CN109792698A - Compensate the receiver and method of the frequency error of reference clock - Google Patents
Compensate the receiver and method of the frequency error of reference clock Download PDFInfo
- Publication number
- CN109792698A CN109792698A CN201680089895.8A CN201680089895A CN109792698A CN 109792698 A CN109792698 A CN 109792698A CN 201680089895 A CN201680089895 A CN 201680089895A CN 109792698 A CN109792698 A CN 109792698A
- Authority
- CN
- China
- Prior art keywords
- frame
- signal
- offset
- estimated value
- clock
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W56/00—Synchronisation arrangements
- H04W56/0035—Synchronisation arrangements detecting errors in frequency or phase
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/26—Systems using multi-frequency codes
- H04L27/2601—Multicarrier modulation systems
- H04L27/2647—Arrangements specific to the receiver only
- H04L27/2655—Synchronisation arrangements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/26—Systems using multi-frequency codes
- H04L27/2601—Multicarrier modulation systems
- H04L27/2647—Arrangements specific to the receiver only
- H04L27/2655—Synchronisation arrangements
- H04L27/2656—Frame synchronisation, e.g. packet synchronisation, time division duplex [TDD] switching point detection or subframe synchronisation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W56/00—Synchronisation arrangements
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Synchronisation In Digital Transmission Systems (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
Receiver has reception chain, and the reception chain is used to be formatted as the signal of frame from transmitter receipt, wherein each frame has synchronizing sequence.The receiver further includes reference clock, for generating the clock signal for coordinating the operation for detecting the signal of the reception chain.The reference clock has the frequency error relative to the clock at the transmitter, the transmission of signal described in clock coordination when described.The receiver further includes synchronization unit, for detecting operation described in the frequency error compensation for the reference clock: the receiver realizes above-mentioned compensation by following operation: being directed to each receiving frame, determine between following items with the presence or absence of offset: (i) is when by the reference clock universal time coordinated, the frame time started of the detection operation and/or carrier frequency；(ii) estimated value of the frame time started and/or carrier frequency obtained from the received synchronizing sequence of the frame, the offset is for synchronizing the detection operation and the candidate item for receiving signal.The receiver averages to the determination candidate offset of multiple frames.Then, the receiver is determined according to the average value of the determining candidate offset of the multiple frame to be applied to the compensation of the coordination of the detection operation, to synchronize the detection operation and the reception signal.It averages to the determination candidate offset of multiple frames and improves its accuracy, to more likely properly compensate for the frequency error of the reference clock.The probability that the frequency error improves the receiver and the transmission signal successful synchronization is correctly compensated, especially in the case where poor signal to noise.Set based on the timing to each frame is averaged, and the receiver can also optimize its timing estimated value to the frame boundaries in the reception signal.
Description
Technical field
The present invention relates to the technologies of the frequency error of the reference clock at compensated receiver.
Background technique
For user equipment reference clock is with respect to its nominal frequency, usually there is frequency error.This is mainly due to cost
Consider: because of point-device clock reference somewhat expensive.Base station and other centralized transmitters usually have than user equipment
More accurate clock.User equipment can be coordinated by generating the clock signal of oneself its local sample frequency, carrier frequency,
Timing etc., thus by oneself and base station synchronization.The clock signal of user equipment is usually all to obtain from its reference clock, therefore join
It examines the offset between the nominal frequency of clock and its actual frequency and shows as frequency error between user equipment and base station.Frequency
Error influence coordination reception device attempt detection base station signal all clock signals so that user equipment be more difficult to successfully from
Base station received signal.
A kind of method for alleviating these problems is base station synchronous signal transmission, and synchronization signal, which can be used, in user equipment comes really
The coordination and answer effective gap that its fixed detection operates.User equipment determines its timing off-set usually using synchronization signal
The estimated value of the offset of estimated value and its carrier frequency.When poor signal to noise, the quality of these estimated values can be reduced because of noise, and
And it will lead to and incorrect compensation is carried out to detection operation.A possibility that failing which increase receiver and base station synchronization.
Improved object of the present invention is to the frequency error to compensated receiver reference clock and to estimation frame timing into
Row improves, especially in the case where poor signal to noise.
Aforementioned and other targets are realized by the feature of independent claims.Further form of implementation is in appurtenance
It is required that, illustrate with it is apparent in attached drawing.
Summary of the invention
According in a first aspect, providing a kind of receiver comprising reception chain, for being formatted as frame from transmitter receipt
Signal, wherein each frame have synchronizing sequence.The receiver further includes reference clock, coordinates the reception for generating
The clock signal of the operation for detecting the signal of chain.The reference clock has relative to the clock at the transmitter
Frequency error, the transmission of signal described in clock coordination when described.The receiver further includes synchronization unit, for being directed to the ginseng
It examines and detects operation described in the frequency error compensation of clock.The receiver realizes above-mentioned compensation by following operation: being directed to
Each receiving frame determines between following items with the presence or absence of offset: (i) is when by the reference clock universal time coordinated, the detection operation
The frame time started and/or carrier frequency；(ii) since the frame that the received synchronizing sequence of the frame obtains when
Between and/or carrier frequency estimated value, it is described offset for for synchronize it is described detection operation and it is described receive signal candidate item.
The receiver averages to the determination candidate offset of multiple frames.Then, the receiver is according to the multiple frame
The compensation of the coordination of the detection operation will be applied to by determining that the average value of candidate offset determines, to synchronize the detection behaviour
Make and the reception signal.It averages to the determination candidate offset of multiple frames and improves its accuracy, thus more likely just
Really compensate the frequency error of the reference clock.Correctly compensate the frequency error improve the receiver with it is described
The probability of the signal successful synchronization of transmitter, especially in the case where poor signal to noise.
The synchronization unit can be used in the candidate offset group of multiple frames, determines and can be considered appointing for exceptional value in described group
What candidate offset.The synchronization unit can also be used to ignore the exception when generating the average value of the candidate offset group
Candidate offset.When the received signal of the receiver is very weak, single frame can generate incorrect candidate offset.It determines
Exceptional value and being ignored during averaging can prevent this candidate offset influence from averaging process.
