The present disclosure relates to estimating conditions of a wireless channel in a wireless communication system.
Radio frequency wireless communication between two wireless communication devices involves certain challenges associated with the conditions of the over-the-air wireless channel between the two devices. For applications where the information to be transmitted requires a relatively high transmission rate and the devices may be moving with respect to each other, it is important that each device on the link be able to compute accurate information representing the current conditions of the channel.
One technique that has been in use in wireless communication systems is to include pilot signals in the transmissions between devices. Generally, the channel information can be estimated based on the pilot signals if there are sufficient number of pilot signals in a transmission. However, in certain practical applications, there is limited availability for pilot signals. Moreover, for certain applications, the transmissions are relatively short in time duration, making it challenging to accurately estimate the channel conditions.
There is room for improving channel estimation in wireless communication devices, and in particular, to make channel estimation more dynamically adaptable to changing conditions of the wireless channel.
Overview
A dynamically adaptive channel estimation process is provided for use in a wireless communication device. At a first device, a wireless transmission is received from a second device. The transmission comprises a plurality of successive symbols each of which comprises a plurality of subcarriers. The first device is configured to compute channel characterizing information for a wireless channel between the first device and the second device based on the received values at subcarriers of the successive symbols. The first device is configured to select one of a plurality of channel estimation schemes based on the channel characterizing information, and to compute an estimate of channel information for the wireless channel using the selected one of the plurality of channel estimation schemes.
Referring first to
The devices 10 and 20 may be configured to perform a broadband wireless communication signaling protocol, such as that used in WiMAX™ systems or 3G Long Term Evolution (LTE) systems. More generally, the second device 20 sends a wireless transmission via its antennas 22(1)-22(M) to the first device 10. The transmission comprises multiple symbols, and each symbol comprises multiple subcarriers. In an example described herein, the transmission is formatted according to orthogonal frequency division multiple access (OFDMA) techniques, such as, but not limited to, those OFDMA techniques used in a WiMAX system.
The first device 10 receives the transmission from the second device and is configured to compute information that characterizes the wireless channel between the first device 10 and the second device 20. The first device 10 is further configured to select one of a plurality of channel estimation schemes based on the characterizing information for the channel, and to compute a channel estimate for the transmission using the selected channel estimation scheme. While the foregoing channel estimation techniques described herein are particularly useful for a channel between two devices that have multiple antennas, this is only an example. Thus, the first device 10 may have a single antenna and the second device 20 may have a single antenna.
To this end, reference is now made to
As shown in
The transmitter 12 may comprise individual transmitter circuits that supply respective upconverted signals to corresponding ones of a plurality of antennas 18(1)-18(M) for transmission. For simplicity, these individual transmitter circuits are not shown. The receiver 14 receives the signals detected by each of the antennas 18(1)-18(M) and supplies corresponding antenna-specific receive signals to controller 16. It is understood that the receiver 14 may comprise a plurality of individual receiver circuits, each for a corresponding one of a plurality of antennas and which outputs a receive signal associated with a signal detected by a respective one of the plurality of antennas 18(1)-18(M). For simplicity, these individual receiver circuits are not shown.
The controller 16 comprises a memory 17 or other data storage block that stores data used for the techniques described herein. The memory 17 may be separate or part of the controller 16. In addition, logic instructions for performing an adaptive channel estimation process 100 may be stored in the memory 17 for execution by the controller 16. The controller 16 may be configured to generate a beamforming weight vector using the channel estimate information that is produced by the adaptive channel estimation process 100. The beamforming weight vector is supplied to the transmitter 12.
The functions of the controller 16 may be implemented by logic encoded in one or more tangible media (e.g., embedded logic such as an application specific integrated circuit, digital signal processor instructions, software that is executed by a processor, etc.), wherein the memory 17 stores data used for the computations described herein (and/or to store software or processor instructions that are executed to carry out the computations described herein). Thus, the process 100 may be implemented with fixed logic or programmable logic (e.g., software/computer instructions executed by a processor) and the controller 16 may be a programmable processor, programmable digital logic (e.g., field programmable gate array) or an application specific integrated circuit (ASIC) that comprises fixed digital logic, or a combination thereof.
