This invention is directed to optimizing timing properties for a communication scheme, such as scheduling aspects relating to time division duplex (TDD), TDMA (Time Division Multiplex Access), VVCDMA (Wideband Code Division Multiplex Access) or OFDMA (Orthogonal Frequency Division Multiplex Access)) systems. The invention relates generally to communication systems in which a receiver is controlling the timing of reverse link transmissions to the receiver from a transmitter and wherein predefined transmission timing time slots or opportunities must or should be observed. More particularly, the invention relates to managing and allocating timing properties for uplink communication to a base station to individual user entities.
According to the 3rd Generation Partnership Project (3GPP) a third generation (3G) mobile phone system, Wideband Code Division Multiple Access (WCDMA), is standardized within the International Telecommunication Union (ITU). WCDMA is based on three main units, the Radio Network Controller (RNC), the Node B and the User Equipment (UE). The communication between the Node B and the UE is based on a common timing scheme between the two units. This timing scheme is defined with a timing offset from the main time (reference) in the Node B. The value of this offset is called the chip offset and is defined individually for each UE, The chip offset is a value between 0 and 38 144 chips rounded to the closest 256 chip border. 38 400 chips correspond to 10 ms. Rounding off to the closest 256 chip border is performed to keep the downlink channels orthogonal.
The Enhanced Uplink features in WCDMA is one example of a communication system in which the base station is controlling, via a downlink control channel, the timing of reverse link transmissions to the base station from at least one user entity and wherein predefined transmission timing slots or opportunities must be observed for the user entity and also on the downlink control channel to the user entity. For VVCDMA, Node B schedules uplink transmissions from users via a downlink control channel. The uplink transmission must adhere to a discrete timing scheme in order to operate in compliance with the HARQ process mentioned above on the up-link. In WCMDA, Node B schedules uplink transmissions such that HARQ process downlink control signalling is carried out in a timely manner.
When a UE is connected to the network, the RNC is responsible for choosing and as signing a chip offset for the UE. Choosing different chip offsets for different UE's will spread the processing load in the Node B over time, so that load peaks can be avoided. When a UE is connected to additional Node Bs, the UE will inherit the timing from the connections it previously has made. The standard supports a reconfiguration of the chip offset, but only for one step. i.e. +−256 chips, which corresponds to steps of 0.0667 ms.
It is noted that e.g. a cell range of 150 km gives a round trip time of around 1 ms. Hence, in practical cell applications the maximum chip offset can easily accommodate the variations in UE locations.
In release 6 of the WCDMA a specification, the standard was extended to include the set of features denoted the Enhanced Uplink (EUL) that increase the uplink speed and reduce the delays in the uplink. EUL is based on an uplink transport channel, the Enhanced Dedicated Channel (E-DCH).
The E-DCH uses soft combining and Hybrid Automatic Repeat Request (HARQ) process which implies that the Node B is transmitting Acknowledgements (ACK) and Negative Acknowledgements (NACK) back to the UE to indicate to the UE if it's transmission on the E-DCH was successful or not. If the UE does not receive an ACK it has to retransmit the data. The standard defines exactly when relative to the original transmission, the ACK or NACK shall be transmitted and when the UE shall retransmit its data, since the HARQ handling and the processing of a HARQ process must be ready within a certain time. Consequently, there is an implicit processing maximum delay requirement for the Node B and the UE.
There is a demand for higher throughput in e.g. VVCDMA systems. Higher throughput gives the operator of the network the possibility to serve more users and hence render the system more profitable. In order to achieve a higher throughput, more advanced receivers are introduced in the base station. These advanced receivers use technologies like for instance Interference Cancellation (IC), Multi-User Detection (MUD) and Generalized Rake Receiver (GRAKE), all complex units that require extra processing power in the base station for the uplink. Processing capacity is expensive and it is a problem to fit in all the new technologies. This together with the maximum processing delay requirements in 3GPP constitutes a tough requirement on the Node B that can be hard to fulfil.
It is a first object of the invention to set forth a method for utilizing computational resources in a receiver or a base station more effectively.
