EP1228581A1

EP1228581A1 - Downlink method to compensate for loss of signal orthogonality in multiuser wireless communication systems

Info

Publication number: EP1228581A1
Application number: EP00964567A
Authority: EP
Inventors: Juha Ylitalo; Marcos Katz
Original assignee: Nokia Oyj
Current assignee: Nokia Oyj
Priority date: 1999-11-08
Filing date: 2000-10-09
Publication date: 2002-08-07
Also published as: AU7548900A; WO2001035547A1

Abstract

A method of compensating for loss of orthogonality in beam transmission in a wireless communications network to improve signal downlink performance. The method includes the determination of two or more transmission directions for transmitting signal beams to a destination. Once transmission directions are determined, time delays associated with transmitting a beam along each determined direction are calculated and a reference direction is obtained as the direction corresponding to a maximum of the calculated delays. Each transmitted signal is delayed by the proper time delay so that multiple beams transmitted to the destination are simultaneously received at the destination.

Description

Downlink Method to Compensate for Loss of Signal Orthogonality in Multiuser Wireless Communication

Systems

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates to a method of compensating for loss of signal orthogonality which occurs during signal transmission over multiuser wireless communication systems.

2. Description of the Related Art

In code division multiple access (CDMA) systems, each user is distinguished and characterized from other system users by a specific and unique signature known as the spreading code. In multiuser environments such as wide band code division multiple access ( CDMA) systems, spreading codes are chosen to fulfill a condition of mutual orthogonality. In other words, the presence of any unwanted user should not adversely affect the signal processing from a particular target user by virtue of the orthogonality of the codes employed in the signal processing. Orthogonalilty is preserved in flat fading channels where the delay spread does not exist or is negligible. However, in frequency selective channels where there exists a considerable delay spread of signatures or spreading codes among multiple users, the orthogonality between co-channel users is lost. The coding sequences typically employed in

WCDMA systems are characterized by non-ideal cross- correlation functions (CCF) which have a low or zero value only for a given phase relation between spreading codes. For other phase relations, the CCF functions are nonzero values. For such phase relations, signals will arrive at a receiving end, such as at a mobile station (MS) at different times. In a typical base station an entire sector is serviced by a single beam and, therefore, there is no selection among possible transmission directions. In multiple beam systems, optimal beam selection with respect to delays can be applied among multiple beams which are formed, for example, using fixed beams or digital beam forming. In other words, a signal intended for a specific user that is transmitted over a multipath channel will undergo different delays. Because the value of the CCF at a given out-of-phase position is typically non-zero and different from position to position, the effect of the different path delays imposed by the radio channel on the transmitted signal will be to destroy the orthogonality between the codes used for signal transmission in the multiple directions, e.g., when transmitting from the base station (BS) to the mobile station (MS). The loss of orthogonality is reflected as a clear deterioration in the link performance between a base station and a mobile station in a wireless communication system.

The orthogonality problem occurs in part due to the delays imposed by the radio channel and its affect on a single user. In other words, replicas of the same desired signal will be received by the receiver (due to multipath propagation) and the signals will not be orthogonal with respect to each other. The other users in the network will (individually) suffer from the same problem, each with its own signals. In addition, there is crosstalk between users, known as multiple access interference. An objective of the present invention is to combat the described orthogonality problems occurring for a given user, without necessarily considering co- channel users who share the common sector. In addition, by employing the method disclosed herein it is possible to reduce or avoid crosstalk interference between users by selecting, whenever possible, convenient directions for signal transmission.