The synchronization unit can be used in the candidate offset group of multiple frames, determine in described group in predetermined threshold each other
Candidate offset in maximum quantity.The synchronization unit can also be used in by being determined as in predetermined threshold each other to described
Candidate offset averages to generate the average value of the candidate offset group.To which the synchronization unit can efficient, system
It eliminates exceptional value and generates the average value in ground.
The synchronization unit can be used for sample position corresponding with correlation peak in frame being determined as candidate offset.The phase
Close the possibility test point in peak value instruction frame, the sampling instant corresponding to the test point be beneficial to indicate the receiver when
Sequence and carrier frequency shift, the offset are caused by the frequency error of the reference clock.
The synchronization unit can be used for generating the described flat of the candidate offset by the mean value for calculating the candidate offset
Mean value.This provides a kind of simple mechanisms of average value for obtaining candidate offset.
The synchronization unit can be used for by determining that the slope for the linear regression model (LRM) for being fitted to the candidate offset generates
The average value of these candidate offsets.To which the synchronization unit can determine the timing off-set of each frame easily, because with
The timing off-set of each frame constantly accumulate, the absolute timing off-set of each frame can change.
What the synchronization unit can be used for the frequency error according to the average generation of the candidate offset first estimates
Evaluation.The synchronization unit can be used for being generated according to the slope for the linear regression model (LRM) for being fitted to the candidate offset
Second estimated value of the frequency error.The synchronization unit can also be used in first estimated value and second estimated value
Overall estimated value of the weighted sum as the frequency error.By this way, the synchronization unit can obtain the totality of enhancing
Estimated value.
The synchronization unit can be used for the mean value according to the candidate offset, generates and is generated according to the clock signal
Carrier frequency and the signal carrier frequency between frequency shift (FS) the first estimated value.The synchronization unit can be used for root
According to the slope for the linear regression model (LRM) for being fitted to the candidate offset, the load generated according to the clock signal is generated
Second estimated value of the frequency shift (FS) between wave frequency rate and the carrier frequency of the signal.The synchronization unit can also be used to determine
Difference between first estimated value and second estimated value；And by by the difference and system frequency excursion of the determination
Predetermined set is compared to verify first estimated value of the frequency shift (FS) and second estimated value.Agree in transmitter
When having used a frequency in carrier frequency predetermined set surely, this provide a kind of estimation of individual authentication frequency shift (FS) whether
Reasonable mechanism.
The synchronization unit can be used for: if the difference between first estimated value and second estimated value exists
In the predetermined threshold of an offset in the system frequency excursion predetermined set, the estimated value of the frequency shift (FS) is determined
Effectively.A frequency in carrier frequency predetermined set has been used certainly in transmitter and this group of carrier frequency is according between known quantity
When separating, this provides a kind of whether reasonable mechanism of frequency shift (FS) of individual authentication estimation.
The candidate offset of the frame can indicate the timing estimated value of the frame.The synchronization unit can be used for by with lower section
The timing estimated value of formula refinement frame: the estimated value is adjusted to estimate the tropic pair of value set with the timing for being fitted to multiple frames
Together.The timing estimated value to framing that this obtains the synchronization unit is more accurate.
It according to second aspect, provides a method, comprising: the signal of frame is formatted as from transmitter receipt, wherein every
A frame has synchronizing sequence.The described method includes: generate the operation for detecting the signal for coordinating the reception chain when
Clock signal, wherein the reference clock has the frequency error relative to the clock at the transmitter, clock coordination institute when described
State the transmission of signal.The method also includes: operation is detected described in the frequency error compensation for the reference clock.This
Be accomplished by the following way: for each receiving frame, determine between following items with the presence or absence of offset: (i) is when by the reference
When clock coordination, the frame time started of the detection operation and/or carrier frequency；(ii) from the received synchronous sequence of the frame
Arrange the estimated value of the frame time started and/or carrier frequency that obtain, the offset for for synchronize detection operation with
The candidate item for receiving signal.It averages to the determination candidate offset of multiple frames.Then, according to the multiple frame
The compensation of the coordination of the detection operation will be applied to by determining that the average value of candidate offset determines, to synchronize the detection behaviour
Make and the reception signal.
According to the third aspect, provide that a kind of non-transient for being stored thereon with processor-executable instruction is machine readable to deposit
Storage media, described instruction realize the method for being formatted as the signal of frame from transmitter receipt.Each frame has synchronizing sequence.
The described method includes: being directed to the frequency error of reference clock, the operation for being used to detect the signal of reception chain is compensated, wherein
The reference clock is coordinated the operation for detecting the signal of the reception chain and is had at relative to the transmitter
The frequency error of clock, the transmission of signal described in clock coordination when described.This is accomplished by the following way: it is directed to each receiving frame,
Determine between following items with the presence or absence of offset: (i) is when by the reference clock universal time coordinated, the frame time started of the detection operation
And/or carrier frequency；(ii) the frame time started obtained from the received synchronizing sequence of the frame and/or carrier frequency
The estimated value of rate, the offset is for synchronizing the detection operation and the candidate item for receiving signal.Really to multiple frames
Determine candidate offset to average.Then, it will be answered according to the average value determination of the determining candidate offset of the multiple frame
The compensation of coordination for the detection operation, to synchronize the detection operation and the reception signal.
Detailed description of the invention
Let us now refer to the figures by way of example that present invention is described.In the accompanying drawings:
Fig. 1 shows the example of a receiver；
Fig. 2 is the exemplary flow chart of a signal acceptance method；
Fig. 3 is the example for detecting the state transition graph of signal；
Fig. 4 is the histogram of carrier frequency shift (Carrier Frequency Offset, CFO) estimated value；
Fig. 5 is the exemplary flow chart of CFO exceptional value determination process；
Fig. 6 shows the cross-correlation of NPSS signal in low dispersive channel；
Fig. 7 shows the cross-correlation of NPSS signal in high dispersive channel；
Fig. 8 is the exemplary flow chart for the method averaged to timing estimated value；
Fig. 9 shows the example of linear regression line；
Figure 10 is the exemplary flow chart for verifying the method for timing and frequency offset estimate f；And
Figure 11 shows the sketch plan of the NPSS testing process of enhancing.
