WO2003107702A1

WO2003107702A1 - Method for allocating channels during the transmission of useful data between an emitting radio station and a receiving radio station via a radio interface

Info

Publication number: WO2003107702A1
Application number: PCT/EP2003/006070
Authority: WO
Inventors: Carsten Ball; Kolio Ivanov; Robert Müllner
Original assignee: Siemens Aktiengesellschaft
Priority date: 2002-06-14
Filing date: 2003-06-10
Publication date: 2003-12-24

Abstract

According to the invention, at least one channel is allocated to a subscriber connection, on a carrier which is selected from at least two carriers, said carriers respectively having a different frequency repetition pattern and/or being defined by various frequency hopping systems. At least one channel is allocated to the subscriber connection, on the carrier on which an optimum data rate is obtained during given radio conditions, especially taking into account the instantaneous occupancy by other packet data connections, and the entire system load.

Description

description

Method for channel assignment in a user data transmission via a radio interface between a transmitting and a receiving radio station

Radio communication systems are used to transmit information, voice or data using electromagnetic waves via a radio interface, also known as an air interface, between a transmitting and a receiving one

Radio station. An example of a radio communication system is the known GSM mobile radio system and its further development with a packet data service GPRS or EGPRS, the architecture of which is described, for example, in B. Walke, "Mobile radio networks and their protocols", Volume 1, Teubner-Verlag Stuttgart, 1998, page 138 to 151 and pages 295 to 311. In this case, a channel formed by a narrow-band frequency range and a time slot is provided for the transmission of a subscriber signal. Since a subscriber signal in one channel differs in frequency and time from the other subscriber signals, the radio station can detect the data of the subscriber signal. In newer radio communication systems, such as the UMTS system, the individual subscribers are also differentiated by different spreading codes.

In the case of packet-switched data transmission, the data transmission for several participants takes place over one and the same physical channel. Each participant can occupy several physical channels at the same time (multi-slot). For each participant, a packet data flow (Temporary Block Flow TBF) is broken down into time-limited radio blocks that are transmitted. Depending on the modulation / coding a set, these radio blocks have different payloads with the same length. The possible modulation / coding schemes differ with regard to the division of the radio block into payload and error protection information. Depending on the radio a coding scheme with higher or lower error protection is assigned to a connection. In the case of poor radio conditions, a coding scheme with higher error protection is assigned, and in good radio conditions a coding scheme with less error protection.

The time-limited radio blocks from different users or the time-limited packet data flows from different users are transmitted over the channel in a nested manner. The quality of the channel is determined for the individual participant on the basis of measured values. The received field strength, the block error rate, the interference level of the channel and the bit error probability serve as measured values. The measured values are used to set various radio parameters. In particular, the modulation / coding scheme is dynamically adapted to the channel conditions on the basis of the measured values. In this way, the participant receives the error protection that he needs based on the radio conditions. This is generally associated with a changing data rate, which can be compensated for, for example, by reallocating the assigned channels.

In known radio communication systems according to the GSM / GPRS standard, packet-switched data transmission today mainly takes place via the broadcast channel (broadcast channel BCCH carrier), which is radiated continuously by all base stations in all time slots with constant power. Operators are increasingly moving to allocating packet data channels to non-BCCH carriers, which usually have a different frequency reuse pattern and a different frequency hopping method. In the packet data channels must be reserved channels between f _o r GPRS / EDGE, which differ approval sen no circuit-switched speech and mixed channels for GPRS / EDGE and language. The invention is based on the problem of specifying a method for channel allocation in a user data transmission via a radio interface between a transmitting radio station and a receiving radio station, with which the capacity of the radio interface can be better utilized.

This problem is solved according to the invention by a method according to claim 1. Further developments of the invention emerge from the remaining claims.

