Patent Application
20100238980
The present invention relates generally to implementing a receiving device that is capable of receiving composite signals, and more specifically by way of example but not limitation, to adjusting a grid having temporal points that are used to place processing delays for receiving a composite signal.
Many specialized terms and abbreviations are used in the communications arts. At least some of the following are referred to within the text that follows, such as in this background and/or the subsequent description sections. Thus, the following terms and abbreviations are herewith defined:
BER Bit Error Rate
BLER Block Error Rate
COG Center of Gravity
MS Mobile Station
PDP Power Delay Profile
RBS Radio Base Station
RMS Root-Mean-Square
SINR Signal-to-Interference-plus-Noise Ratio
WCDMA Wideband Code Division Multiple Access
Electronic communication forms the backbone of today's information-oriented society. Electronic communications are transmitted over wireless or wired channels using electromagnetic radiation, such as radio frequency (RF) transmissions, light waves, and so forth. The availability and capacity of electronic communications is frequently limited by the interference and noise inherent in the communications channel between a transmitting device and a receiving device.
The utilization of a communications channel may be increased by adopting any of a number of different schemes. These schemes can enable more information to be communicated in a given spectrum allocation. Efficient utilization of spectrum can reduce the cost of communication services being provided, can enable richer communication services to be provided, or both. Such schemes can also strengthen or otherwise improve signal reception at a receiving device.
An example scheme involves receiving a composite signal that includes multiple signal images. The multiple signal images are combined or otherwise processed in a manner that increases the likelihood of correctly interpreting the transmitted information. In a spread spectrum system, the signal images may be combined before or after the received signal has been despread. The signal images are combined at the chip level if the combining is performed prior to despreading. They are combined at the symbol level if the combining is performed after the despreading. A Rake receiver, for example, can collect signal energy from different signal images using “fingers” to strengthen reception and demodulation of the actual information carried by the composite signal.
Generally, signal images may be combined using an interference suppression scheme. With regard to Wideband Code Division Multiple Access (WCDMA), for example, linear interference suppression has been standardized for both type 2 (single antenna) and type 3 (two antenna) receivers. Linear interference suppression can be achieved by either chip level or symbol level equalization. A problem with such equalizers involves deciding how to place filter taps (e.g., at the chip level) or fingers (e.g., at the symbol level) in order to maximize the received symbol Signal-to-Interference-plus-Noise Ratio (SINR).
In this Background section, the description focuses on finger placement for symbol level equalization in order to illustrate the problem. However, the filter tap placement problem for chip level equalizers is analogous. With symbol level equalization, a finger typically includes a delay element and a correlator. The delay element delays the received signal by a given amount before the correlator despreads the signal by correlating it with the combined spreading/scrambling code. The term “placing a finger” therefore refers to setting a respective finger delay period for a respective finger delay element.
Several approaches have been proposed for finger placement. Generally, these proposed approaches can be categorized as path-oriented or grid-oriented. Path-oriented approaches are described immediately below. Grid-oriented approaches are described thereafter.
Path-Oriented Approaches: With path-oriented approaches, the receiver relies on information developed by the path searcher/delay estimator. Generally, some form of a mirroring strategy is employed. A common and widely used variant of the mirroring strategy is described in U.S. Pat. No. 6,922,434. U.S. Pat. No. 6,922,434 to Y.-P. Wang et al. is entitled “Apparatus and methods for finger delay selection in RAKE receivers” and was issued on 26 Jul. 2005 to the same Assignee, Telefonaktiebolaget LM Ericsson, as the Instant Patent Application.
An example of this mirroring strategy includes two phases. In a first phase, the set of finger delays is set to the channel delays. If the receiver is a Rake receiver, the process may be terminated after the first phase. Otherwise, in a second phase, the set of channel delays is ranked in decreasing order of channel coefficient magnitude. The ranked channel delay set is then considered in a pair-wise fashion to create candidate finger delays.