Specific embodiment
Fig. 1 shows the receiver including reception chain 101, reference clock 102 and synchronization unit 103.Reception chain 101 is used for
Receive the signal from transmitter.These signal formats turn to frame, and each frame has synchronizing sequence, and synchronization unit 103 uses should
Synchronizing sequence is come the signal that synchronizes reception chain and transmitter is being sent.
The detection operation of reception chain 101 is coordinated by reference clock 102 " priori ".The practical examples of the coordination include reception chain
For receiving the sample frequency and local carrier frequency of signal.The two frequencies are usually all obtained from reference clock.Another example
It is internal " frame time " clock, reception chain primarily determines that frame is from certain certain base station using the clock.Reception chain is based on
The initial assessment of the possibility time started of frame relative to " frame time " clock is determined.Equally, " frame time " clock is logical
Often obtained from reference clock.Reference clock is the local clock usually provided by crystal oscillator.
If clock reference frequency is equal to its nominal value, the various clock signals of the detection operation at coordination reception device
Frequency is accurate, but such case seldom occurs.Especially relative to the clock for coordinating signal transmission, reference clock usually has
There is frequency error.Therefore, receiver is difficult to carrier frequency and/or timing signal-based receiver itself is synchronous with signal,
Unless receiver also knows the offset between its own frequency and timing estimated value and the frequency and timing estimated value of transmitter
It is how many.
Synchronization unit 103 is for determining how the frequency error for reference clock compensates the detection operation of reception chain.Inspection
Surveying operation can be any function of a part as signal receive process of reception chain execution, including mixing, demodulation, decoding
Deng.The general type of compensation is the coordination mode of adjustment detection operation.These adjustment would generally directly apply to coordination reception chain
The various clock signals, such as local carrier frequency, sample frequency, inside " frame time " clock etc. of middle detection operation.
Synchronization unit 103 determines how the frequency error of compensation reference clock it is preferable to use synchronizing sequence.Synchronizing sequence
It is the signal transmitted in each frame, the signal can be used to calculate receiver journey synchronous with the signal of transmitter in receiver
Degree.There are many synchronisation measures that synchronization unit can be used.One is whether there is offset between following two: receiver is to frame
At the beginning of initial estimate and the frame based on synchronizing sequence the practical time started update estimated value.Under another kind is
It states between two with the presence or absence of offset: local carrier frequency and the carrier frequency for being used for transmission signal.In some embodiments,
Offset can obtain from the particular sample moment in each frame or by its expression, which indicates received synchronous sequence
The possibility arrival time of column.When receiving signal to its own when multiple time lags are related, possible arrival time can be receiving frame
In correspond to correlation peak sampling instant.When receiving the local replica cross-correlation of signal and ideal synchronisation sequence, may arrive
It is also possible to the sampling instant corresponding to correlation peak up to the time.The angle that synchronization unit is also based on plural number obtains offset,
The complex representation receive signal to its own or it is related with the local replica of ideal synchronisation sequence.It is reached by the possibility to framing
Time or related angle determination or any offset indicated are properly termed as " candidate offset ".
The determination candidate offset that synchronization unit is not merely based on single frame determines the compensation to detection operation.On the contrary, synchronous
Unit can average to the candidate offset of multiple frames.The mode of the detection operation of compensation reception chain is determined according to the average value.
This improves the quality of gained timing and carrier shift estimated value, because the variance of the candidate offset after average is lower than single frame
The variance of candidate offset.This for that can barely be even more important with the user equipment (user equipment, UE) of base station synchronization,
For example, being located at the UE of cell extension or by biggish noise or the UE of other interference effects.
Synchronization unit shown in fig. 1 includes multiple optional sub-blocks of the different phase for realizing synchronizing process, including inclined
Move determiner 104, timing unit 105, frequency shift (FS) unit 106, offset validator 107 and estimated value enhancement unit 108.Offset
The candidate offset of determiner output is used by timing unit and frequency shift (FS) unit, with obtain just be included into detection operation when
The corresponding estimated value of sequence offset and carrier frequency shift.Then these corresponding estimated values can be tested together by offset validator
Card.The operation of these sub-blocks is more fully hereinafter described.
Structure shown in Fig. 1 (and all block assembly figures for including herein) is intended to correspond to multiple functional blocks.This is only used
In illustration purpose.Fig. 1 is not intended to strictly divide the different components of hardware or distinct program, process or function in software on chip
Energy.In some embodiments, some or all of signal processing technologies described herein may be by processing under software control
Device executes completely or partially.This piece of example can be synchronization unit.Processor can be mobile phone, smart phone, put down
The DSP of plate computer, general purpose I oT equipment or any other equipment with the ability of reception.Some or all of signals described herein
Processing operation can be executed entirely or partly with hardware.This is particularly suitable for the technology for repeating arithmetical operation, such as phase
It closes.Reception chain may include for executing mixing, code coverage mixing, symbol de-maps, frequency transformation, subcarrier demapping etc.
The specialized hardware of function.
Specific component in reception chain depends on specific waveform and telecom agreement that receiver is realized.Below with reference to receiver
Application for being run according to NB-IoT standard describes one or more embodiments of the invention.This is given for example only mesh
's.It should be understood that the scope of the present invention is not limited to any specific waveform or telecom agreement.
Fig. 2 shows the examples of the detection operation of synchronous reception chain and the technology of input signal.Process starts from step
S201: signal is received.Signal includes frame sequence, and each frame has synchronizing sequence.Preferably, synchronizing sequence known to receiver.
