Patent Application
20080151813
This invention relates to air interface communication systems synchronization between base stations and mobile devices and more particularly to initial acquisition and synchronization.
In wireless (air interface) communication systems, when a new user attempts to access a network, such as by powering on a subscriber unit, the subscriber unit needs to determine the frame/symbol timing of the downlink transmission from the network, along with the frequency offset and an identifying code (“cell ID”). Methods for determining this information include energy gap detection, auto-correlation, cyclic prefix detection, and cross-correlation.
The initial acquisition of WiMAX signals based in the IEEE 802.16e standard, however, pose challenges. Some of the reasons for the challenges include low operating signal-to-noise ratios (SNRs), lack of a repetition structure in the preamble, inconsistent repetition of the cyclic prefix, a large number of possible cell IDs, and potentially large frequency offsets. The low SNR affects the reliability of energy gap detection, the lack of a repetition structure in the preamble reduces the effectiveness of autocorrelation, inconsistencies in repetition reduce the effectiveness of cyclic prefix detection, and the large frequency offsets coupled with the number of cell IDs creates a significant computational burden for cross-correlation. The problem is compounded by the fact that at startup the subscriber unit is “blind” at power up (since it has little, if any, prior knowledge of the network, including timing and basestation codes). As a result the subscriber unit performs exhaustive searching, which is computationally expensive.
Time division duplexing (TDD) systems alternate between periods of transmission and reception. Energy gap detection attempts to ascertain the different periods by determining the level of energy in the appropriate frequency ranges. The boundaries of this gap then allow determination of timing information, which is used for acquisition. However, in low SNR environments, such as approximately 0 dB or lower, high noise or interference levels could prevent a detector from identifying a cessation in transmission by the subscriber unit. This is because the signal and interference levels are approximately the same when the SNR is close to 0 dB.
Autocorrelation schemes take advantage of identical segments in a transmission. That is, when a signal is correlated with a time-delayed version of itself, the identical segments, typically in a preamble, will produce a peak in the autocorrelation. This peak can then be used in the initial acquisition. If, however, the preamble does not have a repetition structure, the autocorrelation scheme may not provide the necessary peak. An alternative would be to leverage cyclic prefix timing detectors. A cyclic prefix is a common technique used in orthogonal frequency division multiplexing (OFDM) systems, which for each OFDM symbol is coded. This provides a form of a repetition structure, but it only occurs in part of a transmission frame and is not consistent in a TDD system.
Cross-correlation schemes are common techniques used for timing synchronization in which locally-stored versions of reference signals are correlated with the received signal in order to determine which one matches the basestation's transmitted cell ID. Each Basestation's cell ID is selected by the network planning from a set of predefined codes. However, WiMAX signals based in the IEEE 802.16e standard have a large code space consisting of 114 possible predefined codes. In a typical cross-correlation scheme, the received signal is correlated with all possible codes, along with various timing offsets. That is, a typical cross-correlation scheme attempts to solve:
where m is the timing offset, k is the reference code index, n is the sample index, L is the preamble length, r(n) is the received signal, and ck(n) is the kth local reference.
This is a two-dimensional (2D) search problem, which attempts to determine both the identifying code and the timing offset simultaneously. The m and k which maximize the correlation value are solved for simultaneously. The computational requirement may become burdensome for systems that may have a large number of matched filters, such as 114, where each one requires testing with multiple time shifts.
A received signal may be correlated with combinations of reference signals, each combination representing a subset of the reference signals, rather than being correlated with each reference signal individually. This reduces the number of computations needed to determine the timing offset. Once the timing offset has been found, the specific reference signal matching the received signal may be found with a one-dimensional (1D) search. If one of the representations produced a clear enough correlation peak, only the reference signals in the subset corresponding to that representation need to be correlated with the received signal in order to determine the transmitted code. A large 2-D search is thus reduced to a smaller 2-D search followed by a 1-D search.
Embodiments of the invention use representations for subsets of reference signals, such as a summation of each reference signal in the subset, and cross-correlate a received signal with the representations to determine the timing offset and a candidate subset. The candidate subset corresponds to the representation which produces the highest correlation value, and is expected to contain the reference signal which matches the code in the received signal. After the timing offset and candidate subset have been found, embodiments of the invention then cross-correlate only those reference signals in the candidate subset with the received signal preamble in order to determine which of the predefined identifying codes is present.