For each pair of channel delays from the channel delay set that is ranked in decreasing order of channel coefficient magnitude, the following steps are performed:
Grid-Oriented Approaches: With grid-oriented approaches, finger assignment is contemplated at regularly spaced intervals (e.g., from a grid). An example of a grid-oriented approach is described in U.S. Pat. No. 7,469,024. U.S. Pat. No. 7,469,024 to A. Khayrallah et al. is entitled “Method and apparatus for finger placement in Rake receiver” and was issued on 23 Dec. 2008 to the same Assignee, Telefonaktiebolaget LM Ericsson, as the Instant Patent Application.
Grid points of the grid define potential finger locations, with the grid being of essentially infinite extent. The actual finger assignment is based on information such as the Power Delay Profile (PDP), the maximum number of fingers to place, thresholding level(s), assigned finger quality, and so forth. An applicable assumption is that one can identify regions of energy from the PDP and that covering these regions with sub-Nyquist-spaced fingers allows for accurate demodulation.
Grid spacing 105 is defined as the distance (e.g., time) between two consecutive grid points. With grid 101, grid spacing 105 is fixed. As noted above, grid 101 may be of an essentially infinite size. Consequently, a finger is not assigned to each and every grid point. Instead, fingers are assigned selectively, such as to the grid points corresponding to times of relatively high PDP. With this grid-oriented approach, there are issues regarding how the grid is aligned, regarding grid spacing, and regarding whether there are one or multiple such grids operative at the same time. U.S. Pat. No. 7,469,024 addresses many of these issues.
Unfortunately, there are deficiencies in the foregoing state of the art for interference suppression with regard to both the path-oriented and the grid-oriented approaches. There are two significant limitations of the path-oriented approaches. First, a practical path searcher/delay estimator makes errors. These errors in the path estimation result in sub-optimal finger placement. The severity of this type of error depends, in part, on what type of G-Rake receiver is used. Misidentification of paths can be very serious for a parametric receiver. Such receivers depend upon a model of the channel, and any mismatch between the model and the true channel causes performance degradation. The error is less severe for a nonparametric G-Rake receiver. These receivers blindly learn channel characteristics, and are therefore less affected by misidentification of channel paths. However, the misidentification of paths may result in fewer (or more) fingers being placed than is required for interference suppression. The placement of fewer (or more) fingers may result in decreased receiver performance (e.g., reduced throughput and/or higher BER/BLER).
Another problem with the path-oriented approaches relates to the frequency of finger repositioning. Paths may appear to drift over time due to receiver imperfections (e.g., oscillator drift and/or frequency error). Paths may disappear/reappear as well. These path-related issues can lead to the frequent repositioning of fingers with no apparent performance gain. This can result in significant overhead in receiver software and/or hardware.
With regard to grid-oriented approaches generally, the basic concept is sound. Grid-oriented approaches can therefore serve as a foundation for additional development. However, the existing grid-oriented approaches are relatively static even in the face of changing environmental conditions. Consequently, there is a need to address the deficiencies that exist in the current state of the art with regard to both path-oriented and grid-oriented approaches. Such deficiencies and other needs are addressed by one or more of the various embodiments of the present invention.
It is an object of the present invention to remedy or at least ameliorate one or more of the deficiencies that are identified hereinabove. It is an object of certain embodiments of the present invention to dynamically adjust a grid having temporal points based on current channel conditions. It is another object of certain embodiments of the present invention to implement a grid-oriented approach using an interference suppression receiver.
In an example embodiment, a method entails receiving a composite signal with adjustable processing delay placements. A composite signal having multiple signal images corresponding to multiple reception delays is received. A root-mean-square (RMS) delay spread is ascertained for the multiple reception delays that correspond to the multiple signal images of the composite signal. A set of temporal points is produced responsive to the RMS delay spread. Multiple processing delays are placed based on the set of temporal points. In different example implementations, the set of temporal points may be produced by adjusting the spacing between temporal points, by adjusting a total number of temporal points in the set of temporal points, or by changing the center location of the set of temporal points. The temporal point spacing and the number of temporal points may be adjusted responsive to the RMS delay spread. The center location of the set of temporal points may be adjusted responsive to at least one calculated delay.
