GB2547724A - Adapting reference signal density - Google Patents

Abstract

A method of control for a base station 21 in a wireless communications system wherein a signal 213 is transmitted which contains pilot signals for use in channel assessment by mobile stations 20. Data related to the position of the base station 21 relative to other base stations is used to make a selection from a predetermined set of reference signal distribution patterns. The density of the reference signals is adjusted to the size of the cell, the cell size being directly related to the cells proximity to other cells in the same system. Each pattern in the set contains a respective signal density and the density may be lower in regions of the transmitted data frame that are remote from the master information block. A network management system (NMS) 22 may be used to hold the topological data in a database 220 and the density adjustment 221/214 may be made by the NMS 22 or the base station 21. The system may include an eNodeB and multiple small cells such as femtocells.

Description

ADAPTING REFERENCE SIGNAL DENSITY

This invention relates to the operation of cellular radio systems with multiple small cells, such as femtocells, typically installed inside buildings where the distances between the base station (eNodeB) and the user terminals are not expected to be greater than about 100m and the maximum speed of the user terminals is not expected to be greater than about 30km/hr.

In the “long term evolution” (LTE) standard, reference signals (also known as pilot signals) are constantly broadcast by the base-stations. The user terminals receiving them use them to assess the properties of the downlink channel. The LTE system uses different kinds of reference signals, sent from different antenna ports and performing different tasks. These include cell-specific reference signals (or common reference signals), user terminal-specific reference signals (or demodulation reference signals), positioning reference symbols and channel state reference signals.

User terminals estimate the downlink channel characteristics in order to perform equalisation before demodulation of the data symbols. If the UE is moving, there will be time and frequency domain effects as follows:

- The time domain effect is caused by multipath reflections, whereby the reflections cause disturbances in the amplitude and phase of the signals. Such reflections come from surfaces near to the UE and more distant from the UE, and the difference in time of arrival of the reflected signals generally increases with high eNodeB transmit powers and longer distances between the eNodeB and UE - The frequency domain effect is caused by Doppler shift, whereby the frequency of the received signal at the eNodeB will be shifted down if the UE is moving away from it and shifted up if the UE is moving towards it.

This combination of multipath reflections and the Doppler effect produce an effect known as Doppler spreading. So that the UE can correct for these impairments, the base station repeats the reference signals at regular intervals of time (typically 0.5ms) and of frequency (typically 100kHz), spreading across the transmitted bandwidth, to allow a UE to detect sufficient reference signals to reconstruct the channel. The reference signals can be thought of as sampling the channel in both the time and frequency domain. The user terminal uses the standard properties of the reference signals to calculate the amplitude and phase correction that should be applied to every non-reference symbol, and it interpolates between the reference signals as part of the calculation.

The standard LTE reference signal pattern is suited to macro-cellular systems where the UE can receive reflections from many kilometres away (a 3km round trip is a delay of approximately 1 millisecond), and still rebuild the time domain aspects of the channel, and where the UE can be travelling at 300km/hr and still rebuild the frequency domain aspects of the channel. The reference signals are transmitted all of the time and comprise about one seventh of the maximum output power of the base station.

For small cells, such as indoor femtocells, both the range and speed of user terminals are much smaller. It is therefore possible to reconstruct the channel in both the time and frequency domain with a very much smaller number of samples or reference signals, which reduce the power requirements of the base station, and increases scope to mitigate reference signal pollution, because each base station is using less spectrum for its reference signals.

Significant power savings and efficiency improvements can be realised by adapting the reference symbol density dynamically in this way. However, for this to be done efficiently, the base station needs to be able to determine what reference signal density is appropriate. One factor in determining a suitable reference signal density may be the degree of difficulty experienced by mobile terminals in acquiring the channel transmitted by the base station.

According to a first aspect of the invention, there is provided a method of controlling a base station in a wireless communications network, wherein the base station is configured to transmit signals, the signals comprising a plurality of data units containing reference signals for use in channel assessment by mobile stations receiving the transmissions, the method comprising the steps of: • receiving data relating to topological data relating to the base station’s position in relation to other base stations; • in accordance with the topological data, selecting a distribution pattern from a predetermined set of distribution patterns of reference signals over the plurality of data units, each distribution pattern having a respective density, • generating a signal configured according to a first protocol for transmission to one or more user terminals, the signal comprising reference signals arranged in the selected distribution and density of plurality of data units.

