KR20030055643A - Apparatus for detecting burst signal and method thereof - Google Patents

Abstract

PURPOSE: A device for detecting a burst signal and a method therefor are provided to calculate a high-speed input signal in real time to detect a burst signal, and to decide whether data frames exist as increasing accuracy of deciding a start point, thereby minimizing memory capacity. CONSTITUTION: An oscillator(105) samples input signals at fixed double sampling intervals. The sampled signals are matched by a matching filter(110). A queue(180) stores the matched signals in real time. A switch(170) controls transmission of the matched signals to the queue(180). A correlator(120) compares the matched signals with a preset preamble, and calculates correlation values to output the values. The first adder(123) adds an output of a Q-phase correlator(1212) to an output of an I-phase correlator(1220), and outputs an added value. A subtracter(124) subtracts an output of an I-phase correlator(1210) from an output of a Q-phase correlator(1222), and outputs a subtracted value. The first squarer(125) squares the output of the first adder(123). The second squarer(126) squares the output of the subtracter(124). The second adder(127) adds the output of the first squarer(125) to the output of the second squarer(126), and outputs the added value to a comparator(130). The comparator(130) compares the correlation values of the correlator(120) with a preset threshold value.

Description

Burst signal detection device and method therefor {APPARATUS FOR DETECTING BURST SIGNAL AND METHOD THEREOF}

The present invention relates to digital communications, and more particularly, to an apparatus and a method for detecting a burst signal received at a receiver for digital communications.

In general, in order to improve the performance of a time division duplex (TDD) communication system, it is necessary to quickly and accurately determine the existence of a frame and the starting point of the frame.

In order to minimize the high-speed data transmission and the allocated frequency band in transmitting and receiving digital signals, transmitting frames in a time division multiplexing scheme or transmitting data in a time division multiple access (TDMA) scheme to a plurality of subscribers. For example:

In the two communication transmission schemes, the data frames received by each subscriber station and the base station are not continuous. Therefore, each subscriber station and base station need to find the existence of the frame and the starting point of the frame when receiving the discontinuous frame.

If it does not recognize the existence of a frame, it is inconvenient to wait for data transmission until the next frame transmission. If the starting point of the frame is incorrectly recognized, a malfunction of the terminal / base station system may be caused.

Therefore, a method and apparatus for improving the accuracy of determining whether a data frame exists and determining the starting point of the frame more accurately have always been a research subject.

In general, a method for improving the accuracy of burst signal transmission at a high speed is divided into (1) a method using a large memory and (2) a method of detecting a burst signal by real-time calculation.

In the method using a large memory, a burst signal containing noise is continuously stored in the memory, and then a burst signal is detected by detecting a change in energy value. It is necessary.

A method of detecting a burst signal by real-time calculation is a method of obtaining a correlation value between an input signal and a preamble, comparing the correlation value with a specific threshold value, and detecting the burst signal based on the correlation value.

However, in the burst signal detection method using such a real-time calculation, even when the burst signal is input, the correlation value may be calculated to be smaller than the threshold value or may be calculated as exceeding the threshold value in the data frame section. The development of methods and devices that can be done will be a research subject.

Accordingly, an object of the present invention is to detect a burst signal by calculating a high-speed input signal in real time in order to solve the above problems, and accordingly an apparatus capable of increasing the accuracy of the existence of the data frame and the determination of its starting point and its To provide a way.

1 is a configuration diagram of a burst signal detection apparatus according to an embodiment of the present invention.

2 is a flowchart of a burst signal detection method according to an embodiment of the present invention.

3 is a graph of a correlation coefficient of an input signal extracted by a burst signal detection method according to an embodiment of the present invention over time. For example, a 32 symbol CAZAC (Constant Amplitude and Zero-Auto-Correlation) preamble is input. It is a graph showing the correlation coefficient with the burst signal.

4 is a graph showing the results of testing the performance of the burst signal detection apparatus according to an embodiment of the present invention, the horizontal axis represents the Eb / N value, the vertical axis represents the detection probability (Detection Probability).

Burst signal detection apparatus according to a feature of the present invention for achieving the above object,

A signal receiver for receiving a digital transmission signal; A correlator configured to calculate a correlation value between the transmission signal received by the signal receiver and a preset preamble; A comparison unit comparing the calculated correlation value with a preset threshold value; And extracting a first peak value that is the maximum value among the correlation values continuously calculated by the correlation unit for a first predetermined period when the calculated correlation value exceeds the threshold value, and wherein the first period has elapsed. Extracting a second peak value which is the maximum value among the correlation values continuously calculated by the correlation unit for a preset second period from time; and based on the first peak value and the second peak value, the received transmission signal And a start point calculator for calculating a start point of the preamble.