In another example embodiment, a receiving device adjusts a set of temporal points to be used for placing processing delays for a receiver. The receiving device includes a receiver, a production unit, and a processing delay placement unit. The receiver receives a composite signal having multiple signal images corresponding to multiple reception delays. The production unit ascertains an RMS delay spread for the multiple reception delays corresponding to the multiple signal images of the composite signal. The production unit is to generate a set of temporal points responsive to the RMS delay spread. The processing delay placement unit places multiple processing delays based on the set of temporal points.
In yet another example embodiment, a method entails receiving a signal with adjustable processing delay placements. A composite signal having multiple signal images corresponding to multiple reception delays is received. A set of temporal points is produced by adjusting a spacing between consecutive temporal points of the set of temporal points and by adjusting a center location of the set of temporal points. Multiple processing delays are placed based on the set of temporal points. The composite signal is demodulated by combining the multiple signal images in accordance with the multiple processing delays that are placed based on the set of temporal points.
An advantage of certain embodiments of the present invention is that scarce processing delays may be adaptively utilized responsive to channel conditions by adjusting temporal points on which processing delays are placed. Another advantage of certain embodiments of the present invention is that receiver performance may be enhanced by adaptively allocating a number of processing delays based on whether a channel is relatively more dispersive or relatively less dispersive. Yet another advantage of certain embodiments of the present invention is that processing overhead can be reduced with respect to a path-oriented approach because less repositioning is involved. Additional advantages are noted herein below.
Additional embodiments are also described and/or claimed herein. Example additional embodiments include, by way of example but not limitation, arrangements, memory, systems, and so forth. Additional aspects of the invention are set forth in part in the detailed description, drawings, and claims that follow, and in part may be derived from the detailed description and drawings, or can be learned by practice of the invention. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention as disclosed or as claimed.
A more complete understanding of the present invention may be obtained by reference to the following detailed description when taken in conjunction with the accompanying drawings wherein:
As described herein above, there are drawbacks to the existing approaches for both path-oriented and grid-oriented finger assignment strategies. Because the general underlying concept for grid-oriented approaches is sound, however, example embodiments that are described herein are directed to grid-oriented approaches. In an example embodiment, a grid is dynamically adjusted responsive to current channel conditions. Grid-based approaches usually involve more fingers due to the typical sub-Nyquist grid spacing of the grid points. Grid points represent times along a temporal axis, and they are also termed temporal points herein. In an example implementation, the grid spacing can be adjusted so that it adapts to the channel, thereby effectively utilizing the finite number of processing delays that are available to a receiver. The current channel conditions can also determine how many processing delays are to be used by the receiver. In another example implementation, grid placement is extended in a manner that is appropriate for use when an interference suppression receiver is employed.
More specifically, example embodiments employ a grid-oriented approach for determining the placement of processing delays in which the grid is adjusted based on characteristics of the channel. Different characteristic metrics may be used to adjust the grid. Appropriate placement of the processing delays based on the grid can facilitate interference suppression. In an example embodiment, the width of the grid spacing between consecutive grid points is established responsive to the RMS delay spread. In another example embodiment, the number of processing delays to be employed is set responsive to the RMS delay spread. In yet another example embodiment, a location for the center of the grid is selected responsive to at least one calculated delay, such as (i) a delay corresponding to the center of gravity (COG) of multiple path delays or (ii) a path delay associated with the largest channel coefficient magnitude value or largest power delay profile (PDP).
It should be understood that a single device may function as a transmitting device 202 at one moment and/or with respect to one communication and as a receiving device 204 at another moment and/or with respect to another communication. Examples of transmitting devices 202 and receiving devices 204 include, by way of example but not limitation, network communication nodes, remote terminals, and other devices that are capable of communicating over a channel 206. Network communication nodes may include, for example, a base transceiver station, a radio base station, a Node B, an access point, and so forth. Remote terminals may include, for example, a mobile terminal, a mobile station, a user equipment, a subscriber station, a communication card or module, and so forth. General example device implementations for transmitting/receiving devices 202/204 are described herein below with particular reference to
In operation, channel 206 may be a wired channel or a wireless channel. Regardless, channel 206 affects the transmission, propagation, and/or reception of signal 208. When an interference suppression receiver is used at receiving device 204, the receiver can compensate for or otherwise accommodate the effects of channel conditions, including those caused by interference. One way to accommodate the effects of channel conditions is to combine different signal images of a composite signal.