In current and next generation wireless communication systems, high transmission rates are required to support both stationary and mobile communication devices. As a result, channel estimation for coherent detection of received signals is important. In most of these systems, such as WiMAX and 3G Long Term Evolution (LTE) systems, pilots are inserted in some specific tones or subcarriers of subchannels that also contain data in order to allow a receiving device to estimate or track channel state information. Generally, the channel time response or frequency response can be estimated based on the pilots if enough pilots are used in individual transmissions. However, in certain practical applications, such as traffic transmissions with a small burst size or traffic transmissions with dedicated pilots, there are only a few pilots available in the transmission for channel estimation.
To solve this problem, the adaptive channel estimation process 100 is configured to dynamically select the most appropriate one of a plurality of channel estimation schemes based on some preliminary or pre-processing of a transmission. Each of these schemes is for different channel conditions such as different frequency sensitivity or selectivity of the channel, or different Doppler or time-varying conditions of the channel.
With reference to
The process 100 may be performed for each transmission that the first device receives from the second device. The process 100 can dynamically select the best channel estimation scheme on a transmission by transmission basis depending on the current conditions of the channel. In this sense, the process 100 is dynamically adaptable to changing conditions of the wireless channel. The other information referred to at 120 that may be used to compute the channel characterizing information for a current received transmission may based further on channel characterizing or channel estimate information computed for one or more previously received transmissions, where the previous transmissions comprise subcarriers that may or may not be in the same frequency band as that of the current transmission. Alternatively, the pre-processing computation at 120 and the channel estimation computation at 140 may be performed for two or more received transmissions that are considered together as a single transmission or burst.
As explained above, the channel estimation techniques described herein are useful when the transmission is a relatively short burst that has a minimal number of pilot subcarriers that can be used to derive channel information. However, these techniques are not meant to be limited to such a situation and they can be applied to transmissions with more pilot subcarriers.
Turning now to
Each symbol comprises 18 subcarriers, numbered 1-18 as shown in
As shown in
where hn,m denotes the channel information of the m-th subcarrier in symbol n (for n=0, 1 and 2), yn,m denotes the received value at the m-th subcarrier in symbol n and xn,m represents the transmit value at the m-th subcarrier in symbol n. Thus, x0,2,x0,11,x1,5,x1,14,x2,8 and x2,17 are the transmit values at the pilot subcarriers in Symbols 0, 1 and 2, and are known a priori to the receiving device. The received values at the subcarriers are derived from the received signal.
Next, at 126, using the channel estimates at the subcarriers of pilots, estimates of the transmit values are computed at those data subcarriers that are next to or neighboring (in frequency or time) the pilot subcarriers in the successive symbols from the receive values at those neighboring data subcarriers. For example, as shown in
The transmit values at the neighboring data subcarriers are thus computed as follows:
For example, as shown in the equations above, since subcarrier 1 of Symbol 0 is next to the pilot subcarrier at subcarrier 2 of Symbol 0, then an estimate of the transmit value at subcarrier 1 of Symbol 0, {circumflex over (x)}0,1 is computed from the received value at subcarrier 1 of Symbol 0, y0,1, and the channel estimate at the pilot subcarrier 2 of Symbol 0, ĥ0,2, which was computed at 124 according to the equation
Similar computations are made using the channel information at the pilot subcarriers for computing the transmit values at the neighboring data subcarriers associated with each of the pilot subcarriers in the symbols of the transmission.
At 128, measures are computed that characterize the channel between the two devices based on the estimates of the transmit values at the neighboring data subcarriers. For example, a first measure is computed that represents a degree to which the channel varies with frequency and a second measure is computed that represents a degree to which the channel varies with time.
