CA2193511A1 - Non-transparent data transmission in a digital telecommunications system - Google Patents
Non-transparent data transmission in a digital telecommunications systemInfo
- Publication number
- CA2193511A1 CA2193511A1 CA002193511A CA2193511A CA2193511A1 CA 2193511 A1 CA2193511 A1 CA 2193511A1 CA 002193511 A CA002193511 A CA 002193511A CA 2193511 A CA2193511 A CA 2193511A CA 2193511 A1 CA2193511 A1 CA 2193511A1
- Authority
- CA
- Canada
- Prior art keywords
- data
- data connection
- channel coding
- coding scheme
- transmission
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
- H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
- H04L1/18—Automatic repetition systems, e.g. Van Duuren systems
- H04L1/1812—Hybrid protocols; Hybrid automatic repeat request [HARQ]
- H04L1/1819—Hybrid protocols; Hybrid automatic repeat request [HARQ] with retransmission of additional or different redundancy
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
- H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
- H04L1/18—Automatic repetition systems, e.g. Van Duuren systems
- H04L1/1809—Selective-repeat protocols
Abstract
The invention relates to a digital telecommunications system, and particularly a method for non-transparent data tranmission. There is a non-transparent data connection (V.110+L2R/RLP) between a transmitting party (A) and a receiving party (B), on which channel coding is employed. The transmission buffer buffers the data frames to be transmitted and stores the transmitted data frames until it receives an acknowledgement of successful reception. In accordance with the invention, this is avoided by monitoring the quality of the non-transparent connection, and if the quality of the connection drops to a specific threshold value, by changing the channel coding scheme of the connection for a more efficient one. Simultaneously, however, more channel capacity is allocated for the connection in according with the chosen channel coding scheme so that the original nominal user data rate can be maintained unchanged or nearly unchanged.
Description
WO96/36146 E~~
~ 2, 9351 1 Non-transparent data transmission in a digital telecommunications system Field of the Invention The present invention relates to non-transparent data transmission in a digital telecommunications system, in particular in a radio interface of a mobile communications system.
Ba_~y v~d of the Invention The problem in transmission over transmission channels that are sensitive to interference, such as the radio path, is often the poor quality of the transmissior.. Although severa~ standardiz.ed end-to-er.d data services are actuall~1 provided with error correction protocols, which are applied either to the er.tire end-to-end connection or to some segments of the transmission path, these protocols are designed for error situations that are typical of fi~ed lines, but they are inade~uate or unsuitable for special conditions, such as to a radio link. Therefore, it has been necessary to implement dedicated error correction solutions within the mobile communications .system.
Since no single solution ls suitable for all data services, there are several types of connections available e.g. in the pan-European mobile communications system GSM, and these connections may be divided into two classes. In the first class, error correction is carried out solely by means of channel coding, which in the GSM system is termed as Forward Error Correction (FEC~. In the GSM recommendation this is termed as a transparent transfer mode, which is also abbreviated with "T". In the second connection class, in addition to channel coding, an additional protocol is applied, with retransmission of data which has not been received correctly at the other end. In the GSM
WO~6~614G ~ r~ '~
3~ ~ ~
system this communication protocol is termed as a Radio Link Protocol ~RLP~, and this data transfer mode is referred to as non-transparent asynchronous data transmission, which is also abbreviated with NT. The present invention relates to the non-transparent data transmission.
In the non-transparent asynchronic data transmission over a circuit switched connectior, data is transmitted from a transmitting party A to a recei~ing party B in RLP frames. The RLP frames are channel coded for the duration of transmissior., so that the errors caused by the transmission channel can be corrected by the receiving party B in a channel decoding. ln addition to the actual user data, RLP
frames also contain error correction bits, by means of wnich tke receiving party B can detect the errors that have not been corrected by channel coding. Each R1P
frame is also numbered, or the order of the frames is indicated by another kind of identifier. The correctness of eack. received frame i~ tested at the receiving end. If the frame is detected to be correct, the receiving party acknowledges the reception by using the frame number. If the frame is not detected to be correct, 2 negative acknowledgement is transm,itted.
When the transmitting end recei~es a negative acknowledgement, or no acknowledgement at all, the transmitting party A re-transmits the RLP frame a predetermined number of times. The total number of the retransmissions is limited in order to avoid endless transmission loops in a case of a very poor connection.
The RLP frames are stored, i.e. "buffered" at the trar.smitting end until they have been acknowledged, so that they are available i~ re-transmission i9 required. In order to limit the amour.t of buffering required, a sliding window is employed in the RLP
WO9613C14C I ~ rL '~ :, 2~3571 protocol. Thus, the transmitting party A may transmit several RLP frames before an acknowledgement is required from the receiving party B. Since the errors passing through the channel coding are corrected by means of re-transmission of the defective RLP frames, "surplus capacity" must be reserved for re-transmission. This means that the data rate of the non-transparent connection is higher than the nominal data rate of the user. In a case where a channel has a relatively high quality, that is, when there are few errors passing the channel coding, this surplus capacity allows a sufficient number of re-transmissions without the actual data rate of the connection falling below the nominal user data rate. When the quality of the connection becomes~ poorer, the number of defective and lost RLP frames, and thus the number of re-transmissions increases. On extremely poor connections, buffering of the RLP frames, as well as flow control must be employed, wh.ich limits the amount of incoming data from the user to the transmission buffer. This mears in practice that the actual user data rate is lower than the nominal user data rate, i.e. the throughput of the connection decreases.
It is thus a problem associated with a poor transmission channel that when the number of re-transmissions increases, the throughput of the connection drops below the nominal user data rate. If the rumber of re-transmissions ~urther increases, and the maximum number of re-transmissions set for the connection is reached in some frame, it results in resetting of the RLP protocol, whereby parties A and B
reset their internal variables related to data transmission. In connection with RLP resetting ~ information loss may cccur, and the continuity of data thus cannot be guaranteed. Bata transmission is Wo~6136146 l~lr~
practically impossible in the R~P reset -state, which causes releasing the connection.
The effect of a poQr transmission chanr,el rnay be reduced by changing a better channel coding scheme for the connection, provided that there are several channel coding alternatives availab]e in the system.
E.g. the channel coding scheme ~EC used in th.e GSM
system is convolution coding, whose efficiency may be expressed in terms of a convolution code ratio X/Y, which indicates that in channel coding, X data bits are represented with Y code bits. On a full-rate GSM
traffic ehannel, for instance, the convolutior. code ratios corresponding to the user data rates g.6 kbit/s, 4.3 kbitfs, and 2.4 kbit~s are 1j2 ~1/2 is a so-called buffered ratio, i.e. the ratio is not exactly ltZ~ 3 and 1/6, respectively. On a halE-rate traffic channel the corresponding convolution code ratios for user data rates 4.8 kbit/s and 2.4 kbit/s are 1/2 and 1i3. Taking into use a better and more efficient channel coding 2~ scheme thus requires lowering the nominal user data rate.
n;~l~S~e of the Invention An object of the invention is to improve the throughput of non-transparent data transmission in poor transmission path conditions and maintain tne nomina data rate and/or the connection in conditions in which prior art non-transparent data transmission fails.
This is achieved with a method fc~r nc)n-transparent data transmission in a digital tele-communications system, comprising the steps of:
establishing a non-transparent data connection having a nominal data rate, a first transmission capacity and a first channel coding scheme; transmitting data over a non-transparent data connection in data frames using a cornmunication protocol that acknowledges the WO96/361~6 ~ r~
f ~35~ ~
received correct data frames and re-transmits the received defective data frames; buffering the data frames to be transmitted in a transmission buffer;
storing a transmitted data frame in the transmission S buffer in case of possible re-transmission, until an acknowledgement of successful reception is received.
The invention is characterized by mor.itoring the throughput of the non-transparent data connection on the basis of the number of re-t.ransmi.ssions, requesting more transmission capacity and a more efficient channel coding scheme for said data connection if the throughput of the data connection drops to a predetermined threshold value.
The invention further relates to a telecommunications system in accordance with claim 3.
In the invention, the quality of a non-transparent connection is monitored, and, if the quality of the connection falls to a specific threshold value, a more efficient channel coding scheme is changed for the connection. More channel capacity is simultaneously allocated for the connection, however, in accordance with the chosen channel coding scheme so that the original nominal user data rate can be maintained unchanged or nearly unchanged. A half-rate traffic channel, for instance, may b~ changed for a full-rate traffic channel, and a more efficient channel coding scheme may simultaneously be taken into use.
Correspondingly, in a system in which a multichannel connection can be use-l, the number of the channels rr.ay be increased so that a more efficient channel coding scheme may be taken into use without reducing the nominal data rate.
In an embodirr.ent of the invention, the present channel coding scheme and channel capacity are WO~6136146 ~ IrL ~
~ 2, 9351 1 Non-transparent data transmission in a digital telecommunications system Field of the Invention The present invention relates to non-transparent data transmission in a digital telecommunications system, in particular in a radio interface of a mobile communications system.
Ba_~y v~d of the Invention The problem in transmission over transmission channels that are sensitive to interference, such as the radio path, is often the poor quality of the transmissior.. Although severa~ standardiz.ed end-to-er.d data services are actuall~1 provided with error correction protocols, which are applied either to the er.tire end-to-end connection or to some segments of the transmission path, these protocols are designed for error situations that are typical of fi~ed lines, but they are inade~uate or unsuitable for special conditions, such as to a radio link. Therefore, it has been necessary to implement dedicated error correction solutions within the mobile communications .system.
Since no single solution ls suitable for all data services, there are several types of connections available e.g. in the pan-European mobile communications system GSM, and these connections may be divided into two classes. In the first class, error correction is carried out solely by means of channel coding, which in the GSM system is termed as Forward Error Correction (FEC~. In the GSM recommendation this is termed as a transparent transfer mode, which is also abbreviated with "T". In the second connection class, in addition to channel coding, an additional protocol is applied, with retransmission of data which has not been received correctly at the other end. In the GSM
WO~6~614G ~ r~ '~
3~ ~ ~
system this communication protocol is termed as a Radio Link Protocol ~RLP~, and this data transfer mode is referred to as non-transparent asynchronous data transmission, which is also abbreviated with NT. The present invention relates to the non-transparent data transmission.
In the non-transparent asynchronic data transmission over a circuit switched connectior, data is transmitted from a transmitting party A to a recei~ing party B in RLP frames. The RLP frames are channel coded for the duration of transmissior., so that the errors caused by the transmission channel can be corrected by the receiving party B in a channel decoding. ln addition to the actual user data, RLP
frames also contain error correction bits, by means of wnich tke receiving party B can detect the errors that have not been corrected by channel coding. Each R1P
frame is also numbered, or the order of the frames is indicated by another kind of identifier. The correctness of eack. received frame i~ tested at the receiving end. If the frame is detected to be correct, the receiving party acknowledges the reception by using the frame number. If the frame is not detected to be correct, 2 negative acknowledgement is transm,itted.
When the transmitting end recei~es a negative acknowledgement, or no acknowledgement at all, the transmitting party A re-transmits the RLP frame a predetermined number of times. The total number of the retransmissions is limited in order to avoid endless transmission loops in a case of a very poor connection.
The RLP frames are stored, i.e. "buffered" at the trar.smitting end until they have been acknowledged, so that they are available i~ re-transmission i9 required. In order to limit the amour.t of buffering required, a sliding window is employed in the RLP
WO9613C14C I ~ rL '~ :, 2~3571 protocol. Thus, the transmitting party A may transmit several RLP frames before an acknowledgement is required from the receiving party B. Since the errors passing through the channel coding are corrected by means of re-transmission of the defective RLP frames, "surplus capacity" must be reserved for re-transmission. This means that the data rate of the non-transparent connection is higher than the nominal data rate of the user. In a case where a channel has a relatively high quality, that is, when there are few errors passing the channel coding, this surplus capacity allows a sufficient number of re-transmissions without the actual data rate of the connection falling below the nominal user data rate. When the quality of the connection becomes~ poorer, the number of defective and lost RLP frames, and thus the number of re-transmissions increases. On extremely poor connections, buffering of the RLP frames, as well as flow control must be employed, wh.ich limits the amount of incoming data from the user to the transmission buffer. This mears in practice that the actual user data rate is lower than the nominal user data rate, i.e. the throughput of the connection decreases.
