DESC:FORM 2
THE PATENTS ACT, 1970
[39 of 1970]
&
THE PATENTS RULES, 2003
COMPLETE SPECIFICATION
(Section 10; Rule 13)

METHOD OF PROVIDING LOW-COMPLEXITY LONG TERM EVOLUTION- DEVICE TO DEVICE (LTE-D2D) SYNCHRONIZATION IN LTE D2D COMMUNICATIONS

SAMSUNG R&D INSTITUTE INDIA – BANGALORE Pvt. Ltd.
# 2870, ORION Building, Bagmane Constellation Business Park,
Outer Ring Road, Doddanakundi Circle,
Marathahalli Post,
Bangalore -560037, Karnataka, India
Indian Company

The following Specification particularly describes the invention
and the method it is being performed

RELATED APPLICATION
The present invention claims benefit of the Indian Provisional Application No. 3523/CHE/2015 titled " METHOD OF PROVIDING LOW-COMPLEXITY LONG TERM EVOLUTION- DEVICE TO DEVICE (LTE-D2D) SYNCHRONIZATION ALGORITHMS IN LTE D2D COMMUNICATIONS” by Samsung R&D Institute India – Bangalore Private Limited, filed on 9th July 2015, which is herein incorporated in its entirety by reference for all purposes.

FIELD OF THE INVENTION
The present invention generally relates to wireless communication and more particularly relates to a method of providing low-complexity long term evolution- device to device (LTE-D2D) synchronization algorithms in LTE D2D communications.

BACKGROUND OF THE INVENTION

Long term evolution (LTE) user equipment (UE) supporting D2D (Device to Device) features is expected to both transmit and receive in the legacy uplink band or uplink subframes in contrast to legacy LTE devices that only transmit on uplink resources, as all the D2D communications (sidelink) are expected to occur in legacy uplink resources. Most of the channels and signals defined for the LTE D2D in Release-12 are very similar to legacy uplink transmissions with marginal differences. In contrast to legacy LTE system, for D2D communications, the UE also plays as synchronization reference in multiple scenarios. For the UE, deriving the synchronization reference from eNode B (eNB) is similar to previous versions of 3GPP specifications, whereas deriving the synchronization reference from another UE is new in Release-12 for D2D. This task becomes challenging in two folds, with the new synchronization reference signal design, and increased frequency uncertainty observed by UE.

Figure 1 is a schematic representation 100 illustrating synchronization signals for normal CP case with primary sidelink synchronization signal (PSSS) and secondary sidelink synchronization signal (SSSS), according to an embodiment of the present invention. According to the Figure 1, A D2D sidelink synchronization signal consists of two kinds of signals, namely primary sidelink synchronization signal (PSSS) and secondary sidelink synchronization signal (SSSS). These two signals together indicate number of sidelink identifiers . PSSS for the normal cyclic prefix (NCP) case is transmitted on 1st and 2nd orthogonal frequency division multiplex (OFDM) symbols in the first slot of the sub-frame assigned for sync signal transmission and 0th and 1st symbol for extended CP (ECP) case. SSSS is transmitted on 4th and 5th symbols and 3rd and 4th symbols of the second slot for NCP and ECP case respectively.

The UE interested in D2D communication or discovery needs to find a valid sync reference. The UEs that are in a cellular network coverage take downlink (DL) frame timing as reference for any procedure. The UEs that are in out of coverage area look for any possible sidelink synchronization reference transmitted by other UEs, which are allowed to transmit synchronization signal. In out of coverage cases, the UE uses pre-configured D2D settings for default parameters to be used. Given the reference signals and structure as defined, the UE is able to synchronize time and frequency with in an acceptable accuracy and identify for further communication and estimate S-RSRP for selection or reselection purposes. The detection probability shall be greater than 90% at -4dB. Therefore, it is preferable to complete the synchronization procedure using one reception of PSSS and SSSS signals as shown in figure 1, that are transmitted once every 40ms.

Complexity of algorithm is expected to be minimized, as these synchronization functions are used regularly for selection/reselection of synchronization reference as defined. Given that a UE that is out of network coverage can also transmit synchronization signals, maximum frequency uncertainty expected at UE D2D receiver is significantly higher compared to legacy UEs and hence the challenge in algorithm design.