However, this is merely one example implementation for a WCDMA-based embodiment. Alternatively, the transmitting device may be a remote terminal of any general type, and the receiving device may be a part of the infrastructure of a general wireless network. Also, transmitting devices 202 and/or receiving devices 204 may be devices that operate outside of cellular-type wireless networks or even those that operate within a wired network. Moreover, transmitting devices 202 and receiving devices 204 may operate in accordance with a different network standard.
In an example embodiment, transmitting device 202RBS transmits composite signal 208 (not separately shown in
Hence, receiving device 204MS receives signals via a dispersive, multipath channel. The level or amount of dispersion may be different for different channels and may change over time. Such a dispersive, multipath channel results in different versions of the same communicated signal being received, including at different times. Receiving device 204MS is responsible for combining signal images 208a and 208b to attempt to reconstruct composite signal 208. From composite signal 208, receiving device 204MS may recover the communicated information.
An example mechanism for collecting signal energy from multiple signal images 208a,208b is a G-Rake receiver. Signal images 208a and 208b are despread and combined in accordance with determined combining weights using the G-Rake receiver. With a G-Rake receiver that is operating in conjunction with an interference suppression technique, the determined combining weights may also counteract the effects of interference by assigning processing delays to grid points of a grid.
It should be noted that communications environment 300 represents a relatively simple example. For instance, there may be more than two signal images 208a and 208b received at receiving device 204MS. Also, each signal image may be impacted by the same or different environmental effects such as reflecting structures and/or other signal impairments. Other factors may also impact the signal during transmission, propagation, and/or reception.
Example principles that are described herein for grid-oriented processing delay placement are applicable to symbol level equalization, chip level equalization, combinations thereof, and so forth. For symbol level equalization, placing a processing delay is comparable to placing/locating a finger element. For chip level equalization, placing a processing delay is comparable to placing/positioning a filter tap position. However, by way of example only, the equalizers that are described herein below with particular reference to
These units may be interconnected with a given logic flow as shown in
The following parts of receiver 304A may generally function in a substantially conventional fashion: traffic processing delays unit 408, traffic combiner 410, pilot processing delays unit 412, impairment covariance estimation unit 414, combining weight computation unit 416, and channel coefficient estimation unit 418. Accordingly, operational details of these parts are not described herein.
Instead, the description focuses on searcher 402, channel delay estimation unit 404, and processing delay placement unit 406A. It is given that processing delay placement unit 406A has access to both estimated channel delays (as shown from channel delay estimation unit 404) as well as estimated channel coefficients (not explicitly shown from channel coefficient estimation unit 418).
For example embodiments, processing delay placement unit 406A is adapted to produce a grid to be used for placing processing delays. One or more parameters of the grid may be dynamically adjusted responsive to current channel conditions. For example, the spacing between consecutive temporal points may be adjusted, and a center location of the grid may be adjusted. Also, a number of temporal points for the grid may be adjusted. These adjustments may be made responsive to an RMS delay spread and/or at least one calculated delay. With processing delay placement unit 406A, the RMS delay spread is calculated using estimated net channel coefficients.
For example embodiments, processing delay placement unit 406B is adapted to produce a grid to be used for the placement of processing delays. One or more parameters of the grid may be adjusted responsive to current channel conditions. For example, the spacing between consecutive temporal points and a number of temporal points for the grid may be adjusted. Also, a center location of the grid may be adjusted. These adjustments may be made responsive to an RMS delay spread and/or at least one calculated delay. In contrast with processing delay placement unit 406A (of
Thus, processing delay placement unit 406B operates at least partially differently from processing delay placement unit 406A. These difference are described further herein below, especially with reference to
As shown, graphs 500 include a grid 502 and a grid 504. The grids for each graph include five temporal points 508 with five corresponding processing delays 510. However, a grid may have a different number of temporal points. Grid 502 shows the temporal points with relatively wider grid spacing. Grid 504 shows the temporal points with relatively narrower grid spacing. Thus, the spacing between consecutive temporal points is greater for grid 502 than for grid 504.
