KR20090012299A - Channel profiler and method of profiling an incoming signal - Google Patents

Abstract

A method for profiling an incoming signal and a channel profiler are provided to reduce latency and power consumption by cross-correlating the incoming signal with the extracted reference waveform. A channel profiler(200) includes an autocorrelator(210), an inverse discrete Fourier transform module(230) and a cross correlator(220). The autocorrelator distinguishes the preamble of the incoming signal. The inverse discrete Fourier transform module provides the reference waveforms. The cross correlator cross correlates a reference waveform corresponding to the preamble. The inverse discrete Fourier transform module stores the time domain reference wavelengths and converts the reference waveform to a time domain. The cross correlator calculates the threshold value and has processing functions for distinguishing all peaks of the added results with the amplitude exceeding the threshold value.

Description

CHANNEL PROFILER AND METHOD OF PROFILING AN INCOMING SIGNAL}

The present invention relates to the profiling of incoming signals, and more particularly, to a method and a channel profiler for profiling incoming signals.

In the last decades, interest in wireless communications has increased greatly. This interest has led to the development and refinement of wireless protocols and technologies. All types of wireless communications have one thing in common: they allow data transmission over the air. However, data transmission over the air causes problems such as interference, distortion and multipath. To solve these problems, a number of techniques for processing received data signals have been developed in combination with more robust modulation techniques. Some of the most used modulation techniques include orthogonal frequency division multiplexing (OFDM) and orthogonal frequency division multiple access (OFDMA).

More specifically, OFMA and OFDMA modulation techniques include built-in mechanisms for reducing multipath effects. In these modulation schemes, the concept of guard interval has been introduced. The guard interval is located at the beginning of the OFDM symbol, and the length of the guard interval can be adjusted to overcome the distance effects between the two transmitters. The guard interval includes a cyclic prefix, and the cyclic prefix corresponds to the repetition of the end of the corresponding OFDM symbol copied in the guard interval. Although the length of the guard interval can be adjusted to optimize its effect on multipath effects, most standards set the length of the guard interval to increase interoperability. Thus, multipath and intersymbol interference remain important issues in standards that rely on OFDM and OFDMA modulation techniques.

Many prior art documents suggest methods and apparatus for improving the quality of data received over the air using modulation techniques such as OFDM and OFDMA. These documents generally propose techniques for profiling incoming signals to perform synchronization, adjustment and correction. Typically, this profiling is performed while the incoming signal is decoded.

US published patent 2004/0105512, published on June 3, 2004 from Nokia, describes a two-step procedure for OFDM receivers. The first step of the procedure relies on autocorrelation of the received data signal to locate the cyclic prefixes. The second step uses a process of averaging correlation results for a certain number of symbols to improve the accuracy of the synchronization. Although the invention describes some improvements, it is limited to average results, which in some cases are not good enough.

Another document of interest to this topic is entitled "Timing Recovery for OFDM Transmission," a document published by Baoguo Yang et al. In November 2000 as the IEEE Journal of Selected Areas of Communications, Volume 18, No. 11. The document describes a two-step method. The first step of this method relies on auto-correlate of the received data, more specifically cyclic prefixes. Once the cyclic prefixes are located, the second step is performed in the frequency domain. Performing profiling in the frequency domain is more complex and adds latency to the decoding processor, and typically requires more power, which is of no interest in wireless applications such as wireless broadband (WiBro). You will be asked.

Motorola U.S. Patent No. 6,959,050, filed October 25, 2005, describes a method and apparatus for profiling while synchronizing OFDM signals at time, frequency, and per-subcarrier rotation. have. More specifically, the patent also describes a two-step process, where the first step includes performing symbol timing synchronization and partial frequency synchronization in the time domain. Thereafter, a second step of the process proceeds to perform per carrier-per-rotation synchronization in the frequency domain. The patent is an interesting solution, but the proposed solution still relies on the frequency domain to perform per-carrier-rotation synchronization to reduce multipath and intersymbol interference. The frequency domain is more complex than the time domain and also consumes more power from a hardware point of view. Thus, the solution is not an acceptable solution for wireless applications such as WiBro or IEEE Standard 802.11.

Therefore, there is a need for a method and apparatus for efficiently performing channel profiling without adding any latency or requiring more power.

The present invention provides a method and channel profiler for profiling an incoming signal. The method and profiler of the present invention rely on a simpler method than conventional solutions, while reducing latency and power consumption problems.

To do this, the method of the present invention includes identifying a preamble of the incoming signal. The method performs a process of extracting a reference waveform corresponding to the identified preamble and correlating the extracted waveform with an incoming signal.