The invention also extends to a computer program element comprising computer program code to, when loaded into a computer system and executed thereon, cause the computer to perform the steps of the method specified above, and also extends to apparatus configured to perform such a method. A second aspect of the invention provides a wireless communications system comprising a plurality of base stations configured to transmit signals comprising a plurality of data units, and comprising a density selection processing system for selecting the density at which data units in the transmissions of the base stations contain reference signals for use in channel assessment by mobile stations receiving the transmissions, the communications system comprising: • means for receiving data relating to the relative topological positions of the base stations; • density adjustment means for selecting, for one or more of the base stations, a distribution pattern from a predetermined set of distribution patterns of reference signals each distribution pattern having a respective density, and for causing the or each base station to transmit a signal comprising reference signals arranged in the selected distribution and density of plurality of data units.

The signal may include a master information block of a data frame characterising the signal. The reference signal distribution in each of the set of distribution patterns may be arranged to be substantially the same in a region of a data frame adjacent to a master information block, and varies in regions of the data frame remote from the master information block.

The density adjustment means may be comprised in a network management system controlling a plurality of the base stations, or it may be comprised in one or more of the base stations, the base stations having means to retrieve network topology data from a database maintained by a network management system. Alternatively, it may be comprised in one or more of the mobile terminals, the mobile terminals having means to retrieve network topology data from a database maintained by a network management system.

The invention therefore allows adaptive changes to the density of reference signals according to cell size, determined by the proximity of the cell to others in the system. Small cells are typically associated with indoor locations, or venues where users tend to spend some time, where the users are unlikely to be moving at speeds likely to make Doppler shifts significant. Moreover, because the cells are small, the path lengths are small and any echos causing variations in path lengths and therefore time synchronisation also correspondingly small.

An embodiment of the invention will now be described, by way of example and with reference to the drawings, in which:

Figure 1 is a schematic illustrating the physical elements which co-operate in the performance of this embodiment of the invention

Figure 2 is an Illustration of synchronisation signals in a physical broadcast channel containing a master information block (MIB) in an LTE frame according to section 5.2 of TS36.331, release 13 of the 3GPP standard.

Figures 3A, 3B and 3C are enlarged views of part of the frame depicted in Figure 1, illustrating different densities of reference signals.

Figure 4 is a flow chart depicting the process of configuring the reference signal density

Figure 1 depicts the functional elements operable in the physical components which co-operate in this embodiment. It will be understood that the functional elements may be embodied in software or firmware in the physical devices, and the various functions may share physical elements with each other or with other functions not depicted herein. A user terminal 20 is in wireless communication with a base station (eNodeB) 21, which is under the control of a network management system 22. It will be understood that in a typical system several base stations will be managed by a network manager, and several user terminals can be in communication with each base station. Only the functions relevant to the operation of this embodiment are depicted in Figure 1: there would also be standard functions such as power management, coding and decoding, call logging and, in the user terminal, a user interface.

The user terminal functions include an RF interface 200 through which the terminal communicates with base stations such as the one depicted at 21. The functions of the terminal include a synchronisation module 201 which monitors transmissions received over the RF interface 200 from the base station 21, to detect synchronisation data, and controls transmissions from the user terminal such that they are synchronised with the operation of the base station. There is also a channel estimation function 202 which monitors reference signals and control signals received over the RF interface 200 from the base station 21 in order to determine properties such as signal quality, Doppler shift, etc. The results of such analysis are used by a report generation function 203 to generate a report for transmission via the RF interface 200 to the base station 21. In addition, a physical cell ID interference checking function 204 is provided, which also provides an input to the report function 203.

The base station 21 also comprises an RF interface 210 which communicates with user terminals such as the one depicted at 20. The functional components which contribute to the operation of this embodiment include a location reporting system 211 which co-operates with the network management system 22 to determine the physical location of the base station, in particular in relation to other base stations. The reporting system may make use of absolute position measurement systems like GPS, also detection of other wireless transmitters such as other base stations, and movement detectors which may, for example, indicate that the base station is in a moving vehicle.

