WO1999067971A2

WO1999067971A2 - Method and apparatus for estimating the channel quality in a time division multiple access based network

Info

Publication number: WO1999067971A2
Application number: PCT/SE1999/000970
Authority: WO
Inventors: Daniel Bringby; Patrik Karlsson
Original assignee: Telefonaktiebolaget Lm Ericsson (Publ)
Priority date: 1998-06-22
Filing date: 1999-06-04
Publication date: 1999-12-29
Also published as: AR018914A1; CA2335594A1; WO1999067971A3; BR9911399A; AU4810799A

Abstract

A technique for estimating the quality of an unmeasureable channel in a time division multiple access radio telecommunication network is disclosed. In an embodiment of the invention, the estimate of the unmeasurable channel is based on samples from other time slots or channels on the same carrier frequency in the same cell. By way of example, a plurality of quality measurements, which may include the interference, bit error rate (BER), frame erasure rate (FER), and burst quality (BQ) are performed on associated channels of the same carrier frequency and cell as the unmeasurable channel. The quality is based on the suitable maximum value of the estimated values derived from the associated channels. The method is applicable to either uplink or downlink signal measuring techniques and provides an effective means for estimating intrafrequency channels that are closely correlated.

Description

Method and Apparatus for Estimating the Channel Quality in a Time Division Multiple Access Based Network

BY INVENTORS

Daniel Bringby and Patrik Karlsson

FIELD OF THE INVENTION

The present invention relates generally to wireless telecommunication networks, and more specifically to a technique for providing improved channel quality estimation in such networks.

BACKGROUND OF THE INVENTION

In wireless telecommunication networks, service providers are very interested in providing high quality, reliable services for their customers in today's highly competitive environment. The quality of the perceived service by the end user is directly associated with the quality of the transmitted signals between a mobile station (MS) and its proximately located serving base station (BS). The signal quality can be affected by a number of factors such as path loss, which occurs when the signal becomes weaker due to increasing distance between the MS and the BS when the user moves toward the cell border, for example. Another factor affecting the quality can be log normal fading, which can occur when there are physical obstacles such as hills and buildings that obstruct the signal and causes fluctuations in the received signal strength. Furthermore, multi-path fading is another problem that can occur when the signal takes multiple paths to the receiver, for example, by bouncing off buildings which causes them to be received at the receivers from different directions thereby causing dips in the signal strength. A further significant factor affecting signal quality, which is one of a number of quality parameters that is typically closely monitored by operators, is that from interference. The surrounding noise level may comprise interference generated by mobile stations operating on co-channels in neighboring cells and interfere with channel quality.

In a time division multiple access (TDMA) based network, such as digital advanced mobile phone system (D-AMPS) IS- 136.2, a TDMA channel is limited in the frequency domain and temporal domain. Signal quality measurements are conducted and then transmitted within specific allocated time slot(s) within a given frequency carrier. In the D-AMPS protocol, data is transmitted in bursts of digital packets within a 40 millisecond frame (25 frames/sec) containing six equally sized time slots of 6.67 milliseconds. A "full-rate" digital traffic channel (DTC) uses two time slots for transmitting data associated with speech information in which the channel spacing between channels in the frequency domain is 30 kHz. In the context of the present invention, the term time slot is used interchangeably with channel in reference to the DTC. Enhanced versions of the standard are designed to operate in "half-rate" mode wherein only one time slot is used for speech information thereby doubling carrier capacity.

Shown in Fig. l is an exemplary TDMA frame transmitted in accordance with the D-AMPS protocol wherein which contains six time slots indicated by SI through S6. Speech data is transmitted on time slots SI and S4 for both uplink and downlink transmissions. It should be understood that other TDMA based systems, such as Global System for Mobile Communication (GSM) and Personal Digital Cellular (PDC) may specify different frame lengths and time slot allocations in which the present invention is equally applicable.

In D-AMPS, the signal quality in a particular cell can be estimated by measuring a number of characteristics which are indicative of the level of interference such as bit error rate (BER), burst quality (BQ), frame erasure rate (FER). The estimates are typically performed for each DTC or, for example, each dedicated control channel (DCCH). On downlink signals from the serving BS, signal strength and BER measurements are performed on the speech data by the MS at regular intervals, which are transmitted in associated time slots i.e. SI and S4. In addition, signal strength measurements are performed by the MS on the downlink control channels i.e. digital or analog that are transmitted from BSs of neighboring cells according to a predetermined list of frequencies. On the uplink, the average values of these measurements are periodically reported by the MS to the serving BS at regular intervals on the slow associated control channel (SACCH). The serving BS makes a determination based on the signal quality for making a handoff decision to a more suitable cell. It should be noted that the measurements are not necessarily performed for every frame but may be performed periodically for subsequent frames. Excessive interference may lead the BS into initiating a switchover to another frequency or a handoff to a more suitable cell.