As described above, wireless channels are often dispersive. They can be dispersive to a light, moderate, heavy, etc. degree. In an example embodiment, for a lightly dispersive channel, fewer processing delays are used, and grid spacing is relatively narrower. For a moderately to heavily dispersive channel, more processing delays are used, and grid spacing is relatively wider.
Grid construction unit 902 constructs a grid responsive to RMS delay spread 906. Grid centering unit 904 centers a grid responsive to calculated delay 908. In this manner, grid production unit 802 produces parameters for grid 804. Grid construction unit 902 and grid centering unit 904 are described in greater detail herein below with particular reference to
When implemented in conjunction with processing delay placement unit 406A (of
For an example implementation in which processing delay placement unit 406A is substantially separate from searcher 402 (as illustrated by
where ĥ=[ĥ1,ĥ2,K,ĥL-1]T is a vector of estimated net channel coefficients for the delays reported by the delay searcher, {circumflex over (τ)}j is the jth estimated path delay, and L is the number of estimated paths. The estimated net channel coefficients include the effects of transmit and receive filters as well as the effects that are represented by the radio channel coefficients. The variable τCOG represents the center of gravity (COG) of the path delays, and the variable τRMS, is the RMS delay spread.
Given the RMS delay spread τRMS grid construction unit 902 establishes temporal point spacing 1002 as follows:
Given the RMS delay spread τRMS, grid construction unit 902 sets the number of temporal points 1004 for a grid (or set of temporal points) as follows:
For an example implementation in which processing delay placement unit 406A is substantially integrated with searcher 402 (as illustrated by
where P represents power, PDPk is the kth sample of the PDP, {circumflex over (τ)}k is the delay associated with the kth PDP sample, and Np is the number of PDP samples. The variable τCOG represents the COG of the delays, and the variable τRMS is the RMS delay spread.
Given the RMS delay spread τRMS, grid construction unit 902 establishes temporal point spacing 1002 as follows:
Given the RMS delay spread τRMS, grid construction unit 902 sets the number of temporal points 1004 for a grid (or set of temporal points) as follows:
When implemented in conjunction with processing delay placement unit 406A (of
With the grid spacing determined, for an example implementation in which processing delay placement unit 406A is substantially separate from searcher 402 (as illustrated by
With the grid spacing determined, for an example implementation in which processing delay placement unit 406B is substantially integrated with searcher 402 (as illustrated by
In an example embodiment, flow diagram 1200 depicts a method for a receiving device to dynamically adjust the parameter(s) of a grid having temporal points that are to be used for placing processing delays. The placed processing delays are then used to demodulate a signal. Although particular example elements from other FIGS. are referenced to describe the steps of
At block 1202, a composite signal having multiple signal images corresponding to multiple reception delays is received. For example, a receiver 304 may receive a composite signal 208 having multiple signal images 208a,208b . . . corresponding to multiple reception delays.
At block 1204, an RMS delay spread is ascertained for the multiple reception delays corresponding to the multiple signal images of the composite signal. For example, a production unit 802 may ascertain an RMS delay spread 906 for the multiple reception delays corresponding to the multiple signal images 208a,208b . . . of composite signal 208.
At block 1206, a grid formed by a set of temporal points is produced responsive to the RMS delay spread. For example, production unit 802 may produce or generate a grid (or set of temporal points) 804 having multiple temporal points 508 responsive to RMS delay spread 906. Although not shown in flow diagram 1200, the RMS delay spread may be smoothed to determine a smoothed RMS delay spread (e.g., smoothed using filtering). When smoothed, the smoothed RMS delay spread may be used to produce the grid having the multiple temporal points.
At block 1208, multiple processing delays are placed based on the set of temporal points. For example, processing delay placement unit 406 may place multiple processing delays 510 at the multiple temporal points 508 of grid 804.
At block 1210, the composite signal is demodulated using the multiple processing delays. For example, traffic combiner 410 may combining the multiple signal images 208a,208b . . . in accordance with the multiple processing delays 510 that are placed based on temporal points 508 of the set of temporal points to thereby demodulate composite signal 208.