SUMMARY OF THE INVENTION

A method is disclosed for compensating for loss of signal orthogonality occurring between a plurality of beams transmitted to a select destination over a plurality of transmission directions in a wireless communications system. For transmission diversity, each beam in the plurality is transmitted over a corresponding direction and, as a result, the signal associated with each beam will be received at the select and desired destination (e.g. a mobile station) at a different time delay than other received beams in the plurality. This results in deterioration of downlink performance between a base station and the select destination. In accordance with the inventive method, preferred transmission directions are determined from among possible transmission directions. The delays associated with transmitting a beam over each selected direction are then calculated and compared to a reference delay. The plurality of beams are then coded with codes that are mutually orthogonal to codes used for transmitting beams intended for receipt at destinations other than the select destination and each beam in the plurality is delayed by a corresponding calculated delay with respect to the reference beam. The delayed beams are then transmitted along corresponding transmission directions. In this manner, the delayed transmitted beams will be simultaneously received at the destination for correlation.

In a preferred embodiment, the directions corresponding to minimal or no delay spread, minimal produced interference and an even distribution of power in the transmit power amplifiers are selected for directions of transmission.

Other objects and features of the present invention will become apparent from the following detailed description considered in conjunction with the accompanying drawings. It is to be understood, however, that the drawings are designed solely for purposes of illustration and not as a definition of the limits of the invention, for which reference should be made to the appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGs. la and lb illustrate single and double beam downlink transmissions, respectively, between a base station (BS) and a mobile station (MS) ;

FIG. 2 is a graphical representation of the response of a frequency-selective radio channel with a given delay spread;

FIG. 3a depicts graphical representations of a downlink method in accordance with the present invention;

FIG. 3b is a schematic representation of an implementation of the downlink method of FIG. 3a;

FIG. 4 is a graphical representation of a two- dimensional channel impulse response characterizing the radio channel between the BS and MS in which signal paths are illustrated as a function of delay and azimuth direction; and

FIG. 5 is a graphical representation of an auto-correlation function for an m-sequence. DETAILED DESCRIPTION OF THE PRESENTLY PREFERRED EMBODIMENTS

FIG. la depicts a base station (BS) and a mobile station (MS) operating in a radio environment such as a WCDMA system with two clusters of scatterers. As is known in the art, for WCDMA downlink, different system users experience the same transmitted signal. In other words, the composite signal comprising the signals to all the present users is available at the antenna of each particular MS. The clusters are appropriately positioned such that if the BS transmits with a single beam covering an entire sector serviced by the BS, a signal will be received by the MS in two delayed versions. This result is graphically depicted in FIG. 2 which illustrates the two signals arriving at the MS with a mutual delay of Xi between the two received signals. Employing efficient transmission direction diversity, under ideal conditions a conventional Rake receiver will process the two received signals by their corresponding correlators. However, when non-ideal sequences are used for transmission, the loss of orthogonality between different users will decrease the performance of the system due to the non-zero cross-correlation function values, as discussed above. The performance degradation results in increased transmission power and decreased spectral efficiency. In the event a single narrow beam is used for transmission through a 1-tap signal path which has negligible delay spread as shown in FIG. la, the problem of loss of orthogonality is avoided because the single signal is directed to the MS through a single direction shown as direction D₆. This alternative, however, is not optimal because diversity in transmission direction is not effectively exploited, thus resulting in decreased system performance.

FIG. lb depicts a situation in which a signal is transmitted from a BS to a MS through two beams, each transmitted along a separate direction shown as directions D₂ and D₆. In this manner, diversity of transmission is exploited due to the use of two separate orthogonal beams. However, the problem of loss of orthogonality between the two separate beams will, as discussed above, detrimentally affect link performance between the BS and MS.

In the case of a large delay spread among beams transmitted over multiple resolvable paths, the magnitude of the orthogonality problem becomes increased which leads to unacceptable levels of crosstalk among multiple users in communication with a common BS . If L transmission directions or paths are available, there are L-l out-of-phase or delayed nonorthogonal signal components which will affect the processing of the received signal.