According to the invention, a subscriber connection is assigned a channel on a carrier, which is selected from a set of at least two carriers that are planned with different frequency repetition patterns and / or have different frequency hopping systems. Frequency hopping with different frequencies is used in the different frequency hopping systems. This also includes the case of frequency hopping with zero frequencies, i.e. H. no frequency hopping method. For example, the carrier is selected from a set of at least two carriers, of which one carrier provides a frequency hopping method and the other carrier does not provide a frequency hopping method or another frequency hopping method. The frequency repetition pattern of the carriers is also taken into account. The channel on the carrier on which an optimal data rate is achieved under given radio conditions is assigned to the subscriber connection. The method therefore optimizes a subscriber's data throughput.

The method is suitable for the transmission of user data of packet data as well as of voice data. In this case, more than one channel can be assigned to a connection, the method then being applied to the selection of all channels.

The invention makes use of the knowledge that different coding schemes used in packet data transmission and with some voice data transmissions, eg when using adaptive multirate codecs, behave differently with regard to the achievable data rate depending on whether they are assigned to a connection on a carrier with frequency hopping or a connection on a carrier without

Frequency hopping. Furthermore, the achievable data rate depends on the frequency repetition pattern used. This effect is exploited by assigning a connection to one or more channels on the carrier on which the optimal data rate is achieved under the given boundary conditions.

When connecting to a coding scheme with a high level of error protection, such as CS1 or CS2 for GPRS or MCS1 or MCS2 for EGPRS, the data rate achieved is higher if the connection is assigned a channel on a carrier with frequency hopping than if the same connection was assigned a channel on a Carrier without frequency hopping is assigned. On the other hand, when connecting to a coding scheme with a low level of error protection, e.g. CS3 or CS4 for GPRS or M-CS4 to M-CS9 for EGPRS the data rate is higher when the connection is assigned a channel on a carrier without frequency hopping than when the same connection is assigned a channel on a carrier with frequency hopping.

The invention also makes use of the knowledge that, with a high system load (high fractional load), the C / I distribution in the cell is better planned for a carrier with a large frequency repetition interval than for a carrier with a small frequency repetition interval. Therefore meet connections with coding scheme. With a high level of error protection, the requirements placed on them will probably still be satisfactory on carriers with a small frequency repetition distance, while uncoded or almost uncoded connections only run reasonably on carriers with a large frequency repetition distance. INS the BCCH carrier in particular is planned with a sufficiently large frequency repetition interval.

In contrast, with a low system load, the fulfillment of the requirements on non-BCCH carriers is better than on BCCH carriers, since the BCCH carriers always emit with constant output power and thus generate interference. With a small system load, the interference on a non BCCH carrier with the same frequency repetition distance is therefore drastically less. The packet data connections should therefore preferably be allocated to non-BCCH carriers.

The allocation of the available channel resources to a new or running packet data connection also depends on the current occupancy of the resources by other packet data flows, since multiplexing can drastically reduce the throughput of a connection. The number of available time slots for a multislot mobile also plays a role.

A channel on a carrier with frequency hopping is preferably assigned to a subscriber connection to which a coding scheme with a high level of error protection is assigned on the basis of the radio conditions. A carrier is preferred that has been planned with a small frequency repetition distance. In contrast, a subscriber connection that is assigned a coding scheme with a low level of error protection on the basis of the radio conditions is assigned a channel on a carrier without frequency hopping, preferably on a carrier that generally has a large frequency repetition distance, such as the BCCH carrier.

The connection is thus always assigned a channel in which the highest possible data rate can be achieved. In this way, the capacity of the radio interface is better used. If participants with low data rates (due to the coding scheme with high error protection) are kept away from the channels with maximum throughput, then these are only available to participants with high data rates. Losses occur with this little, since the participants with small data rates due to the phy ^¬ sik on their assigned channels with the requirements even better meet than on channels with maximum throughput.

It is within the scope of the connection to measure a reception field strength, a ratio of useful to interference signal (C / I) and / or a time reserve at the start of a connection. Taking into account the measured values, the coding scheme and the channel to be assigned for the connection are then determined.