The base station includes a signal quality reporting function 212 which, in co-operation with the channel estimation function 202 in the co-operating mobile stations (e.g 20) can be used to determine handover decisions, selection of appropriate bandwidth and error checking modes etc. The signal quality reporting function also provides an output to the network manager 22. A further function of the base station 21 is a density adjustment control 214. The base station 21 may control this autonomously in response to the signal quality reports 212, or it may be controlled by instructions received from a density adjustment function 221 of the network manager 22. The density adjustment control system controls the signal generation system 213, specifically by giving instructions on the density of reference symbols to be used in transmission to the user terminals 20. The signal generation system 213 is responsive to this input firstly to generate reference symbols at the required density, and secondly to include a control signal in the master information block (MIB) indicating what density is being used.

The network management system 22 comprises a store 220 of the network topology. This may be maintained manually for planned network changes, such as the erection of a new fixed base station, or a change of channel allocations. However, modern networks include a large number of unstructured changes as users connect and disconnect portable base stations (femtocells) which are not under the direct control of the network operator, and as base stations on moving vehicles such as railway trains move relative to the fixed network. The topology store 220 therefore also uses reports received from the location reporting systems 211 of such base stations 21 to update the topology record.

The network management system also comprises a density adjustment function 221 which receives signal quality and interference reports from the various base stations and returns instructions to their density adjustment functions 214 when a change of reference symbol density is required.

Figure 2 illustrates a typical LTE data frame. Time is represented on the horizontal axis and frequency on the vertical axis. In the frequency domain, the frame has a bandwidth of 3MHz divided into fifteen resource blocks (RBs), each resource block containing twelve subcarriers. In the time domain, the frame has a duration of 10ms divided into ten subframes, each containing two time slots (so twenty slots in total), each subframe containing seven symbols. A master information block 110 contains data defining the channel, in particular the physical cell ID (PCI). This block 110 takes up part of seven resource blocks (labelled 101) over part of the second time slot. An inset to Figure 1 depicts this block in more detail.

Synchronisation symbols 111,112 occur in the first and eleventh time slot, in the same resource blocks as the MIB.

Reference signals are distributed in a regular pattern over the frame, typically two in each time slot in each resource block. The reference signals are shown in Figure 2 as solid elements (three of them indicated at 120), which occur every seventh symbol in the frequency axis (vertical) and every third and fourth symbol alternately in the time axis (i.e symbols 1,4,8,11,15 etc) It will be seen that this pattern continues throughout the frame, including within the block 110 otherwise occupied by the MIB.

Figure 3A is an enlarged view of the area of Figure 1 marked 100, including part of the MIB 110 and the synchronisation symbols 111. Figures 3B and 3C depict the corresponding area of modified blocks in which the density of reference signals has been reduced. In Figure 3B the reference signals have been reduced by 25% and in Figure 3C they have been reduced by 75%. In both cases, the reduction in density only takes place in the resource blocks (frequency bands) 102 that are furthest from the centre frequency. The blocks 101 closest to the centre frequency contain the MIB 110, and the reference symbols in that region are required for the UE to estimate the channel in order that it can decode the MIB. The reference symbol density does not need to be uniform across the non-central symbol area 102, as different UEs can be served from different parts. As depicted in Figures 3B and 3C, the density is reduced further in subcarriers more remote from the MIB.

It is envisaged that a few patterns of reference signals, say 1/2, 1/10, 1/20, 1/100 of the maximum density would be available, and a signalling protocol developed to identify the pattern currently in use by the base station, to which the user terminals would operate. In this embodiment, the MIB contains a message informing the terminals whether it is using a sparse or a full set of cell-specific or other reference signals. The MIB as defined in the LTE standard (section 5.2 of 3GPP TS36.331, release 13) has ten spare bits. Four patterns (including the standard density) can be represented by two binary digits in the MIB. A third digit would allow up to eight patterns.