It has been discovered that the loss of orthogonality among a plurality of orthogonal beams transmitted to a user (e.g., to a MS, mobile phone or other receiving device) over a plurality of directions can be compensated for by selecting appropriate transmission directions and delaying the plurality of signals relative to a select signal so that the plurality of beams formed from the signals are received by the MS at substantially the same time. For example, FIG. 3a graphically depicts the transmission of two orthogonal beams (Beam 1 and Beam 2) intended for receipt by a particular MS and which are generated from a common signal and transmitted along two directions. The signal is formed into beams by a beamformer, as is known in the art, for transmission by a BS . As shown in FIG. 3a, a transmitted signal formed into Beam 1 which is transmitted over direction D₅ is directed to an intended recipient (i.e. to an intended receiving MS) and is received at a time represented by nT, and the signal formed into Beam 2 is transmitted over a direction D₂ and is received by the MS at a later or delayed time, shown as nT+τ, wherein T represents the chip duration or a multiple or sub-multiple thereof, n is an integer and τ is the delay time. Note that the time delay τ can be a fraction of T and therefore, the present method can combat the detrimental effects of inter-path interference. In certain cases, the time delay τ can be arbitrarily small or zero. By adding the delay amount experienced by Beam 2 to Beam 1, i.e. the relative time delay between the receipt of both beams by the MS, both beams will be received by the MS at the same time. This will, in effect, maintain the orthogonality condition between the two beams while allowing the use of beam diversity. In a multi-user system where signals other than the select signal are transmitted by the BS, the downlink performance for the select signal will be improved provided that the transmission code for the select signal is mutually orthogonal to the transmission codes used for the remaining signals. As used herein, the terms "first plurality of beams" and "first plurality of signals" refer to signals intended for receipt by an intended recipient or receiving device, and the terms "second plurality of beams" and "second plurality of signals" refer to signals not intended for receipt by the intended recipient. Delay compensation in accordance with the inventive method may be accomplished, as shown in FIG. 3b, by the addition of a delay in the amount of τi to the signal applied to Beam 1 which, as shown, is directed to the appropriate input port of the beamformer for transmitting Beam 1 in direction D₆.

The inventive method disclosed herein requires the selection of transmission directions to be used for transmitting the generated beams to the intended destination. The directions are selected based upon associated delay spreads present for each beam. In particular, directions resulting in minimal or no delay spread for a transmitted beam are used for transmission. In principle, the approach does not rely on a feedback channel from MS to BS . Directions of transmission can be selected by the BS solely from uplink measurements since the uplink and downlink channels are reciprocal in the long-term average sense. However, in frequency division duplex (FDD) systems, such a feedback measurement could provide improved results, however, with increased system complexity.

Rather than have the BS perform the calculations required for direction selection and delay, the MS can participate in or perform these functions. This requires use of an identifier on each beam, such as mutually orthogonal pilot/training sequences or spreading codes. Based on the channel impulse response measured in each direction, the MS will inform the BS of the preferred directions and delays for transmission.

With reference to FIG. 4, the delay spreads as measured by a base station from the channel impulse responses of each beam are plotted as a function of time delay or phase displacement versus azimuth or angular direction of signal paths. The directions having minimal delay spread are shown as directions D₄, D₇ and D_i3. In the event several potential directions are available, preferred directions among the available directions are selected based on additional criteria, such as the whitening of generated interference and the even distribution of power in the plurality of power amplifiers used by the base station. In some favorable situations, beam hopping can also be applied in order to achieve more effective interference whitening.