For dynamic adaptation of the connection, it is advantageous to continuously measure the useful to interference signal ratio (C / I), the bit error rate (BET), the block error rate (BLER) and / or the time reserve (timing advance TA) during the connection. The measured values are then used to determine the coding scheme and the channel or channels with which a maximum data rate is achieved, the different data rates being taken into account when assigning a channel on a carrier with frequency hopping and a carrier without or other frequency hopping. It may be necessary to change the channel from a carrier with frequency hopping to a carrier without or another frequency hopping method or vice versa when adapting the coding scheme. The same applies to frequency repetition pattern planning of the carrier.

It is within the scope of the invention, depending on the utilization of the system, the packet data channels with a small load preferably on a non BCCH carrier with frequency hopping and possibly a small frequency repetition distance and the packet data channels with a large load on a carrier without frequency hopping and a large frequency repetition distance such as to configure on the BCCH broadcast channel. The connection is preferably assigned a coding method with high error protection and a channel on a carrier with frequency hopping for a useful to interference signal ratio between 6 and 15 dB.

The method is explained in more detail below with reference to figures in which exemplary embodiments are shown.

FIG. 1 shows the data rate as a function of the useful data-to-interference signal ratio for two coding schemes with little error protection for transmission on a carrier with frequency hopping or a carrier without frequency hopping.

FIG. 2 shows the data rate m as a function of the useful data-to-interference signal ratio for two coding schemes with high error protection for transmission on a carrier with frequency hopping or without frequency hopping.

FIG. 3 shows a useful data to interference signal ratio distribution for an effective frequency repetition factor of 12.

FIG. 4 shows a useful data-to-interference signal ratio distribution for a frequency repetition factor of 1.

When establishing a connection, a field strength, a ratio of useful data to interference signal (C / I) and a time reserve (timing advance TA) are measured. Depending on these measurands, a coding scheme is defined that is assigned to the connection. This is done, for example, using a table. The table is structured in such a way that it specifies the coding scheme for the respective radio conditions with which the highest data rate can be achieved. GPRS is one of the coding schemes CS1, CS2, CS3 or CS4. EGPRS is one of the coding schemes MCS1, MCS2, MCS3, MCS4, MCS5, MCS6, MCS7, MCS8 or MCS9. The coding schemes CS1 to CS4 or MCS1 to MCS4 are used with the GMSK modulation, the coding schemes MCS5 to MCS9 with the 8PSK modulation.

The coding schemes have different error protection, which is caused by redundancy bits for error detection and error correction. The coding schemes CS1 and MCS1 have the highest error protection, the coding schemes CS4 and MCS9 the lowest error protection or are even uncoded. In the coding schemes, the relationship between information and redundancy bits shifts from CS1 to CS4 or MCS1 to MCS9 in favor of the information bits. This makes the transmission more error-prone. It is therefore sensible to use the coding schemes CS4 or MCS9 in good radio conditions and the coding schemes CS1 or MCS1 in bad radio conditions.

For example, if the useful data to interference signal ratio C / I is 24 dB at the start of the connection, the connection is assigned, for example, a channel on a broadcast channel BCCH that continuously transmits on the same frequency with the coding scheme CS4, which has little error protection (see Point I in Figure 1). 1 shows the data rate as a function of the useful data-to-interference signal ratio C / I for the coding scheme CS4 when transmitted on a carrier without frequency hopping (reference symbol 4), for the coding scheme CS4 when transmitting to a carrier with frequency hopping method (reference symbol 4 '), for the coding scheme CS3 on a carrier without frequency hopping (reference symbol 3), and for the coding scheme CS3 on a carrier with frequency hopping method (reference symbol 3 '). It can be seen that the data rate 4 for the coding scheme CS4 is highest on a carrier without frequency hopping in the range above approximately 9 dB. During the connection, the useful data to interference signal ratio C / I, the bit error rate BER, the block error rate BLER and the time reserve TA are measured continuously, that is periodically at specific times. On the basis of these measured values, the coding scheme and the channel with which a maximum data rate is achieved is determined. This is done, for example, on the basis of a table which, for the given measured values, indicates for which coding scheme on which channel the maximum data rate is achieved. The table takes into account that when assigning a channel on a carrier with frequency hopping and on a carrier without frequency hopping, different data rates are achieved.