In view of the foregoing, there is a need of a method of providing low-complexity long term evolution- device to device (LTE-D2D) synchronization algorithms in LTE D2D communications.

The above mentioned shortcomings, disadvantages and problems are addressed herein and which will be understood by reading and studying the following specification.

SUMMARY OF THE INVENTION

The various embodiments of the present invention disclose method of providing low-complexity Long Term Evolution-Device too Device (LTE-D2D) Synchronization in LTE D2D Communications. The method comprises of steps of estimating, by a User Device, a symbol time based on Primary Sidelink Synchronization Signal (PSSS), verifying if the PSSS symbol start timing is exhausted, refining a PSSS correlation if the PSSS symbol start timing is not exhausted, correcting a frequency offset for the PSSS symbols based on a selected PSSS symbol start timing, estimating a channel on the PSSS based on a detected PSSS ID, equalizing a secondary side link synchronization signal (SSSS) symbols to determine a frequency offset having a high value, determining a sidelink identity (NID_SL) along with a symbol time and the frequency offset, validating the NID_SL, adding the validated NID_SL to a list of optimum frequency offset, and terminating synchronization of the LTE D2D devices at the validated NID_SL frequency offset.

In an embodiment of the present invention, the PSSS correlation is performed on a set of received signal samples using a set of reference signals with a set of predefined frequency offset hypothesis. In another embodiment of the present invention, the received signal at a selected PSSS peak location is re-correlated with the reference signal with at least 3 frequency offset hypotheses. In an embodiment of the present invention, the frequency offset hypothesis selection is determined based on a detection probability, wherein the frequency hypothesis need not be equally spaced to cover a full frequency uncertainty range. In another embodiment of the present invention, the frequency offset hypothesis selection is based on a maximization of an observed correlation magnitude to cover a full range of frequency offsets.

In an embodiment of the present invention, the PSSS correlation is implemented using an overlap and discard method of correlation for correlation by one part or more for partial correlation, wherein the overlap and discard method computes the correlation of the received signal with the PSSS reference signal.

In another embodiment of the present invention, the frequency estimation is obtained by using a value of frequency hypothesis and the re-correlation of the received signal with the PSSS signal with a preset frequency offset. In an embodiment of the present invention, a symbol domain correlation is defined to estimate a fractional frequency offset.

In another embodiment of the present invention, the received signal at a selected symbol location is compensated with an estimated finer frequency offset before continuing with SSSS detection. In another embodiment of the present invention, one or more PSSS peaks is selected for further processing based on a threshold obtained by taking an average of a local maxima extracted from one or more peak locations.

In another embodiment of the present invention, the method further comprises of refining the peak location based on an estimated frequency offset. In another embodiment of the present invention, the detected side link identity is verified by confirming a Cyclic Redundancy Check (CRC) pass on a Physical Sidelink Broadcast Channel (PSBCH) using an estimated time and frequency offset.

The foregoing has outlined, in general, the various aspects of the invention and is to serve as an aid to better understand the more complete detailed description which is to follow. In reference to such, there is to be a clear understanding that the present invention is not limited to the method or application of use described and illustrated herein. It is intended that any other advantages and objects of the present invention that become apparent or obvious from the detailed description or illustrations contained herein are within the scope of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS

The other objects, features and advantages will occur to those skilled in the art from the following description of the preferred embodiment and the accompanying drawings in which:

Figure 1 is a schematic representation illustrating synchronization signals for normal CP case with primary sidelink synchronization signal (PSSS) and secondary sidelink synchronization signal (SSSS), according to an existing art.

Figure 2 is a flowchart diagram illustrating a device to device (D2D) synchronization procedure, according to one embodiment.

Figure 3 is a schematic block diagram illustrating a user device or user equipment (UE) providing low-complexity Long Term Evolution-Device-to- Device (LTE-D2D) Synchronization in LTE D2D communication, according to an embodiment of the present invention.

Although specific features of the present invention are shown in some drawings and not in others, this is done for convenience only as each feature may be combined with any or all of the other features in accordance with the present invention.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides a method of providing low-complexity Long Term Evolution-Device to Device (LTE-D2D) Synchronization in LTE D2D Communications. In the following detailed description of the embodiments of the invention, reference is made to the accompanying drawings that form a part hereof, and in which are shown by way of illustration specific embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention, and it is to be understood that other embodiments may be utilized and that changes may be made without departing from the scope of the present invention. The following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present invention is defined only by the appended claims.