It is thus a problem associated with a poor transmission channel that when the number of re-transmissions increases, the throughput of the connection drops below the nominal user data rate. If the rumber of re-transmissions ~urther increases, and the maximum number of re-transmissions set for the connection is reached in some frame, it results in resetting of the RLP protocol, whereby parties A and B
reset their internal variables related to data transmission. In connection with RLP resetting ~ information loss may cccur, and the continuity of data thus cannot be guaranteed. Bata transmission is Wo~6136146 l~lr~
practically impossible in the R~P reset -state, which causes releasing the connection.
The effect of a poQr transmission chanr,el rnay be reduced by changing a better channel coding scheme for the connection, provided that there are several channel coding alternatives availab]e in the system.
E.g. the channel coding scheme ~EC used in th.e GSM
system is convolution coding, whose efficiency may be expressed in terms of a convolution code ratio X/Y, which indicates that in channel coding, X data bits are represented with Y code bits. On a full-rate GSM
traffic ehannel, for instance, the convolutior. code ratios corresponding to the user data rates g.6 kbit/s, 4.3 kbitfs, and 2.4 kbit~s are 1j2 ~1/2 is a so-called buffered ratio, i.e. the ratio is not exactly ltZ~ 3 and 1/6, respectively. On a halE-rate traffic channel the corresponding convolution code ratios for user data rates 4.8 kbit/s and 2.4 kbit/s are 1/2 and 1i3. Taking into use a better and more efficient channel coding 2~ scheme thus requires lowering the nominal user data rate.
n;~l~S~e of the Invention An object of the invention is to improve the throughput of non-transparent data transmission in poor transmission path conditions and maintain tne nomina data rate and/or the connection in conditions in which prior art non-transparent data transmission fails.
This is achieved with a method fc~r nc)n-transparent data transmission in a digital tele-communications system, comprising the steps of:
establishing a non-transparent data connection having a nominal data rate, a first transmission capacity and a first channel coding scheme; transmitting data over a non-transparent data connection in data frames using a cornmunication protocol that acknowledges the WO96/361~6 ~ r~
f ~35~ ~
received correct data frames and re-transmits the received defective data frames; buffering the data frames to be transmitted in a transmission buffer;
storing a transmitted data frame in the transmission S buffer in case of possible re-transmission, until an acknowledgement of successful reception is received.
The invention is characterized by mor.itoring the throughput of the non-transparent data connection on the basis of the number of re-t.ransmi.ssions, requesting more transmission capacity and a more efficient channel coding scheme for said data connection if the throughput of the data connection drops to a predetermined threshold value.
The invention further relates to a telecommunications system in accordance with claim 3.
In the invention, the quality of a non-transparent connection is monitored, and, if the quality of the connection falls to a specific threshold value, a more efficient channel coding scheme is changed for the connection. More channel capacity is simultaneously allocated for the connection, however, in accordance with the chosen channel coding scheme so that the original nominal user data rate can be maintained unchanged or nearly unchanged. A half-rate traffic channel, for instance, may b~ changed for a full-rate traffic channel, and a more efficient channel coding scheme may simultaneously be taken into use.
Correspondingly, in a system in which a multichannel connection can be use-l, the number of the channels rr.ay be increased so that a more efficient channel coding scheme may be taken into use without reducing the nominal data rate.
In an embodirr.ent of the invention, the present channel coding scheme and channel capacity are WO~6136146 ~ IrL ~
3 5 f 1 maintained if there is no more channel capacity available. One or more new attempts are made for increasing the channel capacity and changing the channel coding scheme if the quality of the connection is still poor after a predetermined waiting time. In an embodiment of the invention, the channel codlng scheme is changed for a more efficient one although there is no more channel capacity available after the new atterrpts either, if the throughput of the connection still falls below the predetermined threshold value. If the conditions of the connection improve, the surplus channel capacity may be removed from the connection and the oriqinal channel coding scheme may be restored.
This releases channel capacity for other connections.
Brief Description of the ~r~- n~c In the following, the invention will be disclosed by means of the preferred embodiments with reference tc the attached drawinss, in which Figure l illustrates a part of a mobile communications system in which the invention may be applied on a single-channel non-transparent connection, Figure 2 is a block diagram showins the functional uni.s of a single-channel non-transparer,t ~S~ traffic channel in different protocol layers, Figure 3 shows a L2R PDU, Figure ~ an RLP frame, Figure S illustrates a part of the G.'~ mc,bile communications system in which the invertion may be applied on a m~ltichannel nor.-transparent connectior P ~f~r ed r - ts of the Invention The present invention may be used in all digital data transmission systems on a non-transparent data connection on which. transmission errors are corrected, in addition to channel codin5, by means of a transmission protocol that re-transmits the frames WO96136146 l~lr~ l ~ ~ ~ 9 3 ~ I ~
~ 7 and causes the actual data rate to drop below the nominal data rate or the connection to be interrupted when the quality of the connection becomes poorer.
The present ir.vention is suitable particularly for data transmission applications in digital CDMA or TDMA mobile communications systems, such as the pan-European digital mobile communications system GSM, DCSl800 (Digital Communication System), a mobile communications sy.stem ir accordance with EIA/TIA
Interim Standard IS/41.3, etc. The invention will be disclosed below by way of example of the GSM mobile communications system, but the invention is not limited thereto, however. The basic structure of the GSM system will be disclosed briefly below, but no closer attention will be paid to its features or other aspects of the system. The GSM system is described in more detail in "The GSM System for Mobile Communications", M. Mouly, M. Pautet, Palaiseau, France, 1992, ISBN:2-9507190-0-7, which ls ir,corporated herein by reference.
A mobile services .switching centre MSC
switches incoming and outgoing calls. It carries out tasks similar to the exchange of the public switched telephone network ~PST~. ln addition, it also carries out functions typical of mobile telecommunication only, such as location management of the subscribers in cooperation with the subscriber registers ~not shown) of the network. Mobile stations (MS) are connected to the MSC via base statlon systems BSS. A base station system BSS consists of base station controllers BSC and base stations BTS.
The GSM system is a time division multiple access system in which commur.ication on the radio path takes place in successlve TD~ frames, each of wh.ich consists of several time-slots. Ir. each time-slot a short information package is sent in form of a radio ~096~6146 l~l/r.,L,:
~3~
~ i ~3~ ~ I
frequency burst having a finite duration and consisting of a group of modulated bits. The time-slots are mainly used for transm~tting control char.nels and traffic channels. On traffic channels, speech or data is transmitted. On control channels, signalling is carried out between the base station and the mobile statlon.
The channel structures used in the radio interface of the GSM system are defined ln closer detail ir. the GSM
recommendation 05.02. In the operation in accordance lG with the recommendation, one time-slot is assigned to a mobile station MS as a traffic channel at the beginning of the call (Single Slot Access~. The mobile station MS is synchronized with this time-slot for transmitting and receiving radio frequency bursts.
In the GSM system, a data connection is established between a termir.al adaptation function (TAF~ 31 of the mobile station MS and an interworking function (IT~F) 42 of a fixed network. (usually associated with the MSC). A data cormectior; is a circuit switched connection, which reserves one (or more) traffic channels from the radio interface for the duration of entire call. In the data transmission of the GSM network, the data conrection is a V.110 rate-adapted, V-24 interface-adaptable, ~DI coded digital 2S full duplex connection. A V.110 connection is a digital transmission channel that has originally been developed for ISDN technology (integrated Services Data Network), which is defined in CCITT Recommendation Blue Book V.110. A terminal adaptation function TAF adapts a data terminal TE connected to the mobile station MS to the V. 110 conrection, which i5 established in Figure 1 over a circuit switched conr.ection that employs one traffic channel ci-0. The interworking function IWF adapts the V.110 connection to some other V.110 network, suc.h as ISDN or another GSM network, or to some other tran.sit WO96/36146 P_llr~ ~ , ~ 1 9 3 ~ 1 ~
network, such as the public switched telephone network PSTN.
In addition, channel coding FEC (Forward Error Correction) is employed on the traffic channel for reducing the effect of transmission errors on the radio path. In the GSM system corvolution coding in accordance with GSM recommendation 05.03 is employed, and its efficiency can be expressed by means of a convolution coding ratio, which indicates that in channel coding X data bits are represented with Y code bits. On a full-rate GSM traffic channel, for instance, the convolution code ratios corresponding to the user data rates 9.6 kbit~s, 4.3 kbit~s, and 2.4 kbit~s are 1~2 ~buffered), 1~3 and 1~, respectively.
The non-transparent circuit switched connection between the terminal adaptation function ~TAF) and the interworking function ~IWF) on a GSM
traffic channel comprises several protocol layers.
The terminal interface between the terminal adaptation function TAF of the mobile station MS and the data terminal equipment, as well as the interface between the interworking function IWF and e.g. an audio modem MODEM comply with CCITT V.24, and it is marked with a symbol I.2 ir the figure. erotocols that are essential to the invention are L2R ~Layer 2 Relay~ and RLP ~Radio Link Protocol), which are both located between the terminal adaptation function TAF and the interworking function IWF at both ends of tne connection. In accordance with the invention, the connection also includes several rate adaptation RA
functions, such as RAl between the terr;,inal adaptation function TAF and a CCU unit (Channel Codec Unit) placed in the BSS, RAI between the CCU unit and the interworking function IWF, RAA betweer the CCU unit and a transcoder unit TRAII placed separate from the base =
WO 96136146 P~, J I r ~
3 ~ 1 ~
station, and RA2 bet.ween the transcoder unit TRAU and the interworking function I~F. The rate adaptatio functlons RA are defined in GSM specificatior,s 04.21 and oa.20. Communication between the CCU unit and the transcoder unit TRAU is defined in GSM recommendation 08.60.
Information that has been rate adapted at the radio interface RA1 has also been channel coded ir, the way specified in GSM Recommendation 5.03. This is illustrated by blocks FEC in the mobile station MS and in the CCU unit.
The object of the invention is, however, onl, the L2R~RLP operation of the terminal adaptation function TAF and the interworking adaptation functior, IhTF and the communication between them. The other lower-layer protocols, function,s and units mentioned above only provide a transmission path in accordance with the GSM recommendations between the L2R~RLP units, and they are not relevant to the invention except for channel coding. Therefore, they are not paid closer attention to herein.
L2R (layer 2 Relaying) functionality for non-transparent character-oriented protocols is defined e.g. in GSM Recommendation 07.02. L2R packs the user data and status information received from the terminal interface in protocol data units PDU having a length of 200 bits, or 25 octets. A PDU is illustrated in Figure 3. The octets are numbered from 0 to 24, whereby octet 0 is transmitted first. The bits in octets are numbered from l to 8, and bit 1 is transmitted first. An octet in a PDU may be a status octet, a character ~higher layer data~, or dummy bits. Octet 0 is always a status octet. One status octet contains 3 bits: SA, SB, and Y~.
for the status of the V.24 connection, and 5 bits for indicating the number of the data octets subsequent to WO96i36l46 ~l/rL6~ , '' ' ~' '' :L 1 the status octet, as well as special indications, such as an empty PDU. In Figure 3, status octet 0 is followed by 3 data octet.s in which the word "GSM" is packed, whereafter a r,ew status octet 4 follows.
L2R PDU's are packed into a frame in accordance with the RLP protocol, which is shown in Figure 4. The RLP protocol is defined in GSM
Reco.~mendation 04.22. The RLP is a balanced (HDLC type) data transmission protocol with a frame structure, in which error correction is based on re-transmission of defective frames on request of the receiving party. The RLP extends from t.he terminal adaptation function TAF
of the MS to the interworking function IWF, which is usually placed in the MSC. In accordance with Eigure 4, the RLP frame structure contains a header field (16 bits), an information field ~200 bits) and a frame check sequence (24 bits). A 200-bit L2R PDU is packed into the information field. The net RLP data rate is thus distinctively above the highest possible user rate g.6 kbi.t~s allowed for one channel, which enables a certain r.umber of re-transmissions without reducing the nomir.al user rate. Wherl the user rate. in the terminal interface is 9600 kbit/s, for instance, and the data rate in the radio interface is 12 kbit/sr this "surplus capacity" is about 12.5 %.