In another aspect, the present invention provides a channel profiler. The channel profiler includes an autocorrelator for identifying the preamble of the incoming signal. The channel profiler also includes a discrete Fourier transform module for providing reference waveforms and a cross-correlator for correlating the preamble with a corresponding reference waveform.

In the following description, the following figures are used to explain and illustrate the invention.

1A and 1B are a flowchart of a method for profiling an incoming signal in accordance with an aspect of the present invention.

2 is a block diagram of a channel profiler in accordance with an aspect of the present invention.

3 is a graph showing exemplary autocorrelation results of the method and channel profiler of the present invention.

4 is a graph showing exemplary cross-correlation results of the method and channel profiler of the present invention.

5A, 5B and 5C are graphs showing summation results of exemplary cross-correlation according to the method and channel profiler of the present invention.

6 shows a signal and echoes of the signal in the time domain.

The present invention relates to a method and channel profiler for profiling an incoming signal. The present invention is characterized by many advantages over prior art solutions. First, the present invention typically provides a solution to the prior arts that cause additional latency while profiling the incoming signal. To do this, the present invention provides a method for performing channel profiling while an incoming signal is processed by a transceiver in the time domain.

Since wireless transmissions cause some degradation of the signal, it is important to profile the incoming signal to apply the necessary corrections to compensate for the resulting degradation. Typically, in conventional solutions, profiling is performed in the frequency domain. If the incoming signal is transformed into the frequency domain using the Discrete Fourier Transform module, the specificities of the incoming signal are visually recognizable. However, the time required to change the incoming signal in the frequency domain causes some significant latency and requires increased power consumption. Even trade-offs where latency and increased power consumption are acceptable, in some applications such as wireless broadband (WiBro) systems, such trade-offs are not possible.

Reference is now made to FIGS. 1A and 1B which illustrate a method for profiling an incoming signal in the time domain according to the present invention. The method 100 comprises six main sections: a section 105 for identifying a preamble, a section 110 for extracting a reference waveform, a section 115 for correlating an incoming signal with the extracted reference waveform, It is divided into a section 120 for summing the results to reduce multipath effects, a section 125 for identifying the maximum peak, and a section 130 for calculating the back-off.

The first section of the method relates to identifying a preamble. To do this, any known method can be used, such as autocorrelating an incoming signal. Alternatively, in order to obtain more detailed results, identifying the preamble comprises four sub-steps: autocorrelating the incoming signal 135, determining the location of the plateau 140. , Autocorrelating a portion of the plateau 145 and determining the location of the peak in the obtained curve 150.

For example, in an OFDMA system such as WiBro, there is (N / 3) -pattern repetition in the time domain because the frequency pattern of the preamble has only one third of the sub-carriers. This (N / 3) -pattern repetition can be identified by autocorrelation (Equations 1 and 2).

This autocorrelation 135 outputs a plateau of 682 samples within the preamble region as shown in FIG. 3. This window includes a synchronization point, which can be used once identified to train the Discrete Fourier Transform module used to decode the incoming signal in an optimized manner. Without prior knowledge of synchronization points, the Discrete Fourier Transform module is trained blindly, i.e. sub-optimally. Therefore, it is very important to correctly identify the synchronization point. To do this, the method performs step 140 of determining the position of the plateau in the autocorrelation results of the incoming signal. More specifically, in modulation techniques such as OFDM and OFDMA, symbols include a cyclic prefix (CP), where the cyclic prefix is a copy of the last M samples of the symbol in front of the symbol. Thus, a good estimate of the synchronization point can be given by the second autocorrelation for the cyclic prefix (Equations 3 and 4).

As an example, consider a WIBRO application. In this technique, N = 1024, N / 3 = 341 and M = 128. In an ideal channel, the two autocorrelation outputs of the incoming signal are shown in FIG. 3, where the curve with the plateau represents the results of the first autocorrelation 135 and the other curve is the result of the second autocorrelation 145. Indicates. Thus, the precise synchronization point given by the peak index of the second autocorrelation corresponding to the beginning of the preamble, as well as the window given by the first autocorrelation (here 682 2N / 3 samples), is clearly known. Can be. The first section of this profiling method is particularly fairly accurate for critical frequency offsets since only autocorrelation is used.