The specification may refer to “an”, “one” or “some” embodiment(s) in several locations. This does not necessarily imply that each such reference is to the same embodiment(s), or that the feature only applies to a single embodiment. Single features of different embodiments may also be combined to provide other embodiments.

As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless expressly stated otherwise. It will be further understood that the terms “includes”, “comprises”, “including” and/or “comprising” when used in this specification, specify the presence of stated features, integers, steps, operations, elements and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. As used herein, the term “and/or” includes any and all combinations and arrangements of one or more of the associated listed items.

Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

The present invention provides a method of providing low-complexity Long Term Evolution- Device-to-Device (LTE-D2D) Synchronization in LTE D2D Communications. Various embodiments are described in the present disclosure to describe the working of the method, but not limiting to the scope of the present invention.

The embodiments herein and the various features and advantages details thereof are explained more fully with reference to the non-limiting embodiments that are illustrated in the accompanying drawings and detailed in the following description. Descriptions of well-known components and processing techniques are omitted so as to not unnecessarily obscure the embodiments herein. The examples used herein are intended merely to facilitate an understanding of ways in which the embodiments herein can be practiced and to further enable those of skill in the art to practice the embodiments herein. Accordingly, the examples should not be construed as limiting the scope of the embodiments herein.

According to an embodiment of the present invention, a method of providing low-complexity Long Term Evolution-Device to Device (LTE-D2D) Synchronization in LTE D2D Communications is described herein. The method comprises step of estimating a symbol timing based on Primary Sidelink Synchronization Signal (PSSS). In a data frame transmitted by a user device or user equipment (UE) with LTE D2D communication capability, primary sidelink synchronization signal (PSSS) and secondary sidelink synchronization signal (SSSS) are present. The UE can estimate the symbol timing of the data frame based on the PSSS.

Further, the method comprises step of verifying if the PSSS symbol starts exhausted. Based on the data frame, the UE verifies whether the PSSS symbol is starting exhausted. Further, the method comprise step of refining a PSSS correlation if the PSSS symbol is not exhausted. If the UE identifies that the PSSS symbol is not exhausted, then the UE performs the refining of the PSSS correlation, wherein the PSSS correlation is performed on a set of received signal samples using a set of reference signals with a set of predefined frequency hypothesis. In an embodiment of the present invention, the PSSS correlation is implemented using an overlap and discard method of correlation for correlation by one part or more for partial correlation, where the overlap and discard method computes the correlation of the received signal with the PSSS reference signal. In an embodiment of the present invention, the received signal at a selected PSSS peak location is re-correlated with the reference signal with at least 3 frequency hypotheses. In an embodiment of the present invention, the frequency offset hypothesis selection is determined based on a detection probability; wherein the frequency hypothesis need not be equally spaced to cover a full frequency uncertainty range. In another embodiment of the present invention, the frequency offset hypothesis selection is based on a maximization of an observed correlation magnitude to cover a full range of frequency offsets, without departing from the scope of the invention. In an embodiment of the present invention, the received signal at the selected symbol locations is compensated with estimated finer frequency offset before continuing with SSSS detection. In an embodiment of the present invention, the frequency estimation is obtained by using a value of frequency hypothesis and the re-correlation of the received signal with the PSSS signal with a preset frequency offset, wherein a symbol domain correlation is defined to estimate a fractional frequency offset. In another embodiment of the present invention, the received signal at a selected symbol location is compensated with an estimated finer frequency offset before continuing with SSSS detection. Further, the method comprises step of correcting a frequency offset for the PSSS symbols based on a selected PSSS symbol start, and estimating a channel on the PSSS based on a detected PSSS ID. In an embodiment of the present invention, one or more PSSS peaks is selected for further processing based on a threshold obtained by taking an average of a local maxima extracted from one or more peak locations.