More specifically, at 128, the first measure is computed as follows. The first measure representing the frequency variation of the channel is based on distances between the transmit value estimates at the different data subcarriers (neighboring subcarriers) in the same symbol computed at 126 to corresponding symbol constellation points. For example, the first measure is based on the estimated transmit values {circumflex over (x)}n,m in the same symbol (at those subcarriers with vertical cross hatching in
Vf=mean(d0,1,d0,3,d0,10,d0,12,d1,4,d1,6,d1,13,d1,15,d2,7,d2,9, d2,16, d2,18),
where dn,m is the minimum distance of estimated transmit values {circumflex over (x)}n,m at the neighboring subcarriers within the same symbol to the symbol constellation points, i.e.,
and where ek ∈ {e1, . . . , eK} is the set of symbol constellation points for the transmitted symbols. This distance measure is dependent on the type of modulation scheme used for the symbols, e.g., quadrature phase shift keying (QPSK), quadrature amplitude modulation (QAM), etc.
Similarly, at 128, the second measure is computed as follows. The second measure is based on distances between the transmit value estimates at neighboring subcarriers of the pilot subcarriers in different symbols with respect to the corresponding symbol constellation points. For example, the second measure is based on the estimated transmit values {circumflex over (x)}n,m in the different symbols (at those subcarriers with horizontal cross hatching in
Vt=mean(d1,2,d0,5,d2,5,d1,8,d1,11,d0,14,d2,14,d1,17),
again where dn,m is the minimum distance of the estimated transmit values {circumflex over (x)}n,m at the neighboring subcarriers in different symbols to the symbol constellation points, i.e.,
where ek ∈ {e1, . . . , eK} is the set of symbol constellation points for the transmitted symbols.
In the example of
At 132, the channel characterizing information comprising the first and second measures Vf and Vt are obtained as explained above. At 134, the first and second measures are compared with each other. When Vt<Vf, then channel estimation scheme 1 is selected as shown at 136. Otherwise, channel estimation scheme 2 is selected as shown at 138. In addition, depending on the values for the first and/or second measures, one or more parameters for channel estimation scheme 2 may be selected when invoking channel estimation scheme 2. For example, the order of the curve fitting model used in the channel estimation scheme 2 may be selected depending on the first and second measures. Thus, the process 130 involves selecting a channel estimation scheme depending on which of the first and second measures dominates the other. When the frequency variation of the channel dominates the time variation of the channel, the first channel estimation scheme is selected. Conversely, when the time variation of the channel dominates the frequency variation of the channel, the second channel estimation scheme is selected.
Turning now to
where x0,2,x0,11,x1,5,x1,14,x2,8 and x2,17 are transmitted values at the corresponded pilot subcarriers noted above. At 220, channel information at other subcarriers in the symbols of the transmission is derived by extending in time the channel information at the pilot subcarriers to corresponding data subcarriers in other symbols of the transmission. That is, the channel information at subcarrier 2 of Symbol 1 and at subcarrier 2 of Symbol 2 is set to be the same as the computed channel information at the pilot subcarrier 2 of Symbol 0. Similarly, the channel information at subcarrier 5 of Symbol 0 and at subcarrier 5 of Symbol 2 is set to be the same as the computed channel information at the pilot subcarrier 5 of Symbol 1, and so on.
ĥ1,2=ĥ0,2, ĥ2,2=ĥ0,2, ĥ1,11=ĥ0,11, ĥ2,11=ĥ0,11
ĥ0,5=ĥ1,5, ĥ2,5=ĥ1,5, ĥ0,14=ĥ1,14, ĥ2,14=ĥ1,14
ĥ0,8=ĥ3,8, ĥ2,8=ĥ3,8, ĥ0,17=ĥ3,17, ĥ2,17=ĥ3,17
Next, at 220, the channel information at the remaining subcarriers of the symbols in the transmission is computed using interpolation and extrapolation with respect the channel information for those subcarriers computed or determined up to this point. That is, extrapolation and interpolation is used to compute the channel information at the remaining subcarriers from the channel information ĥ0,2,ĥ0,5,ĥ0,8,ĥ0,11,ĥ0,14,ĥ0,17; ĥ1,2,ĥ1,5,ĥ1,8,ĥ1,11,ĥ1,17 and ĥ2,2,ĥ2,5,ĥ2,8,ĥ2,14,ĥ2,17 for Symbols 0, 1 and 2 of the transmission.
