdf

611-kol-2005-granted-specification.pdf

611-kol-2005-granted-translated copy of priority document.pdf

611-KOL-2005-PA.pdf