The starting point calculation unit may include: a first window storing a correlation value continuously calculated by the correlation value during the first period when the calculated correlation value exceeds the threshold value; And a second window for storing correlation values continuously calculated by the correlator for the second period from when the first period elapses, wherein the first peak value and the second peak value are the first peak value. The correlation values stored in the window and the correlation values stored in the second window may be extracted.

The start point calculator comprises a memory for storing the received transmission signal in a FIFO method; And a switching controller for terminating the storage of the transmission signal to the memory at a specific set time point, wherein calculation of the preamble starting point of the transmission signal based on the first and second peak values is performed by the switching controller to the memory. After the storage of the transmission signal is ended, it is calculated based on the LSB area of the memory, the length of the preamble pattern, and the generation position of the first and second peak values.

Burst signal detection method according to another aspect of the present invention,

a) receiving a digital transmission signal; b) calculating a correlation value between the received transmission signal and a preset preamble; c) comparing the calculated correlation value with a predetermined threshold value; d) if the calculated correlation value exceeds the threshold, extracting a first peak value that is the maximum of the correlation values continuously calculated in step b) for a first predetermined period; e) extracting a second peak value which is the maximum value of the correlation values continuously calculated in step b) for a preset second period from when the first period elapses; And f) calculating a starting point of the preamble of the received transmission signal based on the extracted first and second peak values.

The preamble pattern is added to the digital transmission signal at the beginning of the frame twice in succession.

The preamble pattern is a constant amplitude and zero-auto-correlation (CAZAC) preamble pattern.

The method comprises the steps of: i) storing the correlation value continuously calculated in step b) during the first period if the calculated correlation value exceeds the threshold; And ii) storing the correlation values continuously calculated in step b) for the second period from when step i) ends, wherein the first peak value and the second peak value are stored in the step. The correlation values stored in i) and the correlation values stored in step ii) may be extracted.

The first period may be set to a period corresponding to the preamble pattern length.

The method comprises the steps of: iii) storing the input transmission signal in a memory in a first in first outpur (FIFO) manner; And iv) terminating the storage of the transmission signal to the memory at the set time point, wherein after step f), after the step ii), the LSB area of the memory, the length of the preamble pattern, and the first and the first The start position of the preamble is calculated based on the occurrence position of the 2 peak value.

The set time point is set when the step i) is terminated when the first peak value is greater than the second peak value, and the step ii) is performed when the second peak value is larger than the first peak value. It is characterized in that it is set from when it ends to when the set period has elapsed.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a configuration diagram of a burst signal detection apparatus according to an embodiment of the present invention.

As shown in FIG. 1, an apparatus for detecting a burst signal according to an exemplary embodiment of the present invention includes an oscillator 105, a matched filter 110, a correlator 120, a comparator 130, and a first window. (135), the second window 145, the first counter 140, the switching controller 150, the second counter 155, the OR gate 160, the switch 170 and the queue 180. Include.

The oscillator 105 samples the input signal at twice the fixed sampling interval per symbol to speed up data processing.

The input signal input to the oscillator 105 may be a preamble pattern having good correlation characteristics two times in a row at the beginning of the frame. As the added preamble pattern, for example, 32 symbols of CAZAC (Constant Amplitude) and Zero-Auto-Correlation).

The signal sampled at the oscillator 105 is matched by the matched filter 110 to increase its Signal to Noise Ratio (SNR).

The signal matched by the matched filter 110 is stored in the queue 180 in real time, and the switch 170 controls the transmission of the matched signal to the queue 180.

Since the matched input signal is stored in the queue 180 in real time, the size of memory required to store the input signal until the burst signal is detected can be reduced, and the queue 180 is input from the matching filter 110. The signal is processed in a first-in first-out (FIFO) manner until the switch 170 is turned off.

The length of the queue 180 stores all of the preamble symbols, and is obvious to those skilled in the art in view of the symbol length considering the total length of the correlation result, the set length of the preamble pattern, and the size of the first and second windows 135 and 145. Can be set.

The correlator 120 calculates and outputs a correlation value by comparing the signal matched by the matched filter 110 with a preset preamble.

In more detail, the correlator 120 correlates an input signal with a conjugate value of a predetermined preamble. For this purpose, the first correlator 121 and Q phase for performing a correlation operation on the I phase are performed. And a second correlator 122 that performs a correlation operation on.

The first correlator 121 and the second correlator 122 include I phase correlators 1210 and 1220 and Q phase correlators 1212 and 1222, respectively.

The phase I correlator 1210 of the first correlator 121 and the Q phase correlator 1222 of the second correlator 122 output a real number, and the Q phase correlator 1212 and the second correlator of the first correlator 121. I phase correlator 1220 at 122 outputs an imaginary number.