In instances where the channel profile in the time delay domain and angle domain are complex and the problems of loss of orthogonality have a considerable impact on system performance, the distribution profile (such as is shown in FIG. 4) should be reduced, whenever possible, to a simple or canonical case (such as is shown in FIGS. 1 and 2) where two or more appropriate transmission directions are selected. Once the appropriate transmission directions are determined, the relative delay compensation for each transmitted beam/direction pair is calculated. For a number of directions of N > 2, the delays of signals corresponding to N-l directions are compensated. This is accomplished by calculating the delay associated with each beam for a corresponding direction relative to a reference direction and delaying each beam by its corresponding calculated delay, so that the beams arrive at a destination simultaneously. In a preferred embodiment, the reference direction is defined as a direction having the smallest delay among a beam plurality. In other words, the direction of transmission among the plurality of selected directions having the smallest delay is designated as a reference direction and the delays associated with the remaining directions in the plurality are calculated relative to the reference direction. In practice, the reference direction is associated with the largest calculated delay and the other selected directions are associated with smaller (or even negligible) delays. The delayed signals are then coded utilizing mutually orthogonal codes relative to codes used for coding signals intended for subscribers other than the select subscriber, as is known in the art, and formed into beams for transmission in the determined transmission directions to the select subscriber. In this manner, all the beams will be received at an intended destination, e.g., at an intended MS, simultaneously. This will improve the link performance between the signal source (BS) and the signal destination (MS) for the select system subscriber.

In principle, orthogonal spreading codes are required in each beam to allow signal separation by the receiver. Typically, the number of beams is N = 2. As an alternative to orthogonal spreading codes, the use of different orthogonal pilot sequences may be more practical in some circumstances. Thus, only the identifier of a particular beam will be orthogonal to an identifier of another beam. Of course, the spreading code for a particular user must be orthogonal to the spreading codes of other users of the system.

In a preferred embodiment, the delay profile for each beam is calculated by correlating the known CDMA pilot signal (for a WCDMA system) or training sequence (for a TDMA system) with a received signal. This results in a channel impulse response which is the equivalent of a power-delay curve. The delay spread for each beam is determined as the delay span over which the power of the impulse response is above a certain threshold.

It will be appreciated that the inventive method described above can be employed in CDMA and TDMA systems. Moreover, the method can be used in the transmission of complex beams, where each complex beam, for example, is made up of two or more overlapping sub- beams in orthogonal relationship to each other. For example, the sub-beams in each complex beam beams can be orthogonally polarized by 90° with respect to each other. In a situation where each complex beam includes two orthogonally-disposed beams, a reference complex beam is selected and a delay of another complex beam is calculated relative to the reference complex beam. Signals can then be coded using orthogonal spreading codes or orthogonal pilot sequences in a four-branch diversity scheme, with each branch representing a sub- beam. Moreover, the signals can be coded on sub-beams from non-common complex beams, (e.g. a first sub-beam from one complex beam and a second sub-beam from another complex beam, etc.) and the sub-beams can have the same or different transmission directions.

For a two-branch diversity scheme, a complex beam having the shortest delay spread can be selected. As the selected complex beam has orthogonal components (two sub-beams) which arrive at a desired MS simultaneously with each other, there is no need to delay one sub-beam relative to the other in a single complex beam case. Thus, diversity transmission using orthogonal codes can be realized through the use of orthogonal sub- beams without delaying one beam relative to another.

The inventive method can be used in conjunction with individual beam power control. The power of each beam can be controlled based on measurements obtained from the BS (open loop) and/or based on . feedback information received by the BS from the MS (closed loop) . In addition, complex coefficient weighting can be applied to individual beams attempting coherent combining at the receiving end (e.g. at the MS). These weights can be set by the BS based on information received from the MS. The criteria used for selecting appropriate transmission directions discussed above has a positive effect on the link performance to combat loss of orthogonality as well as an additional benefit for combating problems resulting from mulitpath or interpath interference (IPI) . In an ideal single user case, the signature code for a particular user should have an autocorrelation function equal to the impulse function δ. Thus when correlating a sequence with its replica the result should be high (or 1) when both sequences are in phase or synchronized, otherwise the correlation will be zero. FIG. 5 shows a typical auto-correlation of an m- sequence, where "a" represents the auto-correlation function amplitude for the in-phase case and "b" represents the amplitude for the out-of-phase case. The b/a ratio strongly depends on the length of the sequence. The longer the sequence, the closer the ratio will be to a zero value. For shorter sequences, the ratio will increase considerably. The problem with the non-ideality of the auto-correlation function (ACF) for real sequences (e.g. most of the typical sequences used in telecommunications) is that in frequency selective channels where there is a considerable delay spread, the signal will suffer from interpath interference. This is due to the receipt at the receiving end (MS) of the original signal and one or more delayed replicas. Since "b" is not a zero value, the processing of a particular signal component by a conventional Rake receiver will be affected by the nonzero component of the ACF. Thus, in many situations the decision as to the importance of a given received information bit could be wrongly calculated due to the interfering effect of the nonzero components. The performance degradation will increase in radio environments with considerable multipath spread as the ratio of b/a differs from its ideal value of zero. For example, the b/a ratio can be H with a particular high bit rate mode of WCDMA. By selecting those directions as described above, the occurrence of IPI is avoided.