Referring to Figure 1 and Figure 4, the process operates as follows:

When a User Terminal 20 synchronises with a base station 21 (step 50), the synchronisation function 201 locks onto the Primary and Secondary Synchronisation signals (111,112) transmitted by the base station, in order to deduce the physical cell ID (PCI). After that, the user terminal can make use of the cell-specific reference signals 120 because these reference signals can be descrambled using the PCI. The reason the reference symbols are scrambled for transmission is so that the UE can discriminate between cell-specific reference signals even in areas of overlap between two cells, provided that the PCIs are different.

The user terminal then performs a channel estimation 202 in the region of the Master Information Block (MIB) on the Broadcast Channel (BCH) - step 51. The MIB carries information relating to the bandwidth used by the channel and this is read by the terminal. In this embodiment the MIB includes an indicator of the signal density, and the channel estimation function 202 reads this. From the indicated density, the channel estimation function 202 can determine which symbols in the frame may be used as reference symbols (step 52). It will be understood that, as a symbol location may be used as a reference symbol in some density patterns and not in others, it is necessary for the mobile terminal to identify which pattern is in use in order to determine where the reference symbols are to be found.

Several mechanisms can be used, together or in combination, to set the required density, and these are the subject of the present application and the applicant’s two co-pending applications filed contemporaneously with this one and having the applicant’s references A32871 and A32911.

The reference symbol density can be controlled by a density adjustment function 214 in the base station 21 which is responsive to signal quality measurements 212 to control the signal generation function 213 which generates the data frames to be transmitted over the RF interface 210. In particular the density adjustment function 214 instructs the signal generation system 213 where to place the reference symbols, and to apply the required control signal in the MIB to indicate the density in use. A density adjustment function 221 in the network management system 22 can provide more coordinated control than is possible if each base station acts autonomously, by acting in response to data 220 relating to the deployment topology and reports transmitted from the base station’s signal quality monitoring systems 212 relating to the radio environment (step 56). The topology data 220 can be updated from time to time using a reporting function 211 (step 570), or by manual intervention if the topology changes for example by addition or removal of a cell.

The wireless environment monitoring function 212 monitors signals received over the RF interface 210 and reports changes in quality to the density adjustment function 221 in the network management system 22. The network manager 22 queries the radio environment database 220 periodically and selects a suitable reference signal density (step 57). Changes can be triggered by changes in the topology identified by the cell location database or by the mobility management entity (MME). Such changes would include addition or removal of cells, whether by physical movement of a device or by switching it on or off. As has already been discussed, a higher density is required in conditions where Doppler spreading is to be expected, to ensure that the mobile terminal can lock on to one of them. In particular, where base stations are widely spaced and / or in outdoor locations, there are likely to be more long-path reflections and fast-moving mobile terminals, both of which can contribute to channel-spreading effects, in time and frequency respectively.

Changes in reference symbol density selected by the density adjustment function 221 in the network management system (step 57) are transmitted as instructions to the base station (step 58), where they are used by the density adjustment function 214 to control the signal generation function 213 which generates the data frames to be transmitted over the RF interface 210. In particular the density adjustment function 214 instructs the signal generation system 213 where to place the reference symbols, and to apply the required control signal in the MIB to indicate the density in use.

Another mechanism (steps 53 - 56) is operated by co-operation between terminals and the network. Changes using this mechanism can be triggered if the channel estimation system 202 in a User Terminal 20 is itself able to detect when it is not able to estimate the channel accurately enough, by comparing the BLER in relation to RSRP and RSRQ.

Terminal performance is monitored, for example by detecting an inability to adequately estimate the channel, which can be measured either by the terminal itself or by the network (step 53). One way that the UE or the network can detect when the UE is not able to estimate the channel accurately enough is a high level of block error rate (BLER) when the received signal is strong enough to give a much lower BLER. The symptoms of a high BLER while the RSRP and RSRQ measurements are healthy may indicate an insufficient density of reference signals.

If the error rate is well within acceptable limits the symbol density may be reduced (step 555)

If the UE is failing to estimate the channel, remedial action can be taken, but the cause needs to be identified to determine the appropriate action to take. In particular, a PCI conflict can occur when overlapping cells have the same PCI (physical cell identity), and this can also be a cause of a high BLER, since it would impair the UE from properly estimating the channel as the reference signals are scrambled by another signal using the same PCI. In this embodiment, having identified a high error rate (step 54) the system checks that identity conflict is not the cause (step 55) before the base station is reconfigured to increase the reference signal density (step 56). If a physical cell identity conflict is identified, a self-organising network (SON) process is invoked (step 560) to initiate a change to the identity code, and the monitoring process 53 is repeated.

