Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
END-TO-END QOS AND FLOW CONTROL FOR ADAPTIVE CHANNELS
Document Type and Number:
WIPO Patent Application WO/2010/032106
Kind Code:
A1
Abstract:
A method for dynamically controlling user data throughput at the quality of service (QoS) enforcers output for at least the purpose of regulating it according to the available bit rate over the transmission channel. The method includes an adaptive transmission channel comprising one quality of service enforcer, one encapsulation device, and one modulator, wherein the encapsulation device or the modulator or both are configured to use an adaptive transmission mechanism.

Inventors:
SNAPIR DOR (IL)
LEVER DUBI (IL)
BEN LAISH ISAAC (IL)
Application Number:
PCT/IB2009/006834
Publication Date:
March 25, 2010
Filing Date:
September 14, 2009
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
GILAT SATELLITE NETWORKS LTD (IL)
SNAPIR DOR (IL)
LEVER DUBI (IL)
BEN LAISH ISAAC (IL)
International Classes:
H04J3/14; H04L47/20
Foreign References:
US20060045128A12006-03-02
US6163766A2000-12-19
US20040062268A12004-04-01
US20040008736A12004-01-15
US6163766A2000-12-19
Other References:
See also references of EP 2327176A4
Attorney, Agent or Firm:
REINHOLD COHN & PARTNERS (Tel-Aviv, IL)
Download PDF:
Claims:
JMS

We claim:

n a communication system which includes an adaptive transmission channel comprised >f at least:

one quality of service enforcer;

one encapsulation device; and

one modulator,

wherein the encapsulation device or the modulator or both are configured to use an adaptive transmission mechanism;

a method for dynamically controlling the user data throughput at the QoS enforcer's output for at least the purpose of regulating it according to the available bit rate over the transmission channel.

The method of claim 1 , wherein:

at least one encapsulation device is configured to send available bit rate information to at least one QoS enforcer;

at least one QoS enforcer is configured to receive available bit rate information; and

at least one QoS enforcer is configured to reconfigure one or more internal software or hardware modules for at least the purpose of regulating its output bit rate according to received available bit rate information.

The method of claim 1, wherein the communication system is a satellite communication ystem and the adaptive transmission mechanism follows the DVB-S2 standard (EN 302 107).

n a communication system which includes an adaptive transmission channel, a method or calculating an available bit rate over the transmission channel, wherein said method is iomprised of: measuring the total amount of user data being transmitted over the transmission channel during a predefined interval and calculating a corresponding user traffic bit rate;

estimating average spectral efficiency for the user data transmitted during the same interval;

calculating a null packet bit rate using the estimated average spectral efficiency; and

calculating an available bit rate as the sum of the calculated user traffic bit rate and the calculated null packet bit rate.

The method of claim 4 , wherein estimation of average spectral efficiency is further comprised of:

determining the total amount of user data transmitted over the transmission channel during said predefined measurement interval;

determining the amount of user data out of the total amount of user data that have been transmitted at each modulation and coding combination;

multiplying each amount of user data by a predefined spectral efficiency figure corresponding to the modulation and coding combination used for modulating that amount of user data; and

summing all said multiplication products and dividing the result by said total amount of user data transmitted.

The method of claim 4 wherein calculating a null packet bit rate is further comprised of:

determining the total number of null packets transmitted during said predefined measurement interval and calculating a corresponding bit rate;

dividing said corresponding bit rate by the spectral efficiency corresponding to the modulation and coding option used for transmitting said null packets; and

multiplying the division product by said average spectral efficiency. The method of claim 4, wherein the communication system is comprised of at least:

one quality of service enforcer;

one encapsulation device; and

one modulator,

and wherein the encapsulation device or the modulator or both are configured to use an adaptive transmission mechanism.

Tie method of claim 7, wherein the communication system is a satellite communication ystem and the adaptive transmission mechanism follows the DVB-S2 standard (EN 302 07).