The RLP protocol is also conr.ected with flow control, which is used for controlling the fill level of the receptior. and transmission buffers. This flow control is described e.g. in GSM Recommendation 0~.02.
The L2R~RLP urit oE the receiving party B immediately transmits a "flow control active" indication to the transmitting party A if the reception buffer seen from the direction of the radio path reaches a predetermined threshold value or if the terminal equipment TE has initiated local flow control at the terminal interface.
Wo96/36146 ,_I/r~
2 ~ ~3~ 1 7 In the latter case, upon receiving the flow control indication from the terminal inter~ace, L2R/RLP
interrupts data transmission from the reception buffer to the terminal interface. When the congesti~n of the buffer or local flow control is removed, L2RiRLP sends a "flow control inactive" indication. In addition, data transmission from the reception buffer is restarted.
The L2R~RL~ unit of the transmitting party A activates local flow control immediately wher the transmission buffer reaches a predeterrmined threshold value or when L2R/R1P receives the "flow control active" indication from the receiving party B. When the congestion of the buffer is removed, or when the "flow control inactive"
indication is received, L2RiRLP will cancel local flow control. The transmission buffer buffers data received from the V.24 interface so that no data will be lost in case the mobile station MS is not able to transmit data immediately over the radio path. The reception bu~fer buffers data, which is transmitted to the V.24 interface so that no data received from the traffic channel is lost in case the data cannot be forwarded immediataly via the V.24 interface e.g. to the terminal equipment TF. Said predetermined threshold value that activates the flow control can be e.g. a half-full transmission or reception buffer.
As it was stated above, the RLP protoco1 is attended by the problem that when the quality of the connection becomes poorer, the number of defective and lost frames, and thus the number of re-transmissions increases. On extremely poor connection.s, buffering of the RLP frames, as well as flow control described above must be employed. In practice this means that the actual use rate is lower than the nominal u.ser data rate, that is, the throughput of the connection decreases.
WO96~6146 .. .........................................r ~ .,r~
;~ 35 In accordance with a preferred embodiment of the inventior., it has been endeavoured to solve this problem so that the 12R~RLP unit of the transmittins party monitors the quality of the connectior. by reducing the throughput of the connection, and advantageously also by monitoring th~e fill level of the transmission buffer. The number of the re-transmissior.s of the RLP frames is not limited either, unlike in the RLP protocol in accordance with the GSM
recommendations, so that resetting of the RLP
connection is avoided. As stated above, resetting takes place in the present GSM system when the number of re-transmissions reaches the maximum value.
An embodiment will be described herein by wa;
of example, in which the transmitting party monitors the quality of the connectior., changes the channel coding scheme and requests additional capacity. These procedures, however, may alternatively be carried out by the receiving end according to the same principles.
In such a case, either the transmission buffer is nct monitored, or the fill level of the transmission buffer is concluded from the "flow control active" indicatior.
obtained from the transmitting end.
If the fill level of the transmission buffe-reaches the predetermined value and the throughp~t drops to the predetermined value, the chanr.el codina scheme FEC of the connection is changed for a more efficient one, and more channel capacity is requested.
Due to a better channel coding scheme EEC the residual error rate that remains to be corrected by the RLP
protocol is lower than prior to changing the channel coding scheme. Since radio channel capacity has possibly beer increased to meet the needs of the new channel coding scheme, the nominal data rate can be maintained uncr.anged or nearly unchanged.
WO 96~36146 . ~ rL - ~ 7 35 l t If there is no additional channel capacity available, however, L2R/RLP continues transmitting with the present transmission configuration. A L2R~RLP unit contains a retry timer, however, which is activated at this stage. If the quality of the connection i5 stili the same after the timer has expired, a new attempt is made for increasing the channel capacity, and in case this new attempt fails, the timer is activated again etc. If the throughput of the connectior. f~rther decreases to a lower predetermined value during this procedure, the channel coding scheme is changed to a more efficient one in ary case without additional channel capacity in order to improve error correction.
It is thus po.ssible to maintain the connection, even i~
with a lower user rate. ~ sensible lower value predetermined for the throughput is some value beiow the following higher nominal data rate of the channel coding scheme FEC.
If the radio conditions improve agair., additional channel resources may be returned and the channel coding scheme may be changed back to the original. The decision on returning the origina' configuration may be based Cll a predetermineo throughput value.
The channel coding scheme is changed by means of a Channel Mode Modify (CMM) procedure, which lS
described e.g. in GSM Recommendation 04.0R, pp. 53 - 54 and 181 - 182. Both channel allocation and the CMM
procedure are controlled in the preferred embodiment of the inver.tion by the MSC, in particular its call control CC ~42).
9 3 P l/rl - ~ -1.O D~i~;on criterion for requesting channel ,a~,.. ~r ~ pon monitoring the fill level of the transmission buffer, the activating threshold of the flow control described above is used for activating the request state. In that case, hysteresis is used in order to prevent the flow control from deactivating the request state, that is, once the request state has been activated, it is deactivated only after the fill level of the buffer has decreased to the lower threshold value.
Monitoring of the throughput is based on calculating the re-transmission rate of the frames.
Since the transmission buffer is half-full, the RLP
continuously transmits at the full rate. This rate is known on the basis of the channel capacity, e.g. in the case of one full-rate GSM traffic channel the rate is one frame~20 ms ~50 frames~second). If, for instance, 55 ~ of the last transmitted frames are re-transmissions, 45 % of the capacity will be used for new frames.
Monitoring the throughput may also be based on calculating the "sliding or floating" of the success ratio e.g. with an equation:
2S AVE(n+1) = AVE(r)*(1-D)fMEAS*D
where AVE(i) is an average value at instant i, D is a "history co-efficient" (0 < D < 1), MEAS is the last "measurement", MEAS = 1 represents a successfully acknowledged transmission, MEAS = 0 represents re-transmission of a frame, 0 ~ AVE~i) ~ 1.
W0 96/3C146 ~ ~ ~ ' ' r~ r~
2~93~ ~
2.0 ~icion criterion for r~1o-~ing channel ,.~,,., ., _.
Additional channel resources are released wherl user data transmission has continued at a full rate and with an averaqe success ratio AVE(i) which is greater than the predetermined value ~e.g. 0.8~ for a predetermined period of time, e.g. 10 seconds or 100 seconds. This criterion maintains the additional allocated channel resources even ir, a case where the terminal equipment has temporarily stopped transmitting data. A pendulum between releasing channel resources and requesting channel resources is thus avoided, and channel resources are only released after actual improvement of the radio conditions.
The present invention may be applied both to single-channel service and multichannel high-speed circuit switched data service (HSCSD). In the following, both cases will be described by means of examples.
3.0 Application to a sing1e-channel service, based on a half-rate channel Ir. the GSM system, it is possible to choose a so-called half-rate (max. ~.8 kbit~s) traffir. channel (H~ for low rates. A half-rate traffic channel is established when a mobile station MS operates ir a certain time-slot only in every second frame, i.e. at the half rate. There is another mobile station operating in every second frame of the same time-slot.
Let us assume that the mobile station Ms has a non-transparent circuit switched connection on a half-rate GSM traffic channel. Data transmission in accordance with the invention will be disclosed in the following assuming that the terminal adaptation function TAF is the transmitting party A and the interworking function IWF is the receiving party S. It Wo 96/36146 L i 9 3 5 1 1 1~ l/r~. ~
~ j must be noted, howevex, that data transmission takes place in the same way in the opposite directior, IWF-TAF, as well.
The L2R~RLP unit of the transmitting party A
continuously monitors the fill level of the transmission buffer and the throughput of the connection. When the quality of the connection becomes poorer and the number of error increases, the following events take place:
1) the number of re-transmissions of the RLP
frames increases, 2~ the fill level of the t.ransmission buffer reaches the flow control activation level, whereby the flow control is activated, 3) the number of the re-transmissions of single frames reaches a predetermined maximum value, but,in accordance with the invention, this is disregarded, 4) The throughput of the connection drops to a decision value, which is a value between the original nominal user rate and a half of the original nominal user rate.
When conditions 2 and 4 are both valid, the L2R~RLP unit of the trarsmitting party A requests a change of the channel coding scheme by means of a CMM
procedure, and changing the traffic channel to a full-rate traffic channel in order to maintain the original nominal data rate. After a successful CMM procedure and allocation of a full-rate traffic channel, data transmission continues at the origir.al data rate, but with a better channel coding scheme FEC.
When the channel coding scheme has been changed for a more efficient one, the L2R~RLP unit of the transmitting party A continuously monitors th.e throughput of the connection for returning the original WO 96/36146 ~ /r~ '~ :, 3~ ~ 1 channel coding scheme and channel resources if the radio conditions change.
Example 1. Communication takes place on a half-rate GSM traffic channel on which the original user rate is 4.8 kbit~s and the radio interface rate is 6 kbit/s. ~hen the throughput of the connection drops below the decision value, e.g. to 4 kbitJs, the iill level of the transmission buffer reaches the flow control activation level, and the flow control is activated. The L2RJRLP unit then requests the CMM
procedure for changing for a full-rate traffic channel having a user rate of 4.8 kbit/s and the radio interface rate of 12 kbitJs. Data transmissior, thus continues at the original user rate, but wlth a more efficlent channel coding scheme FEC, as the convolution code ratio of a full-rate channel is l/3 with user rate 4. a kbitJs, whereas that of a half-rate channel is 1~2.
If the requested full-rate chanrlel is not available, data transmission continues on the existir;c, half-rate traffic channel. The L2RJRLp activates the retry timer. If the present conditions are still valid after the timer has expired, a new atterr.pt is made for increasing the channel capacity, and, if the new attenpt fails, th.e timer is re-activated et.c. If the Z5 throughput of the connection still drops to the lowest predetermined decision value, the channel coding scheme of a half-rate traffic channel is changed for a more efficient one with the CMM procedure. A sensible value for said lower predetermined decision ~alue is below a half of the original rominal user rate.
Example 2. Data transmission takes place on a half-rate traffic channel on which the original user rate i5 4.8 kbit~s, the radio interface data rate i.s 6 kbit/s and ~F.c=lJ2. ~hen the quality of the cor.nectior.
becomes poorer, the throughput drops below 2.4 kbit~s, W096136146 1_l/ri1~
2 1 ~ 3 5 1 1 ., ,. ~ i. ~, lg and the fill level of the transmission buffer reaches the flow control activation level, and the flow control is activated. Simultaneously, the number of the re-transmissions of si.ngle frames may reach the predetermined maximum value, but this is disregarded in accordance with the invention. The L2R~R1P unit requests a CMM procedure for changing the traffic chanrel for a configuration having a user rate of 2.4 kbit/s and the radio interface rate of 3.6 kbitis.
Thereafter, data transmission takes place on a half-rate traffic channel at a lower nominal data rate 2.4 kbitls, but, correspondingly with a more efficient channel coding scheme FEC=l~3.
4.0 Application to a single-channel service based on a full-rate channel Let us assume that a mobile station MS has an ongoing non-transparent circuit switched data connection on a full-rate traffic channel. Th.e L2R~RLP
unit of the transmitting party ~i~ continuously monitors the fill level of the transmission buffer, as well as the throughput of the connection. When the quality of the connection becomes poorer and the.number of errors decreases, the following events take place:
l) the number of re-transmissions of the RLP
frames increases, 2) the fill l.evel of the transmission buffer reaches the flow control activation level, and the flow control is activated, 3) the number of the re-transmissions of single frames reaches a predetermined maximum value, but, in accordance with the invention, this is disregarded, 4) The throughput of the connection drops to a decision value, which is below a half of the original nominal user rate.
WO96~6146 ~ !j , r~
; ~ _ ~ i q35 1 i When conditions 2 and 4 are both valid, the L2R/RLP unit of the transmitting party A requests a change of the channel coding scheme by means of a C~M
procedure. After a successful CMM procedure, data transmission continues on th.e same single traffic channel at a lower nominal data rate ~half of the original~, but with a better channel coding scheme FEC.
After charging for a more efficient channel coding scheme, the ~2R/RLP unit continuously monitors:
- fill level of the transmissior buffer as well as the throughput of the connect.ion for detecting a possible need to change for an even better channel codinq scheme, - the throughput of the connection for 15 returning the faster transmission rate if the radio conditions change.