Further, the method comprises step of equalizing a secondary side link synchronization signal (SSSS) symbols to determine frequency offset having a high value. In another embodiment of the present invention, the method further comprises step of refining the peak location based on an estimated frequency offset. After correlating the PSSS symbols, the UE equalizes the SSSS symbols to determine frequency offsets with higher values. Further the method comprises step of determining a sidelink identity (NID_SL) along with a symbol time and frequency offset. In an embodiment of the present invention, the detected side link identity is verified by confirming a CRC pass on a PSBCH using an estimated time and frequency of the case. Upon determining the sidelink identity, the method comprises step of validating the NID_SL, and adding the validated NID_SL to a list of optimum frequency offset. Further, the method comprises step of terminating synchronization of the LTE D2D devices at the validated NID_SL frequency offset.

Figure 2 is a schematic flowchart 200 diagram illustrating a device to device (D2D) synchronization procedure, according to one embodiment. According to the flow diagram 200, at step 202, a User Device or user equipment (UE) estimates a symbol time based on Primary Sidelink Synchronization Signal (PSSS). At step 204, the user device verifies if the PSSS symbol start timing is exhausted. If the PSSS symbol start timing is not exhausted, then at step 206 the user device refines a PSSS. At step 208, the user device corrects a frequency offset for the PSSS symbols based on a selected PSSS symbol start timing. Further, at step 210, the user device estimates a channel on the PSSS based on a detected PSSS ID. At step 212, a secondary side link synchronization signal (SSSS) symbols are equalized to determine a frequency offset having a high value. At step 214, a sidelink identity (NID_SL) along with a symbol time and the frequency offset are determined. At step 216, the user device validates the NID_SL. Further, at step 218, the user device adds the validated NID_SL to a list of optimum frequency offset. Further, at step 220, the user device determinates synchronization of the LTE D2D devices at the validated NID_SL frequency offset.

Figure 3 is a schematic block diagram 300 illustrating a user device or user equipment (UE) providing low-complexity Long Term Evolution-Device-to- Device (LTE-D2D) Synchronization in LTE D2D communication, according to an embodiment of the present invention. According to the block diagram, the user device or UE 300 comprises of one or more antennas 302, a transceiver 304, a processor 306, wherein the processor 304 comprises of a verification module 308, a correlation module 310, a correction module 312, an estimation module 314, an equalizer 316, and a validator 318. Further, the UE 300 comprises of a storage unit 320.

According to the present invention, the one or more antennas 302 of the UE 300 transmit or receive data to the other UE over a network. The transceiver 304 receives the data and processes the data after receiving from the one or more antennas 302. Further, the processor 306 of the User Device or UE 300 estimates a symbol time based on Primary Sidelink Synchronization Signal (PSSS). The verification module 308 verifies if the PSSS symbol start timing is exhausted. Once the verification of the PSSS symbol is done, the correlation module 310 refines a PSSS correlation if the PSSS symbol start timing is not exhausted.

The correction module 312 corrects a frequency offset for the PSSS symbols based on a selected PSSS symbol start timing. The estimation module 314 estimates a channel on the PSSS based on a detected PSSS ID. Further, the equalizer 316 equalizes a secondary side link synchronization signal (SSSS) symbols to determine a frequency offset having a high value. Further, once the frequency offset with high value is equalized, a sidelink identity (NID_SL) along with a symbol time and the frequency offset are determined. Further, the validator 318 validates the NID_SL, and adds the validated NID_SL to a list of optimum frequency offset. Upon adding the validated NID_SL to the list of optimum frequency offset, synchronization of the LTE D2D devices at the validated NID_SL frequency offset can be terminated. In an embodiment of the present invention, the synchronization of the LTE D2D devices at the validated NID_SL frequency offset can be stored in the storage unit 320.

The present invention discloses an efficient synchronization method considering all receiver design constraints including received signal dynamic range, initial and remnant frequency uncertainty, complexity and wireless channel propagation effects. The present method proposes a primary sidelink synchronization signal (PSSS) correlation with low complexity using fast convolution methods, novel frequency hypothesis selection for improved performance, and sample normalization for PSSS correlation for robustness. The present method also discloses improvements in combining methods for coherent secondary sidelink synchronization signal (SSSS) detection with frequency offset correction. The present method discloses reduction in complexity of PSSS correlation without any loss in performance. The present invention also describes a fine frequency offset localization method using the PSSS properties.

