Note: Descriptions are shown in the official language in which they were submitted.
CA 02212072 1997-07-31
METHOD FOR DYNAMIC CHANNEL ALLOCATION IN RADIO SYSTEMS,
ESPECIALLY FOR WIRELESS LOCAL LOOP (WLL) SYSTEMS, AND
DEVICES FOR CARRYING OUT THE METHOD
The invention relates to a method for dynamic
channel allocation in radio systems, especially for wireless
local loop (WLL) systems, and devices for carrying out the
method.
Radio systems have become much more significant in
Germany's communications and data technology, especially
since the Federal Post Office has lost its
telecommunications monopoly. The reason for this is that
private providers do not have access to a line plant network
to the individual households. The creation of such a line
plant network, however, would require considerable
investment costs so that the private providers would not be
competitive. The data are therefore transmitted from a base
station to the user unit by means of radio. Such wireless
local loop (WLL) systems allow centralized communication
with households without a corresponding creation of an
extensively-branching line plant.
In radio systems, dynamic channel allocation is
required for adaptation to changing transmission rates since
the number of available channels is greatly limited. The
data produced must therefore be distributed over the
individual channels so that the available channels are
utilized optimally. At the same time, the radio system must
be able to respond flexibly to changes in the transmission
rates. The cause for changing transmission rates can be the
data service itself or the setting-up and clearing-down of
simultaneous connections with constant data rate such as,
for example, voice transmissions.
In the known radio systems which transmit in
parallel in a plurality of channels between transmitter and
receiver, the data produced are mapped into the allocated
channels in accordance with a certain algorithm. The
CA 02212072 1997-07-31
disadvantage of the known radio systems is the lack of
synchronization of the activation or deactivation of
channels in transmitter and receiver so that a changed
channel allocation does not become effective simultaneously
in the transmitter and receiver. The consequence of the
lack of synchronization is that the mapping algorithm is
disturbed so that data errors occur during the transmission.
If, for example, the mapping algorithm allocates additional
channel capacity too early to the receiver, this can lead to
reception of pseudo data. If this additional channel
capacity is included too late in the mapping algorithm, in
contrast, this will lead to data losses. The reverse
applies in the case of a reduction of channel capacities,
i.e. too early a reduction leads to data losses and too late
a reduction can lead to the occurrence of pseudo data. In
bidirectional transmission, each transmitter at the same
time acts as a receiver so that the above-mentioned errors
can occur at each transmission end.
The invention is therefore based on the technical
problem of creating a method for dynamic channel allocation
in radio systems, and devices for carrying out the method,
in which the changed channel allocation becomes effective
simultaneously for transmitter and receiver.
The solution of the problem is found by
establishing that an involved station is the master which
derives new channel and link tables from the data transfer
load and transfers these to master and slave. The system is
thereby placed into a state of expectation, and in the case
of a duplex radio transmission link, the master can be
defined independently of the initiating station. The final
reconfiguration is executed by sending out a reconfiguration
signal from the master to the slave and by acknowledgement
of the reception by the slave, the reconfiguration and
acknowledgement signals being ignored in each case by the
mapping algorithm. At the same time, the reconfiguration
signal and the acknowledgment signal overlap in time with
CA 02212072 1997-07-31
the allocation or withdrawal procedures for the channels on
the side of the master and of the slave, which avoids data
losses or the injection of pseudo data.
In a further embodiment, there exists a single
virtual connection module for each active slave in the
mapping module of the master, and also a single virtual
connection module in the mapping module of the slave. The
additional method step of checking whether the demanded
channels are available or the connection capacity is
adequate ensures that no physically-impossible configuration
is set at the master or the slaves. When a plurality of
channels are allocated or withdrawn, the same methods can be
carried out in parallel. When expected signals are not
received, the master or the slave can return to the initial
state after a presettable number of cycles or time, which
prevents the system from being in an unstable state over a
prolonged period of time.
In another embodiment with an instruction for
reducing the channel capacity, the channel to be deactivated
is taken out of the mapping algorithm at the master end in
the direction of transmission whilst it remains tied in in
the receiving direction. At the slave end, the relevant
channel remains tied into the mapping algorithm. The master
can then send the reconfiguration signal to the slave via
the channel to be deactivated. After the reconfiguration
signal has been received, the slave sends an acknowledgment
signal to the master and unties the channel from the mapping
algorithm at the transmitting and receiving end. After
receiving the acknowledgement signal, the master then also
unties the channel from the mapping algorithm at the
receiving end.