This can be illustrated with example 3. The o.riginal user rate of a full-rat.e channel is 9.6 kbit/s, the radio interface rate is 17 kbit~s, and 20 FEC=1~2. The thloughput of the connection drops below 4.8 kbitis and the fill level oE the transmission buffer reaches the flow control activation level, and flow control is activated. Thus, the L2R/RLP unit requests a CMM procedure, which changes the user rate 25 to 4.8 kbit/s and the radio interface rate to 12 kbit~s. Thereafter, data transmission continues at a lower nominal user rate, but with a more efficient channel coding FEC=l/3.
This releases channel capacity for other connections.
Brief Description of the ~r~- n~c In the following, the invention will be disclosed by means of the preferred embodiments with reference tc the attached drawinss, in which Figure l illustrates a part of a mobile communications system in which the invention may be applied on a single-channel non-transparent connection, Figure 2 is a block diagram showins the functional uni.s of a single-channel non-transparer,t ~S~ traffic channel in different protocol layers, Figure 3 shows a L2R PDU, Figure ~ an RLP frame, Figure S illustrates a part of the G.'~ mc,bile communications system in which the invertion may be applied on a m~ltichannel nor.-transparent connectior P ~f~r ed r - ts of the Invention The present invention may be used in all digital data transmission systems on a non-transparent data connection on which. transmission errors are corrected, in addition to channel codin5, by means of a transmission protocol that re-transmits the frames WO96136146 l~lr~ l ~ ~ ~ 9 3 ~ I ~
~ 7 and causes the actual data rate to drop below the nominal data rate or the connection to be interrupted when the quality of the connection becomes poorer.
The present ir.vention is suitable particularly for data transmission applications in digital CDMA or TDMA mobile communications systems, such as the pan-European digital mobile communications system GSM, DCSl800 (Digital Communication System), a mobile communications sy.stem ir accordance with EIA/TIA
Interim Standard IS/41.3, etc. The invention will be disclosed below by way of example of the GSM mobile communications system, but the invention is not limited thereto, however. The basic structure of the GSM system will be disclosed briefly below, but no closer attention will be paid to its features or other aspects of the system. The GSM system is described in more detail in "The GSM System for Mobile Communications", M. Mouly, M. Pautet, Palaiseau, France, 1992, ISBN:2-9507190-0-7, which ls ir,corporated herein by reference.
A mobile services .switching centre MSC
switches incoming and outgoing calls. It carries out tasks similar to the exchange of the public switched telephone network ~PST~. ln addition, it also carries out functions typical of mobile telecommunication only, such as location management of the subscribers in cooperation with the subscriber registers ~not shown) of the network. Mobile stations (MS) are connected to the MSC via base statlon systems BSS. A base station system BSS consists of base station controllers BSC and base stations BTS.
The GSM system is a time division multiple access system in which commur.ication on the radio path takes place in successlve TD~ frames, each of wh.ich consists of several time-slots. Ir. each time-slot a short information package is sent in form of a radio ~096~6146 l~l/r.,L,:
~3~
~ i ~3~ ~ I
frequency burst having a finite duration and consisting of a group of modulated bits. The time-slots are mainly used for transm~tting control char.nels and traffic channels. On traffic channels, speech or data is transmitted. On control channels, signalling is carried out between the base station and the mobile statlon.
The channel structures used in the radio interface of the GSM system are defined ln closer detail ir. the GSM
recommendation 05.02. In the operation in accordance lG with the recommendation, one time-slot is assigned to a mobile station MS as a traffic channel at the beginning of the call (Single Slot Access~. The mobile station MS is synchronized with this time-slot for transmitting and receiving radio frequency bursts.
In the GSM system, a data connection is established between a termir.al adaptation function (TAF~ 31 of the mobile station MS and an interworking function (IT~F) 42 of a fixed network. (usually associated with the MSC). A data cormectior; is a circuit switched connection, which reserves one (or more) traffic channels from the radio interface for the duration of entire call. In the data transmission of the GSM network, the data conrection is a V.110 rate-adapted, V-24 interface-adaptable, ~DI coded digital 2S full duplex connection. A V.110 connection is a digital transmission channel that has originally been developed for ISDN technology (integrated Services Data Network), which is defined in CCITT Recommendation Blue Book V.110. A terminal adaptation function TAF adapts a data terminal TE connected to the mobile station MS to the V. 110 conrection, which i5 established in Figure 1 over a circuit switched conr.ection that employs one traffic channel ci-0. The interworking function IWF adapts the V.110 connection to some other V.110 network, suc.h as ISDN or another GSM network, or to some other tran.sit WO96/36146 P_llr~ ~ , ~ 1 9 3 ~ 1 ~
network, such as the public switched telephone network PSTN.
In addition, channel coding FEC (Forward Error Correction) is employed on the traffic channel for reducing the effect of transmission errors on the radio path. In the GSM system corvolution coding in accordance with GSM recommendation 05.03 is employed, and its efficiency can be expressed by means of a convolution coding ratio, which indicates that in channel coding X data bits are represented with Y code bits. On a full-rate GSM traffic channel, for instance, the convolution code ratios corresponding to the user data rates 9.6 kbit~s, 4.3 kbit~s, and 2.4 kbit~s are 1~2 ~buffered), 1~3 and 1~, respectively.
The non-transparent circuit switched connection between the terminal adaptation function ~TAF) and the interworking function ~IWF) on a GSM
traffic channel comprises several protocol layers.
The terminal interface between the terminal adaptation function TAF of the mobile station MS and the data terminal equipment, as well as the interface between the interworking function IWF and e.g. an audio modem MODEM comply with CCITT V.24, and it is marked with a symbol I.2 ir the figure. erotocols that are essential to the invention are L2R ~Layer 2 Relay~ and RLP ~Radio Link Protocol), which are both located between the terminal adaptation function TAF and the interworking function IWF at both ends of tne connection. In accordance with the invention, the connection also includes several rate adaptation RA
functions, such as RAl between the terr;,inal adaptation function TAF and a CCU unit (Channel Codec Unit) placed in the BSS, RAI between the CCU unit and the interworking function IWF, RAA betweer the CCU unit and a transcoder unit TRAII placed separate from the base =
WO 96136146 P~, J I r ~
3 ~ 1 ~
station, and RA2 bet.ween the transcoder unit TRAU and the interworking function I~F. The rate adaptatio functlons RA are defined in GSM specificatior,s 04.21 and oa.20. Communication between the CCU unit and the transcoder unit TRAU is defined in GSM recommendation 08.60.
Information that has been rate adapted at the radio interface RA1 has also been channel coded ir, the way specified in GSM Recommendation 5.03. This is illustrated by blocks FEC in the mobile station MS and in the CCU unit.
The object of the invention is, however, onl, the L2R~RLP operation of the terminal adaptation function TAF and the interworking adaptation functior, IhTF and the communication between them. The other lower-layer protocols, function,s and units mentioned above only provide a transmission path in accordance with the GSM recommendations between the L2R~RLP units, and they are not relevant to the invention except for channel coding. Therefore, they are not paid closer attention to herein.
L2R (layer 2 Relaying) functionality for non-transparent character-oriented protocols is defined e.g. in GSM Recommendation 07.02. L2R packs the user data and status information received from the terminal interface in protocol data units PDU having a length of 200 bits, or 25 octets. A PDU is illustrated in Figure 3. The octets are numbered from 0 to 24, whereby octet 0 is transmitted first. The bits in octets are numbered from l to 8, and bit 1 is transmitted first. An octet in a PDU may be a status octet, a character ~higher layer data~, or dummy bits. Octet 0 is always a status octet. One status octet contains 3 bits: SA, SB, and Y~.
for the status of the V.24 connection, and 5 bits for indicating the number of the data octets subsequent to WO96i36l46 ~l/rL6~ , '' ' ~' '' :L 1 the status octet, as well as special indications, such as an empty PDU. In Figure 3, status octet 0 is followed by 3 data octet.s in which the word "GSM" is packed, whereafter a r,ew status octet 4 follows.
L2R PDU's are packed into a frame in accordance with the RLP protocol, which is shown in Figure 4. The RLP protocol is defined in GSM
Reco.~mendation 04.22. The RLP is a balanced (HDLC type) data transmission protocol with a frame structure, in which error correction is based on re-transmission of defective frames on request of the receiving party. The RLP extends from t.he terminal adaptation function TAF
of the MS to the interworking function IWF, which is usually placed in the MSC. In accordance with Eigure 4, the RLP frame structure contains a header field (16 bits), an information field ~200 bits) and a frame check sequence (24 bits). A 200-bit L2R PDU is packed into the information field. The net RLP data rate is thus distinctively above the highest possible user rate g.6 kbi.t~s allowed for one channel, which enables a certain r.umber of re-transmissions without reducing the nomir.al user rate. Wherl the user rate. in the terminal interface is 9600 kbit/s, for instance, and the data rate in the radio interface is 12 kbit/sr this "surplus capacity" is about 12.5 %.
The RLP protocol is also conr.ected with flow control, which is used for controlling the fill level of the receptior. and transmission buffers. This flow control is described e.g. in GSM Recommendation 0~.02.
The L2R~RLP urit oE the receiving party B immediately transmits a "flow control active" indication to the transmitting party A if the reception buffer seen from the direction of the radio path reaches a predetermined threshold value or if the terminal equipment TE has initiated local flow control at the terminal interface.
Wo96/36146 ,_I/r~
2 ~ ~3~ 1 7 In the latter case, upon receiving the flow control indication from the terminal inter~ace, L2R/RLP
interrupts data transmission from the reception buffer to the terminal interface. When the congesti~n of the buffer or local flow control is removed, L2RiRLP sends a "flow control inactive" indication. In addition, data transmission from the reception buffer is restarted.
The L2R~RL~ unit of the transmitting party A activates local flow control immediately wher the transmission buffer reaches a predeterrmined threshold value or when L2R/R1P receives the "flow control active" indication from the receiving party B. When the congestion of the buffer is removed, or when the "flow control inactive"
indication is received, L2RiRLP will cancel local flow control. The transmission buffer buffers data received from the V.24 interface so that no data will be lost in case the mobile station MS is not able to transmit data immediately over the radio path. The reception bu~fer buffers data, which is transmitted to the V.24 interface so that no data received from the traffic channel is lost in case the data cannot be forwarded immediataly via the V.24 interface e.g. to the terminal equipment TF. Said predetermined threshold value that activates the flow control can be e.g. a half-full transmission or reception buffer.
As it was stated above, the RLP protoco1 is attended by the problem that when the quality of the connection becomes poorer, the number of defective and lost frames, and thus the number of re-transmissions increases. On extremely poor connection.s, buffering of the RLP frames, as well as flow control described above must be employed. In practice this means that the actual use rate is lower than the nominal u.ser data rate, that is, the throughput of the connection decreases.
WO96~6146 .. .........................................r ~ .,r~
;~ 35 In accordance with a preferred embodiment of the inventior., it has been endeavoured to solve this problem so that the 12R~RLP unit of the transmittins party monitors the quality of the connectior. by reducing the throughput of the connection, and advantageously also by monitoring th~e fill level of the transmission buffer. The number of the re-transmissior.s of the RLP frames is not limited either, unlike in the RLP protocol in accordance with the GSM
recommendations, so that resetting of the RLP
connection is avoided. As stated above, resetting takes place in the present GSM system when the number of re-transmissions reaches the maximum value.
An embodiment will be described herein by wa;
of example, in which the transmitting party monitors the quality of the connectior., changes the channel coding scheme and requests additional capacity. These procedures, however, may alternatively be carried out by the receiving end according to the same principles.
In such a case, either the transmission buffer is nct monitored, or the fill level of the transmission buffer is concluded from the "flow control active" indicatior.
obtained from the transmitting end.
If the fill level of the transmission buffe-reaches the predetermined value and the throughp~t drops to the predetermined value, the chanr.el codina scheme FEC of the connection is changed for a more efficient one, and more channel capacity is requested.
