This application claims the benefit, under 35 U.S.C. §365 of International Application PCT/US2006/002934 filed Jan. 27, 2006 which was published in accordance with PCT Article 21(2) on Aug. 2, 2007 in English.
The present invention generally relates to multipath searchers and methods and, more particularly, to a system and method where multiresolution searching is performed to identify valid multipath energy peaks in a wireless environment.
A mobile communication signal in wireless communication systems is transmitted between a base station (BS) to a mobile station (MS), e.g., a mobile phone or device. This signal may be reflected off numerous scattering objects, such as buildings or structures, before reaching the MS. As a result of these reflections, the wireless communication signal may propagate over many paths before reaching the MS or BS, for example. These reflected signals or multipath replicas represent replicas of the original wireless communication signal and arrive at the MS with variations in signal power levels, and times, due to varying signal propagation lengths. Upon receipt, the originally transmitted communication signal and the multipath replicas are filtered, despread, recombined and decoded to give the voice or data signal originally sent.
To search for the wireless communication signal, a multipath signal searcher is employed, along with a rake receiver in a CDMA communication system with the multipath signal searcher. The wireless communication signal is searched over a search window having a plurality of time offsets. The multipath signal searcher includes numerous search paths, each separately searching for the wireless (RF) communication signal at a particular time offset within the search window. The time offsets correspond to pseudo noise offsets in CDMA systems. In typical systems, the search paths within a multipath signal searcher function in either serial or parallel modes.
In serial operation, the multipath signal searcher searches through the first half of the search window for the mobile communication signal by having all the search paths within the multipath signal searcher search only those time offsets over the first half of the search window. The first half is searched in one search time slot and the remaining portion of the search window, or second half of the search window, is searched in the next search time slot. An integration period of a number of power control-groups (PCGs) in length-represents the entire non-coherent accumulation of a search metric over the integration period or the integration period can represent that a search metric may be coherently accumulated within different PCGs and non-coherently accumulated over the entire integration period. It takes a second integration period to search the remaining time offsets. With parallel operation, the searcher is able to search the entire window in one search time slot.
A multipath searcher in a code division multiple access (CDMA) system functions to identify valid multipath energy peaks in a wireless environment. Multipath searcher resolution is a critical parameter in mobile equipment designs. The higher the resolution, the better the multipath searcher performance. Higher resolution multipath searches will take more time and use more hardware resources than low resolution searches. Longer search times and more hardware resource requirements result in greater power consumption.
A multipath searcher and method includes a programmable decimation filter configured to adjust a sample rate of a received pilot signal. A plurality of correlators is configured to compare the received pilot signal to a reference code in a first mode and in a second mode. The first mode includes a low resolution search of a search window performed such that the plurality of correlators includes a number of correlators configured to cover an entire search window concurrently (or at least cover a significant fraction of the search window, e.g. half the search window, so that the search could be completed in two iterations) and the plurality of correlators receives a delayed reference code delayed to correspond with a portion of the search window in which a corresponding correlator performs correlation to identify peaks in the received pilot signal. The second mode includes a high resolution search of a refined search window only at or near identified peaks discovered in the first mode. The high resolution search is focused at the peak location by adjusting delays in the plurality of correlators.
In an alternate embodiment of the present invention, a multipath searcher includes a programmable decimation filter configured to adjust a sample rate of a received pilot signal, and a sequence generator configured to provide a reference signal for comparison to the received pilot signal. A plurality of correlators is configured to compare the received pilot signal to the reference code in a first mode and in a second mode, and a plurality of sequence delays are coupled to the plurality of correlators. The plurality of sequence delays are configured to delay the reference code sequence such that correlation activities are adjusted in accordance with a position in a search window. The first mode includes a low resolution search of the search window performed such that the plurality of correlators includes a number of correlators configured to cover the entire search window or portion thereof concurrently, and the plurality of correlators receives a delayed reference code delayed to correspond with a position in the search window corresponding to a given correlator to identify peaks in the received pilot signal by identifying matches between the received pilot signal and the reference signal. The second mode includes a high resolution search of a refined search window only at or near identified peaks discovered in the first mode. The high resolution search is focused at the peak locations by adjusting delays in the plurality of correlators using the plurality of sequence delays.