In the case of an allocation of channel capacity,
the newly-allocated channel is activated in the transmitting
and receiving directions on the master side; however, the
mapping algorithm is not considered. On the slave side, the
channel being newly-set-up is blocked in the transmitting
CA 02212072 1997-07-31
direction and activated in the receiving direction without
being taken into consideration by the mapping algorithm.
The master can then send a reconfiguration signal to the
slave on the channel. After receiving the reconfiguration
signal, the slave sends an acknowledgment signal to the
master via the newly-allocated channel, and in the receiving
direction the channel is taken into consideration by the
mapping algorithm on the slave side. After the
acknowledgement signal has been received by the master, the
newly-set-up channel can then be tied into the mapping
algorithm of the master at the transmitting and receiving
end, and the transmission of useful data can be begun.
After receiving the first useful data, the newly-allocated
channel is also tied into the mapping algorithm at the
transmitting end by the slave.
To carry out the method, a base station may be
used that acts as master with a bidirectional interface
which is connected to the telecommunication network and
which is connected to a mapping module. On the radio side,
the mapping module is connected to a RF module via a data
bus, the mapping module being driven by a protocol
controller connected to the bidirectional interface. In
another embodiment, the mapping module is formed by a
multiplexer on the network side and by virtual connection
modules on the radio side. The user unit acting as slave
comprises a mapping module which is connected to a RF module
arranged at the input of the user unit on the radio side.
On the user side, the mapping module is connected to the
respective terminals of the users. In another embodiment,
the mapping module can be formed by a virtual connection
module on the radio side and by a multiplexer on the user
side.
In the text which follows, the invention will be
explained in greater detail with reference to a preferred
exemplary embodiment. In the drawings:
Figure 1 is a block diagram of the radio system;
CA 02212072 1997-07-31
Figure 2 is a layer model of the radio system;
Figure 3 is a state machine for a base station in
the case of the allocation of channels;
Figure 4 is a state machine for a user unit in the
case of the allocation of channels;
Figure 5 is a state machine for the base station
in the case of a reduction in channels; and,
Figure 6 is a state machine for a user unit in the
case of a reduction in channels.
The radio system 1 comprises a base station 2 and
a plurality of independent user units 3. The base station
2 comprises a bidirectional interface 4 connected to the
telecommunication network, a protocol controller 5, a
mapping module 6, a data bus 7 and a radio-frequency (RF)
module 8 with an air interface 9. The mapping module 6
comprises a multiplexer 10 and a multiplicity of virtual
connection modules 11. The user units 3 comprise a radio-
frequency (RF) module 12 and a mapping module 13. Like the
mapping module 6 of the base station 2, the mapping module
13 of the user unit 3 comprises a multiplexer 14 and a
virtual connection module 15. At the user end, the
multiplexer 14 is connected to subscriber interface circuits
16 of connected terminals. The protocol controller 5 is
connected bidirectionally to the bidirectional interface 4.
Furthermore, a signal output of the protocol controller 5 is
connected to the control inputs of the multiplexer 10 and of
the virtual connection module 11. The RF module 8 is
connected on the radio side to the virtual connection
modules 11 of the mapping module 6 via the data bus 7. The
RF module 12, equipped with an air interface 17, of the user
unit 3 is connected to the virtual connection module 15 of
the mapping module 13.
In the radio system 1 shown, the base station 2 is
the higher hierarchy level in relation to the user units 3
and therefore assumes the function of the master. The data,
arriving, for example, from the telecommunication network,
CA 02212072 1997-07-31
are present at the data inputs of the multiplexer 10 of the
mapping module 6 via the bidirectional interface 4. The
protocol controller 5 derives appropriate control signals
for the multiplexer 10 and the virtual connection modules 11
of the mapping module 6 from the data. The multiplexer 10
allocates the data streams for a respective active user unit
3 to one of the virtual connection modules 11 without
specifically identifying the individual connections of the
user unit. Such a connection module 11 is generated and
continues in existence for as long as the user unit 3 is
active. The virtual connection module 11 correlates with
the respective virtual connection module 15 of the mapping
module 13 of the user unit 3. In this process, the data are
forwarded by the virtual connection module 11 to the RF
module 8 via the data bus 7 and transmitted to the air
interface 17 of the RF module 12 of the respective user unit
3 via the air interface 9. The RF module 12 forwards the
received data to the virtual connection module 15. The data
are then present at the input of the multiplexer 14 from
where they are multiplexed to the respective active
terminals of the users by the multiplexer 14.