Due to a better channel coding scheme EEC the residual error rate that remains to be corrected by the RLP
protocol is lower than prior to changing the channel coding scheme. Since radio channel capacity has possibly beer increased to meet the needs of the new channel coding scheme, the nominal data rate can be maintained uncr.anged or nearly unchanged.
WO 96~36146 . ~ rL - ~ 7 35 l t If there is no additional channel capacity available, however, L2R/RLP continues transmitting with the present transmission configuration. A L2R~RLP unit contains a retry timer, however, which is activated at this stage. If the quality of the connection i5 stili the same after the timer has expired, a new attempt is made for increasing the channel capacity, and in case this new attempt fails, the timer is activated again etc. If the throughput of the connectior. f~rther decreases to a lower predetermined value during this procedure, the channel coding scheme is changed to a more efficient one in ary case without additional channel capacity in order to improve error correction.
It is thus po.ssible to maintain the connection, even i~
with a lower user rate. ~ sensible lower value predetermined for the throughput is some value beiow the following higher nominal data rate of the channel coding scheme FEC.
If the radio conditions improve agair., additional channel resources may be returned and the channel coding scheme may be changed back to the original. The decision on returning the origina' configuration may be based Cll a predetermineo throughput value.
The channel coding scheme is changed by means of a Channel Mode Modify (CMM) procedure, which lS
described e.g. in GSM Recommendation 04.0R, pp. 53 - 54 and 181 - 182. Both channel allocation and the CMM
procedure are controlled in the preferred embodiment of the inver.tion by the MSC, in particular its call control CC ~42).
9 3 P l/rl - ~ -1.O D~i~;on criterion for requesting channel ,a~,.. ~r ~ pon monitoring the fill level of the transmission buffer, the activating threshold of the flow control described above is used for activating the request state. In that case, hysteresis is used in order to prevent the flow control from deactivating the request state, that is, once the request state has been activated, it is deactivated only after the fill level of the buffer has decreased to the lower threshold value.
Monitoring of the throughput is based on calculating the re-transmission rate of the frames.
Since the transmission buffer is half-full, the RLP
continuously transmits at the full rate. This rate is known on the basis of the channel capacity, e.g. in the case of one full-rate GSM traffic channel the rate is one frame~20 ms ~50 frames~second). If, for instance, 55 ~ of the last transmitted frames are re-transmissions, 45 % of the capacity will be used for new frames.
Monitoring the throughput may also be based on calculating the "sliding or floating" of the success ratio e.g. with an equation:
2S AVE(n+1) = AVE(r)*(1-D)fMEAS*D
where AVE(i) is an average value at instant i, D is a "history co-efficient" (0 < D < 1), MEAS is the last "measurement", MEAS = 1 represents a successfully acknowledged transmission, MEAS = 0 represents re-transmission of a frame, 0 ~ AVE~i) ~ 1.
W0 96/3C146 ~ ~ ~ ' ' r~ r~
2~93~ ~
2.0 ~icion criterion for r~1o-~ing channel ,.~,,., ., _.
Additional channel resources are released wherl user data transmission has continued at a full rate and with an averaqe success ratio AVE(i) which is greater than the predetermined value ~e.g. 0.8~ for a predetermined period of time, e.g. 10 seconds or 100 seconds. This criterion maintains the additional allocated channel resources even ir, a case where the terminal equipment has temporarily stopped transmitting data. A pendulum between releasing channel resources and requesting channel resources is thus avoided, and channel resources are only released after actual improvement of the radio conditions.
The present invention may be applied both to single-channel service and multichannel high-speed circuit switched data service (HSCSD). In the following, both cases will be described by means of examples.
3.0 Application to a sing1e-channel service, based on a half-rate channel Ir. the GSM system, it is possible to choose a so-called half-rate (max. ~.8 kbit~s) traffir. channel (H~ for low rates. A half-rate traffic channel is established when a mobile station MS operates ir a certain time-slot only in every second frame, i.e. at the half rate. There is another mobile station operating in every second frame of the same time-slot.
Let us assume that the mobile station Ms has a non-transparent circuit switched connection on a half-rate GSM traffic channel. Data transmission in accordance with the invention will be disclosed in the following assuming that the terminal adaptation function TAF is the transmitting party A and the interworking function IWF is the receiving party S. It Wo 96/36146 L i 9 3 5 1 1 1~ l/r~. ~
~ j must be noted, howevex, that data transmission takes place in the same way in the opposite directior, IWF-TAF, as well.
The L2R~RLP unit of the transmitting party A
continuously monitors the fill level of the transmission buffer and the throughput of the connection. When the quality of the connection becomes poorer and the number of error increases, the following events take place:
1) the number of re-transmissions of the RLP
frames increases, 2~ the fill level of the t.ransmission buffer reaches the flow control activation level, whereby the flow control is activated, 3) the number of the re-transmissions of single frames reaches a predetermined maximum value, but,in accordance with the invention, this is disregarded, 4) The throughput of the connection drops to a decision value, which is a value between the original nominal user rate and a half of the original nominal user rate.
When conditions 2 and 4 are both valid, the L2R~RLP unit of the trarsmitting party A requests a change of the channel coding scheme by means of a CMM
procedure, and changing the traffic channel to a full-rate traffic channel in order to maintain the original nominal data rate. After a successful CMM procedure and allocation of a full-rate traffic channel, data transmission continues at the origir.al data rate, but with a better channel coding scheme FEC.
When the channel coding scheme has been changed for a more efficient one, the L2R~RLP unit of the transmitting party A continuously monitors th.e throughput of the connection for returning the original WO 96/36146 ~ /r~ '~ :, 3~ ~ 1 channel coding scheme and channel resources if the radio conditions change.
Example 1. Communication takes place on a half-rate GSM traffic channel on which the original user rate is 4.8 kbit~s and the radio interface rate is 6 kbit/s. ~hen the throughput of the connection drops below the decision value, e.g. to 4 kbitJs, the iill level of the transmission buffer reaches the flow control activation level, and the flow control is activated. The L2RJRLP unit then requests the CMM
procedure for changing for a full-rate traffic channel having a user rate of 4.8 kbit/s and the radio interface rate of 12 kbitJs. Data transmissior, thus continues at the original user rate, but wlth a more efficlent channel coding scheme FEC, as the convolution code ratio of a full-rate channel is l/3 with user rate 4. a kbitJs, whereas that of a half-rate channel is 1~2.
If the requested full-rate chanrlel is not available, data transmission continues on the existir;c, half-rate traffic channel. The L2RJRLp activates the retry timer. If the present conditions are still valid after the timer has expired, a new atterr.pt is made for increasing the channel capacity, and, if the new attenpt fails, th.e timer is re-activated et.c. If the Z5 throughput of the connection still drops to the lowest predetermined decision value, the channel coding scheme of a half-rate traffic channel is changed for a more efficient one with the CMM procedure. A sensible value for said lower predetermined decision ~alue is below a half of the original rominal user rate.
Example 2. Data transmission takes place on a half-rate traffic channel on which the original user rate i5 4.8 kbit~s, the radio interface data rate i.s 6 kbit/s and ~F.c=lJ2. ~hen the quality of the cor.nectior.
becomes poorer, the throughput drops below 2.4 kbit~s, W096136146 1_l/ri1~
2 1 ~ 3 5 1 1 ., ,. ~ i. ~, lg and the fill level of the transmission buffer reaches the flow control activation level, and the flow control is activated. Simultaneously, the number of the re-transmissions of si.ngle frames may reach the predetermined maximum value, but this is disregarded in accordance with the invention. The L2R~R1P unit requests a CMM procedure for changing the traffic chanrel for a configuration having a user rate of 2.4 kbit/s and the radio interface rate of 3.6 kbitis.
Thereafter, data transmission takes place on a half-rate traffic channel at a lower nominal data rate 2.4 kbitls, but, correspondingly with a more efficient channel coding scheme FEC=l~3.
4.0 Application to a single-channel service based on a full-rate channel Let us assume that a mobile station MS has an ongoing non-transparent circuit switched data connection on a full-rate traffic channel. Th.e L2R~RLP
unit of the transmitting party ~i~ continuously monitors the fill level of the transmission buffer, as well as the throughput of the connection. When the quality of the connection becomes poorer and the.number of errors decreases, the following events take place:
l) the number of re-transmissions of the RLP
frames increases, 2) the fill l.evel of the transmission buffer reaches the flow control activation level, and the flow control is activated, 3) the number of the re-transmissions of single frames reaches a predetermined maximum value, but, in accordance with the invention, this is disregarded, 4) The throughput of the connection drops to a decision value, which is below a half of the original nominal user rate.
WO96~6146 ~ !j , r~
; ~ _ ~ i q35 1 i When conditions 2 and 4 are both valid, the L2R/RLP unit of the transmitting party A requests a change of the channel coding scheme by means of a C~M
procedure. After a successful CMM procedure, data transmission continues on th.e same single traffic channel at a lower nominal data rate ~half of the original~, but with a better channel coding scheme FEC.
After charging for a more efficient channel coding scheme, the ~2R/RLP unit continuously monitors:
- fill level of the transmissior buffer as well as the throughput of the connect.ion for detecting a possible need to change for an even better channel codinq scheme, - the throughput of the connection for 15 returning the faster transmission rate if the radio conditions change.
This can be illustrated with example 3. The o.riginal user rate of a full-rat.e channel is 9.6 kbit/s, the radio interface rate is 17 kbit~s, and 20 FEC=1~2. The thloughput of the connection drops below 4.8 kbitis and the fill level oE the transmission buffer reaches the flow control activation level, and flow control is activated. Thus, the L2R/RLP unit requests a CMM procedure, which changes the user rate 25 to 4.8 kbit/s and the radio interface rate to 12 kbit~s. Thereafter, data transmission continues at a lower nominal user rate, but with a more efficient channel coding FEC=l/3.
5.0 Application to a mul~ n~ol hiyl. ~ed 30 data ~ r~-n~ j~n The highest allowable user data rate on one GS~I traffic channel is limited to 9.6 kbit/s.
The pre~ious applicatior.s FI-942,150 and FI-945,817 of the applicant disclose a procedure in which 35 two or more time-slots are allocated ~rom the same TDMA
W096136146 2 ~ 935 ~ 1 P~l/r~L~ .
~ r;~
~,, ! ~ ., .~
frame to a mobile station MS that requires data transmission at a higher transfer rate than what can be provided on one traffic channel. The maximum user rate on a multichannel data connection is the number of parallel traffic channels x the user rate 9.6 kbit/s of one traffic channel. Thus, e.g. a user rate of lg.2 kbit~s can be provided by two traffic channels at the minimum. This procedure is disclosed in this application as an example of carrying out high-speed data transmission based on a plurality of parallel traffic channels in a radio system. As to the closer details of this procedure, reference is made to the above-mentioned applications. It must be noted, however, that the only thing relevant to the invention is the possibility to establish a multichannel transmission connection, and the object of the invention is thus primarily to change the efficiency of the channel capacity and the channel coding scheme when the quality of the connection varies.
Figure 5 illustrates the architecture of the GSM network implementing this kind of data transmission employing a plurality of parallel traffic channels.
Figure 5 is similar to Figure l except for the fact that in Figure 5, between the terminal adaptation function TAF and the interworking function IWF there is a circuit switched non-transparent connection consisting of N parallel traffic channels chO-chn, wherein N=l, 2, 3, ... In the mobile station, the TAF
acts as a divider, distributing the incoming high-speed data signal DATA IN from the data terminal equipment into parallel traffic channels chO-chn, as well as a corr,biner, which combines the low-speed subsignals received from the parallel traffic channels chO-chn again with the high-speed data signal DATA OUT.
Correspondingly, at the second end of the multichannel W096~361~ ~ r~
22 ~3 ) 9 3S ~ t data connection, the IWF acts as a distributor distributing the incoming high-speed data signal DATA
I~ to parallel traffic chanr.els chO-chn, and as a combiner, which combines the low-speed data signals received from the parallel traffic ch.annels again with a high-speed data signal DATA OUT.
The protocol structure shown in Figure 2 also holds for multichannel connection architecture shown in Figure 5. In the preferred embodiment of the invention, the L2R~RLP unit is common to al1 traffic channels allocated for the same connection, also acting as the distributor and combiner mentioned above. Each traffic channel has, however, rate adaptation ~RA~ and channel coding (FEC) functions of their own in accordance with Figure 2. A multichannel data conr.ection is thus substantially similar to the single-channel data connection from the point of view oE the L2R~RLe unit.