It is noted that the inventive method described herein does not attempt to solve the problem of loss of orthogonality from a general standpoint. The orthogonality problems are more severe when short sequences are used. In general, the larger the sequence, the more immune a receiver is to the degrading effects of orthogonality loss, and visa versa . Thus, the proposed method is to operate on few very high bit rate users using very short sequences to provide an otherwise resulting significant performance drop. Other (low bit) users (illuminated by the same delayed beam) may see a loss in orthogonality. However, due to the other lower bit user's larger spreading codes, such users will not be adversely effected. Thus, the higher the bit rate, the shorter the spreading codes, resulting in improved results from the proposed method.

Thus, while there have shown and described and pointed out fundamental novel features of the invention as applied to a preferred embodiment thereof, it will be understood that various omissions and substitutions and changes in the form and details of the devices illustrated, and in their operation, may be made by those skilled in the art without departing from the spirit of the invention. For example, it is expressly intended that all combinations of those method steps which perform substantially the same function in substantially the same way to achieve the same results are within the scope of the invention. Moreover, it should be recognized that the method steps described in connection with any disclosed form or embodiment of the invention may be incorporated in any other disclosed or described or suggested form or embodiment as a general matter of design choice. It is the intention, therefore, to be limited only as indicated by the scope of the claims appended hereto.

Claims

1. A method of compensating for loss of orthogonality between a first plurality of beams formed from a data signal and transmitted over directions of a plurality of possible transmission directions to an intended receiving device in a wireless communications system having a second plurality of beams, the second plurality of beams being intended for receipt by receiving devices other than the intended receiving device, comprising the steps of: selecting a transmission direction for each beam in the first plurality of beams, each transmission direction having a corresponding time delay at which each said beam in said first plurality of beams is received by the intended receiving device; calculating the corresponding time delay for each selected transmission direction relative to a reference direction; delaying each beam in said first plurality of beams by said calculated corresponding time delay for each beam so that the first plurality of beams will be simultaneously received at the intended receiving destination device; and coding each beam in the first plurality of beams with mutually orthogonal codes relative to codes used for transmitting beams in the second plurality of beams .

2. The method of claim 1, wherein said reference direction is a direction having a smallest associated delay from among the calculated corresponding time delays.

3. The method of claim 1, wherein said step of selecting a transmission direction for each beam in the first plurality of beams comprises the step of determining, for each of said possible transmission directions, the directions having negligible associated delay spreads .

4. The method of claim 3, wherein said step of selecting a transmission direction for each beam in the first plurality of beams further comprises the steps of determining an amount of interference caused by each beam in said first plurality of beams relative to the beams in said second plurality of beams and a power level associated with a plurality of transmit power amplifiers in a base station.

5. The method of claim 1, wherein said selecting step comprises selecting two directions.

6. The method of claim 1, wherein said step of calculating the corresponding time delay comprises obtaining uplink measurements for each beam in the first plurality of beams.