According to the present invention, the received signal at the UE Rx antenna port on the D2D sidelink sub-frame after the receiver sample processor and down conversion be denoted by with the effect of channel and transmitter and receiver frequency uncertainties.
(1)
where denotes linear convolution, denotes the multipath fading channel, the transmitted signal, the receiver noise and and denote the normalized frequency offset transmitter and receiver respectively, normalized by subcarrier-frequency separation, . For the synchronization processing, time domain signal corresponding to central 6 RBs is considered with 1.92MHz sampling rate.

The present invention proposes multiple D2D synchronization methods that range from initial timing acquisition to final detection methods. According to an embodiment of the present invention, a PSSS correlation is performed on the received samples using the reference signals with a set of predefined frequency hypothesis using low complexity method. Further, a sample normalization method is disclosed minimizes the peak detection uncertainty in scenarios with high dynamic input signal range. Further, PSSS correlation outputs from multiple frequency hypotheses are used to find the symbol start locations and likely frequency hypothesis. For each shortlisted PSSS peak, the received signal at the selected PSSS peak location is further re-correlated with reference signal with 3 more frequency hypothesis and use the results to localize the frequency offset and finer symbol start location. Further from the finer symbol start location, finer frequency offset is estimated using symbol correlation method. Received signal at the selected symbol locations is compensated with estimated finer frequency offset, before continuing further with SSSS detection using coherent detection method using the signal properties of SSSS. Further, the detected SSSS is verified by confirming the CRC pass on PSBCH, using the timing and frequency information estimated during the process.

According to the present invention, Primary Sidelink Synchronization Signal (PSSS) Correlation and various other methods are described herein that reduces the implementation complexity without compromising performance. One of the methods used for PSSS correlation is partial correlation, which comprises of the following equation:
(2)
Wherein represents the reference signal generated at UE for D2D PSSS for . denotes the correlation output on receive antenna and can take value of 1 for single part correlation, which is same as matched filter correlation. In this paper, we consider values of M to be either 1 or 2 for complexity and performance comparisons.

The PSSS correlation method further comprises of i) overlap and discard method of correlation for reduced complexity, ii) frequency offset hypothesis selection for improved performance, and iii) sample normalization for improved robustness.

i) Overlap and Discard method of correlation
Based on (2), it is evident that PSSS correlation is of the order . According to the present invention, the Overlap and Discard (OaD) method can be used to implement M-part partial correlation method for all values of including value of 1 with significantly smaller complexity, . Overlap and Discard or Overlap and Save can be defined for implementation of linear convolution using the fast convolution methods using transform methods. Suggested method uses modified fast convolution method to achieve -part correlation. Linear convolution of two sequences and represented by can be implemented using the fast convolution method OaD. Overlap and discard methods have been famous for evaluating convolution of two functions. The same method can be used for computing the correlation of the received signal with PSSS reference signal. Given the reference sequence , correlation can be obtained by computing the linear convolution of received signal and , represented by , where superscript ‘*’ represents complex conjugate operation. For evaluating the correlation using OaD method, the FFT of the zero padded reference signal can be stored to obtain length transform domain reference signal. This will reduce the complexity of repetitive reference signal processing.

Let OaD based correlation of received signal and reference sequence , be denoted by , for OaD implementation with FFT/IFFT size of . M-part correlation can be implemented using OaD method, as follows. For the brevity of description without loss of generality, let us assume =2. For -part partial correlation method, reference signal shall be partitioned in to parts. Let them be denoted by and for =2, which are obtained as follows,
(2)
Then -part partial correlation output for =2, can be obtained by OaD method, as follows
(3)

Frequency offset hypothesis selection method:
The PSSS correlation is very sensitive to frequency offset between transmitted signal and received signal. This is because of out-of phase addition of correlation output from different segments of reference signal, which reduces the correlation output to a fraction of maximum correlation output obtained at near zero frequency offset values. To retain the detection probability for a range of frequency offsets, it is required to correlate the received signal with reference signal shifted in frequency by multiple offset hypotheses and choose the best. Though the implementation complexity increases with higher number of frequency hypothesis, it is required to meet the detection probability requirements. Therefore, for different values of M in partial correlation method, it is possible to find a set of optimum frequency offset hypothesis, which will maximize the detection probability. Based on simulation results, maximum of 3 frequency offset hypothesis will provide good detection probability of PSSS peaks.