The method performs a second section 110 to extract the reference waveform. Extraction 110 of the reference waveform may be performed in alternative ways. For example, reference waveforms representing all possible preambles in the frequency domain and / or time domain may be stored in memory. Once the preamble is identified in section 105, the corresponding reference waveform can then be identified. Thus extracting section 110 accesses the waveforms stored in step 155, and the corresponding reference waveform is identified in step 160. If the stored waveforms are in the frequency domain, since the incoming signal is still in the time domain, the identified reference waveform is transformed into the time domain at step 165, for example using an inverse discrete Fourier transform module. Since an Inverse Discrete Fourier Transform module is already present in OFDM systems, and the conversion of the reference waveform from frequency domain to time domain is performed only once, this aspect of the invention does not add latency to the normal operations of OFDM systems. There is a significant advantage in this regard.

The method proceeds to section 115 using the identified preambles to correlate the identified reference waveforms in the time domain (Equations 5 and 6).

For example, with OFDMA modulation, since there are 3x-repeats of the same pattern in the preamble, the result for cross-correlation in step 115 will correspond to the graph shown in FIG.

The method then proceeds to step 120, where the results of cross-correlation step 115 are summed, the results of which are shown in FIGS. 5A, 5B and 5C. The three peaks of FIGS. 5A, 5B and 5C correspond to three different paths with amplitudes proportional to the power of the path. For example, the largest amplitude in FIG. 5A is the second peak, the largest amplitude in FIG. 5B is the third peak, and the largest peak in FIG. 5C is the first peak. In general, the path with the largest amplitude is selected as the best synchronization point, but there are other factors that can affect selecting the best synchronization point. Therefore, the method proceeds to section 125 for identifying the optimal peak.

For example, in multiple environments, the choice of synchronization point is important to minimize intersymbol and intercarrier interferences. This also affects interpolation capabilities for the calculation of the correction vector. Reference is now made to FIG. 6, which illustrates an incoming signal in a multipath environment in the time domain. In FIG. 6, each of the three signals corresponds to one of the peaks of the cross-correlation results. By selecting the second cross-correlation peak, the obtained synchronization point is no longer within the cyclic prefix of the first signal causing intersymbol interferences. This analysis can be generalized to other echoes. However, by selecting the first signal (ie, the first peak), the synchronization point is placed within the cyclic prefix for the three replicas of the signal. Therefore, to reduce intersymbol and intercarrier interferences, the method proceeds to calculating a threshold. The threshold may preferably correspond to the amplitude of the maximum peak of the cross-correlation divided by a constant, such as k. The constant k may be set to different values to correspond to various environments. For example, in the case of urban and rural environments, the k parameter can vary greatly.

The method then performs step 175 of identifying peaks above the threshold. Typically, identification of peaks will be performed in the set window. The method then proceeds to step 180 of selecting one of the peaks above the threshold. In a preferred manner, the first peak above the threshold is selected as the synchronization point. In the particular case where the delay of the main signal is greater than the cyclic prefix, the main peak is selected as the synchronization point. Also in this case, the selection of the first peak will add intersymbol interference and intercarrier interference. By identifying the synchronization point in the proposed manner, more weight is given to possible synchronization points that keep all signal copies in the cyclic prefix. In addition, since the intersymbol interference and intercarrier interference added by other signal copies will be limited, additional flexibility is provided by the method when the maximum peak is much stronger than the other peaks.

Finally, the method ends at 130 calculating the back-off. The back-off corresponds to the difference between the selected synchronization point and the index of the peak with the largest amplitude. For example, if the selected synchronization point corresponds to the first peak, which exceeds the threshold, but the first peak is not the peak with the largest amplitude, the back-off will correspond to 16 samples. Since the synchronization point may not correspond to the first peak, the selection of the synchronization point can cause significant phase rotation. Back-off may be used to improve the performance of different blocks of the receiver for channel interpolation, decoding, and the like. More specifically, for channel interpolation, back-off can be used to calculate phase rotation pre-correction values and phase rotation post-correction values. The phase rotation pre-correction value can be calculated using the following formula:

C _pc = C _p × e ^{j θp}

θ = d × 2π / N

Where C _pc is the phase rotation pre-correction value for pilot p; C _p is the original correction value for pilot p; d is back-off; N is the DFT size; p is the index of the pilot associated with the direct current.

After channel interpolation is performed, the correction vector is rotated as follows:

C _m = C _rm × e ^-jθm

Where C _rm is the phase rotation post-correction value for pilot p; C _m is the final correction vector for equalization; m is the subcarrier index associated with direct current.

Thus, the profiling method of the present invention provides a simple and efficient method of identifying synchronization points in view of multipath effects while reducing intersymbol and intercarrier interferences.

The invention also describes a channel profiler for profiling an incoming signal. The channel profiler of the present invention is described in FIG. In general, the channel profiler 200 includes an autocorrelator 210, a correlator 220, an inverse discrete Fourier transform module 230, a summation module 240, and an identification module 250.