Each frame is handled to determine receiver appointing to the offset between priori estimates and the frame practical time started of frame time started
What candidate item.Also (or) processing frame to be to determine whether frame is that offset between local carrier frequency and transmitter carrier frequency mentions
For any candidate item (step S202).There are various well known signal processing technologies to extract timing and frequency from synchronization signal
Information.A kind of concrete signal processing technique is described below --- the detail of NPSS testing process, this is for illustrative purposes only.
Any suitable technology can use.
In step S203, multiple candidate offsets are collected together.Candidate offset in set is from reception signal
What different frame generated.It is alternatively possible to some possible candidate offsets be excluded from set, because entry evaluation shows these candidates
Offset is exceptional value, therefore cannot indicate the true excursions of coordination reception chain detected in the clock signal operated.Then to time
Offset collection is selected to average (step S203).Then, the frequency error that reference clock is directed to using average candidate offset is mended
Repay detection operation (step S204).Synchronization unit directly can carry out adequate compensation or synchronization unit passes through into receiver
The instruction of another functional block output adequate compensation to carry out adequate compensation indirectly.
A kind of possible application of above-mentioned receiver is that receiver is used to receive the narrowband according to 3GPP NB-IoT standard transmission
Signal.The standard specifies the receiver of referred to as user equipment (User Equipment, UE) for realizing with base station (base
Station, BS) initial synchronisation synchronization signal.The synchronization signal is known as narrowband primary synchronization signal (Narrowband
Primary Synchronisation Signal, NPSS).UE estimates its carrier frequency shift using NPSS signal
The timing of (carrier frequency offset, CFO) and BS signal frame boundaries.
Other embodiments are described with reference to concrete application, in the specific application, receiver is by for according to 3GPP
The UE of NB-IoT standard operation is realized.This is for illustrative purposes only.Receiver described herein and signal processing technology are not limited to
Any specific telecommunication standard or agreement.Particularly, many signal processing technologies described below have general applicability, can be with
It is applied below applying in different scene and embodiment from 3GPP NB-IoT for specific descriptions.
NPSS testing process is summarized
NPSS signal is by the every 10ms frame transmission primaries of BS.NPSS signal is the concatenation based on short Zadoff-Chu sequence, short
Zadoff-Chu sequence all has good correlation properties in time domain and frequency domain.
Receive process R1-161981 " NB-PSS and NB-SSS design (revised edition) (NB-PSS in Qualcomm of NPSS
And NB-SSS Design (Revised)) " (RAN1NB-IoT Ad-hoc#2, French Sofia peace mention Pohle this, 2016 3
Month) in be described later in detail.The process is referred to herein as " NPSS testing process ".The general introduction of the process is shown in FIG. 3, figure
3 show the example of the state transition graph of detection NPSS signal.
NPSS testing process is two stages process.Substantially, the process is by auto-correlation stage 301 and cross-correlation stage 302
Composition.Compared with the process subsequent more accurate stage, the sampling rate that the early stage of testing process uses is lower, to be simple
UE reduces the complexity of detection NPSS signal.The process is related to:
A, by calculating received synchronous sequence with the sampling rate Continuous plus ' cost function ' of 240kHz or 120kHz
The auto-correlation of column.It will be more than that the cost function peak value of threshold value is considered as possible NPSS and detects and submit it processing step b.
B, the cross-correlation of calculating NPSS signal and ideal signal is assumed according to multiple CFO and timing, wherein according to associated
The timing selection of cost function peak value receive the appropriate part of signal for cross-correlation.It equally, will be more than all peaks of threshold value
Value is considered as possible NPSS detection, selects peak-peak as most promising candidate item.Fig. 6 and Fig. 7 are shown according to multiple
The example of the cross-correlation of hypothesis.
The reason of testing multiple hypothesis in stepb is to solve any uncertainty in original estimated value.Peak value
Cost function angle (as generated in step a) at position is CFO (CFOpeak) provide the estimated value that range is ± 7kHz.
In order to solve the CFO for being greater than ± 7kHz, receiver must test multiple ' integer CFOs ' it is assumed that must test n*14kHz+
CFOpeakA CFO.Therefore, the purpose of NPSS process is to obtain the refinement frequency and timing off-set of each frame, but require cost letter
Number peak value and cross-correlation peak value are above respective detection threshold value in the frame.
The hypothesis for generating highest cross-correlation is considered as correct hypothesis.This solves uncertain in the frequency shift (FS) of estimation
Property.Sampling instant corresponding to the peak value is the possibility test point of the frame.The sampling instant indicates " candidate offset ", can be used
" candidate offset " is somebody's turn to do to obtain timing off-set and local carrier frequency offset.And not all frame all have possibility test point (for example,
Since correlation peak is unsatisfactory for threshold value), the candidate offset for being output to the process follow-up phase is not necessarily successive frame.
Receiver shown in FIG. 1 will realize NPSS testing process or be related to relevant similar testing process, then the receiver
In offset determiner 104 will preferably include one or more (hardware or software realization can be used) correlators.
Timing principle summary
In UE, receive signal sample frequency (sampling rate) and local carrier frequency all with usually crystal reference
Local clock is proportional.If the frequency of clock reference is equal to its nominal value, these frequencies are accurate.But usual feelings
Condition is really not so.Usual situation can indicate are as follows:
Wherein, ε is frequency error of the UE clock reference relative to its nominal frequency, usually with part per million (parts-
Per-million, ppm) it indicates, and be between several ppm to tens ppm in UE.
UE local carrier frequency is proportional to UE clock reference, so that:
UE is the load that UE is used for transmission these signals to BS transmitter based on the CFO that received NPSS signal is observed is analyzed
Difference between the estimated value and UE local carrier frequency of wave frequency rate:
It is usually 0.1ppm grades or better frequency tolerance that clock reference used in BS, which has, this is than typical UE clock
With reference to more acurrate.Therefore, for the sake of simplicity, BS carrier frequency can be considered as accurately.