If the radio conditions deteriorate during the connection, for example from 24 dB to 17.5 dB (point II in FIG. 1), a point is reached at which, based on the bit error rate BER and the block error rate BLER, the data rate 3 for the coding scheme CS3 a carrier without frequency hopping represents the maximum data rate for the connection. Therefore, the connection in point II is assigned the coding scheme CS3 on the broadcasting channel BCCH, which transmits continuously without frequency hopping (see point III in FIG

1) •

If the radio conditions deteriorate further, point IV is reached, at which the useful data to interference signal ratio C / I is 12.5 dB. At this point IV, taking into account the bit error rate and the block error rate, the maximum data rate 2 for the coding scheme CS2 is reached on a carrier with frequency hopping (see FIG

2).

In FIG. 2, the data rates are a function of the useful data to inference signal ratio C / I for the coding scheme CS2 on a carrier without frequency hopping (reference symbol 2), for the coding scheme CS2 on a carrier with frequency hopping method (reference symbol 2 '), for the coding scheme CS1 on a carrier without frequency hopping method (reference symbol 1) and for the coding scheme CSl on a carrier with frequency hopping method (reference symbol IM shown.

The assignment of the coding scheme CS2 is associated with the simultaneous assignment of another carrier using frequency hopping, for example a traffic channel TCH.

If the radio conditions deteriorate from point V to point VI, at which the useful data to interference signal ratio is 10 dB, there is a change to the coding scheme CS1 on the carrier with frequency hopping (point VII in FIG. 2). If the radio conditions deteriorate further to about C / I = 6 dB, the radio connection is broken off in point VIII.

An example of the algorithm for changing the coding scheme is explained in more detail below:

1) Measure the system load by determining the occupied channels in the cell and, if necessary, by determining the occupied channels in the neighboring cells.

2) Measure the interference on all channels of the cell, which are activated for packet data traffic.

3) Determine the other packet data transfers that are already running on all channels.

4) Calculate the combination of channels for all possible channel combinations according to the multislot class and the possible coding scheme (based on the radio conditions / interference) of the subscriber station and the set frequency hopping method, so that a) the throughput is optimal, b) the number of assigned channels Time slots is maximum, c) as few other subscribers as possible are multiplexed, d) a maximum number of reserved GPRS packet data channels is used, d) depending on the coding scheme CS / MCS, a carrier with the corresponding appropriate frequency repetition pattern or frequency hopping system is selected, e) depending on the coding scheme CS / MCS, a preference for BCCH or non-BCCH carriers is fulfilled.

If there are several carriers with different frequency repetition patterns in a radio cell, it can be used - as can be seen in the algorithm description above - that the achievable data rate also depends on the frequency repetition pattern and the system load. FIG. 3 shows the value CDF (1) as a function of the useful data to interference signal ratio C / I for a cell with a frequency repetition pattern 4x3 at 100% load. For comparison, the value CDF (1) in FIG. 4 is a function of the useful data to interference signal ratio C / I

Cell with frequency repetition pattern shown 1x1 at 30% load. The value CDF (1) means “Cumulative Distribution Function *” and describes the integral distribution of the C / I ratios in the cell for the wearer with the respective frequency repetition pattern.

The comparison of FIGS. 3 and 4 shows that the probability that the useful data to interference signal ratio is higher is greater with a frequency repetition pattern 4x3 even at 1001 load than with the carrier with 1/1 frequency repetition pattern at 30% load. In this case, the radio conditions are statistically better. In particular, the carrier with 4x3 frequency repetition patterns has for the most part C / I ratios which allow the operation of the uncoded GPRS CS4 or EDGE MCS9 codings. The carrier with 1/1 frequency repetition pattern, on the other hand, will only allow small coding schemes such as CS1 / 2 or MCS1 / 2 at 301 load. With a very low load, however, a carrier with a 1/1 frequency repetition pattern will also be able to offer very good ratios, since the interference does not occur if there is no load. Therefore, the above algorithm includes both the consideration of the current system load and the consideration of the frequency repetition pattern per carrier.