Let denote the frequency offset hypothesis and M denote the number of parts in M-part partial correlation method. Then the frequency offset hypothesis that optimizes the detection probability is the one that maximizes the correlation magnitude for the full range of frequency offsets. Such a frequency offset can be found by maximizing the minima of correlation across the frequency range as,
(4)
wherein
(5)
and . Where represents cross correlation of reference signal ( ) with frequency offset f with reference signal with frequency offset . Closed form expression for this maximization is found to be non-trivial, hence optimum values of frequency offset hypothesis can be searched for different odd values of M using the symmetry property of cross correlation with frequency offsets. Other optimization criteria, that would help improve the performance is to maximize , which is even more complex.

Table 1 shows the optimal frequency hypothesis for 3 hypothesis case, based on the numerical search that satisfies the criteria in (4). It is required to choose the minimum number of frequency hypothesis, with which the requirements to minimize the total complexity of implementation can be met. Based on simulation results, it is found that with 3 frequency hypothesis, the requirements can be met.

Table 1: Optimal Frequency hypothesis for 3 hypothesis case at different bands based on (4), by numerical offline search.
800MHz Band for +/-20ppm (+/-16kHz) 2GHz band for +/-20ppm (+/-40kHz)
M=1 0, ±10.7kHz 0, ±9.3kHz
M=2 0, ±10.7kHz 0, ±7.2kHz

According to the herein above mentioned table, it can be observed that the choosing the frequency locations from optimal frequency hypothesis as described here, over uniform spacing across the maximum frequency offset range, can provide significant gain. The optimal frequency values found for a given band on adjacent bands can be reused, as the expected loss is not significant within the range of few 100MHz from center frequency.

According to the present method, it can be observed that M=2 in partial correlation is expected to perform better based on its higher correlation value across the range of frequency offsets. But performance also depends on noise rejection, which need not be same for M=1 and M=2 cases.

Sample normalization for robust peak detector
Identifying a valid PSSS peak and location from the correlation output is not obvious as it may look, because of different levels of signals received at the receiver and cross correlations, possibly from other user equipment’s (UEs) in the vicinity from the same or other cell area or from other UESyncRef UEs. To overcome this problem, the present invention discloses normalizing each sample received at the antenna port with its magnitude for PSSS correlation procedures, as defined below.
(6)

With this change on the received signal and regular PSSS correlation, valid PSSS peaks can be easily isolated irrespective of dynamic range of the input signal. This is achieved because of correlation output value being agnostic to magnitude of received signal, but dependent only on signal to noise and interference ratio observed. Other ways of achieving the same is by normalizing the regular correlation output with norm of received sample vector. Such a method is complex to implement compared to the method proposed here. Sample normalization method also helps in optimizing the implementation of PSSS correlation method in fixed point design because of its deterministic range of values. Using the simulation results, the normalization on input samples will cause SNR loss of ~1dB, with a significant reduction in uncertainty with peak detection, which reduces the total complexity. It is observed that, SNR loss is lesser for Zadoff-Chu (ZC) sequences, because of unit norm property, compared to other non-unit magnitude sequences. This method can be used only, if high dynamic range in the input signal is expected and is otherwise optional and can be bypassed.

PSSS Correlation parameters and combining method
For D2D synchronization and search use cases, CP length is known in advance, so one can prepare the reference signal for OaD method as defined in (0, including CP of the SCFDMA symbol. To minimize the storage requirement, it is preferable to store the sequences corresponding to NCP length symbols ( =137). Given there are two consecutive symbols with same PSSS transmitted for D2D synchronization purposes, it is possible to combine the correlation output such that correlation values separated by one SCFDMA symbol are added coherently. But this may hinder the performance of PSSS detector with large frequency offsets, as discussed in previous sections. So, it can combine the correlation output as follows
(7)
where is obtained as defined in (3) for parts, frequency offset hypothesis and for receive antenna . Correlation output shall be combined over multiple Rx antennas if used, to obtain as
(9)