Fig. 2 illustrates a situation where the quality of the connection between a mobile station and its serving base station is affected by a mobile station in a neighboring cell operating on the same DTC. MSI and MS2 are shown having their primary connections, as indicated by reference numerals 200 and 210, to their respective BSs i.e. BSl and BS2 in their respective cells. In the example, both MSI and MS2 are transmitting on the same DTC thus BSl will receive signals simultaneously from both mobile stations i.e. 200 from MSI and 220 from MS2. The signal strength of signal 200 received at BSl from MSI is referred to as the carrier signal strength. Conversely, the signal strength from signal 220 from MS2 is called the interference signal strength. Consequently, the relative interaction between the carrier signal strength and the interference signal strength will affect the quality of the connection between BSl and MSI.

Normally the BER, BQ, and FER measurements are performed when the DTC is busy i.e. when the DTC is carrying digital traffic. On the other hand, interference measurements are typically performed when the channel is idle, i.e. , when the channel is not carrying traffic. The quality measurements are usually performed for a period of time in order to generate a distribution of samples for each DTC. There are situations where the current procedure for estimating interference may yield inadequate results. By way of example, if the DTC is busy most of the time, there may not be a sufficient number of samples associated with the interference to yield a statistically reliable estimate. Similarly, when the channel is idle for long periods, obtaining an accurate estimate of the BER may be difficult thus yielding an unreliable estimate of the quality. Another difficulty can occur when the DTC is blocked i.e. all traffic is halted on channel due conditions affecting the transceiver such as excessive interference, hardware problems, software maintenance or software loading, restarts, etc. Since no measurements can be made while the channel is blocked, the quality cannot be estimated during that time.

Complications in estimating the quality for digital control channels may also occur when mobile stations in the same cell access the DCCH at the same time. The simultaneous access collisions may create an momentary disturbance that is perceived as interference instead of carrier signal strength by the BS, which in turn registers a corresponding drop in quality. It is necessary to exclude access collisions in order avoid skewing the quality estimation. This type of error is typically reviewed and adjusted for manually by a technician in a tedious and time consuming process.

In view of the foregoing, it is an objective of the present invention to provide a technique for providing improved quality estimation without experiencing many of the aforementioned disadvantages.

SUMMARY OF THE INVENTION

Briefly described, and in accordance with an embodiment thereof, the invention discloses a method of estimating the quality of an unmeasurable channel in a time division multiple access radio telecommunication system having a multi-channel configuration and operating within a plurality of carrier frequencies. In the present invention, the estimate of the unmeasurable channel is based on samples from other channels (time slots) on the same carrier frequency in the same cell. In an embodiment, the quality of the unmeasurable channel is estimated by measuring a plurality of signal quality parameters of associated channels in order to obtain a sufficient quantity of measurement samples suitable for statistical analysis. Measurements on, for example, the interference, are performed on the associated channels on the same carrier frequency as the unmeasurable channel. The interference of the unmeasurable channel is based on the maximum value of the interference estimate derived from a statistical evaluation of the measurement samples of the associated channels. In accordance with another aspect of the embodiment, additional parameters such as the bit error rate (BER), frame erasure rate (FER), and burst quality (BQ) are suitable for measurement for which the inventive concept may be applied in estimating other aspects of signal quality.

The present invention provides an effective means of quality estimation of unmeasurable channels by performing intrafrequency channel estimation of correlated channels. The method is applicable to either uplink or downlink signal measuring techniques which can be performed in either the base station, the mobile station or both. These and other advantages of the present invention will become apparent upon reading the following detailed descriptions and studying the various figures of the drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention, together with further objects and advantages thereof, may best be understood by reference to the following description taken in conjunction with the accompanying drawings in which:

Fig. 1 illustrates a TDMA frame as transmitted in accordance with the D- AMPS protocol;

Fig. 2 depicts an interference situation caused by mobile stations operating on the same DTC; - o -

Fig. 3 is a cumulative density function plot showing the level of interference for the ninety percentile region; and

Fig. 4 is a graph showing the correlation between time slots for single and multi-carrier uplink interference.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

In a basic cellular telecommunication system, a mobile switching center (MSC) is linked to a plurality of base stations by a series of digital transmission links. The base stations are geographically dispersed to form an area of coverage for the system. Each base station (BS) is designated to cover a specified area, known as a cell, in which a two way radio communication connection can take place between a mobile station (MS) and the BS in the associated cell. One of the most significant factors that can affect the quality of the connection perceived by the end user comes from interference. Therefore, the ability to accurately estimate the quality, and thus the accurate measurement of the associated parameters, is extremely important in order to provide the high level of service that the user has come to expect.