The first adder 123 adds and outputs the output of the Q phase correlator 1212 of the first correlator 121 and the output of the I phase correlator 1220 of the second correlator 122.

The subtractor 124 subtracts and outputs the output of the I phase correlator 1210 of the first correlator 121 from the output of the Q phase correlator 1222 of the second correlator 122.

The first squarer 125 squares the output of the first adder 123, and the second squarer 126 squares the output of the subtractor 124.

The second adder 127 adds the output of the first squarer 125 and the output of the second squarer 126 and outputs the output to the comparator 130.

The comparator 130 compares the correlation value output from the correlator 120 with a predetermined threshold value. As a result of the comparison, when the correlation value is larger than the threshold value, the first window 135 is driven.

The threshold value is set in consideration of the length of the preamble correlated with the signal-to-noise ratio (Eb / N) as a reference in the system.

The first window 135 is driven according to the result signal of the comparator 130 and stores the correlation value continuously output from the comparator 130 in a specific register. As described above, while the first window 135 operates, the queue 180 continuously processes data in a FIFO manner.

When the operation period of the first window 135 elapses, the operation is stopped by outputting an off signal to prevent detection of repeated burst signals in the same burst.

The first counter 140 counts the sampling clock by the length of the preamble pattern so as to detect the next expected peak point of the correlation value of the preamble at the same time as the operation of the first window 135 is stopped.

Like the operation of the first window 135, the first counter 140 also generates an off signal to stop detection of repeated burst signals and stops the operation.

As soon as the counting of the first counter 140 ends, a driving signal for the second window 145 is transmitted to the second window 145, whereby the second window 145 has a threshold value in the comparator 130. The correlation value is stored regardless of the comparison result with.

As in the first window 135, the operation of the second window 145 also generates an off signal after a predetermined time, and the operation ends.

The switching controller 150 may include a peak value among the correlation values stored in the first window 135 (hereinafter referred to as "first peak value") and a peak value among the correlation values stored in the second window 145 (hereinafter referred to as "second peak"). Value ", and the switch 170 is controlled according to the comparison result.

In more detail, when the first peak value is larger than the second peak value, as shown in the accompanying FIG. 3, the first peak point is missed at three peak points, and the capture of the burst signal at the second peak point is lost. In this case, it is assumed that the correlation between the preamble in the burst signal to which the current correlation value is input and the conjugate value inside the correlator 120 is completed, and the data is input to the queue 180 to prevent data from being stored. Outputs a signal 151 for controlling the switch 170.

On the contrary, when the second peak value is larger than the first peak value, as shown in FIG. 3, it is determined that both the first and second peak values are stored in the first and second windows 135 and 145. Assuming that the current time point is before the correlation between the preamble and the preamble in the burst signal is finished, a signal that starts a sampling time count equal to the length of the preamble pattern is output to the second counter 155.

Therefore, the second counter 155 counts the sampling time as long as the length of the preamble pattern and then outputs a signal 153 for controlling the switch 170 to prevent the data from being stored by being input to the rule 180. do.

The OR gate 160 receives the signals 151 and 153 output from the switching control unit 150 and outputs a signal for turning off the switch 170 by one of the two signals. When the first peak value becomes larger than the second peak value, or when the counting period of the second counter 155 elapses from the time when the second peak value is larger than the first peak value, it is turned off. 180 ends the FIFO operation.

The starting point of the actual preamble in the queue 180 that has finished the FIFO operation can be determined by calculating the distance from the LSB (Least Significant Bit) of the queue by the length of the preamble pattern and the address value of the peak value in the window.

2 is a flowchart illustrating a burst signal detection method performed in the burst signal detection apparatus according to the embodiment of the present invention.

First, the queue 180 starts by turning on the input signal so that the input signal can be stored (S210).

Therefore, when an input signal is received by the oscillator 105 (S220), the input signal is continuously stored in the queue 180 in real time.

After the input signal is received (S220), the correlation between the input signal and the preset preamble is taken by the correlator 120 and the correlation value is calculated (S230).

The calculated correlation value is compared or greater than the threshold value in the comparator 130 (S240), and if it is not larger than the threshold value, the signal reception (S220) to the correlation value calculation step (S230) is repeated.

When the calculated correlation value is larger than the threshold value, the first window 135 is driven to store the correlation value continuously output for a predetermined first period (S250).

After the first period has elapsed, the second window 145 is driven to store the calculated correlation value for a second predetermined period (S260).

After the correlation values are stored in the first window 135 and the second window 145, the first peak value is selected from the correlation values in the first window 135, and the second peak value is selected from the correlation values in the second window 145. The peak value is extracted (S270).

The starting point of the preamble of the input signal is calculated based on the first peak value and the second peak value, which is specifically as follows.