It will be noted that if PCI conflict is the cause, increasing the density of the reference symbols would be likely to increase the number of such conflicts, thereby exacerbating the problem. It is therefore desirable to determine whether a PCI conflict is the cause before increasing the reference symbol density.

In the absence of any PCI impairment reports, the report generator 203 transmits a report through the RF interfaces 200, 210 to the signal quality report generator 212, allowing the base station 21 to report to the network manager 22 that an increase in density should be attempted (step 56). The network manager’s density adjustment management system 221 can then select a suitable reference signal density (step 57).

When the network manager determines that a change in reference signal density is required (step 57) it passes an instruction to the base station 21 (step 58) causing its density adjustment management system 214 to reduce the number of reference signals generated by the signal generation system 213 according to a predetermined pattern, and to cause the signal generation system 213 to amend the MIB 110 to indicate the pattern of reference signals now in use (step 58). The channel estimation system 202 in each User terminal 20 connected to the base station 21 periodically reads the MIB and so will identify the revised reference signal pattern (step 59), and read the reference signals from the pattern currently being used (step 52).

The process of changing channel density is of course reversible, as changes in network topology or channel estimation values indicate that an increase in density is necessary to maintain signal quality, or that a decrease in density can be achieved without unacceptable impairment in signal quality.

Claims (10)

1. A method of controlling a base station in a wireless communications network, wherein the base station is configured to transmit signals, the signals comprising a plurality of data units containing reference signals for use in channel assessment by mobile stations receiving the transmissions, the method comprising the steps of: • receiving data relating to topological data relating to the base station’s position in relation to other base stations; • in accordance with the topological data, selecting a distribution pattern from a predetermined set of distribution patterns of reference signals over the plurality of data units, each distribution pattern having a respective density, • generating a signal configured according to a first protocol for transmission to one or more user terminals, the signal comprising reference signals arranged in the selected distribution and density of plurality of data units.

2. A method according to claim 1, wherein the signal includes a master information block of a data frame characterising the signal.

3. A method according to Claim 2, wherein the reference signal distribution in each of the set of distribution patterns is substantially the same in a region of a data frame adjacent to the master information block, and varies in regions of the data frame remote from the master information block.

4. Apparatus configured to perform a method in accordance with any of the preceding claims.

5. A computer program element comprising computer program code to, when loaded into a computer system and executed thereon, cause the computer to perform the steps of a method as claimed in any of claims 1 to 3.

6. A wireless communications system comprising a plurality of base stations configured to transmit signals comprising a plurality of data units, and comprising a density selection processing system for selecting the density at which data units in the transmissions of the base stations contain reference signals for use in channel assessment by mobile stations receiving the transmissions, the communications system comprising: • means for receiving data relating to the relative topological positions of the base stations; • density adjustment means for selecting, for one or more of the base stations, a distribution pattern from a predetermined set of distribution patterns of reference signals each distribution pattern having a respective density, and for causing the or each base station to transmit a signal comprising reference signals arranged in the selected distribution and density of plurality of data units.

7. A wireless communications system according to claim 6, wherein the transmitted signal includes a master information block of a data frame characterising the signal.

8. A wireless communications system according to Claim 6 or Claim 7, wherein the reference signal distribution in each of the set of distribution patterns is substantially the same in a region of a data frame adjacent to a master information block, and varies in regions of the data frame remote from the master information block.

9. A wireless communications system according to Claim 6 or Claim 7 or Claim 8, wherein the density adjustment means is comprised in a network management system controlling a plurality of the base stations

10. A wireless communications system according to Claim 6 or Claim 7 or Claim 8, wherein the density adjustment means is comprised in one or more of the base stations, the base stations having means to retrieve network topology data from a database maintained by a network management system.

11 .A wireless communications system according to Claim 6 or Claim 7 or Claim 8, wherein the density adjustment means is comprised in one or more of the Mobile terminals, the mobile terminals having means to retrieve network topology data from a database maintained by a network management system.