Embodiments of the invention provide for a method of signal detection comprising: determining a timing offset of a received signal using representations of reference signals, wherein at least one of the reference signals is expected to correlate with at least a portion of the received signal, and wherein at least one of the representations is a representation of two or more of the reference signals; and determining which of the reference signals matches said received signal using the timing offset. The received signal may be a WiMAX signal, which comprises a preamble containing an identifying code from a set of predefined codes. The set of reference signals would then correspond to the set of predefined codes, and the representations may be created by summing or averaging the reference signals in each subset. The determination of the timing offset may comprise cross-correlating the received signal with the representations. This allows identification of a candidate subset using the highest value of the cross-correlation results. If desired, embodiments of the invention may reconfigure the subsets and create new representations.
The foregoing has outlined rather broadly the features and technical advantages of the present invention in order that the detailed description of the invention that follows may be better understood. Additional features and advantages of the invention will be described hereinafter which form the subject of the claims of the invention. It should be appreciated by those skilled in the art that the conception and specific embodiment disclosed may be readily utilized as a basis for modifying or designing other structures for carrying out the same purposes of the present invention. It should also be realized by those skilled in the art that such equivalent constructions do not depart from the spirit and scope of the invention as set forth in the appended claims. The novel features which are believed to be characteristic of the invention, both as to its organization and method of operation, together with further objects and advantages will be better understood from the following description when considered in connection with the accompanying figures. It is to be expressly understood, however, that each of the figures is provided for the purpose of illustration and description only and is not intended as a definition of the limits of the present invention.
For a more complete understanding of the present invention, reference is now made to the following descriptions taken in conjunction with the accompanying drawing, in which:
It is not necessary to resolve both the timing offset and identify the specific reference code simultaneously. Rather, it is computationally advantageous to break the search problem into two steps: First the timing offset is found using a smaller search space, and then the specific reference code is identified. For example, if a system has 100 possible codes and 100 possible timing offset increments, the 2-D search must solve 100×100=10,000 correlation values. However, if the codes are divided into 10 subsets of 10 codes each only 10×100=1,000 correlation values for the first step. Once the 10 candidate subset has been identified and the timing offset has been resolved, only the 10 reference codes in the candidate subset will need to be correlated with the received signal, bringing the total number of correlations to only 1,000+10=1,010 rather than 10,000. The above numbers were selected for illustrative purposes only.
Once the set I of reference codes are partitioned into J subsets, denoted Ij, representations for the subsets are generated. One way to form a representation, denoted c′j(n), is to sum the codes as shown:
Cross-correlation is then done using the representations, rather than all of the individual codes:
where the m and j which maximize the correlation value are solved for simultaneously. While this remains a 2-D search problem, it may be considerably smaller. At this point, with m known, (denoted as {tilde over (m)}), the correlations need only be maximized for the k's within the candidate Ij as shown:
This is a 1D search problem, since the timing offset is fixed. Even if there is ambiguity regarding which Ij contains the matching reference signal, reference signals in multiple subsets, possibly even all the reference signals may be tested in a 1D problem. This would still provide computational savings over a single, large 2D search. Using the notional numbers given above, the 100 reference codes tested in a 1D search would bring the total number of correlations to 1,000+100=1,100, which is still fewer than 10,000. Such ambiguity is not normally expected to be present, however it is considered acceptable where it allows for broader coverage.
The basestation signal is received by the subscriber unit 44 in box 106, and is correlated with the representations in box 107 to determine the timing offset. With this information, and possibly a candidate subset also selected, the subscriber is able to determine the specific code transmitted by the basestation in box 108. At this point, initial acquisition and synchronization have been largely completed. It should be noted that, even if two or more subsets produce similar correlation results, such that there is ambiguity regarding which subset contains the matching reference code, the entire set of reference codes may be searched as a 1D problem. Thus, while identifying a candidate subgroup may speed computations, it is not necessary to identify a candidate subset in order to obtain significant computation savings.
Generally, the larger the subsets the better the computational savings may be. However, this comes at a cost. The more reference codes there are in each subset that must be represented simultaneously, the lower the probability of detection will be when attempting to solve for the timing offset. Thus, there is a trade-off between speed and accuracy. Criteria for selecting the operational point may include a required detection probability, SNR and computational burdens.
Although the present invention and its advantages have been described in detail, it should be understood that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the invention as defined by the appended claims. Moreover, the scope of the present application is not intended to be limited to the particular embodiments of the process, machine, manufacture, composition of matter, means, methods and steps described in the specification. As one of ordinary skill in the art will readily appreciate from the disclosure of the present invention, processes, machines, manufacture, compositions of matter, means, methods, or steps, presently existing or later to be developed that perform substantially the same function or achieve substantially the same result as the corresponding embodiments described herein may be utilized according to the present invention. Accordingly, the appended claims are intended to include within their scope such processes, machines, manufacture, compositions of matter, means, methods, or steps.