In an alternate embodiment of the present invention, a method for multipath searching includes filtering a received pilot signal using a programmable decimation filter configured to adjust a sample rate of the received pilot signal, and comparing the received pilot signal to a reference code in a first mode and in a second mode using a plurality of correlators, The comparing includes, in the first mode, searching a search window by performing a low resolution search of the search window such that the plurality of correlators includes a number of correlators configured to cover the entire search window or portion thereof concurrently and the plurality of correlators receives a delayed reference code delayed to correspond with a portion of the search window in which a corresponding correlator performs correlation to identify peaks in the received pilot signal. In the second mode, the searching employs one or more refined search windows in a high resolution search where the refined search windows exist only at or near identified peaks discovered in the first mode, the high resolution search being focused at the peak locations by adjusting delays in the plurality of correlators.
The teachings of the present invention can be readily understood by considering the following detailed description in conjunction with the accompanying drawings, in which:
It should be understood that the drawings are for purposes of illustrating the concepts of the invention and are not necessarily the only possible configuration for illustrating the invention. To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures.
The present invention advantageously provides a system and method for multipath searching in wireless communication environments. Although the present invention will be described primarily within the context of a multipath searcher for CDMA applications, the specific embodiments of the present invention should not be treated as limiting the scope of the invention. It will be appreciated by those skilled in the art and informed by the teachings of the present invention that the concepts of the present invention can be advantageously applied in communication system which is capable of wireless transmitting and receiving of signals over a network. In addition, although several concepts of the present invention are described in terms of a cellular network, the concepts of the present invention may be extended to other wireless network types, such as satellite networks and the like.
It should also be understood that the elements shown in the figures provided herewith may be implemented in various forms of hardware, software or combinations thereof. Preferably, these elements are implemented in a combination of hardware and software on one or more appropriately programmed general-purpose devices, which may include a processor, memory and input/output interfaces.
The multipath signal searcher 26 searches the mobile communication signal 16. The multipath signal searcher 26 functions to search for signals at various time offsets by measuring a search metric, such as energy, of the received signal. From this measurement, the multipath signal searcher 26 creates or updates a listing of those time offsets which have a search metric above a certain threshold level, thereby identifying those time offsets which potentially include the desirable mobile communication signal 16.
The multipath signal searcher 26 of
Typically a multipath dispersion specification will determine a maximum searching window. A searching window may include a time dimension equal to a number of time offsets and an energy dimension within a preferred range of energy levels for the signal 16. A reasonable design has to cover the worst case scenario, for example, a desirable resolution within the maximum searching window. Accordingly, necessary hardware and software resources are allocated in advance to accommodate the specified searching window. The worst case needs to be handled due to the lack of knowledge of a multipath dispersion map before multipath searching.
In accordance with an embodiment of the present invention, a wide multipath searching range is provided with low resolution. As such, the size of the searching window can be expanded. Then, a high resolution search for neighbors of the energy peaks discovered in the low resolution search is performed. Because the number of multipaths is limited, it is possible to share resources between these operations. For example, if it is assumed that the resources needed for two times chip rate multipath searcher resolution is N, then 2*N resources are needed for four times chip rate resolution, and 4*N resources is needed for 8 times chip rate if the same searching method is used. Sample buffer size however will become very large when the searching resolution increases.
Another factor is that after the first searching (coarse searching), the energy peaks are scattered in the searching window. For the second search, the method is only interested in the neighbor points of the first searching result and not interested in the rest of searching window at all.
The typical approach is to store the sample data for correlation at a desirable rate. There are two disadvantages for this approach. One is that the sample buffer memory size is large. Another is that read data rate could be very high depending on a total number of correlators. The present invention will advantageously reduce memory usage and access clock rate while reducing power consumption.
In the coarse searching mode, all registers 208 are serially connected. The correlators 212 are evenly placed for a whole searching space of the searching window. For example, the correlators 212 are sufficient in number to cover the time duration of the searching window. In the fine searching mode, the sample registers 208 have inputs directly connected with to the input data. The configuration of the registers 208 and the correlators 212 is controlled by the control block 210 based on the result of the coarse searching.
The programmable sample buffer 200 can be adapted to perform the searching methods of the present invention. Assuming the system works at clock rate R, a total number of the registers 208 is W*C and the total number of the correlators is W*C/R, where W is the length of the searching window. The decimation filter 202 at the front of a multipath searcher 201 has a rate programmed between C to R. Registers 208 may include modified shift registers. These registers 208 can work in two different modes: shift register mode and sparse segment shift register mode. Shift register mode is used for coarse searching and sparse segment shift register mode is used for fine searching. Sparse segment shift mode is adapted to focus only on peaks in the received pilot signal. In this way, delays are introduced in the registers 208 until areas of interest (peaks) are encountered.
In coarse searching mode, a total of W*C/R correlators 212 are placed evenly every R registers. The data rate is the coarse searching resolution rate C. So there are a total of W*C multipath searching points. After a first pass, a multipath profile is obtained (a map of delay dispersion). Available correlators (W*C/R) are preferably larger in number than the multipath number (N).