That is, at the set time determined based on the first peak value and the second peak value, the real-time storage of the input signal to the queue 180 is terminated (S280), and input from the data array in the queue 180 determined accordingly. The starting point of the preamble of the signal is calculated (S290).

The setting time point is when the second storage step is terminated when the first peak value is larger than the second peak value. When the second peak value is larger than the first peak value, the set period elapses from the end of the second storage step. It is decided when it is done.

The preamble start point of the input signal may be calculated based on the LSB region of the memory, the length of the preamble pattern, and the generation position of the first and second peak values.

As mentioned above, although the preferred embodiment regarding the burst signal detection method and apparatus of this invention was described, this invention is not limited to the said embodiment, A person of ordinary skill in the art from the embodiment of this invention belongs. It includes all changes in the range which are easily changed by and deemed to be equal.

According to the present invention, not only the presence or absence of a frame is transmitted in real time, but also the start position of the frame can be accurately calculated, and the memory capacity used for this can be minimized.

In addition, the signal-to-noise ratio is improved and the functions for the carrier frequency, phase error, timing error, and multipath channel are enhanced.

In addition, the embodiment of the present invention can find the exact starting point of the preamble based on the output signal of the matched filter stored in the queue can be used in the DA (Digital Access) synchronization parameter extraction circuit.

In addition, acquiring the correct starting point of a preamble preceding a data frame in a frame is a major factor in determining the direction in which the synchronization algorithm of the data frame can be determined. Increase the usefulness of the algorithm.

Claims (10)

A signal receiver for receiving a digital transmission signal; A correlator configured to calculate a correlation value between the transmission signal received by the signal receiver and a preset preamble; A comparison unit comparing the calculated correlation value with a preset threshold value; And When the calculated correlation value exceeds the threshold value, the first peak value, which is the maximum value among the correlation values continuously calculated by the correlation unit for the first predetermined period, is extracted, and when the first period has elapsed. Extracts a second peak value which is the maximum value among the correlation values continuously calculated by the correlation unit for a preset second period, and based on the first peak value and the second peak value, Starting point calculation unit for calculating the starting point of the preamble Burst signal detection device comprising a. The method of claim 1, The starting point calculation unit, A first window for storing a correlation value continuously calculated by the correlation value during the first period when the calculated correlation value exceeds the threshold value; And A second window for storing correlation values continuously calculated by the correlation unit during the second period from when the first period elapses Including, The first peak value and the second peak value are respectively extracted from the correlation values stored in the first window and the correlation values stored in the second window. Burst signal detection device, characterized in that. The method of claim 1, The starting point calculation unit, A memory for storing the received transmission signal in a FIFO method; And Switching control unit for terminating the storage of the transmission signal to the memory at a particular set time More, The calculation of the preamble starting point of the transmission signal based on the first and second peak values After the storage of the transmission signal to the memory is terminated by the switching controller, it is calculated based on the LSB area of the memory, the length of the preamble pattern and the occurrence position of the first and second peak values. Burst signal detection device, characterized in that. The method of claim 1, And a matching filter for matching the transmission signal received by the signal receiving unit and providing the matching signal to the correlation unit. a) receiving a digital transmission signal; b) calculating a correlation value between the received transmission signal and a preset preamble; c) comparing the calculated correlation value with a predetermined threshold value; d) if the calculated correlation value exceeds the threshold, extracting a first peak value that is the maximum of the correlation values continuously calculated in step b) for a first predetermined period; e) extracting a second peak value which is the maximum value of the correlation values continuously calculated in step b) for a preset second period from when the first period elapses; And f) calculating a starting point of the preamble of the received transmission signal based on the extracted first and second peak values Burst signal detection method comprising a. The method of claim 5, And a preamble pattern having good correlation characteristics in the digital transmission signal is added twice at the beginning of the frame. The method of claim 5, i) if the calculated correlation value exceeds the threshold, storing the correlation value continuously calculated in step b) during the first period; And ii) storing the correlation values continuously calculated in step b) for the second period from when step i) ends; More, The first peak value and the second peak value are respectively extracted from the correlation values stored in step i) and the correlation values stored in step ii). A burst signal detection method, characterized in that. The method of claim 5, And the first period is set to a period corresponding to the length of the preamble pattern. The method of claim 7, wherein iii) storing the input transmission signal in a memory in a first in first outpur (FIFO) manner; And iv) terminating the storage of the transmission signal to the memory at the set time point More, After step f), after step ii), the start position of the preamble is calculated based on the LSB region of the memory, the length of the preamble pattern, and the occurrence positions of the first and second peak values. A burst signal detection method, characterized in that. The method of claim 9, The set time point is If the first peak value is larger than the second peak value, the step is set at the end of step i). If the second peak value is larger than the first peak value, the second peak value is set to the time elapsed from the end of the step ii). A burst signal detection method, characterized in that.