As part of the grid production step of block 1206, a grid may be constructed at block 1206a and/or a grid may be centered at block 1206b. For instance, the step of producing may include constructing a grid (at block 1206a) by spacing the multiple temporal points responsive to the RMS delay spread and/or by determining a number of temporal points of a set of temporal points for the grid responsive to the RMS delay spread. It should be understood that the RMS delay spread may be smoothed for the production of the grid. Also, the step of producing may include centering the multiple temporal points of the grid (at block 1206b) responsive to at least one calculated delay. Example implementations for block 1206a are described below with particular reference to
Although the establishment of the temporal point spacing and the setting of the number of temporal points have been described herein above using a single predetermined spread threshold, they may alternatively be determined using two or more predetermined spread thresholds. For example, there can be one threshold for determining when a channel transitions from relatively low dispersion to relatively high dispersion (e.g., Ψthresh
As another example, the grid spacing and the number of temporal points may be determined as follows:
For example embodiments, device 1502 may represent any processing-capable device. Processor 1504 may be implemented using any applicable processing-capable technology, and one may be realized as a general-purpose or a special-purpose processor. Examples include, but are not limited to, a central processing unit (CPU), a digital signal processor (DSP), a microprocessor, some combination thereof, and so forth. Memory 1506 may be any available memory that is included as part of and/or is accessible by device 1502. It includes volatile and non-volatile memory, removable and non-removable memory, hard-coded logic, combinations thereof, and so forth.
Interconnection 1514 interconnects the components of device 1502. Interconnection 1514 may be realized as a bus or other connection mechanism and may directly or indirectly interconnect various components. I/O interfaces 1508 may include (i) a network interface for monitoring and/or communicating across network 1516, (ii) a display device interface for displaying information on a display screen, (iii) one or more human-device interfaces, and so forth. Example network interfaces include, but are not limited to, a radio or transceiver (e.g., a transmitter and/or receiver), a modem, a network card, some combination thereof, and so forth. Human-device interface equipment 1512 may be integrated with or discrete from device 1502.
Generally, processor 1504 is capable of executing, performing, and/or otherwise effectuating processor-executable instructions, such as processor-executable instructions 1510. Memory 1506 is comprised of one or more processor-accessible memories. In other words, memory 1506 may include processor-executable instructions 1510 that are executable by processor 1504 to effectuate the performance of functions by device 1502. Processor-executable instructions 1510 may be embodied as software, firmware, hardware, fixed logic circuitry, some combination thereof, and so forth. Processor 1504 and processor-executable instructions 1510 of memory 1506 may be realized separately (e.g., as a DSP executing code) or integrated (e.g., as part of an application-specific integrated circuit (ASIC)).
In example implementations, one device 1502 may comprise a transmitting device 202, and another device 1502 may comprise a receiving device 204 (both of
Different embodiment(s) of the invention can offer one or more advantages. Generally, an adaptive mechanism is implemented that responds to channel conditions to better utilize scarce processing delays (e.g., fingers, filter taps, etc.) of the receiver. More specifically, in certain embodiments a more dispersive channel uses a grid with a larger separation between temporal points (e.g., chip spaced separations) while a less dispersive channel uses a grid with a smaller separation between temporal points (e.g., sub-chip spaced separations). This switching mechanism can improve performance given a finite number of receiver processing delays.
As another advantage, in certain embodiments, receiver performance can be enhanced by adaptively allocating processing delays. Less dispersive channels utilize fewer processing delays while more dispersive channels utilize more processing delays, so the number of temporal points is adjusted. Simulations have indicated that this strategy can enhance receiver performance. Additionally, for at least some embodiments, the overhead associated with conventional path-based approaches can be reduced. Because the COG and/or the strongest path typically change relatively slowly over time, the receiver can reposition the temporal points of the grid, as well as the corresponding processing delays, less frequently.
The devices, features, functions, methods, steps, schemes, data structures, procedures, components, etc. of
Although multiple embodiments of the present invention have been illustrated in the accompanying Drawings and described in the foregoing Detailed Description, it should be understood that the invention is not limited to the disclosed embodiments, for it is also capable of numerous rearrangements, modifications and substitutions without departing from the scope of the invention as set forth and defined by the following claims.