Turning now to
At 330, the estimate of the channel information at other subcarriers is computed using a curve fitting model (at the selected order). This involves computing a curve fitting coefficient vector at the selected order. An example of a second order (two-dimensional) curve fitting model is h=c0+c1f+c2t+c3f2, where f is frequency index variable, and t is the time (symbol) index variable.
The following is an example of a two-dimensional (second order) curve fitting channel estimation technique. Given that the channel information at the pilot subcarriers can be computed, i.e., [ĥ0,2,ĥ1,5,ĥ2,8,ĥ0,11,ĥ1,14,ĥ3,17], a curve fitting coefficient vector C, C=[c0 c1 c2 c3]T, is computed for the two-dimensional fitting curve h=c0+c1f+c2t+c3f2, such that C=B−1Z , where B is a 4×4 matrix and Z is a 4×1 vector. The B and Z matrices may be generated according to the following pseudo-code.
Once the curve fitting coefficient vector is computed, then the channel information at the other subcarriers in the symbols of the transmission may be computed from the equation ĥn,m=[1 m n m2]C, where m and n are the subcarrier and symbol indices, respectively, in the current transmission.
The curve fitting technique described above can be extended to a higher order computation depending on the channel characterization information. For example, a third order computation would be h=c0+c1f+c2t+c3f2+c4f3, and a fourth order computation would be h=c0+c1f+c2t+c3f2c4f3+c5f4, and so on. The computation of the coefficient vector C would then be expanded to higher matrix computations involving the B and Z matrices referred to above. Thus, if the channel characterization information indicates that the channel is varying rapidly (in time and/or frequency), a parameter, e.g., the order, of the curve fitting computation may be adjusted or selected, e.g., increased, to account for such faster channel variations.
Although the apparatus, system, and method are illustrated and described herein as embodied in one or more specific examples, it is nevertheless not intended to be limited to the details shown, since various modifications and structural changes may be made therein without departing from the scope of the apparatus, system, and method and within the scope and range of equivalents of the claims. Accordingly, it is appropriate that the appended claims be construed broadly and in a manner consistent with the scope of the apparatus, system, and method, as set forth in the following claims.
Number | Name | Date | Kind |
---|---|---|---|
6961364 | Laroia et al. | Nov 2005 | B1 |
7388979 | Sakai et al. | Jun 2008 | B2 |
7389086 | Toda et al. | Jun 2008 | B2 |
7733989 | Guey | Jun 2010 | B2 |
7778211 | Hedayat et al. | Aug 2010 | B2 |
7792352 | Sakai et al. | Sep 2010 | B2 |
20030043732 | Walton et al. | Mar 2003 | A1 |
20030123559 | Classon et al. | Jul 2003 | A1 |
20050147287 | Sakai et al. | Jul 2005 | A1 |
20060203932 | Palanki et al. | Sep 2006 | A1 |
20060280262 | Malladi | Dec 2006 | A1 |
20070041311 | Baum et al. | Feb 2007 | A1 |
20080049814 | Jeon | Feb 2008 | A1 |
20080076477 | Hedayat et al. | Mar 2008 | A1 |
20080101489 | Wang et al. | May 2008 | A1 |
20080123516 | Anderson et al. | May 2008 | A1 |
20080159455 | Wang et al. | Jul 2008 | A1 |
20080165901 | Krupka | Jul 2008 | A1 |
20080232674 | Sakai et al. | Sep 2008 | A1 |
20090042530 | Wang et al. | Feb 2009 | A1 |
20090116592 | Namba et al. | May 2009 | A1 |
20100130135 | Wang et al. | May 2010 | A1 |
20100157861 | Na et al. | Jun 2010 | A1 |
20100165851 | Mantravadi et al. | Jul 2010 | A1 |
20100328446 | Sakai et al. | Dec 2010 | A1 |
Number | Date | Country |
---|---|---|
1936901 | Jun 2008 | EP |
Number | Date | Country | |
---|---|---|---|
20100130135 A1 | May 2010 | US |