The requirement for dynamic channel allocation is
triggered by the protocol controller 5. The control signal
of the protocol controller 5 triggers the multiplexer 10 and
the virtual connection modules 11. This controlling
involves the transfer of an updated link table for the
multiplexer 10. The link table contains the active user
links which must be allocated to a certain virtual
connection module 11. The controlling also involves the
transfer of an updated channel table for the virtual
connection modules 11. The channel table contains the
channels which must be utilized for transmitting the data.
Furthermore, the protocol controller 5 controls the virtual
connection module 15 and the multiplexer 14 in the user unit
3 for the purpose of dynamic channel allocation. This
controlling involves the transfer of an updated channel
CA 02212072 1997-07-31
table which corresponds to the channel table of the virtual
connection module 11 of the base station 2 for the virtual
connection module 15. The multiplexer 14 is controlled by
transfer of an updated user link table which contains the
information regarding which link is to be allocated to which
of the subscriber interface circuits 16. In the case of an
active user unit 3, the control is transferred to the user
unit 3 via the channel or channels already set up. If there
are no established channels as yet, the instruction for
generating a virtual connection module 15 and a shortened
channel table containing only one channel is transmitted to
the user unit 3 via a control channel. In accordance with
the procedure for the dynamic channel allocation which will
be explained in greater detail in the text which follows,
all control instructions are transmitted, and, if necessary,
other channel allocations are also carried out, via this
channel. The virtual connection modules 11, 15 utilize the
provided channels as total capacity which is not split in a
user-related manner. For this purpose, a mapping algorithm
exists according to which the data arriving serially from
the multiplexers 10, 14 are mapped into the channels without
reference to user subscriptions or, respectively, according
to which the mapping must be cancelled and a serial data
stream must be generated again in the opposite direction to
the multiplexers 10, 14.
Figure 2 shows the radio system 1 as a layer
model. The protocol controller 5 of base station 2 is an
entity of layer 3 and derives the necessity for changing
transmission capacities from signalling for the setting-up
and clearing-down of connections. As a result, the
instructions for dynamic channel allocation to the virtual
connection modules 11 of the base station 2 and the virtual
connection modules 15 of user units 3 are derived. The
virtual connection modules 11, 15 implement layer-2 and
partly layer-l tasks.
CA 022l2072 l997-07-3l
In the text which follows, the process of dynamic
channel allocation is described in detail with reference to
state machines which are a component of the virtual
connection modules 11, 15.
Figure 3 shows the state machine 20 for a virtual
connection model 11 of base station 2 for the case of an
allocation of channel capacity. The static mode 21 involves
the utilization of 0 5 n 5 nmax channels between base station
2 as master and user unit 3 as slave. In the case where n
> O, the data are transmitted via the allocated and
established channels in accordance with the mapping
algorithm. In this case, n can assume values of between
fractions of a channel up to nmaX ~ 1. Following the
instruction for increasing the channel capacity 22 which is
contained in the channel table transferred by the protocol
controller 5, the transition to the state "Testing the
instruction" 23 takes place. The instructed increase in
channel capacity 22 can assume values of mmin 5 m 5 1. If
the requested increase in channel capacity 22 is greater
than 1 channel, a plurality of state machines 20 become
active in parallel. A negative test result 24 leads back to
the static mode 21 with the starting parameters if the
allocated channel is not available or the capacity of the
virtual connection module 11 would be exceeded. In these
cases, an alarm indication signal is sent to the protocol
controller 5. If the test result is positive 25, the
transition to the state Phase 1 of the dynamic channel
allocation 26 takes place. The state "Phase 1 of the
dynamic channel allocation" 26 involves activation of
transmission and reception in the newly-allocated channel,
although the channel is not yet utilized jointly with
already-established channels in accordance with the mapping
algorithm. In the transmitting direction, an allocation
information item (Info 1) acting as a reconfiguration signal
iS sent out to user unit 3 in the channel to be set up. In
the direction of reception, an acknowledgment for the
CA 02212072 1997-07-31
allocation information item (Info 1) is expected. The state
"Reception without acknowledgement" 27 again leads to the
state "Phase 1 of the dynamic channel allocation" 26, and to
a retransmission of an allocation information item (Info 1).