Post processing and Coarse Frequency offset estimation
PSSS peak post processing
Correlation output from each hypothesis is processed to obtain multiple local maximums using a moving window based search to eliminate the multiple and duplicate peaks within a given window (typically, or ). Any valid local maximum detection method that fits the purpose can be used here, without departing from the scope of the invention. From the list of peak and peak locations, a simple peak classification algorithm can be used to shortlist a smaller set of peaks that have high chances of being a valid PSSS peak. Peak classification can be based on a threshold obtained from the weak peaks in a table. If the initial peak list is of size . Let denote the value of peak in the sorted peak list of size , sorted based on peak value, in the descending order. Then the list of peaks shortlisted for further processing can be obtained by checking each peak against a threshold as,
(10)
(11)
where c denotes a scale value > 1. For each shortlisted peak k, peak location ( ), frequency hypothesis and ZC root index are noted for further processing. For every shortlisted peak, the following processing is required to further continue with detection and validation, such as, but not limited to, Coarse frequency estimation and peak location by re-correlation, Symbol correlation based fractional frequency offset estimation, compensation of frequency offset, and the like.

Coarse frequency estimation and refining peak location
According to the present invention, frequency hypothesis are tried at PSSS correlation stage to maximize the probability of detecting a valid peak. In Coarse frequency estimation and refining peak location method, the frequency hypothesis can be refined using the cross correlation properties of PSSS. In Frequency offset hypothesis selection method, it can be observed that there are frequency offsets other than 0 (e.g.: at 30kHz, -30kHz), at which PSSS can peak. Each PSSS peak observed could have peaked because of frequency offset value being close to 0 or 30kHz or -30kHz. So, it is required to further find which of these frequency offsets caused the peak. For every PSSS peak shortlisted, it is possible to further find the coarse frequency offset by determining which of these modes (0, ) on the detected frequency hypothesis, as tagged to the peak shortlisted, is closer to actual frequency offset. By re-correlating the stored received signal around each of the shortlisted peak taking samples around each peak, with reference signals with frequency offsets , and where is the frequency hypothesis tagged to the peak being processed. This will not only provide the coarse frequency localization, but also provide a better symbol start estimate. Re-correlation output shall be processed to find frequency mode and finer location of strongest peak, in the vicinity of shortlisted peak k, using the methods described in Overlap and Discard method of correlation methodError! Reference source not found.. This effectively means that 9 frequency hypothesis have been tried at the cost of 3 frequency hypothesis, as the re-correlation phase is run only on very short span of samples.

Every peak in the shortlisted PSSS peak list , has the following details tagged to it: a) frequency hypothesis ( ) as listed in Table 1, b) initial location estimate ( ), c) finer frequency offset mode, ( 0, ), d) finer location, , e) ZC root-index. Total estimated coarse frequency offset for shortlisted peak till this step is,
(12)

Symbol correlation based fractional frequency offset estimation
Using the estimated coarse frequency offset obtained in (12), significant part of the frequency offset can be compensated. For compensation, it is worth observing that, there are only few possible for any value of and , and hence it is possible to optimize the computation sequence generation by storing few offline computed sequences. There are two consecutive symbols of PSSS and two consecutive symbols of SSSS that need to be compensated for each peak , assuming the start of first PSSS symbol start from the .

Let the coarse frequency offset compensated, received symbols for PSSS and SSSS, for the peak be denoted by and respectively for .

Finer frequency offset can be obtained from the angle of inner product of two symbols as,
(13)
where and can be coherently combined to obtain , to obtain the remnant frequency offset as,
(14)
Note that maximum remnant frequency offset acquisition range using this method is within , which is ±7kHz (approx.) for NCP and ±6kHz for ECP case. Based on the frequency hypothesis method described above, remnant frequency is ensured to be with in these limits.

This shall be compensated on and for to further continue with SSSS detection as described in next section. Total compensated frequency offset till now in the process is
(15)

SSSS detection and Validation
Structure of SSSS is very similar to 3GPP Rel-8 Secondary Synchronization signal (SSS), with two m-sequences interleaved. For D2D, same SSSS is transmitted twice, and there is only one kind of SSSS, in contrast to Rel-8 LTE downlink SSS that is different in SF0 and SF5. The present invention discusses two kinds of detection methods, a) Coherent detection method based on channel estimate from PSSS and b) Non-coherent method based on differential correlation.