The incoming signal 205 is received at the autocorrelator 210. The autocorrelator autocorrelates the incoming signal 205 with itself to identify the preamble of the incoming signal 205. Alternatively, the autocorrelator 210 additionally determines the location of the plateau of the autocorrelated incoming signal, autocorrelates a portion of the plateau, and determines the peak position of the autocorrelated portion of the plateau. The positioned peak corresponds to the beginning of the preamble. The autocorrelator 210 provides the information of the preamble to the crosscorrelator 220.

Cross-correlator 220 performs cross-correlation with the identified preamble and the corresponding reference waveform. The crosscorrelator 220 may extract a reference waveform from the inverse discrete Fourier transform module 230 or extract a reference waveform from a memory (not shown). The stored reference waveforms may be in time domain format or in frequency domain format. If the stored reference waveform is in the frequency domain, inverse discrete Fourier transform module 230 may additionally be used to convert the reference waveform to time domain.

The cross-correlated signal is then provided to a summing module 240 for summation. Summing module 240 may also be embedded in cross-correlator 220. The aggregated results are provided to the identification module 250 via the cross-correlator 220 or directly to the identification module 250 (not shown). Alternatively, identification module 250 may also be embedded within cross-correlator 220.

Crosscorrelator 220 may additionally include processing functions for calculating a threshold and identifying all peaks of the summed results having an amplitude above the threshold. Also, crosscorrelator 220 may select a first peak that exceeds a threshold as a synchronization point. Crosscorrelator 220 may also calculate the back-off of the incoming signal, as previously described.

The invention has been illustrated by the preferred embodiments. It will be apparent to those skilled in the art that the preferred embodiments described are for illustrative purposes only and should not be construed to limit the scope of the invention. The method and apparatus described in the preferred embodiments can be modified without departing from the scope of the present invention. The scope of the invention should be defined with reference to the appended claims, which clearly describe the scope to be protected.

Claims (19)

A method for profiling an incoming signal, Identifying a preamble of the incoming signal; Extracting a reference waveform corresponding to the identified preamble; And Correlating the extracted waveform with the incoming signal. The method of claim 1, Summing up the results of the cross-correlation; And Identifying a peak of the summed results as a synchronization point. The method of claim 1, Wherein said extracting includes accessing stored time domain reference waveforms to identify said reference waveform corresponding to said identified preamble. The method of claim 1, The extracting step, Accessing stored frequency domain waveforms; Identifying a reference waveform corresponding to the identified preamble; And Converting the identified reference waveform to a time domain reference waveform. The method of claim 2, Identifying the synchronization point, Calculating a threshold value; Identifying all peaks having an amplitude above the threshold; And Selecting a first peak above the threshold as a synchronization point. The method of claim 5, wherein The threshold is equal to (amplitude of maximum peak) / k, wherein k is a predefined constant. The method of claim 1, Identifying the preamble, Autocorrelating the incoming signal; Determining a position of a plateau in the autocorrelated incoming signal; Autocorrelating a portion of the plateau; And Determining a location of said autocorrelated portion of said plateau, said peak corresponding to the beginning of said preamble. The method of claim 5, wherein Calculating a back-off of the incoming signal equal to (synchronization point-index of main peak). The method of claim 8, The back-off is used to calculate a phase correction value. An autocorrelator for identifying a preamble of the incoming signal; An inverse discrete Fourier transform module for providing reference waveforms; And And a cross-correlator for correlating the reference waveform corresponding to the preamble. The method of claim 10, The cross-correlator, A summing module for summing results obtained by the cross-correlation; And And an identification module for identifying a peak of the summed results as a synchronization point. The method of claim 10, And said inverse discrete Fourier transform module additionally stores time domain reference waveforms. The method of claim 10, And said inverse discrete Fourier transform module further converts said reference waveform into a time domain. The method of claim 11, And the cross-correlator further has processing functions for calculating a threshold and identifying all peaks of the summed results having an amplitude above the threshold. The method of claim 14, And the cross-correlator further selects a first peak above the threshold as a synchronization point. The method of claim 14, Wherein the threshold is equal to (amplitude of maximum peak) / k, where k is a predefined constant. The method of claim 10, The autocorrelator, Determine the location of the plateau of the autocorrelated incoming signal; Autocorrelate a portion of the plateau; And Determine a position of said autocorrelated portion of said plateau, said peak corresponding to the beginning of said preamble. The method of claim 10, And the cross-correlator further calculates a back-off of the incoming signal equal to (synchronization point-index of main peak). The method of claim 18, The back-off is used to calculate a phase rotation correction value.

Images