A transition 28 to static mode 21 with the starting
parameters can take place if no acknowledgement has arrived
for a presettable number of cycles or a presettable time, or
a received acknowledgement is invalid. In these cases, an
alarm indication signal is sent to the protocol controller
5. The arrival of an acknowledgement 29 leads to the state
~Phase 2 of the dynamic channel allocation" 30. The state
"Phase 2 of the dynamic channel allocation" 30 involves that
the newly-allocated channel is used jointly with the
already-established channels in accordance with the mapping
algorithm. In the transmitting direction, the transmission
of useful data is begun. In the direction of reception, the
failure of the acknowledgement to appear is expected.
Reception with acknowledgements 31 again leads to state of
"Phase 2 of the dynamic channel allocation" 30. The
acknowledgements are ignored by the mapping algorithm. When
the acknowledgements 32 fail to appear, transition to static
mode 21 with the new parameters takes place.
Figure 4 shows the state machine 40 for the
virtual connection module 15 of the user unit 3 for the case
of the allocation of channel capacity. The static mode 41
state involves utilization of 0 5 n ~ nmaX channels between
the base station 2 as master and the user unit 3 as slave.
In the case where n ~ 0, data are transferred via the
allocated and established channels in accordance with the
mapping algorithm. In this process n can assume values of
between fractions of a channel up to nmaX >> 1. With the
instruction for increasing the channel capacity 42 which is
contained in the channel table transferred by the protocol
controller 5, the transition to the state "Testing the
instruction" 43 takes place. The instructed increase in
channel capacity 42 can assume values of from mmin ~ m ~ 1.
CA 02212072 1997-07-31
If the demanded increase in channel capacity 42 is greater
than 1 channel, a plurality of state machines 40 become
active in parallel. A negative result of the test 44, if
the maximum transmission capacity of the user unit 3 would
be exceeded, again leads to static mode 41 with the starting
parameters. In this case, an alarm indication signal is
sent to the protocol controller 5. In the case of a
positive result of the test 45, a transition to the state
"Phase 1 of the dynamic channel allocation" 46 occurs. The
state "Phase 1 of the dynamic channel allocation" 46
involves that the channel being set up is blocked in the
transmitting direction, and receiving is activated and an
allocation information item (Info 1) is expected. The
channel is not yet being used jointly with already-
established channels in accordance with the mappingalgorithm. Reception without an allocation information item
47 (Info 1) again leads to the state of "Phase 1 of the
dynamic channel allocation" 46. A transition 48 to static
mode 41 with the starting parameters can take place if no
allocation information item (Info 1) has arrived for a
presettable number of cycles or a presettable time or a
received allocation information item is invalid. In these
cases, an alarm indication signal is sent to the protocol
controller 5. The arrival of an allocation information item
49 leads to the state of "Phase 2 of the dynamic channel
allocation" 50. The state of "Phase 2 of the dynamic
channel allocation" 50 involves that an acknowledgement is
sent to base station 2, and the receiving direction is
utilized jointly with already-established channels in
accordance with the mapping algorithm. The reception of
further allocation information items 51 again leads to the
state of "Phase 2 of the dynamic channel allocation" 50 and
a retransmission of an acknowledgement. The allocation
information items are ignored by the mapping algorithm. The
failure of a further allocation information item 52 to
appear leads to the static mode 41 with the new operating
CA 02212072 1997-07-31
parameters. This ensures that, in the case of the
allocation of channels, pseudo data are not included in the
mapping in a receiver before the beginning of a transmission
of useful data, and that the beginning of the transmission
of useful data is included in a defined manner in the
mapping without loss of data.