Referring back to
The programmable decimation filter 302 reduces the sample rate for coarse correlations. An input sample rate 301 can be, for example, 4 or 8 times a chip rate, while in accordance with one illustrative embodiment, the desired coarse correlation would be 1 or 2 times the chip rate. Note that with a processing clock, C, of 64 times the chip rate, only 8 correlation bins per sample can be computed, whereas at 1 times the chip rate, 64 correlation bins per sample could be computed.
The correlators 304 receive data samples from the filter 302 and correlate the data samples to a code sequence 402 generated by a sequence generator. The code sequences are appropriately delayed by the sequence delays 308 to provide alignment between the data samples and the code sequences 402. Trigger delays 310 provide synchronization to properly synchronize the operation of the correlators 304. Multiplexers, 312 are employed to ensure that the trigger delays ˜310 are activated together and in accordance with an initial_search signal 314. A frame synch signal 316 activates the trigger delays 310 and the sequence generator 306 to maintain the synchronization of operations of the multipath searcher circuit 300.
The sequence generator 306 produces a local (reference) copy of a pilot signal. The pilot signal is employed for detecting multipath signals and the reference copy will be compared to the sampled incoming signals (input signals 301). The local pilot signal generated by the sequence generator 306 is synchronized to the incoming input signal using known means.
The sequence delays 308 time shift the reference signal output from the sequence generator 306. These delays 308 are programmed to the time span of the correlators 304 in coarse mode such that the entire search window is covered by the correlators 304. While in fine mode, the sequence delays 308 are adjusted to match the time delays of the peaks discovered in the coarse mode. Adjustment of the sequence delays 308 is preferably based on a review of the correlator results after the coarse search mode. In one embodiment of the present invention, a software module 320 receives the results from the correlators and determines the energy peaks of the received pilot signal by reviewing the accumulated sums from the correlators 304. These sums are associated with a position (time) in the searching window. When a peak region is determined in module 320, sequence delays are adjusted in preparation of the fine search mode that succeeds the coarse search. The time shift of the sequence delays 308 focuses the fine mode search only on the discovered peaks determined in the initial coarse search.
The reference data (code sequence) 402 is presented to the correlators 304 at the computation rate (e.g., 8× the data sample rate) by a shift register, multiplexer or other device (not shown). For example,
The initial search signal (multiplexer) 314 is an optional operation, which forces all correlators 304 to start at the same time when performing the initial (coarse) search. The sequence delays 308 would be established so that a first correlator 304 receives code sequence 0 to 31, a next correlator (304) receives code sequence 32 to 63, etc. so that a continuous span of time would be covered by the correlators 304. The span of time corresponds with the size of the search window as described above.
The trigger delays 310 of
The trigger delays 310 decode time with respect to a reference (local) copy of the pilot signal. This is implemented by comparison to a counter, for example. The trigger delays 310 decode both a start signal for the correlators 304, as well as a write enable (“start” signal 411,
The correlators 304 correlate a code sequence to input data samples as filtered by the filter 302 and in accordance with the appropriate sample rate. The correlators 304 are described in greater detail with reference to
Referring to
The input 402 is from a sequence delay 308. The sequence data 402 is cycled through the delays of the sequence which the correlators 304 use. This can be accomplished using a shift register or multiplexer, as known in the art, to ensure proper synchronization between the input data 303 and the code sequence 402. The sequence delay is applied to the reference sequence, in one example, for WCDMA, such that the delay may include a single bit representing +/−1 to describe the sequence. The code sequence is a binary signal (+/−1) synchronized to the pilot signal component of the received signal (input data samples). Over the course of a chip period, different shifts of the code sequence are presented to the correlator and the input signal is multiplied by each shift of the code sequence and the product is accumulated in the register file. In practice, the multiplier can be implemented as a multiplexer, and the code sequence can be a bit signal (0 or a 1) driving the select of a multiplexer choosing either the input signal or −1 times the input signal.
The input data samples 303 and the code sequences 402 (delayed if needed) are multiplied by a multiplier 404 to provide alignment of time shifts between the reference sequence and the received pilot signals. The product signal is input to a summer 405 which adds zero or the accumulated value in the register file 406. In a first instance, the sum is added to zero so that the register file stores code sequence 0 corresponding to input sample zero (see
Referring to
In step 504, a coarse searching is performed by the multipath searcher at a lower resolution to find approximate energy peak locations of a received pilot signal within the searching window. This is performed by using the correlators to correlate a delayed reference code signal to the received pilot signal to identify code matches. The identified code matches are assumed to be peak or near peak signals. In one embodiment of the present invention, a peak is identified as approximately 3 times the average noise level of the input signal. The result is identified portions which are candidate peak locations.