The above method can be generalized from packet data to voice data. With adaptive multirate codecs (AMR) in particular, there are coding schemes with low and high error protection, which are adapted to the radio channel conditions using a link adaptation process. With the introduction of wideband AMR in accordance with 3GPP Release 5 Standard, even speech codecs based on 8PSK modulation are introduced, which offer optimal quality, but also only operate satisfactorily under very good radio channel conditions (large C / I). Here, too, an attempt must be made to use clever resource management or an optimal time slot allocation

Based on the algorithm presented above to maximize system capacity.

Claims

claims

1. A method for channel assignment in a user data transmission over a radio interface between a transmitting and a receiving radio station, in which a channel is assigned to a subscriber connection on a carrier, the carrier being selected from a set of at least two carriers, of which the carriers have different Frequency repetition patterns are planned and / or have different frequency hopping systems, in which the subscriber connection is assigned the channel on the carrier on which an optimal data rate is achieved under given radio conditions.

2. The method of claim 1, wherein a subscriber connection, which is assigned a coding scheme with a high error protection on the basis of the radio conditions, is assigned a channel on a carrier with frequency hopping, in which a subscriber connection, which is assigned a coding scheme with a on the basis of the radio conditions low error protection is assigned, a channel is assigned on a carrier without frequency hopping.

3. The method as claimed in claim 1 or 2, in which the channel on the carrier is assigned to the subscriber connection on which an optimum data rate is achieved, taking into account the occupancy by other packet data connections and taking into account the system load.

4. The method as claimed in one of claims 1 to 3, in which a channel on a BCCH carrier is assigned to a subscriber connection if the system load is high, in which a channel on a non-BCCH carrier is preferably assigned to a subscriber connection if the system load is low.

5. The method according to claims 1 to 4, wherein a subscriber connection, which is assigned a coding scheme with a high level of error protection on the basis of the radio conditions, is assigned a channel on a carrier with a small frequency repetition interval, - in the case of a subscriber connection, which is based on the

A coding scheme with a low level of error protection is assigned to radio conditions, a channel on a carrier with a large frequency repetition interval is assigned.

β. Method according to one of Claims 1 to 5, in which, at the start of a connection, a reception field strength, a ratio of useful data to interference signal (C / I) and / or a time reserve (TA) are measured, taking into account the measured reception field strength , the useful data to interference signal ratio and / or the time reserve, the coding scheme to be assigned and / or the channel to be assigned for the connection are determined, in which channels with the least possible interference / interference are used.

7. The method according to any one of claims 1 to 6, in which the minimum occupancy of the channels to be allocated for the connection is taken into account by other packet data connections, in which in the case of a multislot connection the number of assigned channels is maximized, in which the Use of reserved packet data channels is maximized.

Method according to one of claims 1 to 7, - in which the channel assignment is carried out at the start of a new packet data connection,

- in which the procedure for channel assignment is called up several times or periodically during the ongoing packet data connection, in which the procedure for channel assignment is called during the ongoing packet data connection as soon as the radio conditions of the connection or the coding scheme change, - in which the procedure for Channel assignment is called during the ongoing packet data connection as soon as the allocation status of the channels in the cell changes.

9. The method according to any one of claims 1 to 8, - in which, during a connection, the reception field strength, the useful data to interference signal ratio, a bit error rate, a block error rate and / or the time reserve are measured, in which the coding scheme and / or or the channel or the channels with which a maximum data rate is achieved, the different data rates being taken into account when assigning a channel on a carrier as a function of the frequency hopping method and frequency repetition pattern.

10. The method according to claim 6 or 9, wherein the connection for a useful data to interference signal ratio between 6 and 15 dB is assigned a coding method with high error protection and a channel on a carrier with frequency hopping.

11. The method according to any one of claims 1 to 10, wherein the channels are configured on a carrier with frequency hopping on a traffic channel and the channels on a carrier without frequency hopping on a radio channel.