Figure 5 shows the state machine 60 for a virtual
connection module 11 of the base station 2 for the case of
a withdrawal of channel capacity. Static mode 61 involves
the utilization of nmin 5 n ~ nmaX channels between base
station 2 and the user unit 3. The data are transmitted via
the allocated and established channels in accordance with
the mapping algorithm. With the instruction for withdrawing
channel capacity 62 which is contained in the channel table
transferred by the protocol controller 5, the transition to
the state "Testing the instruction" 63 takes place. If the
required reduction in channel capacity is greater than 1
channel, a plurality of state machines 60 will act in
parallel. A negative result of the test 64, if this is
justified by plausibility checks not directly relevant for
the method, again leads to static mode 61 with the starting
parameters. In this case, an alarm indication signal is
sent to the protocol controller 5. A positive result of the
test 65 leads to the state of "Phase 1 of the dynamic
channel allocation" 66. The state of "Phase 1 of the
dynamic channel allocation" 66 involves that the
transmitting direction is no longer used in accordance with
the mapping algorithm in the channel to be deactivated. In
the transmitting direction, a withdrawal information item
(Info 2) is sent in the channel to be deactivated. In the
receiving direction, the channel remains tied into the
mapping algorithm. An acknowledgement for the withdrawal
information item is expected and is ignored by the mapping
algorithm. "Reception without acknowledgement" 67 again
leads to the state of "Phase 1 of the dynamic channel
allocation" 66. A withdrawal information item is again
CA 022l2072 l997-07-3l
transmitted. A transition 68 to static mode 61 with the
starting parameters can take place if no acknowledgement has
arrived for a presettable number of cycles or a presettable
time, or if a received acknowledgement is invalid. In these
cases, an alarm indication signal is sent to the protocol
controller 5. The arrival of an acknowledgement 69 leads to
the state of "Phase 2 of the dynamic channel allocation~' 70.
The state of "Phase 2 of the dynamic channel allocation" 70
involves that the channel to be deactivated is deactivated
at the transmitting end. Further reception of
acknowledgement 71 again leads to the state of "Phase 2 of
the dynamic channel allocation" 70. Incoming
acknowledgements are ignored by the mapping algorithm. The
failure of acknowledgements 72 to appear leads to the static
mode 61 with the new parameters.
Figure 6 shows the state machine 80 for a virtual
connection module 15 of user unit 3 for the case of a
reduction in channel capacity. Static mode 81 involves the
utilization of nmin S n 5 nmax channels between base station
2 and user unit 3. The data are transmitted via the
allocated and established channels in accordance with the
mapping algorithm. With the instruction for reducing the
channel capacity 82 which is contained in the channel table
transferred by the protocol controller 5, the transition to
the state "Testing the instruction~' 83 takes place. The
instructed reduction in channel capacity can assume values
of from mmin S m s 1. If the required reduction in channel
capacity is greater than 1 channel, a plurality of state
machines 80 will act in parallel. A negative result of the
test 84, if this is justified by plausibility checks not
directly relevant for the method, leads to the static mode
81 with the starting parameters. In this case, an alarm
indication signal is sent to the protocol controller 5. A
positive test result 85 leads to the state of "Phase 1 of
the dynamic channel allocation" 86. The state of "Phase 1
of the dynamic channel allocation" 86 involves that a
CA 02212072 1997-07-31
withdrawal information item (Info 2) is expected in the
channel to be deactivated. The channel is still being used
jointly with the already-established channels in accordance
with the mapping algorithm. Reception without a withdrawal
information item 87 (Info 2) again leads to the state of
"Phase 1 of the dynamic channel allocation" 86. A
transition 88 to static mode 81 with the starting parameters
can take place if no withdrawal information item has arrived
for a presettable number of cycles or a presettable time, or
if a received withdrawal information item is invalid. In
these cases, an alarm indication signal is sent to the
protocol controller 5. The arrival of a withdrawal
information item 89 leads to the state of "Phase 2 of the
dynamic channel allocation" 90. The state of "Phase 2 of
the dynamic channel allocation" 90 involves that an
acknowledgement is sent to the base station 2. The
reception of further withdrawal information items 91 again
leads to the state of "Phase 2 of the dynamic channel
allocation" 90, and an acknowledgement is sent again. The
withdrawal information items are ignored by the mapping
algorithm. The failure of the withdrawal information item
92 to appear leads to the static mode 81 with the new
parameters. This ensures that when channels are withdrawn,
the end of the transmission of useful data is included in a
defined manner in the mapping without loss of this data, and
pseudo data are not included in the mapping after the end of
a transmission of useful data in a receiver.