At step 506, it is determined if any correlation exist (if peaks are found). If no correlation exists (no peaks have been found) the method proceeds to step 508 to set new searching parameters for another iteration or, alternatively, stops searching if programmed to do so. However, if at step 506 correlation is found (peaks exist), the method proceeds to block 510 to begin fine searching.
At step 510, once the approximate peak locations have been identified in the coarse search, a high resolution search (or fine search) is configured to get more accurate multipath profiles. Parameters for the fine search include fixing the fine resolution search window in an area close to or at the candidate peak locations to further search for multipath signals. The fine search is performed at a higher resolution to find more details around the result of the first pass (coarse). The fine search resolution could be, for example, as high as the sample rate, S, (and the coarse searching resolution could as low as the chip rate, R).
Other parameters set for the multipath searcher for the high resolution pass (fine pass) may include resolution (e.g., select a sample rate), thresholds (e.g., energy thresholds for determining whether a peak has been discovered), a number of iterations, etc.
In block 512, the fine search is performed and the searching results are obtained and sent to appropriate components in a receiver, for example a cellular telephone or other wireless device. Such components may include a rake receiver or any other receiver or component that needs to eliminate or determine an appropriate signal from a plurality of replica multipath signals. The method is then resent for the next searching operation.
It should be understood that coarse searching needs to encompass an entire searching window while fine searching only needs to encompass a very small searching segment (normally one chip time duration or period). Furthermore, a number of paths supported may be limited (e.g., approximately 6 paths). During fine searching, only locations around peaks (identified searching results) are needed to be further searched. If the code sequence is matched at N (odd number) points, the points are arranged as (N−1)/2 for early points and (N+1)/2 for late points around the assumed peak. Such regions are focused on in the fine search. This reduces the memory usage by 2, 4, 8, etc. times.
In an alternate embodiment of the present invention, the searcher can be configured as an independent high resolution searcher when the searching window is relatively small. The threshold of the searcher profile could be decided through calculation and experiments. Normally, a searching profile includes the results of limited single multipath searcher iterations.
Having described preferred embodiments for a multiresolution/multipath searcher (which are intended to be illustrative and not limiting), it is noted that modifications and variations can be made by persons skilled in the art in light of the above teachings. It is therefore to be understood that changes may be made in the particular embodiments of the invention disclosed which are within the scope and spirit of the invention as outlined by the appended claims. As such, the appropriate scope of the invention is to be determined according to the claims, which follow.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/US2006/002934 | 1/27/2006 | WO | 00 | 7/17/2008 |
Publishing Document | Publishing Date | Country | Kind |
---|---|---|---|
WO2007/086858 | 8/2/2007 | WO | A |
Number | Name | Date | Kind |
---|---|---|---|
5784364 | Ahn et al. | Jul 1998 | A |
6055264 | Kenney et al. | Apr 2000 | A |
6157820 | Sourour et al. | Dec 2000 | A |
6693955 | Arimitsu | Feb 2004 | B1 |
20020094017 | Wang | Jul 2002 | A1 |
20030012312 | Gerhards et al. | Jan 2003 | A1 |
20030235238 | Schelm et al. | Dec 2003 | A1 |
20040076225 | You et al. | Apr 2004 | A1 |
20050047488 | Sugahara | Mar 2005 | A1 |
20050185697 | Gargin | Aug 2005 | A1 |
20060007895 | Coralli et al. | Jan 2006 | A1 |
Number | Date | Country |
---|---|---|
1146133 | Mar 1997 | CN |
1307413 | Aug 2001 | CN |
1318921 | Oct 2001 | CN |
1492610 | Apr 2004 | CN |
1526206 | Sep 2004 | CN |
1079537 | Feb 2001 | EP |
2001-069568 | Mar 2001 | JP |
2003-158504 | May 2003 | JP |
2005-535880 | Nov 2005 | JP |
2005-333456 | Dec 2005 | JP |
2005-539442 | Dec 2005 | JP |
2006-005390 | Jan 2006 | JP |
WO0176087 | Oct 2001 | WO |
03007498 | Jan 2003 | WO |
2004015444 | Feb 2004 | WO |
2004028017 | Apr 2004 | WO |
WO2004077692 | Sep 2004 | WO |
Number | Date | Country | |
---|---|---|---|
20080317182 A1 | Dec 2008 | US |