This object has been achieved by the subject matter according to claim 1, according to which there is provided
A method for adjusting a timing offset value for a receiver, the receiver being adapted to controlling and receiving reverse radio link transmissions from a transmitter according to predetermined response time requirements. The method comprising the steps of
According to a further aspect of the invention there is provided a method for adjusting a timing offset value for a radio link of a base station, the base station being adapted to receive uplink transmissions from a user entity, UE, on the radio link according to predetermined response time requirements. The method comprises the steps of
According to a first aspect of the invention, the offset value—in some applications also denoted as the chip offset—is used actively to take control of the timing budget in the Node. This can give the base station more available processing time without introducing more system delay or any other drawbacks on the network or in the user entity. This allows the Node B to use more advanced receivers that are needed to be able support more UEs and higher bit rates. It will also make it easier to be able to support time aligned transmissions in the uplink. The demand for more processing capacity in the Node B will also be lowered.
There is moreover provided a radio network control apparatus being adapted for adjusting a timing offset value for a radio link of a base station, the base station being adapted to receive uplink transmissions from a user entity on the radio link according to predetermined response time requirements.
The radio network control apparatus, comprises a time budget map, being adapted for
Further advantages of the invention will appear from the following detailed description of the invention.
As will be understood from the above description of the prior art, in order to obtain synchronisation, the base station imposes a timing regime to which all user entities of the cell must adhere.
As regards enhanced uplink, EUL, there apply specific timing requirements for the E-DCH channel for the user entity, E-DCH at UE (E-DCH@UE) and the E-DCH channel at Node B (E-DCH@NodeB).
In
The following notions will also appear from the above specifications and shall not be dealt with further in this document:
BFN, Base station Frame system clock, which is related to a System Frame Number, SFN, P-CCPCH, E-HICH@NodeB, E-HICH@UE DL, DPCH@NodeB DL, DPCH@UE, E-DCH@UE, E-DCH@NodeB, Tcell.
Tcell is defined in 25.402, Tn,E-HICH in 25.211
The following parameters related to the above notions are shown in FIG. 2:
N*TTI, Tn,E-HICH, Tprop, Tn,DPCH, TUE, TNodeB.
According to the invention, it is reckoned that from 3GPP 25.211, a Node B and UE delay. T_UE and T_NODE B, can be calculated in the following manner:
T
NodeB
=N*TTI−T
prop
−T
HICH
−T
UE
−TTI−T
propτn,E-HICH+Tprop+THICH+TUE=τn,DPCH+Tprop+(1024±148)/3840+10 ms →TUE=τn,DPCH+(1024±148)/3840+10 ms−THICH−τn,E-HICH
Definitions:
Tn,DPCH=Tn×256 chip, Tn ε{0, 1, . . . , 149}. Equals Frame Offset+Chip Offset rounded to closest 256 chip boarder, ref 25.402
TTI=2 or 10 ms
N=4 (10 ms TTI) or 8 (2 ms TTI), Number of HARQ processes.
N*TTI=the overall roundtrip time for the HARQ protocol
TUE=UE processing time.
TNode B=Node B processing time (from last bit in to first bit out)
Tcell=a cell specific offset
Tprop=propagation delay in air interface.
THICH=the length of E-HICH
The table below shows the delay for the UE and Node B, using the calculations above, in case of 50 and 200 km cell radius.
From the table above, it shows that differences may appear for the delay which can be as low as T_NODE_B MINIMUM and as high T_NODE_B_MAXIMUM. It can be concluded that the value of the chip offset affects the maximum allowed Node B processing time and the maximum allowed UE processing time.
As mentioned in the background as explained in relation with
However, instead of arbitrarily choosing a chip offset which can satisfy the basic response time requirements for a cell system, the selection of a chip offset/response timing according to the invention is further selected so as to optimize e.g. computational resources in the base station.
For an EUL application, this means that when Node B has received its last bit in a E-DCH transmission, Node B must be able to process according to the currently selected receiver algorithm until the predetermined E-HICH transmit time occurs. If the RNC were to select a chip offset without any “interaction” with Node B, a limitation would occur as to which receiver algorithms that could be used. This limitation is aggravated with the complexity of the receiver.