The "transmission channel" between the L2R~RLP units only has a larger capacity than before.
2~ In the following, the operation of the L2P.~P~LP
unit will be disclosed in more detail. A mobile static,n has an ongoing non-transparent circuit switched data connection comprising one or more traffic channels in a system that is capable of high-speed multichannel data transmission. The L2R~RLP unit of the transmicting party A continuously monitors the fill level of the transmission buffer and the throughput of the connection. When the quality of the connection becomes poorer and the number of errors increases, the following events take place:
l) the number of re-transmissions of the RLP
frames increases, 2~ the fill level of the transmission buffer reaches the flow control activation level, and the flow control is activated, W096/36l46 ~ ~ 93S ~ ~ E~l~r~
,~" ~. ~
. ~ t' Ir~ t 2 3 3) the number of the re-transmissions of single frames reaches a predetermined maximum value, but, in accordance with the invention, this is disregarded, 4) The th~roughput of the connection drops to a decision value. The decision value may be some value between the current nominal user rate and a half of the current nominal user rate as it is possible to request more channel capacity in order to compensate the 10 decreased data rate on a single traffic channel due to changing for a more efficient channel coding scheme.
~hen both 2~ and 4) are valid, the L2R/RLP
unit of the transmitting party A requests increasing of the channel capacity and charging for a more efficient 15 channel coding scheme by means of a CMM procedure. If e.g. one traffic channel is currently in use, a secor.d traffic channel is requested from the same carrier wave. After a successful channel allocation and CM!~
procedLIre data transmission continues at the same or 20 almost the same data rate as originally, but with a better channel coding scheme FEC.
Additional channel capacity may be requestes depending on the original channel capacity e.g. in the following way:
woscl36l46 .~f~ /rl.~
2 ~ 93~ ~ 1 original capacity requested additional capacity n The decision val~e of the throughput is different in cases in which it is not possible to double the channel capacity, since the original nominal user rate cannot be reached after the CMM proçedure and changing the channel coding scheme. As a rule, it car.
be stated that the lower the percentage of the additional traffic channels, the lower the decisior.
value for the throughput is by means of which the change of the channel coding scheme and additional channel capacity are requested.
After changing for a more efficient channel coding scheme, the L2RfRLP unit of the transmittlng party A continuously monitors:
- the fill level of the transmission buffer and the throughput of the connection for detecting a possible need for changing for a more efficient channet coding scheme and requesting additional capacity, - the throughput of the connectior. fo detecting a possibility to restore the higher speec~
channels and release the additional channel resources if the radio conditions improve.
This can be illustrated with example 4. Let us -WO9C/3614C ~ ~ 935 ~ llr~
assume that a non-tran-;parent data connection has been established on one full-rate traffic channel on which the original user rate is 9.6 kbit/s, the radio interface rate is 12 kbiti and FEC=1/2. The throughput of the connection drops below the predetermined decision value, e.g. 8 kbit~s, and the fill level of the transmission buffer reaches the flow control activation level, and flow control is thus activated.
Simultaneously, the number of re-transmissions of single frames may reach the predetermined maximum value, but this is disregarded in accordance with the invention. The L2R/RLP unit then requests one more full-rate traffic channel hal~ing a user rate of 4.8 kbit/s, a radio interface rate of 12 kbit/s and FEC=
1/3. In addition, the L2R/RLP unit requests a CMM
procedure by means of which the original traffic channel is changed for a configuration having a user rate of 4.8 kbit/s, a radio interface rate of 12 kbit/s, and FEC= 1/3. Thereafter, data transmission continues on two traffic channels at a total nominal data rate o~ 4.8 + 4.8 kbit/s = 9.6 kbit~s (the original nominal user rate), and with a more efficient channel coding scheme FEC = 1~3.
Provided that the requested additional channels are not available, data transmission continues with the present channel configuration. The L2R~RLP
activates the retry timer, however. If the same situation continues still after the timer has expired, a new attempt is made for increasing the channel capacity, and if the new attempt fails, the timer is re-started etc. If the throughput of the connection still weakens to a lower predetermir.ed decision value, the channel coding scheme of each available traffic channel is changed for a more efficient one by means of a CMM procedure. A sensible lower predetermined W096l36l46 ~ . P~l/r~:~ 7 ~ ~ 935 ~ j decision value is a value below a half of the original nominal user rate.
This may be illustrated with example 5. Let us assume that there is an ongoing non-transparent data connection on two parallel traffic channels where the original user rate of the connection is 19.2 kblt/s, the radio interface rate is 12 kbitJs on both traffic channels, and FEC= 1/2. Thereafter, the throughput of the connection drops below 9.6 kbitJs, and the fill level of the transmission buffer reaches the flow control activation level, and flow control ls activated. The number of re-transmissions of individual frames may thus reach a predetermined maximum value, but this is disregarded. There are no additional traffic channels available, in which case the L2R/RLP
unit reouests a CMM procedure, which will change the configuration of the present traffic channels so that the user rate is 4.8 kbit/s, the radio interface rate 12 kbitJs, and FEC=1~3. Thereafter, data transmissio continues on these two traffic channels at a lower combined nomiral data rate 4.8 + ~.8 = 9.6 kbitJs, but ccrrespondingly with a more efficient chanr.el coding scheme FFC=1/3.
6.0 FleYible appli¢ation to a hiqh-speed mult;~-nnol data service Let us assume that a mobile station ~S has an ongoing non-transparent circuit switched connection on one Gr more traffic channels in a system that is capable of high-speed multichannel data transmission.
In such a case, the L2RJRLP unit of the transmitr.ing party A continuously monitors the fill level of the transmission buffer and the throughput. of the connection. ~hen the quality of the connection becomes poorer and the number of errors increases, the nu~ber of re-transmissions of RLP frames increases. At so~e WO96/3614G ~i~ 935 1 ~ L-,'~, 7 t J ~ ' .
t ' 27 stage, the fill level of the transmission buffer reaches the flow control activation level, and flow control is thus activated. The number of re-transmissions of individual ~rames can also reach the predetermined maximum value, but it is disregarded. The throughput of the connection drops below the predetermined maximum decision value. The L2R/R1P unit then requests increasing the number of radio channels.
If the required channel capacity is available, the 10 procedure continues as above in 4Ø If the requested channel capacity is not available, the decision value of throughput is decreased stepwise, and data transmission continues with the present channel configuration. If the combined throughput drops to a 15 new decision value, a corresponding amount of additional capacity is requested, which is less than the additional capacity corresponding to the maximum decision value. If this additional capacity is available, the procedure continues as in the case 20 described above. If the requested additional capacity is not available, the throughput value is further decreased stepwise and data transmission is continued with the present channel configuration until the throughput reaches a new decision value.
Although the invention has been disclosed with reference to some specific embodiments, it is obvious that the description is only made by way of example, the disclosed embodiments allowing alterations and modifications without deviating from the scope and the 30 spirit of the invention set forth in the attached claims.
The pre~ious applicatior.s FI-942,150 and FI-945,817 of the applicant disclose a procedure in which 35 two or more time-slots are allocated ~rom the same TDMA
W096136146 2 ~ 935 ~ 1 P~l/r~L~ .
~ r;~
~,, ! ~ ., .~
frame to a mobile station MS that requires data transmission at a higher transfer rate than what can be provided on one traffic channel. The maximum user rate on a multichannel data connection is the number of parallel traffic channels x the user rate 9.6 kbit/s of one traffic channel. Thus, e.g. a user rate of lg.2 kbit~s can be provided by two traffic channels at the minimum. This procedure is disclosed in this application as an example of carrying out high-speed data transmission based on a plurality of parallel traffic channels in a radio system. As to the closer details of this procedure, reference is made to the above-mentioned applications. It must be noted, however, that the only thing relevant to the invention is the possibility to establish a multichannel transmission connection, and the object of the invention is thus primarily to change the efficiency of the channel capacity and the channel coding scheme when the quality of the connection varies.
Figure 5 illustrates the architecture of the GSM network implementing this kind of data transmission employing a plurality of parallel traffic channels.
Figure 5 is similar to Figure l except for the fact that in Figure 5, between the terminal adaptation function TAF and the interworking function IWF there is a circuit switched non-transparent connection consisting of N parallel traffic channels chO-chn, wherein N=l, 2, 3, ... In the mobile station, the TAF
acts as a divider, distributing the incoming high-speed data signal DATA IN from the data terminal equipment into parallel traffic channels chO-chn, as well as a corr,biner, which combines the low-speed subsignals received from the parallel traffic channels chO-chn again with the high-speed data signal DATA OUT.
Correspondingly, at the second end of the multichannel W096~361~ ~ r~
22 ~3 ) 9 3S ~ t data connection, the IWF acts as a distributor distributing the incoming high-speed data signal DATA
I~ to parallel traffic chanr.els chO-chn, and as a combiner, which combines the low-speed data signals received from the parallel traffic ch.annels again with a high-speed data signal DATA OUT.
The protocol structure shown in Figure 2 also holds for multichannel connection architecture shown in Figure 5. In the preferred embodiment of the invention, the L2R~RLP unit is common to al1 traffic channels allocated for the same connection, also acting as the distributor and combiner mentioned above. Each traffic channel has, however, rate adaptation ~RA~ and channel coding (FEC) functions of their own in accordance with Figure 2. A multichannel data conr.ection is thus substantially similar to the single-channel data connection from the point of view oE the L2R~RLe unit.
The "transmission channel" between the L2R~RLP units only has a larger capacity than before.
2~ In the following, the operation of the L2P.~P~LP
unit will be disclosed in more detail. A mobile static,n has an ongoing non-transparent circuit switched data connection comprising one or more traffic channels in a system that is capable of high-speed multichannel data transmission. The L2R~RLP unit of the transmicting party A continuously monitors the fill level of the transmission buffer and the throughput of the connection. When the quality of the connection becomes poorer and the number of errors increases, the following events take place:
l) the number of re-transmissions of the RLP
frames increases, 2~ the fill level of the transmission buffer reaches the flow control activation level, and the flow control is activated, W096/36l46 ~ ~ 93S ~ ~ E~l~r~
,~" ~. ~
. ~ t' Ir~ t 2 3 3) the number of the re-transmissions of single frames reaches a predetermined maximum value, but, in accordance with the invention, this is disregarded, 4) The th~roughput of the connection drops to a decision value. The decision value may be some value between the current nominal user rate and a half of the current nominal user rate as it is possible to request more channel capacity in order to compensate the 10 decreased data rate on a single traffic channel due to changing for a more efficient channel coding scheme.
~hen both 2~ and 4) are valid, the L2R/RLP
unit of the transmitting party A requests increasing of the channel capacity and charging for a more efficient 15 channel coding scheme by means of a CMM procedure. If e.g. one traffic channel is currently in use, a secor.d traffic channel is requested from the same carrier wave. After a successful channel allocation and CM!~
procedLIre data transmission continues at the same or 20 almost the same data rate as originally, but with a better channel coding scheme FEC.
Additional channel capacity may be requestes depending on the original channel capacity e.g. in the following way:
woscl36l46 .~f~ /rl.~
2 ~ 93~ ~ 1 original capacity requested additional capacity n The decision val~e of the throughput is different in cases in which it is not possible to double the channel capacity, since the original nominal user rate cannot be reached after the CMM proçedure and changing the channel coding scheme. As a rule, it car.
be stated that the lower the percentage of the additional traffic channels, the lower the decisior.
value for the throughput is by means of which the change of the channel coding scheme and additional channel capacity are requested.
After changing for a more efficient channel coding scheme, the L2RfRLP unit of the transmittlng party A continuously monitors:
- the fill level of the transmission buffer and the throughput of the connection for detecting a possible need for changing for a more efficient channet coding scheme and requesting additional capacity, - the throughput of the connectior. fo detecting a possibility to restore the higher speec~
channels and release the additional channel resources if the radio conditions improve.