^. communication system comprised of at least:

one quality of service enforcer;

one encapsulation device; and

one modulator,

wherein the encapsulation device or the modulator or both are configured to use an adaptive transmission mechanism;

at least one encapsulation device is configured to send available bit rate information to at least one QoS enforcer;

at least one QoS enforcer is configured to receive available bit rate information; and

at least one QoS enforcer is configured to reconfigure one or more internal software or hardware modules for at least the purpose of regulating its output bit rate according to received available bit rate information.

"he system of claim 9, wherein at least one encapsulation device is configured to alculate an available bit rate over the transmission channel. The system of claim 9, wherein the communication system is a satellite communication system and the adaptive transmission mechanism follows the DVB-S2 standard (EN 302 307).

Description:
END-TO-END QoS AND FLOW CONTROL FOR ADAPTIVE CHANNELS

FIELD OF THE INVENTION

This invention relates to the field of communication networks in general and more specifically to the fields of traffic engineering and quality-of-service providing and monitoring.

BACKGROUND

High speed data networks may be configured to support quality-of-service sensitive applications and/or to enforce service level agreements (SLA). These requirements may be satisfied through use of tight quality of service (QoS) monitoring and traffic engineering mechanisms. Therefore such networks may often be configured to use schedulers and other methods for distributing available bandwidth and/or capacity between users and applications according to priorities, weights, SLA, etc.

In another aspect, one characteristic of adaptive channels (e.g. DVB-S2 ACM) is dynamic user data throughput. While a channel's rate (i.e. the number of transmitted symbols per second) remains constant, user data throughput constantly changes according to the modulation and coding (MODCOD) combinations being used for modulating the transmission carrier signal. User data throughput may change very rapidly in response to a sudden degradation in link conditions (e.g. rain fades over satellite links).

Introducing adaptive channels to high speed communication networks renders simple flow-control mechanisms and/or algorithms insufficient, especially in scenarios of rapid changes in user data throughput. In order to maintain quality of service (e.g. for real-time applications such as VoIP sessions) even when the total user data throughput has been significantly decreased, at least the QoS mechanism has to be constantly informed, in real-time, regarding the currently available user data throughput. SUMMARY

In aspects of this invention, a feedback mechanism may be used for at least the purpose of informing one or more QoS enforcers regarding the amount of available bandwidth and/or capacity over a transmission link or a channel, regardless of the current amount of user data being transmitted over that link (e.g. a link utilization ratio).

In another aspect of this invention, the available bandwidth and/or capacity may be calculated at any given time while user data may be transmitted over the transmission link or channel. Such calculation may be performed according to the average spectral efficiency by which data is transmitted over a transmission link or a channel at the time of calculation.

In yet another aspect of this invention, a QoS mechanism may be configured to forward traffic according to the available bandwidth and/or capacity it can actually send to the network.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig 1 shows a block diagram of a communication network in accordance with the aspects of this invention.

DETAILED DESCRIPTION

Fig 1 shows a block diagram of a communication network, which includes a transmission chain 100. Transmission chain 100 may be comprised of a boarder router 110, a QoS enforcer 120, one or more data processors 130a to 13On, an encapsulation function or an encapsulation device 140, and a modulator 150. Furthermore, as transmission chain 100 may be a part of a communication network, Fig 1 also shows a network controller 160 and a front end unit 190.

Boarder router 110 may be configured to allow only selected traffic to enter the transmission chain, such as traffic destined to one or more receiving parties at the other end of the transmission channel. QoS enforcer 120 may be configured to at least provide flow control functionality by limiting the total data rate sent towards data processors 130a to 13On and thereafter towards encapsulation function 140. In some embodiments, QoS enforcer 120 may be replaced by a traffic engineering device or a traffic shaping device, which perform similar flow control functions.