The time budget map depends on whether a TTI of 10 ms or a TTI of 2 ms applies. Note that the UE processing time budget increases when the Node B processing time budget is decreasing. The difference between two peaks in the plots is 7680 chips or 2 ms.
According to one embodiment of the invention, shown in
According to one aspect of the invention, the scheduler of Node B is set to operate so as to support an allowed given minimum processing time, 112, in which the receiver in Node B is ‘guaranteed time’ to perform the necessary computations for decoding and processing. If this minimum time is not available, there may not be enough time to process the up-link transmission and proving the required response. Advantageously, the supported minimum processing time 112 is set sufficiently high such as to accommodate the various computational processes required, that is, set with the latter requirements in mind.
Still further and according to the invention, a time budget map 116—according to which the available base station processing time in relation to the assigned chip offset, CO, for at least one given user entity for rather location of the user entity) in the cell is given—is assessed using the calculations mentioned above. One such time budget map 116 for a first given user entity at a given location is shown in
For a second user entity being arranged further away from the first user entity, there exist another time budget map 116′. This latter time budget map 116′ appears parallel displaced downwards in relation to the former time budget map 116.
According to a further aspect of the invention, the chip offset is determined and assigned to a given user entity so that the processing time in the Node B can be predicted (predetermined) and so that there will be enough time in the Node B for the processing in the receiver (guaranteed). Note, that the available time in the UE for processing will decrease if the response time is increased in the Node B, but that it does not matter for the UE since the minimum time the UE has to support is not changed—i.e. the requirement on the UE stays the same even though the available time in the Node B can be increased.
According to one aspect, the following procedure, illustrated in
Adjusting a timing offset value, CO,—such as a chip offset value in a WCDMA system—for a radio link of a base station, Node-B, the base station being adapted to receive up-link transmissions from a user entity, UE, on the radio link according to predetermined response time requirements. The method comprising the steps of
According to one embodiment, a generalized time budget map 116 is established in step 11 for one representative user entity. This will suffice for some applications, for instance small cells.
For instance, the step of determining the optimized chip offset, step 15, may be undertaken after a new use entity is seeking connection to a base station, step 13. Optionally, the radio link set-up may be initiated in step 13.
Optionally, a new offset value may be re-assigned—step 17—after a delay, step 21, adapting for a changed location of a user entity or for obtaining a different distribution of offsets, whereupon a re-configuration 19A, of the radio link set-up for the user entity, UE, is performed.
According to another embodiment, the time budget map is established for various distributed user entities taking into account the propagation delay.
At step 17, NBAP RL SETUP (i.e. at the initial setup for the UE) to Node B and Radio Resource Control (RRC) signalling to the UE, only the chip offset values that give the Node B the longest possible processing time are used, i.e. the values 10, 40, 70, 100 or 130 (in units of 256 chips) are used as candidates for offset assignment. These values will guarantee that the Node B will have as much time for processing as possible for each radio link.
According to a further aspect of the invention, and as illustrated in
In one preferred embodiment, the method moreover comprises the step of optimizing the offset value, CO, step 15, involves comparing the time budget map, TBM, with a minimum supported base station processing time, 112, for the base station. The assignment, 15, of the timing offset value is optimized such that base station processing time is above the minimum supported base station processing time, 112.
Accordingly, a number of recurrent peaks, 125, and their associated timing offset values are estimated in the time budget map, TBM.
The time budget map, TBM, may constitute a generalized map which pertains to a predetermined location, or a set of locations, in a cell associated with the base station.
Alternatively, a plurality of time budget maps, 116, 116′ is provided. Each respective time budget map is established for timing conditions which applies for the given user entity or given set of user entities.