This can be illustrated with example 4. Let us -WO9C/3614C ~ ~ 935 ~ llr~
assume that a non-tran-;parent data connection has been established on one full-rate traffic channel on which the original user rate is 9.6 kbit/s, the radio interface rate is 12 kbiti and FEC=1/2. The throughput of the connection drops below the predetermined decision value, e.g. 8 kbit~s, and the fill level of the transmission buffer reaches the flow control activation level, and flow control is thus activated.
Simultaneously, the number of re-transmissions of single frames may reach the predetermined maximum value, but this is disregarded in accordance with the invention. The L2R/RLP unit then requests one more full-rate traffic channel hal~ing a user rate of 4.8 kbit/s, a radio interface rate of 12 kbit/s and FEC=
1/3. In addition, the L2R/RLP unit requests a CMM
procedure by means of which the original traffic channel is changed for a configuration having a user rate of 4.8 kbit/s, a radio interface rate of 12 kbit/s, and FEC= 1/3. Thereafter, data transmission continues on two traffic channels at a total nominal data rate o~ 4.8 + 4.8 kbit/s = 9.6 kbit~s (the original nominal user rate), and with a more efficient channel coding scheme FEC = 1~3.
Provided that the requested additional channels are not available, data transmission continues with the present channel configuration. The L2R~RLP
activates the retry timer, however. If the same situation continues still after the timer has expired, a new attempt is made for increasing the channel capacity, and if the new attempt fails, the timer is re-started etc. If the throughput of the connection still weakens to a lower predetermir.ed decision value, the channel coding scheme of each available traffic channel is changed for a more efficient one by means of a CMM procedure. A sensible lower predetermined W096l36l46 ~ . P~l/r~:~ 7 ~ ~ 935 ~ j decision value is a value below a half of the original nominal user rate.
This may be illustrated with example 5. Let us assume that there is an ongoing non-transparent data connection on two parallel traffic channels where the original user rate of the connection is 19.2 kblt/s, the radio interface rate is 12 kbitJs on both traffic channels, and FEC= 1/2. Thereafter, the throughput of the connection drops below 9.6 kbitJs, and the fill level of the transmission buffer reaches the flow control activation level, and flow control ls activated. The number of re-transmissions of individual frames may thus reach a predetermined maximum value, but this is disregarded. There are no additional traffic channels available, in which case the L2R/RLP
unit reouests a CMM procedure, which will change the configuration of the present traffic channels so that the user rate is 4.8 kbit/s, the radio interface rate 12 kbitJs, and FEC=1~3. Thereafter, data transmissio continues on these two traffic channels at a lower combined nomiral data rate 4.8 + ~.8 = 9.6 kbitJs, but ccrrespondingly with a more efficient chanr.el coding scheme FFC=1/3.
6.0 FleYible appli¢ation to a hiqh-speed mult;~-nnol data service Let us assume that a mobile station ~S has an ongoing non-transparent circuit switched connection on one Gr more traffic channels in a system that is capable of high-speed multichannel data transmission.
In such a case, the L2RJRLP unit of the transmitr.ing party A continuously monitors the fill level of the transmission buffer and the throughput. of the connection. ~hen the quality of the connection becomes poorer and the number of errors increases, the nu~ber of re-transmissions of RLP frames increases. At so~e WO96/3614G ~i~ 935 1 ~ L-,'~, 7 t J ~ ' .
t ' 27 stage, the fill level of the transmission buffer reaches the flow control activation level, and flow control is thus activated. The number of re-transmissions of individual ~rames can also reach the predetermined maximum value, but it is disregarded. The throughput of the connection drops below the predetermined maximum decision value. The L2R/R1P unit then requests increasing the number of radio channels.
If the required channel capacity is available, the 10 procedure continues as above in 4Ø If the requested channel capacity is not available, the decision value of throughput is decreased stepwise, and data transmission continues with the present channel configuration. If the combined throughput drops to a 15 new decision value, a corresponding amount of additional capacity is requested, which is less than the additional capacity corresponding to the maximum decision value. If this additional capacity is available, the procedure continues as in the case 20 described above. If the requested additional capacity is not available, the throughput value is further decreased stepwise and data transmission is continued with the present channel configuration until the throughput reaches a new decision value.
Although the invention has been disclosed with reference to some specific embodiments, it is obvious that the description is only made by way of example, the disclosed embodiments allowing alterations and modifications without deviating from the scope and the 30 spirit of the invention set forth in the attached claims.
Claims (14)
1. A method for non-transparent data transmission in a digital telecommunications system, comprising the steps of:
establishing a non-transparent data connection having a nominal data rate, a first transmission capacity and a first channel coding scheme;
transmitting data over a non-transparent data connection in data frames using a communication protocol that acknowledges the received correct data frames and re-transmits the received defective data frames;
buffering the data frames to be transmitted in a transmission buffer;
storing a transmitted data frame in the transmission buffer in case of possible re-transmission, until an acknowledgement of successful reception is received, c h a r a c t e r i z e d by monitoring the throughput of the non-transparent data connection on the basis of the number of re-transmissions, requesting more transmission capacity and a more efficient channel coding scheme for said data connection if the throughput of the data connection drops to a predetermined threshold value.
establishing a non-transparent data connection having a nominal data rate, a first transmission capacity and a first channel coding scheme;
transmitting data over a non-transparent data connection in data frames using a communication protocol that acknowledges the received correct data frames and re-transmits the received defective data frames;
buffering the data frames to be transmitted in a transmission buffer;
storing a transmitted data frame in the transmission buffer in case of possible re-transmission, until an acknowledgement of successful reception is received, c h a r a c t e r i z e d by monitoring the throughput of the non-transparent data connection on the basis of the number of re-transmissions, requesting more transmission capacity and a more efficient channel coding scheme for said data connection if the throughput of the data connection drops to a predetermined threshold value.
2. A method as claimed in claim 1, c h a r a c t e r i z e d by monitoring the fill level of the transmission buffer, requesting more transmission capacity and a more efficient channel coding scheme for said data connection if the throughput of the data connection drops to a predetermined threshold value and the fill level of the transmission buffer reaches a predetermined threshold level.
3. A method as claimed in claim 1 or 2, c h a r a c t e r i z e d by changing a more efficient channel coding scheme for the data connection, allocating more transmission capacity for the data connection corresponding to the more efficient channel coding scheme so that the original nominal user rate remains unchanged or nearly unchanged.
4. A method as claimed in claim 1 or 2, c h a r a c t e r i z e d by maintaining the first transmission capacity and the first channel coding scheme if the requested additional capacity is not available, which capacity is required with a more efficient channel coding scheme for maintaining the original nominal user rate unchanged or nearly unchanged.
5. A method as claimed in claim 2, c h a r a c t e r i z e d by maintaining the first channel coding scheme if the requested additional channel capacity is not available, repeating the request for more additional capacity and a more efficient channel coding scheme at regular time intervals if the throughput of the data connection does not increase from said predetermined threshold value and if the fill level of the transmission buffer does not decrease from said predetermined threshold level, changing the channel coding scheme for a more efficient one without allocating more transmission capacity if the throughput of the data connection decreases to a predetermined second threshold level which is lower than the former threshold value.
6. A method as claimed in claim 4, c h a r a c t e r i z e d by maintaining the first channel coding scheme if the requested additional transmission capacity is not available, reducing said predetermined threshold value for the throughput, requesting more transmission capacity and a more efficient channel coding scheme for said data connection if the throughput of the connection drops to said reduced threshold value and if the fill level of the transmission buffer has not decreased below said predetermined threshold value.
7. A method as claimed in any one of the preceding claims 1-6, c h a r a c t e r i z e d by allocating one or more traffic channels to the data connection as additional transmission capacity.
8. A digital telecommunication system comprising a transmitting party (A) having a transmission buffer, a receiving party (B), a non-transparent circuit switched data connection between the transmitting party and the receiving party, channel coding means (FEC) on said data connection, and a communication protocol (RLP) in which data is transmitted over said data connection in data frames so that the received correct data frames are acknowledged and the received defective data frames are re-transmitted, said transmission buffer buffering the data frames to be transmitted and storing the transmitted data frames until it receives an acknowledgement of successful reception, c h a r a c t e r i z e d in that the transmitting party (A) or the receiving party (B) is arranged to monitor the throughput of the non-transparent data connection on the basis of the number of re-transmissions, and that the transmitting (A) or the receiving (B) party is arranged to request more transmission capacity and a more efficient channel coding scheme for the data connection in question if the throughput of the data connection drops to a predetermined threshold value.
9. A method as claimed in claim 8, c h a r a c t e r i z e d in that the transmitting party (A) is arranged to monitor the fill level of the transmission buffer, and that the transmitting party (A) is arranged to request more transmission capacity and a more efficient channel coding scheme for said data connection if the throughput of the data connection drops to a predetermined threshold value and the fill level of the transmission buffer increases to a predetermined threshold value.
10. A method as claimed in claim 8 or 9, c h a r a c t e r i z e d in that the amount of the requested additional capacity is such that the original nominal user data rate remains unchanged or nearly unchanged after changing for a more efficient channel coding scheme.
11. A method as claimed in claim 8 or 9, c h a r a c t e r i z e d in that the method is a multiple access mobile communication system in which the data connection comprises one traffic channel (ch0) on the radio path, and that increasing the transmission capacity of the data connection includes changing a half-rate traffic channel for a full-rate traffic channel.
12. A method as claimed in claim 8 or 9, c h a r a c t e r i z e d in that the system is a multiple access mobile communication system in which the data connection comprises one traffic channel (ch0) or a plurality of traffic channels (ch0-chn) on the radio path, and that increasing the transmission capacity of the data connection includes increasing the number of the traffic channels allocated for the data connection.
13. A method as claimed in any one of the claims 8-12, c h a r a c t e r i z e d in that the transmitting party is the terminal adaptation function (TAF) of the mobile station (MS), and the receiving party is the interworking function (IWF) of the mobile communication network.
14. A method as claimed in any one of the claims 8-13, c h a r a c t e r i z e d in that the transmitting party is the interworking function (IWF) of the mobile communication network, and the receiving party is the terminal adaptation function (TAF) of the mobile station (MS).