According to the agreement reached in 3GPP NB-IoT standardisation process, the carrier frequency that BS is used for transmission NPSS signal can
Be 100kHz exact multiple (at ' independence ') or 100kHz multiple plus set -7.5kHz, -2.5kHz,
2.5kHz, 7.5kHz } in constant offset (at ' band in ' or ' guard band ').Therefore, in general, 100kHz light
The UE receiver that NPSS is searched on grid will be observed that the CFO with offset d.
Wherein, d is in set { -7.5kHz, -2.5kHz, 0kHz, 2.5kHz, 7.5kHz }.
If the CFO estimated value of UE is also considered as accurately:
The sampling rate of receiver are as follows:
Thus, it is supposed that receiver is sampled with the nominal sampling rate of 1.92MHz to signal is received, then receiver is observed
The received NPSS signal arrived had comprising 19200* (1+ ε) a period for receiving sampling.
It averages to CFO estimated value
It has been that any candidate offset that frame determines is output to timing off-set unit by offset determiner 104 in Fig. 1
105 and frequency shift (FS) unit 106.The unit is preferred for storage input candidate offset to form the candidate offset collection of multiple frames
It closes.Frequency shift (FS) unit is for averaging to the candidate offset of multiple frames to obtain the overall estimated value (as described below) of CFO.
It is also likely to be beneficial that any this additional process steps of abnormal candidate item are removed before calculating average value.
Timing off-set unit 105 and frequency shift (FS) unit 106 are usually from offset determiner 104 received " candidate offset "
Sampling instant corresponding to the possibility test point in each frame.In any stage of the process, can by these timing off-sets it
Between difference be converted into frequency values, such as by the set by Regression Model Simulator to such timing off-set, as described below.
CFO estimation of the CFO estimated value as the NPSS detection candidate item in different frame after can simply obtaining averagely
The arithmetic equal value of value set.It averages and improves the quality of CFO estimated value, because the variance of average value is lower than single estimated value
Variance.Fig. 4 illustrates this point.It (is 8550Hz) in this example that Fig. 4, which shows that single CFO estimated value may deviate practical CFO,
200Hz, and mean estimates (in this example be 8524Hz) is closer to right value.
Due to the characteristic of NPSS signal, CFO ' exceptional value is removed before calculating average value ' additional process steps can be
Beneficial (as shown in Figure 5).
As described above, NPSS testing process is related to searching for so-called ' integer CFO ' hypothesis.This is related to that signal and frequency will be received
0kHz is deviated in rate, ± 14kHz is related to the ideal NPSS signal progress of ± 28kHz.If incorrect integer frequency is because mentioning
It has supplied to be erroneously selected to the best related of signal is received, then the processing can generate incorrect NPSS timing and CFO estimates
Evaluation.When low with signal-to-noise ratio under certain channel conditions, it may occur that such case.Fig. 6 and Fig. 7 illustrate this problem.Low
In the channel of dispersion, NPSS signal and related between itself be significantly greater than offset ± 28kHz in NPSS signal and frequency from
Correlation between the copy of body, as shown in Figure 6.However, this surplus can greatly reduce, such as Fig. 7 in the more channels of dispersion
It is shown.In this channel, when SNR is low, noise can further decrease required correlation peak and enhance unwanted phase
Peak value is closed, it is incorrect so as to cause detecting.
If selected incorrect integer frequency it is assumed that if will lead to very big CFO evaluated error and timing is estimated
Count error.For example, in figure 6 and figure 7, relative to the relevant peak value that (correct) 0kHz integer is assumed, integer assumes ±
The relevant peak value of 28kHz is offset in time.If the CFO for corresponding to this incorrect NPSS detection is included in CFO
It averages in processing, then the CFO after being averaged can also malfunction (even if the size of error will be reduced by averaging).
It is eliminated in the CFO set of estimated value preferably by this CFO error after being used to generate averagely.This can by by it
Be determined as " exceptional value " to complete.The process of Fig. 5 shows the example of the suitable procedure of determining exceptional value.In the flow chart
In, ' T ' is the threshold value (such as 250Hz) suitably selected, and CFO (n) indicates the CFO for corresponding to detection serial number ' n '.
Receiver searches for NPSS signal within the period for covering multiple frames.Assuming that (according to NPSS testing process) carries out
Multiple NPSS detection, the then CFO for storing these detections gather to be handled.Once establishing these candidate CFO set, Fig. 5
Process start.The process is primarily to establish maximum CFO group, these CFO are in mutual predetermined threshold, the CFO group
It can gather from candidate CFO and establish.It is executed in the step S503 of Fig. 5 and checks particular candidate CFO whether another candidate CFO's
Process in predetermined threshold.If meeting threshold value standard, another candidate CFO, which is added to, will become the base for operation of averaging
In the summation of plinth, and the quantity for the candidate CFO for including in current " candidate set " is added into 1 (step S504).The process is repeated, until
Particular candidate CFO is compared (step S502 to S506) with all other candidate CFO.If the time at the end of the process
Choosing group is greater than any candidate set that previously one of candidate CFO is determined, then maximum group of (step in candidate set replacement precedence record
Rapid S508).Identical process repeated to all candidate CFO, i.e., all candidate CFO alternately as particular candidate CFO come with own
Other candidate CFO are compared (step S501 to S510).At the end of the process, the average CFO being retained on record will be
The average value of the maximum group of candidate CFO in mutual predetermined threshold, which can gather from original candidates CFO produces
It is raw.(step S508).
Therefore, process shown in fig. 5 is used to eliminate the exceptional value in candidate CFO set, and obtains remaining candidate's CFO group
Average value.
Seek the average value of timing estimated value
It has also been that any candidate offset that frame determines is output to timing off-set list by the offset determiner 104 in Fig. 1
Member 105.Preferably, also storage inputs candidate offset to form the candidate offset set of multiple frames.It averages for it
Operation, timing off-set unit are preferably used for the candidate offset set as frequency shift (FS) unit 106, that is, remove any
The candidate offset set of exceptional value.Timing off-set unit can be used process shown in Fig. 8 oneself to calculate the flat of candidate offset
Mean value.