The following operations are required to be performed for SSSS detection, namely, but not limited to, Channel estimation based on PSSS, Shortlist m0 indices, Compute the metric for all possible m1 indices for each m0 and combine with m0 metric coherently, shortlist top candidates passing a threshold, and the like. The shortlist top candidates passing the threshold can be described as below:

Let the channel estimate obtained by simple linear block averaging (block length 8 tones) across frequency domain and averaging over two PSSS symbols, be denoted by for peak candidate and tone. Let denote the received signal after frequency offset compensation on SSSS in time domain for the peak candidate for tone. Then the metric for shortlisting is found by
(16)
where corresponds to sequence. Top few indices shortlisted based on metric as in (16). For every index shortlisted, compute the metric similar to (16) using the odd tones and appropriate reference sequence. Let that be called for all the possible indices (up to 7) for each . Both and soft values for each value are coherently combined to form a new value as
(17)
Then can be used to find the list of and pairs that top the list, sorted in descending order. Pair is used to lookup . Coherent soft symbols obtained after equalization with channel estimated from PSSS symbols, from multiple Rx shall be combined coherently in (16) and (17).

Root index found for the peak candidate, is used to look up , which is used together with to find as . Validity of each of the detected candidate can be verified by confirming the CRC pass on PSBCH.

In the following detailed description of the embodiments of the invention, reference is made to the accompanying drawings that form a part hereof, and in which are shown by way of illustration specific embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention, and it is to be understood that other embodiments may be utilized and that changes may be made without departing from the scope of the present invention. The following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present invention is defined only by the appended claims.

,CLAIMS:
CLAIMS
We claim:

1. A method of providing low-complexity Long Term Evolution-Device-to- Device (LTE-D2D) Synchronization in LTE D2D communication, the method comprises of:
estimating, by a User Device, a symbol time based on Primary Sidelink Synchronization Signal (PSSS);
verifying if the PSSS symbol start timing is exhausted;
refining a PSSS correlation if the PSSS symbol start timing is not exhausted;
correcting a frequency offset for the PSSS symbols based on a selected PSSS symbol start timing;
estimating a channel on the PSSS based on a detected PSSS ID;
equalizing a secondary side link synchronization signal (SSSS) symbols to determine a frequency offset having a high value;
determining a sidelink identity (NID_SL) along with a symbol time and the frequency offset;
validating the NID_SL;
adding the validated NID_SL to a list of optimum frequency offset; and
terminating synchronization of the LTE D2D devices at the validated NID_SL frequency offset.

2. The method of claim 1, wherein the PSSS correlation is performed on a set of received signal samples using a set of reference signals with a set of predefined frequency offset hypothesis.

3. The method of claim 1, wherein the received signal at a selected PSSS peak location is re-correlated with the reference signal with at least 3 frequency offset hypothesis.

4. The method of claim 1, wherein the frequency offset hypothesis selection is determined based on a detection probability; wherein the frequency hypothesis need not be equally spaced to cover a full frequency uncertainty range.

5. The method of claim 1, wherein the frequency offset hypothesis selection is based on a maximization of an observed correlation magnitude to cover a full range of frequency offsets.

6. The method of claim 1, wherein the PSSS correlation is implemented using an overlap and discard method of correlation for correlation by one part or more for partial correlation, where the overlap and discard method computes the correlation of the received signal with the PSSS reference signal.

7. The method of claim 1, wherein the frequency estimation is obtained by using a value of frequency hypothesis and the re-correlation of the received signal with the PSSS signal with a preset frequency offset.

8. The method of claim 1, wherein a symbol domain correlation is defined to estimate a fractional frequency offset.

9. The method of claim 1, wherein the received signal at a selected symbol location is compensated with an estimated finer frequency offset before continuing with SSSS detection.

10. The method of claim 1, wherein one or more PSSS peaks is selected for further processing based on a threshold obtained by taking an average of a local maxima extracted from one or more peak locations.

11. The method of claim 1, further comprises of refining the peak location based on an estimated frequency offset.

12. The method of claim 1, wherein the detected side link identity is verified by confirming a Cyclic Redundancy Check (CRC) pass on a Physical Sidelink Broadcast Channel (PSBCH) using an estimated time and frequency offset.

Dated this the 8th day of July 2016

Signature

KEERTHI J S
Patent agent
Agent for the applicant