Measurements indicative of the signal quality such as interference (I), bit error rate (BER), frame erasure rate (FER), burst quality (BQ), and carrier signal strength (C) are typically performed for a period of time in order to generate a sufficient number of samples. When the carrier signal strength is known, an estimate of the interference can be based on the BER, FER, or BQ from predetermined tables relating the parameters to carrier-to-interference ratio C/I. Since the traffic level varies throughout the day on each digital traffic channel (DTC), there will typically be a relative distribution of the samples of each DTC. It is then possible to determine the channel quality by analyzing the sampled measurements that are indicative of inadequate quality such as high levels of interference, excessive BER and FER rates or insufficiently low BQ measurements. The distribution spread can be represented graphically in the form of a depiction of the probability. Fig. 3 is a graphical representation of the cumulative density function for an exemplary set of samples taken for one DTC. It can be seen that ninety percent of the interference samples are equal to or below 17 IU, or conversely, ten percent of the samples are above 17 IU. Where an IU refers to an interference unit and wherein each IU represents dB above -118 dBm. Furthermore it should be noted that the probability point of ninety percent is arbitrarily chosen for convenience and may be set for any chosen value but a percentile level in the range of 90-98% is typically selected. Higher percentile levels more closely represents increasingly worse-case interference levels but, since there are fewer measured samples in this range, there will be increased sensitivity to statistical variations thereby making analysis less reliable.

In the case where the number of valid samples may be insufficiently small or not able to be taken at all, the resulting quality estimation may not be statistically reliable. Such situations may include trying to obtain interference samples when the channel is continually busy, or trying to measure the BER when the channel is primarily idle, or during the period when the channel has been blocked. One approach to measuring the interference on an unmeasurable channel has been described in Almgren et al. , International Publication number WO 96/31989 published on 10/10/96 assigned to the present assignee. The approach described is intended for use with an adaptive channel allocation (ACA) scheme wherein channels in different frequency carriers are grouped together in a procedure where they are dynamically allocated to cells in order to accommodate changes in system use that occur over time. The objective is to obtain satisfactory quality for all of the allocated channels by, in part, estimating the interference on unmeasurable channels by measuring channels on other carrier frequencies. This technique relies on the fact that a substantial correlation exists among the grouped frequencies. But this may not always be the case because not all frequencies are re-used in the same procession, for example, a co- channel cell may not have the same number of transceivers than the current cell thereby altering the grouping configuration. Another drawback is that prioritization algorithms that specify the use of carrier frequencies may vary from cell to cell. In accordance with an embodiment of the present invention, the quality of an DTC of, for example, the interference of unmeasurable channels is estimated by estimating the interference of another channel on the same carrier frequency. By way of example, in a system operating in D-AMPS "full-rate" mode, if a channel (time slots 1 & 4) is continually busy or the channel has been blocked such that an insufficient number of samples have been taken, the interference is estimated by estimating the interference of the other two associated channels i.e. time slots 2 & 5 and time slots 3 & 6. Therefore the interference for the first DTC can be estimated as: c(l&4) ^{= maX}(lDTC(2&5)> IDTC(3&6))

where I_DTC(1&4) is the maximum of either the two channels in order to obtain an estimate of the worst case scenario for interference. If, for example, the carrier-to- interference C/I is to be estimated, it is desirable to use the minimum value of C/I_Dτc_(2&5) ^and C/I_DTC(3&6). This may be represented as:

C/IDTC(1&4) ~ ^min(C/lDTC(2&5)> /IoTC(3&6))

It should be noted that in "full-rate" mode, the interference is typically estimated on a per channel basis and thus measurements are taken over two time slots. In another aspect of the embodiment, if only one other DTC on the same frequency is measurable, that interference estimate is then used, for example: c(l&4) ⁼ IDTC(2&5) ^0Γ IDTC(1&4) ⁼ IDTC(3&6)

Analogously, if an estimation of the interference on a control channel (DCCH) is desired for a situation where there are excessive carrier samples making an estimate of the interference difficult, the other two DTCs on the same carrier frequency may be used. Thus the interference of the DCCH can be estimated as:

IDCCH(1&4) ^{= maX}(lDTC(2&5)> c(3&6))

The aforementioned approach is based on the concept that a substantial correlation exists between channels operating on the same carrier frequency. Investigations performed by those skilled in the art have found that this is indeed the case and wherein the correlation between channels in the same frequency are shown to be notably stronger than those between channels on different frequencies.