Fig. 8 shows the example for seeking the process of average value of timing estimated value.In step S801, receiver cover it is more
NPSS signal is searched in the period of a frame.Assuming that having carried out multiple NPSS detection (step S802) within this time, then deposit
Store up that these individually detect when ordered sets to be handled (step S803).Timing can with as unit of the received sample of UE come table
Show.In step S804, linear regression model (LRM) is fitted to the when ordered sets obtained in step S801.The slope and UE of model
Clock reference frequency error ε is proportional.Then the NPSS timing estimated value after obtaining averagely in the following manner: adjustment estimated value
(step S805) is located on the tropic.
Fig. 9 shows an example of linear regression model (LRM).It can be by the step of ordered sets when linear regression model (LRM) is fitted to
It is following to complete:
A, with the sequence for the timing for indicating each detection as unit of the received sample of UE, wherein the nominal sampling rate of UE is
1.92MHz:T (0), T (1) ... T (total_detections-1).(actual sample rate in UE is provided by equation 6).
B, x (n)=round ((T (n)-T (0))/19200) and y (n)=* 19200 sequence of T (n)-T (0)-x (n) are calculated.
For example, Fig. 9 is figure y (n)=f (x (n)) of the set of each detection, the corresponding tropic is also shown.Y (n)=f (x (n)).
C, it calculates summationThe coefficient a and b of minimum.
By using coefficient a and b, the estimation of NPSS signal in ' the timing estimated value after average ' i.e. any frame k is reached
It is expressed as follows as unit of the time sample of the reception signal sampling rate of UE:
Y ' (k)=a*k+b
Wherein, k is the frame index relative to the frame for including the first single detection (k=0 at this time).
The estimated value of UE clock reference frequency error based on the timing regression slope a calculated as described above are as follows:
εT=a/19200
Timing off-set unit (105) indicates a timing estimation from the received each candidate offset of offset determiner (104)
Value.Therefore, timing off-set unit for refining timing estimated value associated with a frame in the following manner: the timing is estimated
It is worth and is aligned with the tropic for being fitted to the timing estimation value set obtained from multiple frames.
Estimated value verifying
Above-mentioned technology can be further improved by verifying CFO and timing estimated value.This can reception as shown in Figure 1
Offset validator 107 in device executes.Deviate validator can process according to Fig.10, execute mutual authentication.
As described above, CFO and UE sampling rate is all related with UE clock reference frequency error ε.Can using this relationship come
Further increase the reliability of NPSS detection and the accuracy of CFO and timing estimated value.
Firstly, the UE clock reference frequency error estimated based on timing regression slope is expressed as εT.The process starts from
Step S1001: the system frequency excursion d indicated by CFO estimated value relative to 100kHz grating is determined:
The difference calculated in equation 7 should be close to one in set { -7.5kHz, -2.5kHz, 0kHz, 2.5kHz, 7.5kHz }
A offset.If UE has with the priori knowledge of operation or independent operation in interior operation, guard band, the set can reduce.
Therefore, by d0(step S1002) is compared with scheduled system frequency excursion set.For example, ifHz, by gathering closest to d0Offset be expressed asIt can announce the invalid (step of the detection
S1003)。
Estimated value enhancing
Optional estimated value enhancement unit 108 is for generating the frequency error of reference clock using timing and CFO estimated value
Improved overall estimated value.The UE clock reference frequency error estimated based on CFO is expressed as εF:
Then, which improves device can calculate the estimation of UE clock reference frequency errors based on timing and CFO two
Value.For example, it may be the weighted mean of two estimated values:
εFT=wF*εF+wT*εT (9)
Then, estimated value, which improves device, can be used εFTCFO estimation after refining the timing estimated value after being averaged and/or being averaged
Value.
Figure 11 shows the general introduction of above-described enhancing NPSS testing process.To the CFO that is obtained in multiple frames and when
Sequence estimated value, which is averaged, improves the accuracy of estimated value, especially when NPSS signal is very weak and UE must be in signal-to-noise ratio
When detecting signal in low-down situation.
Above-mentioned NPSS testing process is the current version of the process, can be changed with the development of 3GPP NB-IoT standard.
It should be understood that the enhancing described herein to existing NPSS testing process is not limited to be used together with existing procedure, can equally fit
Any evolution sooner or later for existing procedure.Particularly, technique described herein is not limited to using any particular sample frequency.
Specific sample frequency mentioned above is merely illustrative.
Applicant is separately disclosed each personal feature described herein and times of two or more this category features herein
Meaning combination.It, can be real as a whole by this category feature or combination based on this specification with the general knowledge of those skilled in the art
It is existing, without considering whether the combination of this category feature or feature can solve any problem disclosed herein；And not to claim
The range of book causes to limit.Applicant points out that each aspect of the present invention can be by the combination structure of any this kind of independent feature or feature
At.In view of being described above, can carry out various modifications within the scope of the invention is apparent for a person skilled in the art
's.
Claims (12)
1. a kind of receiver characterized by comprising
Reception chain, for being formatted as the signal of frame from transmitter receipt, wherein each frame has synchronizing sequence；
Reference clock, for generating the clock signal for coordinating the operation for detecting the signal of the reception chain, wherein institute
Stating reference clock has frequency error relative to the clock at the transmitter, the transmission of signal described in clock coordination when described；
And
Synchronization unit detects operation for being directed to by following operation described in the frequency error compensation of the reference clock:
For each receiving frame, determine between following items with the presence or absence of offset: (i) is described when by the reference clock universal time coordinated
Detect frame time started and/or the carrier frequency of operation；(ii) described in the received synchronizing sequence acquisition from the frame
The estimated value of frame time started and/or carrier frequency, the offset is for synchronizing the detection operation and the reception signal
Candidate item；
It averages to the determination candidate offset of multiple frames；And
The association of the detection operation will be applied to by being determined according to the average value of the determining candidate offset of the multiple frame
The compensation of tune, to synchronize the detection operation and the reception signal.
2. receiver according to claim 1, which is characterized in that the synchronization unit is used for:
In the candidate offset group of multiple frames, any candidate offset that can be considered exceptional value in described group is determined；And
Ignore the abnormal candidate offset when generating the average value of the candidate offset group.