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| FI952253 | 1995-05-09 | ||
| FI952253A FI97927C (en) | 1995-05-09 | 1995-05-09 | Non-transparent data transmission in a digital communication system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA2193511A1 true CA2193511A1 (en) | 1996-11-14 |
Family
ID=8543382
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA002193511A Abandoned CA2193511A1 (en) | 1995-05-09 | 1996-05-08 | Non-transparent data transmission in a digital telecommunications system |
Country Status (8)
| Country | Link |
|---|---|
| US (1) | US5920545A (en) |
| EP (1) | EP0770294A1 (en) |
| JP (1) | JPH10503074A (en) |
| KR (1) | KR970705261A (en) |
| AU (1) | AU708421B2 (en) |
| CA (1) | CA2193511A1 (en) |
| FI (1) | FI97927C (en) |
| WO (1) | WO1996036146A1 (en) |
Families Citing this family (56)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FI101581B (en) * | 1995-11-07 | 1998-07-15 | Nokia Telecommunications Oy | Adaptation of protocols in the fixed network without signaling support to mobile networks |
| FI955944A (en) * | 1995-12-11 | 1997-06-12 | Nokia Telecommunications Oy | Data rate matching procedure and data rate matching unit |
| FI101332B1 (en) * | 1995-12-18 | 1998-05-29 | Nokia Telecommunications Oy | Discontinuous transmission in a multi-channel high-speed data transmission |
| FI103850B1 (en) * | 1996-07-12 | 1999-09-30 | Nokia Mobile Phones Ltd | Automatic control of communication mode |
| US5926468A (en) * | 1997-04-04 | 1999-07-20 | Telefonaktiebolaget L M Ericsson | Wireless communications systems and methods utilizing data link reset |
| SE511881C2 (en) * | 1997-08-08 | 1999-12-13 | Ericsson Telefon Ab L M | Method and arrangement for transmitting packet information in a digital telecommunication system |
| US5909469A (en) * | 1997-08-29 | 1999-06-01 | Telefonaktoebolaget Lm Ericsson | Link adaptation method for links using modulation schemes that have different symbol rates |
| US6134230A (en) * | 1997-08-29 | 2000-10-17 | Telefonaktiebolaget Lm Ericsson | Method for selecting a link protocol for a transparent data service in a digital communications system |
| US6125148A (en) * | 1997-08-29 | 2000-09-26 | Telefonaktiebolaget Lm Ericsson | Method for demodulating information in a communication system that supports multiple modulation schemes |
| US6208663B1 (en) * | 1997-08-29 | 2001-03-27 | Telefonaktiebolaget Lm Ericsson (Publ) | Method and system for block ARQ with reselection of FEC coding and/or modulation |
| US6456627B1 (en) | 1997-08-29 | 2002-09-24 | Telefonaktiebolaget Lm Ericsson (Publ) | Method for communicating information in a communication system that supports multiple modulation schemes |
| US6108348A (en) * | 1997-09-12 | 2000-08-22 | Lockheed Martin Corporation | 2.4-to-3 kbps rate adaptation apparatus for use in narrowband data and facsimile communications systems |
| FI104527B (en) * | 1997-09-17 | 2000-02-15 | Nokia Mobile Phones Ltd | Customizable radio link |
| US6389066B1 (en) * | 1997-09-21 | 2002-05-14 | Lucent Technologies Inc. | System and method for adaptive modification of modulated and coded schemes in a communication system |
| US6028851A (en) * | 1997-09-26 | 2000-02-22 | Telefonaktiebolaget L M Ericsson (Publ) | System and method for mobile assisted admission control |
| FI110048B (en) * | 1998-09-16 | 2002-11-15 | Nokia Corp | Method and apparatus for dynamic control of radio resources |
| FI112425B (en) | 1998-09-16 | 2003-11-28 | Nokia Corp | Method and apparatus for dynamic control of radio resources |
| US6308082B1 (en) * | 1998-10-06 | 2001-10-23 | Telefonaktiebolaget L M Ericsson (Publ) | Cell-level throughput adjustment in digital cellular systems |
| EP1014641A1 (en) * | 1998-12-22 | 2000-06-28 | Telefonaktiebolaget Lm Ericsson | Method and system for reducing the processing time of data in communication networks |
| SE515837C2 (en) * | 1999-01-22 | 2001-10-15 | Ericsson Telefon Ab L M | Adaptable bandwidth |
| US6865233B1 (en) * | 1999-02-19 | 2005-03-08 | Telefonaktiebolaget Lm Ericsson (Publ) | Method and system for control signalling enabling flexible link adaptation in a radiocommunication system |
| FI19992002A (en) * | 1999-09-20 | 2001-03-20 | Nokia Networks Oy | Measurement of cavity on circuit-switched service in a cellular radio network |
| DE10026927B4 (en) * | 2000-05-30 | 2006-07-13 | Siemens Ag | Method for transmitting data packets in a communication system between a transmitter and a receiver and corresponding transmitter and receiver |
| US7120134B2 (en) | 2001-02-15 | 2006-10-10 | Qualcomm, Incorporated | Reverse link channel architecture for a wireless communication system |
| US7403498B2 (en) * | 2001-05-31 | 2008-07-22 | Qualcomm Incorporated | Method and apparatus for selective examination of PPP packets for renegotiation of a PPP link on a Um interface |
| US7372811B1 (en) * | 2001-08-07 | 2008-05-13 | Cisco Technology, Inc. | Load balancing model for multilink frame relay |
| FI112141B (en) * | 2001-08-23 | 2003-10-31 | Nokia Corp | Non-transparent data transfer in a mobile network |
| ATE299630T1 (en) * | 2002-02-14 | 2005-07-15 | Matsushita Electric Ind Co Ltd | METHOD FOR CONTROLLING DATA RATE IN A WIRELESS PACKET DATA COMMUNICATIONS SYSTEM, TRANSMITTER AND RECEIVER FOR USE THEREOF |
| ATE274770T1 (en) * | 2002-06-12 | 2004-09-15 | Cit Alcatel | DE-/ACTIVATION OF TANDEM CONNECTIONS |
| US6804253B2 (en) * | 2002-10-01 | 2004-10-12 | Motorola, Inc. | Method for determining mobile station coding scheme adaptation capability |
| US7961617B2 (en) * | 2002-10-29 | 2011-06-14 | Telefonaktiebolaget Lm Ericsson (Publ) | System and method for wireless network congestion control |
| GB2396088B (en) * | 2002-12-06 | 2005-06-29 | Qualcomm | A data transfer procedure for transferring data of a data sequence between a transmitting entity and a receiving entity |
| US7489703B2 (en) * | 2002-12-20 | 2009-02-10 | Motorola, Inc. | Apparatus and method for a coding scheme selection |
| DE10324470A1 (en) * | 2003-05-30 | 2005-03-10 | Deutsche Telekom Ag | Method and apparatus for controlling data connections in a data network having a plurality of data network nodes |
| CA2510034A1 (en) * | 2003-11-14 | 2005-05-26 | Nokia Corp | Generic trau frame structure |
| GB2414628B (en) * | 2003-11-21 | 2006-05-10 | Motorola Inc | Method for selecting a channel coding scheme and apparatus therefor |
| US8782654B2 (en) | 2004-03-13 | 2014-07-15 | Adaptive Computing Enterprises, Inc. | Co-allocating a reservation spanning different compute resources types |
| IL160921A (en) * | 2004-03-18 | 2009-09-01 | Veraz Networks Ltd | Method and device for quality management in communication networks |
| US20070266388A1 (en) | 2004-06-18 | 2007-11-15 | Cluster Resources, Inc. | System and method for providing advanced reservations in a compute environment |
| DE102004037815B4 (en) * | 2004-08-04 | 2013-01-03 | Infineon Technologies Delta Gmbh | Mobile radio device and method for controlling mobile radio transmitter resources in a mobile radio device |
| US8176490B1 (en) | 2004-08-20 | 2012-05-08 | Adaptive Computing Enterprises, Inc. | System and method of interfacing a workload manager and scheduler with an identity manager |
| CA2586763C (en) | 2004-11-08 | 2013-12-17 | Cluster Resources, Inc. | System and method of providing system jobs within a compute environment |
| US7602714B2 (en) * | 2004-12-30 | 2009-10-13 | Motorola, Inc. | Methods for managing data transmission between at least two different data regions |
| US9075657B2 (en) | 2005-04-07 | 2015-07-07 | Adaptive Computing Enterprises, Inc. | On-demand access to compute resources |
| US8863143B2 (en) | 2006-03-16 | 2014-10-14 | Adaptive Computing Enterprises, Inc. | System and method for managing a hybrid compute environment |
| US9231886B2 (en) | 2005-03-16 | 2016-01-05 | Adaptive Computing Enterprises, Inc. | Simple integration of an on-demand compute environment |
| US9015324B2 (en) | 2005-03-16 | 2015-04-21 | Adaptive Computing Enterprises, Inc. | System and method of brokering cloud computing resources |
| WO2006112980A2 (en) | 2005-03-16 | 2006-10-26 | Cluster Resources, Inc. | Reserving resources in an on-demand compute environment from a local compute environment |
| US8782120B2 (en) | 2005-04-07 | 2014-07-15 | Adaptive Computing Enterprises, Inc. | Elastic management of compute resources between a web server and an on-demand compute environment |
| SE529757C2 (en) * | 2005-07-15 | 2007-11-20 | Teliasonera Ab | Method and radio network controller for adaptive radio channel switching |
| US8189603B2 (en) * | 2005-10-04 | 2012-05-29 | Mammen Thomas | PCI express to PCI express based low latency interconnect scheme for clustering systems |
| US8041773B2 (en) | 2007-09-24 | 2011-10-18 | The Research Foundation Of State University Of New York | Automatic clustering for self-organizing grids |
| CN101345709B (en) * | 2008-08-18 | 2012-02-22 | 中兴通讯股份有限公司 | Method for implementing non-transparent data traffic switch and a TD-SCDMA terminal |
| US11720290B2 (en) | 2009-10-30 | 2023-08-08 | Iii Holdings 2, Llc | Memcached server functionality in a cluster of data processing nodes |
| US10877695B2 (en) | 2009-10-30 | 2020-12-29 | Iii Holdings 2, Llc | Memcached server functionality in a cluster of data processing nodes |
| US10560755B2 (en) * | 2016-09-23 | 2020-02-11 | Verizon Patent And Licensing Inc. | Methods and systems for concurrently transmitting object data by way of parallel network interfaces |
Family Cites Families (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4334306A (en) * | 1978-06-02 | 1982-06-08 | Texas Instruments Incorporated | Transparent intelligent network for data and voice |
| GB2160392B (en) * | 1984-05-02 | 1987-08-19 | Racal Res Ltd | Data transmission |
| US4888767A (en) * | 1984-12-25 | 1989-12-19 | Nec Corporation | Repeat request signal transmission method for multi-station packet communication |
| US4691314A (en) * | 1985-10-30 | 1987-09-01 | Microcom, Inc. | Method and apparatus for transmitting data in adjustable-sized packets |
| US4771417A (en) * | 1986-09-18 | 1988-09-13 | Racal Data Communications, Inc. | High speed modem |
| GB8628821D0 (en) * | 1986-12-02 | 1987-01-07 | Plessey Co Plc | Data transmission systems |
| JPH0216827A (en) * | 1988-07-04 | 1990-01-19 | Omron Tateisi Electron Co | Data transmission equipment |
| DE3933262A1 (en) * | 1989-10-05 | 1991-04-11 | Bosch Gmbh Robert | METHOD AND DEVICE FOR BIDIRECTIONAL TRANSMISSION OF DATA |
| US5130993A (en) * | 1989-12-29 | 1992-07-14 | Codex Corporation | Transmitting encoded data on unreliable networks |
| JPH04276922A (en) * | 1991-03-04 | 1992-10-02 | Mitsubishi Electric Corp | Radio communication equipment |
| US5487068A (en) * | 1994-07-29 | 1996-01-23 | Motorola, Inc. | Method for providing error correction using selective automatic repeat requests in a packet-switched communication system |
| US5627970A (en) * | 1994-08-08 | 1997-05-06 | Lucent Technologies Inc. | Methods and apparatus for achieving and maintaining optimum transmission rates and preventing data loss in a processing system nework |
-
1995
- 1995-05-09 FI FI952253A patent/FI97927C/en active
-
1996
- 1996-05-08 WO PCT/FI1996/000259 patent/WO1996036146A1/en not_active Application Discontinuation
- 1996-05-08 AU AU56505/96A patent/AU708421B2/en not_active Ceased
- 1996-05-08 KR KR1019970700079A patent/KR970705261A/en not_active Application Discontinuation
- 1996-05-08 EP EP96913556A patent/EP0770294A1/en not_active Withdrawn
- 1996-05-08 CA CA002193511A patent/CA2193511A1/en not_active Abandoned
- 1996-05-08 US US08/765,059 patent/US5920545A/en not_active Expired - Lifetime
- 1996-05-08 JP JP8533809A patent/JPH10503074A/en active Pending
Also Published As
| Publication number | Publication date |
|---|---|
| AU5650596A (en) | 1996-11-29 |
| US5920545A (en) | 1999-07-06 |
| FI952253A0 (en) | 1995-05-09 |
| WO1996036146A1 (en) | 1996-11-14 |
| FI97927B (en) | 1996-11-29 |
| KR970705261A (en) | 1997-09-06 |
| FI97927C (en) | 1997-03-10 |
| JPH10503074A (en) | 1998-03-17 |
| AU708421B2 (en) | 1999-08-05 |
| EP0770294A1 (en) | 1997-05-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5920545A (en) | Non-transparent data transmission in a digital telecommunications system | |
| EP0809901B1 (en) | Multichannel high-speed data transfer | |
| EP0826280B1 (en) | Sliding-window data flow control using an adjustable window size | |
| JP3464453B2 (en) | Data transmission method | |
| CA2193379C (en) | Data transmission system with sliding-window data flow control | |
| US6857095B2 (en) | Method for making data transmission more effective and a data transmission protocol | |
| US20020001287A1 (en) | Data transmission in a telecommunications system | |
| US6414938B1 (en) | Method and system for retransmitting data packets in a communication system having variable data rates | |
| MXPA02005637A (en) | A method for non transparent transmission of data in a digital mobile communications system. | |
| US6728218B1 (en) | Method of dynamic rate switching via medium access channel layer signaling |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FZDE | Discontinued |