Fig. 4 is a graph of the correlation of the uplink interference plotted from data measured at a base station operating with typical traffic activity. Curve 400 shows the interference difference (in dB) between two time slots on the same carrier. Curve 410 shows the interference difference between two time slots on different carriers. Curve 400 shows that at a probability level of 95 percent, there is only approximately 2 dB difference between times slots on the same carrier. In contrast, curve 410 shows approximately a 9 dB difference for the time slots between different carriers.

The present embodiment describes a method of estimating the quality of a TDMA channel by estimating the interference on associated channels on the same frequency, but method is also applicable for effectively estimating other quality characteristics such as BER,, FER, and BQ. Furthermore, the estimation can be performed on both uplink (base station based) and downlink (mobile station based) signals. Separated estimations of the uplink and downlink signals are typically performed in which they are not invariably correlated. One advantageous characteristic of base station estimation is that they may use sectorized antennas whereas mobile stations typically use omnidirectional antennas.

Although the invention has been described in some respects with reference to a specified embodiment and various aspects thereof, variations and modifications will become apparent to those skilled in the art. In particular, the application of the invention to conducting estimations over a single time slot for "half-rate" mode operation may be performed as well. It is therefore, the intention that the following claims not be given a restrictive interpretation but should be viewed to encompass variations and modifications that are derived from the inventive subject matter disclosed.

What is claimed is:

Claims

C L A I M S

1. A method of estimating the quality of an unmeasurable channel in a time division multiple access radio telecommunication system, wherein said system having a multi-channel configuration operating within a plurality of carrier frequencies, said method comprising the steps of:

estimating quality parameters of suitably measurable associated channels within the same carrier frequency and cell as said unmeasurable channel by perfo╧Çning a plurality of measurement samples; and

determining an estimate of the quality of said unmeasurable channel based on the estimation of quality of said associated channels.

2. A method according to claim 1 wherein said telecommunication system is operating in accordance with the digital advanced mobile phone system standard.

3. A method according to claim 1 wherein the step of estimating quality parameters includes measuring interference samples on the associated channels.

4. A method according to claim 1 wherein the step of estimating quality parameters includes measuring any one of bit error rate, frame erasure rate, burst quality, and carrier signal strength of the associated channels.

5. A method according to claim 1 wherein said determining step includes basing said quality estimate of said associated channels on a statistical evaluation of said measurement samples.

6. A method according to claim 1 wherein said determining step includes basing the quality of said unmeasurable channel on the maximum quality estimate of said associated channels. WO 99/67971 j , PCT/SE99/00970 ^"

7. A method according to claim 1 wherein said measurement samples may be performed on any one of uplink and downlink signals.

8. A method of estimating the interference of an unmeasurable channel in a time division multiple access radio telecommunication system, wherein said system having a multi-channel configuration operating within a plurality of carrier frequencies, said method comprising the steps of:

estimating the interference of suitably measurable associated channels within the same carrier frequency and cell as said unmeasurable channel by performing a plurality of measurement samples; and

determining an estimate of the interference of said unmeasurable channel based on the estimation of interference of said associated channels.

9. A method according to claim 8 wherein said telecommunication system is operating in accordance with╬¢he digital advanced mobile phone system standard.

10. A method according to claim 8 wherein said estimating step may further include measuring any one of the bit error rate and frame erasure rate and burst quality together with the carrier signal strength of the associated channels.

11. A method according to claim 8 wherein said determining step includes basing said quality estimate of said associated channels on a statistical evaluation of said measurement samples.

12. A method according to claim 8 wherein said determining step includes basing the interference of said unmeasurable channel on the maximum interference estimate of said associated channels.

13. A method according to claim 8 wherein said interference estimation may be performed on any one of uplink and downlink signals.

14. A system for estimating the quality of an unmeasurable channel in a time division multiple access radio telecommunication network, wherein said network having a multi-channel configuration operating within a plurality of carrier frequencies, said system comprising:

means for measuring quality parameters of associated measurable channels within the same carrier frequency and cell as the unmeasurable channel; and

means for evaluating the quality parameters such that a quality estimate of the unmeasurable channel is determined from the measured quality parameters.

15. A system according to claim 14 wherein the quality estimate of the unmeasurable channel is the interference.

16. A system according to claim 15 wherein the quality parameters of the associated channels is the interference.

17. A system according to claim 14 wherein said quality parameters includes any one of the bit error rate, frame erasure rate, burst quality, and carrier signal strength of the associated channels.

18. A system according to claim 14 wherein said measurement means are located in a plurality of base stations within the network.

19. A system according to claim 14 wherein said measurement means are located in the mobile station.