3. receiver according to claim 1 or 2, which is characterized in that the synchronization unit is used for:
In the candidate offset group of multiple frames, the maximum quantity of the candidate offset in described group in predetermined threshold each other is determined；
And
The candidate offset group is generated by averaging to the candidate offset being determined as in predetermined threshold each other
The average value.
4. receiver according to any one of the preceding claims, which is characterized in that the synchronization unit is used for will be in frame
Sample position corresponding with correlation peak is determined as candidate offset.
5. receiver according to any one of the preceding claims, which is characterized in that the synchronization unit is based on passing through
The mean value for calculating the candidate offset generates the average value of the candidate offset.
6. receiver according to any one of the preceding claims, which is characterized in that the synchronization unit is used for by true
Surely the slope for being fitted to the linear regression model (LRM) of the candidate offset generates the average value of these candidate offsets.
7. receiver according to any one of the preceding claims, which is characterized in that the synchronization unit is used for:
According to the first estimated value of frequency error described in the average generation of the candidate offset；
The second of the frequency error is generated according to the slope for the linear regression model (LRM) for being fitted to the candidate offset
Estimated value；And
Using the weighted sum of first estimated value and second estimated value as the overall estimated value of the frequency error.
8. receiver according to any one of the preceding claims, which is characterized in that the synchronization unit is used for:
According to the mean value of the candidate offset, the carrier frequency generated according to the clock signal and the signal are generated
First estimated value of the frequency shift (FS) between carrier frequency；
According to the slope for the linear regression model (LRM) for being fitted to the candidate offset, generate raw according to the clock signal
At carrier frequency and the signal carrier frequency between frequency shift (FS) the second estimated value；
Determine the difference between first estimated value and second estimated value；And
By being compared to the difference of the determination and system frequency excursion predetermined set to verify the institute of the frequency shift (FS)
State the first estimated value and second estimated value.
9. receiver according to claim 8, which is characterized in that the synchronization unit is used for: if first estimation
It is worth the predetermined of an offset of the difference between second estimated value in the system frequency excursion predetermined set
In threshold value, determine that the estimated value of the frequency shift (FS) is effective.
10. receiver according to any one of the preceding claims, which is characterized in that the candidate offset of the frame indicates institute
The timing estimated value of frame is stated, the synchronization unit for refining the timing estimated value of frame in the following manner: adjusting the estimation
Value is to estimate that the tropic of value set is aligned with the timing for being fitted to multiple frames.
11. a kind of method characterized by comprising
The signal of frame is formatted as from transmitter receipt, wherein each frame has synchronizing sequence；
Generate the clock signal of the operation for detecting the signal of coordination reception chain, wherein the reference clock has phase
For the frequency error of the clock at the transmitter, the transmission of signal described in clock coordination when described；And
By detecting operation described in the frequency error compensation of the following operation for the reference clock:
For each receiving frame, determine between following items with the presence or absence of offset: (i) is described when by the reference clock universal time coordinated
Detect frame time started and/or the carrier frequency of operation；(ii) described in the received synchronizing sequence acquisition from the frame
The estimated value of frame time started and/or carrier frequency, the offset is for synchronizing the detection operation and the reception signal
Candidate item；
It averages to the determination candidate offset of multiple frames；And
The association of the detection operation will be applied to by being determined according to the average value of the determining candidate offset of the multiple frame
The compensation of tune, to synchronize the detection operation and the reception signal.
12. a kind of non-transient machinable medium for being stored thereon with processor-executable instruction, which is characterized in that institute
Method of the instruction realization for being formatted as the signal of frame from transmitter receipt is stated, each frame has synchronizing sequence, the method
Include:
For the frequency error of reference clock, the operation for being used to detect the signal of reception chain is compensated, wherein when the reference
The operation for detecting the signal of reception chain described in clock coordination simultaneously has frequency relative to the clock at the transmitter
Error, the transmission of signal described in clock coordination when described:
For each receiving frame, determine between following items with the presence or absence of offset: (i) is described when by the reference clock universal time coordinated
Detect frame time started and/or the carrier frequency of operation；(ii) described in the received synchronizing sequence acquisition from the frame
The estimated value of frame time started and/or carrier frequency, the offset is for synchronizing the detection operation and the reception signal
Candidate item；
It averages to the determination candidate offset of multiple frames；And
The association of the detection operation will be applied to by being determined according to the average value of the determining candidate offset of the multiple frame
The compensation of tune, to synchronize the detection operation and the reception signal.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/EP2016/073934 WO2018065061A1 (en) | 2016-10-06 | 2016-10-06 | Receiver and method for compensating a frequency error in a reference clock |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109792698A true CN109792698A (en) | 2019-05-21 |
| CN109792698B CN109792698B (en) | 2021-02-12 |
Family
ID=57153446