One or more data processors 130a to 130n may be configured to receive user data and process it using one or more methods and/or algorithms, including but not limited to encryption algorithms, compression algorithms, acceleration methods and any other method which may be applicable for optimizing user traffic over the transmission link. Furthermore, data processors 130a to 130n may be configured to send the processed data towards encapsulating device 140, e.g. over an IP protocol. In some embodiments, one or more TCP and/or UDP tunnels may be used by each data processor in order to transfer different types of data (e.g. real-time information, multicast streams, non-real-time data, etc).

Encapsulation device 140 may be configured to at least receive user data information (e.g. over an IP protocol using one or more TCP and/or UDP tunnels), encapsulate it over an applicable transport stream (e.g. Baseband Frames for a DVB-S2 carrier in a satellite communication system) and provide the transport stream to modulator 150. Modulator 150 may be configured to at least modulate the transport stream using the applicable modulation technique and to output a modulated signal.

Though Fig 1 shows encapsulation device 140 and modulator 150 as two different entities or devices, in some embodiments the encapsulation function may be integrated with the modulator in a single device. Both types of embodiments are consistent with the aspects of this invention.

Furthermore, in some embodiments, where no processing of the user data prior to transmission is required, data processors 130a to 130n may be either bypassed or absent from the transmission chain. In such embodiments, user data may be sent from QoS enforcer 120 directly to encapsulation device 140. This type of embodiments is also consistent with the aspects of this invention. In yet further embodiments, the output of modulator 150 may be fed into a front-end 190, which may be configured to do any one or more of amplifying the modulated signal, changing the signal's carrier frequency, converting the signal to a different form (e.g. form electric form to electro-magnetic form or to optic form) and any other function which may be required in order to actually have the signal transmitted over the designated media.

In some communication systems, modulator 150 and encapsulation device 140 may be configured to use one or more adaptive transmission techniques (e.g. DVB-S2 in satellite communication systems), whereby the spectral efficiency of the modulated signal may be modified on a frame-by-frame basis. If a receiving station has good reception conditions (i.e. the transmitted signal is received at high C/N ratio), modulator 150 may be indicated to modulate the data destined to that station using a modulation technique, which maps more bits into each transmitted symbol. Furthermore, encapsulation device 140 may be indicated in such a case to encode said data using weaker forward error correction (FEC) codes for at least the purpose of decreasing the number of transmitted bits used for FEC. On the other hand, if a receiving station has medium or poor reception conditions (i.e. the transmitted signal is received at low or minimal C/N ratio respectively), modulator 150 may be indicated to modulate the data destined to that station using a modulation technique, which maps less bits into each transmitted symbol. Furthermore, encapsulation device 140 may be indicated in such a case to use stronger FEC codes, which may impose higher overhead but also increase the probability of recovering the transmitted data at the receiving side.

In some embodiments, a network controller (e.g. network controller 160) may be configured to send and data processors 130a to 130n may be configured to receive information regarding a most efficient MODCOD combination that may be supported by a receiving party. Data processors 130a to 130n may be further configured to attach MODCOD information received from the network controller to every data packet sent towards encapsulation device 140, wherein said MODCOD information relates to the receiving party which the data included in the transmitted packet is destined for. In yet further embodiments, where data processors are not used, MODCOD information may be sent from the network controller to encapsulation device 140, which may be further configured to use the appropriate MODCOD information based on destination analysis.

Each MODCOD combination, as described above, may be characterized by a different spectral efficiency figure, i.e. different number of user bits per transmitted symbol. Since MODCOD adaptability may be implemented on a frame-by-frame basis and since modulator 150, in most cases, may be configured to transmit at a constant preconfigured symbol rate, the channel's throughput depends on the number of frames transmitted using each MODCOD combination. Throughput may be higher as more frames are transmitted using more efficient MODCOD combinations, or lower as more frames are transmitted using more robust and less efficient MODCOD combinations.

In one aspect of this invention, encapsulation device 140 may be configured to calculate an average spectral efficiency figure for the transmission channel and thereafter use this average spectral efficiency figure for calculating the available user data throughput or bit rate.

Encapsulation device 140 may be configured to measure the actual user data throughput, e.g. by counting and/or calculating the number of user traffic bits or bytes being transmitted over a period of time. Encapsulation device 140 may be further configured to gather information on the current blend of MODCOD combinations, which may be used for transmitting the same user data traffic over the same period of time. As previously described, MODCOD information may be attached to each data packet arriving from the data processors or internally generated by encapsulation device 140 based on information received from a network controller. Using this MODCOD information, encapsulation device 140 may calculate the number of user traffic bits or bytes that were transmitted using each MODCOD combination. Having measured the actual user throughput and determining how many user traffic bits or bytes have been transmitted using each MODCOD combination, encapsulation device 140 may calculate the average spectral efficiency for the transmitted data.

Furthermore, knowing the preconfϊgured channel (symbol) rate, encapsulation device 140 may be configured to calculate the number of unused symbols (which may be filled either by modulator 150 with dummy frames or by encapsulation device 140 with null packets). Knowing the number of unused symbols per the measurement interval, encapsulation device 140 may further be configured to calculate the user traffic bit rate or throughput that may be accommodated by these unused symbols assuming the average spectral efficiency previously calculated is applicable to these symbols as well.

The above mechanism may be further described using the following example. Considering the following traffic being transmitted:

Port #1 of encapsulation device 140 may receive user traffic at a rate of 250Kbytes per second (2Mbps), out of which lOOKbytes per second are transmitted using a QPSK 5/6 MODCOD combination and the remaining 150Kbytes per second are transmitted using an 8PSK 2/3 MODCOD option. Port #2 of encapsulation device 140 may receive user traffic at a rate of 250Kbytes per second (2Mbps), all of which is transmitted using an 8PSK 2/3 MODCOD option.

Encapsulation device 140 generates lMbps worth of null packets, which are transmitted using a QPSK 1/4 MODCOD combination.

Considering only the user traffic on ports #1 and #2, and given that the spectral efficiency figures for QPSK 5/6 is 1.654663 and that of 8PSK 2/3 is 1.980636, the average spectral efficiency may be calculated as follows:

Average Spectral efficiency = (100K*1.654663+400K*1.980636)/500K = 1.9154414

In addition, given a spectral efficiency figure of 0.490243 for the QPSK 1/4 MODCOD combination and the average spectral efficiency for user traffic as calculated above, the available bandwidth represented by the null packets may be calculated as follows: Null Packets Bit Rate = (1M/O.49O243)*1.9154414 = 3.907126Mbps

Therefore:

Available Bit Rate = User Traffic Bit Rate + Null Packet Bit Rate

Available Bit Rate = 2Mbps + 2Mbps + 3.907126Mbps = 7.907126Mbs

In another aspect of this invention, a feedback path may exist between encapsulation device 140 and QoS enforcer 120. Encapsulation device 140 may be configured to use this feedback path for at least the purpose of sending the calculated available bit rate to QoS enforcer 120. Though there might be many appropriate methods for sending the available bit rate information to QoS enforcer 120, in some preferred embodiments this information may be encapsulated into an SNMP trap packet, which may be transmitted over a local LAN to which both devices may be connected.

Furthermore, QoS enforcer 120 may be configured to receive available bit rate information over a feedback channel and to reconfigure any of its internal software and/or hardware modules for at least the purpose of regulating its output bit rate according to the received available bit rate information. In some embodiments, QoS enforcer 120 may first serve preconfigured service level agreements (SLA) and high priority applications and only then allocate the remaining throughput to other applications, streams and/or traffic types. Thus, changes in the transmission channel's throughput may affect low priority traffic while SLA and high priority applications may continue to receive high quality of service (i.e. their required throughput).