7. The method of claim 1, wherein said step of selecting a transmit direction comprises calculating the corresponding time delay comprises obtaining uplink measurements and downlink measurements for each beam in the first plurality of beams.

8. The method of claim 1, wherein said destination device is a mobile telephone.

9. The method of claim 1, wherein said destination device is a mobile station.

10. The method of claim 9, wherein said selecting step is performed by the mobile station.

11. The method of claim 9, wherein said calculating step is performed by the mobile station.

12. The method of claim 1, wherein each beam in said first plurality of beams is associated with an identifier.

13. The method of claim 12, wherein the identifier comprises a sequence of known bits.

14. The method of claim 12, wherein the identifier comprises a sequence of known symbols.

15. The method of claim 12, wherein each identifier is a reference sequence.

16. The method of claim 1, wherein the beams in said first plurality of beams is amplified in accordance with beam power levels measured by the intended receiving device.

17. The method of claim 1, wherein each beam in said first plurality of beams is amplified and phased according to complex weights measured by the intended receiving device.

18. A method of compensating for loss of orthogonality between a first plurality of beams formed from a data signal and transmitted over directions of a plurality of possible transmission directions to an intended receiving device in a wireless communications system having a second plurality of beams, the second plurality of beams being intended for receipt by receiving devices other than the intended receiving device, said method comprising the steps of: selecting a transmission direction for each complex beam in the first plurality of beams, each transmission direction having a corresponding time delay at which each said complex beam in said first plurality of beams is received by the intended receiving device; calculating the corresponding time delay spread for each selected transmission direction relative to a reference direction; designating a direction for transmission corresponding to a transmission direction having a smallest time delay spread among said selected transmission directions; coding each beam in the first plurality of beams with mutually orthogonal codes relative to codes used for transmitting beams in the second plurality of beams; and transmitting the coded beams in the first plurality of beams in said designated direction.

19. The method of claim 18, wherein at least one of said beams in said first plurality of beams is a complex beam comprising a pair of orthogonally arranged sub-beams .

20. The method of claim 18, wherein said step of selecting a transmission direction for each beam in the first plurality of beams comprises the step of determining, for each of said possible transmission directions, the directions having negligible associated delay spreads.

21. The method of claim 18, wherein said destination device is a mobile station.

22. The method of claim 18, wherein said destination device is a mobile telephone.

23. The method of claim 19, wherein orthogonality between sub-beams in the complex beam is achieved by polarizing each sub-beam approximately 90° with respect to each other.

24. The method of claim 19, wherein said step of selecting a transmission direction for each beam in the complex beams further comprises the steps of determining an amount of interference caused by each sub-beam in said complex beam relative to the beams in said second plurality of beams and a power level associated with a plurality of transmit power amplifiers in a base station.

25. The method of claim 18, wherein said step of selecting a transmission direction comprises selecting at least two transmission directions.

26. The method of claim 18, wherein said step of calculating the corresponding time delay comprises obtaining uplink measurements for each beam in the first plurality of beams.

27. The method of claim 18, wherein said step of selecting a transmit direction comprises calculating the corresponding time delay comprises obtaining uplink measurements and downlink measurements for each beam in the first plurality of beams.

28. The method of claim 18, wherein said selecting step is performed by the mobile station.

29. The method of claim 18, wherein said calculating step is performed by the mobile station.

30 The method of claim 18, wherein each beam in said first plurality of beams is associated with an identifier .

31. The method of claim 29, wherein the identifier comprises a sequence of known bits.

32. The method of claim 29, wherein the identifier comprises a sequence of known symbols.

33. The method of claim 29, wherein each identifier is a reference sequence.

34. The method of claim 18, wherein the beams in said first plurality of beams is amplified in accordance with beam power levels measured by the intended receiving device.

35. The method of claim 18, wherein each beam in said first plurality of beams is amplified and phased according to complex weights measured by the intended receiving device.