According to the invention there is provided a radio network control apparatus being adapted for adjusting a timing offset value, CO, for a radio link of a base station, the base station being adapted to receive uplink transmissions from a user entity, UE, on the radio link according to predetermined response time requirements, the radio network control apparatus, RNC, comprising a transmit bit map unit, TBM_U, The transmit bit map unit is being adapted for
It should be noted that the radio network control apparatus could be a UTRAN (UMTS (Universal Mobile Telephony System) Terrestrial Access Network, Radio Network Controller interacting with a Node B, but the radio network control apparatus could also be implemented as a part of the radio base station, such that the two entities the radio network control apparatus and the radio base station are located in the same entity (as sub entities) or in a common housing.
According to a second embodiment of the invention, timing drift is considered.
Timing drift appears because a second radio link may be set up in another Node B, e.g. due to the user entity being in soft handover. The downlink DPCH arrival can start to differ due to the fading and drift in Node B clocks (BFN). The uplink transmission point in time can then start to drift.
From
Therefore, typically advantageous for larger cells, it is estimated how much drift that is supported, e.g. how many “chips” can the timing “frame offset”+“chip offset” be reconfigured without the processing time falls below the minimum supported processing time 112. From
In other words, an initial chip offset allocation, 123, is chosen situated substantially a predetermined number offsets values x from the point, 121, where the minimum supported processing time, 112, is found and substantially a predetermined number of offset values x from the peak, 125. In other words, an offset value is chosen substantially centred between a position where the peak is found and a position where the minimum supported processing time is found.
At NBAP RL SETUP (i.e. at the initial setup for the UE) to Node B and RRC signalling to the UE, only the chip offset values that substantially guarantee the Node B the longest possible processing time with regard to possible timing drift, i.e. the values 10+x (corresponding to peak 125+x), 40+x, 70+x, 100+x or 130+x (in units of 256 chips, where x=max_drift_in_one_direction) are used for assignment. These values will with high likelihood avoid that the processing time for Node B due to timing drift will fall below the minimum supported processing time 112.
Another possibility is to select the timing offset value is such that base station processing time is chosen such as to be arranged at a first given distance (e.g. a distance corresponding to points 125 and 124 in
In an implementation for the enhanced uplink. the radio network control apparatus is being adapted for adjusting the timing offset value, CO, for a radio link of a base station, Node-B, the base station being adapted to receive uplink transmissions from a user entity. UE, on the radio link according to predetermined response time requirements. The radio network control apparatus, RNC, comprises a time budget map unit, TBM_U, and is being adapted for carrying out the method shown in
In
According to a further embodiment of the invention, the actual assignment of offset to an individual user entity on the E-DCH and on the Dedicated Channel (DCH) is performed such that for each new user entity, the actual chip offset is assigned such that the distribution of the offsets for all the user entities are spread evenly. This is done in step 17 of
Such an exemplary distribution has been show in
The spreading could also be performed over a wider time base, such as over different chip offsets of say 10, 20, 40, 80 ms TTI, as the chip offset—as mentioned above—has a range from 0 to 38144 chips (i.e. from 0 to 9.9333 ms). The chip offset allocation between different E-DCH with 10 ms TTI shall be spread evenly, with multiple 2 ms separation. This is also valid for DCH with a TTI of 10, 20, 40 or 80 ms.
The spreading will prevent the Node B from experiencing extreme peaks in the processing, as opposed to a simultaneous offset assignment where all UEs shall be demodulated and decoded simultaneously.
3G Third Generation
3GPP Third Generation Partnership Project
ACK Acknowledgement
DCH Dedicated Channel
E-DCH Enhanced Dedicated Channel
EUL Enhanced Liplink, also called High Speed Uplink Packet Access HSUPA)
GRAKE Generalized Rake Receiver
HARQ Hybrid Automatic Repeat Request
NACK Negative Acknowledgement
NBAP Node B Application Part, i.e. the signalling protocol responsible for the control of the Node B by the RNC
RNC Radio Network Controller
RRC Radio Resource Control
TTI Time Transmission Interval
UE User Entity
UL Uplink
WCDMA Wideband Code Division Multiple Access
Number | Date | Country | Kind |
---|---|---|---|
PCT/EP2009/067339 | Dec 2009 | EP | regional |
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/EP2010/069797 | 12/15/2010 | WO | 00 | 6/18/2012 |