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201680089895.8A Active CN109792698B (en) | 2016-10-06 | 2016-10-06 | Receiver and method for compensating frequency error of reference clock |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109792698B (en) |
| WO (1) | WO2018065061A1 (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112217601A (en) * | 2019-07-12 | 2021-01-12 | 天地融科技股份有限公司 | Data frame communication method and system |
| CN112217600A (en) * | 2019-07-12 | 2021-01-12 | 天地融科技股份有限公司 | Data frame receiving method and device, and communication method and system |
| CN112383498A (en) * | 2020-11-17 | 2021-02-19 | 紫光展锐(重庆)科技有限公司 | Low-frequency clock compensation method and device, storage medium and terminal |
| CN112636859A (en) * | 2020-12-23 | 2021-04-09 | 西安云维智联科技有限公司 | IEEE1588 protocol time calibration method based on linear regression algorithm |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113014344B (en) * | 2019-12-19 | 2022-04-15 | 中国科学院沈阳自动化研究所 | Time synchronization method of redundant network |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1191704A2 (en) * | 2000-09-20 | 2002-03-27 | Nec Corporation | Method and apparatus for acquiring slot timing and frequency offset correction |
| CN101277290A (en) * | 2007-03-26 | 2008-10-01 | 富士通株式会社 | Method and apparatus for synchronization of orthogonal frequency division multiplexing system frequency |
| EP2378822A2 (en) * | 2010-04-16 | 2011-10-19 | Simmonds Precision Products, Inc. | Synchronizing wireless devices |
| US20150139219A1 (en) * | 2012-05-11 | 2015-05-21 | Neul Ltd. | Method and apparatus for estimating frequency errors |
| US20150156009A1 (en) * | 2005-06-27 | 2015-06-04 | John W. Bogdan | Direct Synchronization of Synthesized Clock |
| CN105577351A (en) * | 2015-12-18 | 2016-05-11 | 北京无线电计量测试研究所 | Time and distance measuring method and device |
| CN105959003A (en) * | 2016-04-25 | 2016-09-21 | 华为技术有限公司 | Digital frequency division phase-locked loop |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110026479A (en) * | 2008-07-07 | 2011-03-15 | 스미토모덴키고교가부시키가이샤 | Base station device |
| EP2429105B1 (en) * | 2010-09-13 | 2013-03-27 | Ntt Docomo, Inc. | Node in a wireless system with time and clock frequency synchronizing and corresponding method |
| US8948157B2 (en) * | 2011-09-07 | 2015-02-03 | Qualcomm Incorporated | Method and apparatus for distributed synchronization in femtocell networks |
-
2016
- 2016-10-06 CN CN201680089895.8A patent/CN109792698B/en active Active
- 2016-10-06 WO PCT/EP2016/073934 patent/WO2018065061A1/en active Application Filing
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1191704A2 (en) * | 2000-09-20 | 2002-03-27 | Nec Corporation | Method and apparatus for acquiring slot timing and frequency offset correction |
| US20150156009A1 (en) * | 2005-06-27 | 2015-06-04 | John W. Bogdan | Direct Synchronization of Synthesized Clock |
| CN101277290A (en) * | 2007-03-26 | 2008-10-01 | 富士通株式会社 | Method and apparatus for synchronization of orthogonal frequency division multiplexing system frequency |
| EP2378822A2 (en) * | 2010-04-16 | 2011-10-19 | Simmonds Precision Products, Inc. | Synchronizing wireless devices |
| US20150139219A1 (en) * | 2012-05-11 | 2015-05-21 | Neul Ltd. | Method and apparatus for estimating frequency errors |
| CN105577351A (en) * | 2015-12-18 | 2016-05-11 | 北京无线电计量测试研究所 | Time and distance measuring method and device |
| CN105959003A (en) * | 2016-04-25 | 2016-09-21 | 华为技术有限公司 | Digital frequency division phase-locked loop |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112217601A (en) * | 2019-07-12 | 2021-01-12 | 天地融科技股份有限公司 | Data frame communication method and system |
| CN112217600A (en) * | 2019-07-12 | 2021-01-12 | 天地融科技股份有限公司 | Data frame receiving method and device, and communication method and system |
| CN112383498A (en) * | 2020-11-17 | 2021-02-19 | 紫光展锐(重庆)科技有限公司 | Low-frequency clock compensation method and device, storage medium and terminal |
| CN112636859A (en) * | 2020-12-23 | 2021-04-09 | 西安云维智联科技有限公司 | IEEE1588 protocol time calibration method based on linear regression algorithm |
| CN112636859B (en) * | 2020-12-23 | 2022-11-18 | 西安云维智联科技有限公司 | IEEE1588 protocol time calibration method based on linear regression algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109792698B (en) | 2021-02-12 |
| WO2018065061A1 (en) | 2018-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109792698A (en) | Compensate the receiver and method of the frequency error of reference clock | |
| Rebeiz et al. | Optimizing wideband cyclostationary spectrum sensing under receiver impairments | |
| US7027429B2 (en) | Method and apparatus for time and frequency synchronization of OFDM communication systems | |
| EP1507378A1 (en) | Frame and frequency synchronization for OFDM | |
| TWI454072B (en) | Frequency estimation | |
| CN102025671B (en) | Time domain joint estimation method for time coarse synchronization and frequency fine synchronization | |
| CN109791208A (en) | Received method and apparatus for DS/FH signal | |
| US10855494B2 (en) | Transmitter and receiver and corresponding methods | |
| KR100809020B1 (en) | Apparatus and method for acquiring initial synchronization of mobile in communication system | |
| TW201826855A (en) | System and method for blind detection of numerology, manufacturing method and testing method | |
| CN104093168A (en) | LTE (Long Term Evolution) common-frequency adjacent region detection method and device | |
| TW200541354A (en) | Frequency and timing error estimation and corresponding channel characterization in a communication system | |
| WO2022093858A1 (en) | A hybrid method for time-of-arrival-based ranging | |
| US10652068B2 (en) | Synchronization signal detection | |
| CN107070833A (en) | Towards the detection method of the synchronizing signal with symmetry, synchronous method and terminal | |
| US20070053459A1 (en) | System and method for obtaining accurate symbol rate and carrier phase, frequency, and timing acquisition for minimum shift keyed waveform | |
| CN104734840A (en) | Signal Timing | |
| KR20140050263A (en) | Apparatus and method for acquiring gps signal using multi-stage partial cross-correlator | |
| EP3972212A1 (en) | Adaptive selection of signal-detection mode | |
| CN104980184B (en) | Small region search method and equipment for TD-SCDMA system | |
| RU2425394C2 (en) | Method of detecting distorted pulsed signals | |
| EP1953978B1 (en) | Decoding method and apparatus in communication devices | |
| EP2463685A1 (en) | Method of synchronizing a receiver with a received ambiguous signal having a known number of at least two peaks | |
| KR20180061270A (en) | Enhanced Signal Detection in Dynamic Channels with High Carrier Frequency Offset | |
| CN105188129B (en) | Synchronization acquistion and tracking based on the collaboration of multi-angle fractional order correlation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |