JP2004088276A - Apparatus and method for detecting signal - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an apparatus for detecting a signal for efficiently detecting a pulse position of a received pulse signal in a pulse modulation receiver for receiving a pulse-modulated signal. <P>SOLUTION: A pulse generator 410 of a searcher 207 superimposes a plurality of pulses on corresponding plurality of pulse positions and generates a reference pulse. The reference pulse is multiplied by the received pulse signal, and a detection signal representing a correlation between the reference pulse and the received pulse signal is output based on this multiplier output. In this case, a correlator 402 detects the correlation between the received pulse signal and the reference pulse preferably at a plurality of pulse position units. A code component is removed from the output of the correlator by squaring to obtain a correlation detected value. A window controller 412 shifts the position of the reference pulse. The pulse position having the largest value of the correlation detected values obtained for all the positions of the reference pulses is selected as the pulse position of the received signal. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a signal detection method and apparatus for detecting a pulse position of a received pulse signal in a pulse modulation receiving apparatus that receives a pulse-modulated signal.
[0002]
[Prior art]
Pulse modulation is a method of transmitting information by adding information to a pulse in a baseband, and has an advantage that communication can be performed with a simple circuit configuration because a carrier wave is not used. 2. Description of the Related Art In recent years, as a communication method for performing wireless transmission at a data rate of 100 Mbps or more, pulse modulation using an ultra wideband (Ultra Wideband) has attracted attention.
[0003]
Initial synchronization of a wideband signal receiver using pulse modulation requires detection of the pulse position of the received signal. This is done with a detector commonly called a searcher. The pulse position detection includes a search method of performing a sliding correlation with one correlator and a method of performing a sliding correlation with a plurality of correlators for speeding up.
[0004]
[Problems to be solved by the invention]
When performing wireless transmission at a data rate of 100 Mbps or more, a band of about several GHz is required, and the peak width of the pulse is narrowed to about 100 pico-sec. Therefore, a high-speed and high-efficiency signal detection method of a received pulse signal position is required. .
[0005]
There is a trade-off between the speedup and the circuit scale, and an initial synchronization method that can achieve a high speed on a small scale is desired.
[0006]
The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a signal detection device that efficiently detects a pulse position of a reception pulse signal in a pulse modulation reception device that receives a pulse-modulated signal. It is in.
[0007]
It is another object of the present invention to provide a signal detection device having a relatively small circuit configuration for detecting a pulse position of a received pulse signal in a pulse modulation reception device that receives a pulse-modulated signal.
[0008]
[Means for Solving the Problems]
A signal detection device according to the present invention is a signal detection device that detects a pulse position of a reception pulse signal in a pulse modulation reception device that receives a pulse-modulated signal, and in each symbol period, pulses in a symbol section do not overlap. A pulse generator that superimposes a plurality of pulses corresponding to a plurality of pulse positions to generate a reference pulse, a multiplier that multiplies the reference pulse by a received pulse signal, and the reference pulse based on an output of the multiplier. Correlator for outputting a detection signal representing the correlation between the signal and the received pulse signal, calculating means for squaring the output of the correlator, and generating a correlation detection value by adding the squared signal a predetermined number of times n Adding means for controlling the pulse generator so as to shift the position of the reference pulse; and Storage means for storing the correlation detection values obtained for all positions of the pulse, and means for comparing a plurality of correlation detection values stored in the storage means and selecting a pulse position having the largest value. It is characterized by the following.
[0009]
The pulse generator generates a reference pulse by superimposing a plurality of pulses corresponding to a plurality of pulse positions such that pulses in a symbol section do not overlap in each symbol period. The multiplier multiplies the reference pulse by the received pulse signal. The correlator outputs a detection signal indicating the correlation between the reference pulse and the received pulse signal based on the output of the multiplier. The calculating means squares the output of the correlator. The adding means for generating a correlation detection value generates a correlation detection value by adding the squared signal a predetermined number of times n. The control means controls the pulse generator to shift the position of the reference pulse. The storage unit stores the correlation detection values obtained for all positions of the reference pulse. The selecting means compares a plurality of correlation detection values stored in the storage means and selects a pulse position having the largest value.
[0010]
Preferably, the correlator has a plurality of detectors that output a detection signal indicating a correlation between a received pulse signal and the reference pulse, and each detector has the received pulse signal and the reference pulse at different pulse positions. And the calculating means and the adding means process a plurality of detection signals in parallel, and at least one detector detects a correlation between the reception pulse signal and the reference pulse in a plurality of pulse position units. Is detected. In this specification, "detecting a correlation in a plurality of pulse position units" means that a correlation detection value is obtained as a whole of a plurality of pulse positions, but it is not known mainly which pulse position contributes to the correlation detection value. Means that. This makes it possible to reduce the required number of shifts of the position of the reference pulse as compared with the case where a reference pulse corresponding to a single pulse position is generated.
[0011]
If the correlation detection value of the plurality of pulse positions is larger than the other correlation detection values, the plurality of pulse positions are divided and assigned to the plurality of detectors, and the correlation detection value is obtained again.
[0012]
When the second correlation detection value is obtained, the predetermined number n may be reduced to a smaller number n ′, thereby enabling a higher-speed processing.
[0013]
The correlator may include a single detector that outputs a detection signal representing a correlation between a received pulse signal and the reference pulse, wherein the detector preferably includes a plurality of pulse positions. The correlation between the received pulse signal and the reference pulse is detected in units. After detecting the plurality of pulse position units having the largest correlation detection value, the single detector sequentially calculates correlation detection values for the plurality of pulse positions, and determines the pulse position having the largest correlation detection value as the true pulse position. And
[0014]
Preferably, the pulse generator generates the reference pulse based on a gate signal corresponding to one or a plurality of pulse positions in a symbol section, and the detector is designated by the gate signal according to the gate signal. And an integrator for integrating the output of the multiplier within the set period. Thus, by using the gate signal, generation of the reference pulse and designation of the integration period can be easily realized.
[0015]
A signal detection method according to the present invention is a signal detection method for detecting a pulse position of a received pulse signal in a pulse modulation receiving apparatus that receives a pulse-modulated signal, and in each symbol period, pulses in a symbol section do not overlap. Generating a reference pulse by superimposing a plurality of pulses corresponding to a plurality of pulse positions, multiplying the reference pulse by a received pulse signal, and at least one of all pulse positions based on an output of the multiplier. Outputting a detection signal indicating the correlation between the reference pulse and the received pulse signal in units of a plurality of pulse positions, squaring the output of the correlator, and converting the squared signal to a predetermined value. A correlation detection value obtained by adding the number of times n is used as a plurality of pulse position units for at least a part of all pulse positions. Generating the pulse; controlling the pulse generation to shift the position of the reference pulse; storing the correlation detection values obtained for all positions of the reference pulse; Comparing the correlation detection values of the plurality of pulse positions and selecting a pulse position having the largest value, and when the correlation detection value of the plurality of pulse position units is larger than the other correlation detection values, the plurality of pulse positions are detected by a plurality of detectors. And re-calculating the correlation detection value by dividing and assigning the correlation detection value again.
[0016]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
[0017]
<Schematic configuration of signal detection device>
A schematic configuration of an example of a pulse modulation transmission / reception device using the signal detection device of the present invention will be described with reference to FIGS. In the present embodiment, a pulse waveform as shown in FIG. 3A and a monocyclic pulse having an autocorrelation as shown in FIG. 3B are used.
[0018]
As shown in FIG. 1, the pulse modulation transmission device includes a scrambler (Scrambler) 101, a pulse generator (Pulse generator) 102, a multiplier 103, a transmission RF unit 104, and a transmission antenna 105. This pulse modulation transmission device generates transmission symbols by randomizing transmission data (TX data) by a scrambler 101 and then generates a monocyclic pulse generated by a pulse generator 102 as shown in FIG. Modulation is performed, the signal is amplified to an appropriate value by the transmission RF unit 104, and transmitted from the transmission antenna 105. Note that the scrambler 101 has a function of preventing a transmitted pulse waveform from being biased to one data of both polarities.
[0019]
On the other hand, as shown in FIG. 2, the pulse modulation receiving apparatus includes a receiving antenna 201, a receiving RF unit 202, a pulse generator (Pulse generator) 203, a multiplier 204, an integral dump unit (Integrate & dump) 205, a desk It comprises a rambler (De-scrambler) 206, a searcher (Searcher) 207, a tracking loop (Tracking loop) 208, and a voltage-controlled oscillator (VCO) 209. This receiving apparatus receives a received signal by a receiving antenna 201, shapes the received signal into an appropriate form by a receiving RF section 202, multiplies the signal by a reference pulse generated by a pulse generator 203 of the receiver, and integrates the signal into an integration dump section 205. The data is demodulated by descrambling by the descrambler 206 with respect to the demodulated symbol that has been integrated and dumped for one symbol period in step (1), thereby obtaining demodulated data (Demod.data). Further, in order to synchronize the received signals, a search for a pulse position by a searcher 207 that performs a received signal position search at power-on or the like, and a phase difference between a pulse position of the received signal by a tracking loop 208 and a reference pulse are performed. The synchronous tracking for controlling the VCO 209 is performed by performing the following tracking.
[0020]
<Schematic configuration of searcher>
A schematic configuration of an example of the searcher 207 according to the signal detection device of the present invention will be described with reference to FIG.
[0021]
The searcher 207 in FIG. 4 includes a multiplier (Multiplier) 401, a correlator (Correlator) 402, a square operation unit 403, a pulse position estimator (Pulse position estimator) 404, an adder (adder) 405, a register (register) 406, It comprises a comparator 407, a selector 408, a pulse generator 410, and a window controller 412.
[0022]
In an asynchronous state such as when the power is turned on, the searcher 207 multiplies the received signal by a reference pulse at an arbitrary pulse position generated by the pulse generator 410, and performs integration over one symbol period by the correlator 402. A / D conversion is performed for each symbol rate, the difference between the A / D sample value delayed by one symbol period and the A / D sample value at the current time is squared by the square operation unit 403, and the window is calculated. A search operation is performed such that the addition unit 405 performs symbol addition for the number of additions (n) set by the control unit 412 and stores the result in the register 406 as a correlation detection value. In the full search, correlation detection values are calculated for all pulse positions within one symbol period while changing the pulse position of the reference pulse generator 410 according to the control information of the window control, and stored in the register 406, whereby the full search is performed. Complete. The symbol addition n times is a process for improving the pulse position detection probability by adding correlation detection values over a plurality of symbols for one pulse position, as described later.
[0023]
After the completion of the entire search, the correlation detection values of all the obtained pulse positions within one symbol period are compared by the comparator 407, and the pulse position having the largest correlation value is selected by the selector 408 and tracking is performed as the reception signal pulse position. By outputting to the loop, the initial synchronization is completed.
[0024]
The tracking loop 208 shown in FIG. 2 controls the VCO 209 by tracking the pulse position of the received signal based on the pulse position information obtained from the searcher 207, and controls the demodulation timing and the reference pulse according to the clock generated by the VCO 209. Determine the position. When the tracking loop 208 loses lock due to the influence of noise or the received signal strength becomes weak, loss of lock information is sent to the searcher 207, and the searcher 207 performs the above-described search operation again.
[0025]
<Conventional initial synchronization method>
Here, for comparison with the operation of the embodiment of the present invention, a conventional initial synchronization method will be briefly described with reference to FIGS. The pulse modulation in the present embodiment uses a bi-phase (two-phase) method in which the pulse waveform in FIG. 3A is assigned to the scrambled symbol 1 and its inverted pulse is assigned to symbol 0 at the same pulse position. I have.
[0026]
In the pulse modulation signal used in the present embodiment, as shown by the RX pulse in FIG. 5, six times the pulse period corresponds to the symbol period. However, in the present invention, the ratio between the pulse period and the symbol period is not limited to this. When a reference pulse (searcher pulse) used for correlation detection or demodulation of a received pulse modulated signal is generated at a pulse position equal to the received pulse as in case (case 1) of FIG. 5, lag = 0, and FIG. ) Becomes the maximum value. This maximum value is a value obtained after integration for the pulse period.
In a normal searcher, a pulse is generated at a symbol period using a gate signal 1 (gate signal 1) that generates a pulse in a low period, as in case1, and a period in which this gate is low is defined as all periods in the symbol period. A full search is performed by sliding about the pulse position. In the present embodiment, it is assumed that the pulse is slid every half of the pulse period. In case 1 in FIG. 5, the entire search is completed by sliding the pulse position 6 pulses × 2 = 12 times.
[0027]
At this time, with the correlation detection values of only one symbol for each of the twelve pulse positions, it is not always possible to detect a correct pulse position from the twelve pulse positions due to the influence of noise and signal attenuation even if comparison and selection are performed. Therefore, it is necessary to improve the pulse position detection probability by adding the correlation detection value over a plurality of symbols for one pulse position in accordance with the line condition. As described above, if the addition for n symbols (n is plural) is performed for one pulse position, the time required for the entire search is 12 × n symbol periods. Therefore, when the search reference pulse of case 1 is used, the time required for the entire search is 12 × n symbols.
[0028]
<Search Method According to Embodiment>
Next, an initial synchronization method according to the signal detection device according to the embodiment of the present invention will be described with reference to FIGS.
[0029]
Case 2 in FIG. 5 uses a gate signal 1 at the same timing as case 1 and a gate signal 2 delayed by a pulse period from the gate signal 1, and a pulse superimposed at the timing of the gate signals 1 and 2 is referred to as a search reference pulse ( This is an example of performing a search as a searcher pulse.
[0030]
Further, in case 3 of FIG. 5, the search is performed using the gate signals 1 and 2 and the gate signal 3 delayed by the pulse period from the gate signal 2 and using the pulse superimposed at the timing of the gate signals 1 to 3 as the search reference pulse. It is an example of performing.
[0031]
In cases 2 and 3, in order to shorten the search time as compared with case 1 in which a normal search is performed, the pulse generator 203 is used, and the correlations 402 as shown in FIGS. Use a container.
[0032]
The pulse generator 410 generates a reference pulse in a low period of the gate signal 1 in case 1 according to gate timing information such as a gate signal set in the window control unit 412, and generates a reference pulse in gate 2 in case 2 in case 2. A reference pulse in which two pulses are superimposed is generated in a low section where AND is performed. In case 3, a reference pulse in which three pulses are superimposed is generated in a low section in which gate signals 1 to 3 are ANDed. After performing a search using these reference pulses, the pulse is moved by a predetermined position to perform the next search. As will be described later, the number of the moving positions can be smaller than that in the related art.
[0033]
Next, a specific example of the correlator 402 of the searcher 207 will be described.
[0034]
FIG. 6 shows a three-parallel correlator 402a for independently correlating the pulse positions of the gate signals 1 to 3. The three parallel correlators 402a include three detectors (Detectors 1 to 3) 600 having the same configuration. Each detector 600 includes an integrator (Integrator) 601, an A / D converter (A / D) 602, a one-symbol delay unit 603, and an adder 604. In the three detectors 600, the integrator 601 integrates the output of the multiplier 401 in each of the low periods of the gate signals 1 to 3, and the A / D converter 602 performs A / D conversion for each symbol rate. Then, the adder 604 calculates the difference between the A / D sample value delayed by one symbol period by the delay unit 603 and the A / D sample value at the current time. The subtraction by the adder is for obtaining the amount of change in the integral value in the gate section. With such a configuration, it is not necessary to reset the integrator 601 for each gate section, so that the integrator can be realized by a simple analog circuit. The output of the adder 604 is squared by the squaring operation unit 403 as described above with reference to FIG. 4, and then symbol addition is performed by the adder 405 by the number of additions (n) set by the window control unit 412. It is stored in the register 406 as a correlation detection value. Note that, for each of the parallel detectors 600, the subsequent-stage squaring unit 403 and adder 405 also perform processing on a plurality of outputs in parallel.
[0035]
FIG. 7 shows a two-parallel correlator 402b for independently correlating two pulse positions, that is, a superimposed pulse position where the gate signals 1 and 2 are superimposed, and a pulse position of the gate signal 3. The two-parallel correlator 402b includes two detectors 701 and 702. Each of these detectors has the same configuration as the detector 600 shown in FIG. 6, but one detector 701 receives the output of an AND circuit 705 that takes the logical product of the gate signals 1 and 2 as a gate signal, The other detector 702 receives the gate signal 3. That is, the detectors 701 and 702 integrate the outputs of the multiplier 401 in the low period of the superposed pulse of the gate signals 1 and 2 and the low period of the gate signal 3, respectively, and perform A / D conversion for each symbol rate. After the A / D conversion in 602, the difference between the A / D sample value delayed by one symbol period in the delay unit 603 and the A / D sample value at the current time is obtained in the adder 604. As shown in FIG. 4, these obtained signals are each squared by the square operation unit 403, and then symbol-added by the number of additions set by the window control unit, and are stored in the register 406 as correlation detection values. It is memorized. Also in this case, corresponding to each of the parallel detectors 701 and 702, the subsequent-stage squaring unit 403 and adder 405 also perform the processing in parallel on a plurality of outputs. The specific operation of the two-parallel correlator 402b will be described later. The two-parallel correlator 402b is smaller in size than the three-parallel correlator 402a in FIG. 6, but can still realize a sufficiently high speed compared to the related art.
[0036]
FIG. 8 shows a serial correlator 402c that correlates the superimposed pulse positions of the gate signals 1 to 3. The serial correlator 402c includes a single detector 800. Detector 800 is the same as detector 600 shown in FIG. 6, but integrator 601 receives an output of AND circuit 805 that generates AND outputs of gate signals 1 to 3. The detector 800 integrates the output of the multiplier 401 by the integrator 601 in the low interval, performs A / D conversion by the A / D converter 602 for each symbol rate, and then delays by one symbol period by the delay unit 603. An adder 604 calculates the difference between the delayed A / D sample value and the A / D sample value at the current time. After squaring each output, the output is symbol-added by the number of additions set by the window control unit and stored in a register as a correlation detection value. The specific operation of the serial correlator 402c will be described later. The serial correlator 402c is smaller than the two-parallel correlator 402b of FIG. 7, and can realize a considerably higher speed than the conventional one.
[0037]
FIG. 9 shows a serial correlator 402d that integrates the correlation values of three gate signals having different timings with one integrator and outputs the respective correlation detection values in parallel for the superimposed pulse positions of the gate signals 1 to 3. The serial correlator 402d includes a detector 900 including an integrator 901, an A / D converter 902, a delay unit 904, an adder 905, and a selector 906. In this detector, an AND circuit 903 generates a low section where the gate signals 1 to 3 are superimposed, and integrates the output of the multiplier 401 in this section. Further, after performing A / D conversion by the A / D converter 902 in accordance with the timing signal CKctl for performing A / D conversion at the rise of the gate signals 1 to 3, the integrated value of one sample before stored in the delay unit 904 and the current time Is obtained by an adder 905, and the integrated value of each of the symbols of the gate signals 1 to 3 is sequentially output to the selector, and the integrated values corresponding to the gate signals 1 to 3 are detected by the selector 906 as detect 1 to 3, respectively. Output. These output signals are each squared by the squaring operation unit 403, then symbol-added by the number of additions set by the window control unit, and stored in the register as a correlation detection value. This serial correlator 402d is slightly larger in scale than the serial correlator 402c of FIG. 8, but can realize a sufficiently high speed as compared with the conventional one.
[0038]
<Initial synchronization method in the embodiment>
Next, an initial synchronization method according to the embodiment of the present invention will be described with reference to FIGS.
[0039]
FIG. 10 is a timing chart showing an example of initial synchronization when the three parallel correlators of FIG. 6 are used. FIG. 10A shows the timing of the transmission symbol and the reception signal. In this example, since the period is 6 pulses per symbol and the search interval is 1/2 pulse period, the number of search pulse positions is 12 as shown in FIG. Also, the three parallel correlators are used to independently search for the gate signals 1, 2, and 3 as shown in case 3 of FIG. 5 to obtain the correlation detection values c1 (t), c2 (t), and c3 (t). Thus, the entire search can be performed by shifting the pulse position at time t = 1, 2, 3, 4 and calculating the correlation detection values as shown in FIG.
[0040]
Assuming that the pulse position assigned at the time of transmission is c1 (1) at the settable pulse position in FIG. 11, the integrated value which is the output of the correlator 402 is expressed as shown in FIG. 10 (b) and is a correct pulse position. The integrated value of c1 (1) is the autocorrelation value at lag ≒ 0 in FIG. 3B in the case of symbol 1, and the value obtained by multiplying −1 by that in the case of symbol 0, and the integrated value of other pulse positions is It becomes an uncertain component such as noise. The integral is squared to remove the code component, and n symbols are added as shown in FIG. 10C to obtain a correlation detection value per pulse position.
[0041]
In FIG. 10, since three parallel correlators are used, correlation detection values of three pulse positions are obtained each time n symbols are received, and after 4 × n periods, that is, at time t = 4n + 1 (n = 0, 1, 2) ,...), The search is completed, and the peak value c1 (1) of the correlation detection value is obtained. The pulse position having the largest value among the twelve correlation detection values obtained at this time is detected, and the detected position is output as the pulse position of the received signal, thereby completing the initial synchronization.
[0042]
FIG. 12 shows an example of initial synchronization when the two parallel correlators 402b of FIG. 7 are used. FIG. 12A shows the timing of a transmission symbol and a reception signal. In this example, a pulse configuration similar to that of FIG. 10 is used, and the search is performed for a total of 12 pulse positions as shown in FIG. In the case of this example, as shown in case 2 of FIG. 5, the search is performed independently on the pulse position where the gate signals 1 and 2 are superimposed and the pulse position of the gate signal 3 to detect the correlation detection values c1 (t) and c2 (t). ). Therefore, if the pulse position is shifted four times at t = 1, 2, 3, 4 as shown in FIG. 13, the entire search is completed. That is, when shifting is performed in units of 0.5 pulses, a peak can be detected in a period of 4 symbols in a 2-parallel search.
[0043]
When the full search is completed, comparison is performed for correlation detection values at eight pulse positions of c1 (t), c2 (t); t = 1, 2, 3, and 4, and the pulse position having the largest value is c2 (t). In the case of t), the detected position is output as the pulse position of the received signal, and the initial synchronization is completed. When the pulse position with the largest value is c1 (ta), either of the gate signals 1 and 2 can be regarded as the correct pulse position for the pulse position of ta as in case 2 of FIG. In this case, there is a high probability that the pulse position can be detected without searching for n symbols again. Therefore, the correlation detection is performed for each smaller number of symbols n ', for example, n / 4 symbols, and when any of the correlation detection values is clearly large, the detected position is output as the pulse position of the received signal and the initial synchronization is performed. Finish. By doing so, correlation detection can be performed without obtaining a correlation value of n symbols for each candidate. In the example of the figure, this corresponds to the case where ta = 1.
[0044]
As described above, in the initial synchronization using the superimposed pulse, when there is a true pulse position in the superimposed pulse, after the entire search, the superimposed individual pulse is decomposed, and the correlation detection value is obtained again and compared. There is a need. When a two-parallel correlator 402b as shown in FIG. 7 is provided, after the entire search, the superimposed pulse is decomposed and the correlation detection value for the corresponding pulse position may be obtained by the two detectors. In this case, the maximum number of symbols required for the initial synchronization is (4 + 1) × n symbols. However, as described above, the symbols required for the re-search can be reduced. In the case of the serial correlator 402c as shown in FIG. 8, first, the search is performed n × 4 times to find the largest correlation detection value, and then the three superimposed pulse positions are decomposed to search n symbols each. Then, the maximum correlation detection value may be obtained. In this case, the maximum number of symbols required for the initial synchronization is (4 + 3) × n symbols. However, also in this case, the symbols required for the re-search can be reduced.
[0045]
When the serial correlator 402c in FIG. 8 is used, the delay time is longer than in the two parallel correlators in FIGS. However, one correlator shown in the conventional example requires 12 × n symbols for initial synchronization, and the initial synchronization time can be reduced by adding a little gate circuit as compared with the conventional example.
[0046]
When comparing the correlation detection value between the superimposed pulse and the non-superimposed pulse (single pulse) as shown in FIG. 12, if an appropriate SNR is secured and an appropriate value of n is used, the position other than the true pulse position is used. Since the correlation detection value has an average of 0 according to the Gaussian distribution, the correlation detection value of the pulse position including the true signal indicates the maximum value corresponding to lag ≒ 0 in FIG. Therefore, as shown in FIG. 12, even if the correlation detection values at a plurality of pulse positions based on the two superposed pulses corresponding to the gate signals 1 and 2 and the correlation detection values at the single pulse position corresponding to the gate signal 3 are compared, an error occurs. The detection probability is small.
[0047]
<Application to PPM>
So far, the embodiment of the signal detection apparatus for the bi-phase type pulse modulation method has been described. However, pulse position modulation (PPM: Pulse) in which different pulse positions are assigned to symbols 1 and 0 as shown in FIG. In the Position Modulation method, if the pulse generator 410 (FIG. 4) generates a pulse as shown in FIG. 14B, the same initial synchronization as in the above embodiment can be performed. FIG. 14A shows pulse positions assigned by PPM, and FIG. 14B shows superimposed pulses in the case of PPM corresponding to cases 1, 2, and 3 in FIG. In this example, the pulse period is 6 per symbol, and the pulse position assigned to the PPM is shifted by one pulse period.
[0048]
<Modification>
A determination threshold for the correlation detection value may be provided in advance for each number of superimposed pulses, and the determination threshold may be set according to the number of superimposed pulses. In this case, when the correlation detection value at a certain pulse position exceeds the determination threshold, the position is determined as the received pulse signal position. The determination threshold may be changed so as to be an appropriate threshold according to the SNR or the line condition so as not to cause an erroneous determination. This makes it possible to determine the position of the received pulse signal without performing a search for all pulse positions.
[0049]
Also, for example, the superimposed pulse is a pulse at an adjacent pulse position, but it is not always necessary to be adjacent.
[0050]
Although the example in which the number of superimposed pulses of the reference pulse is up to three has been described, it is also possible to superimpose four or more pulses. In that case, a corresponding gate signal is added. In the two-parallel correlator shown in FIG. 7, one-to-multiple pulse positions are initially assigned to two detectors, but multiple-to-multiple pulse positions can also be assigned. In this case, a plurality of pulse positions having a higher correlation detection value are divided and assigned to a plurality of detectors, a correlation detection value is obtained again, and division is performed until a high correlation detection value is obtained for a single pulse position. repeat.
[0051]
In addition, various modifications and changes are possible within the technical scope of the present invention.
[0052]
【The invention's effect】
According to the signal detection device and the signal detection method of the present invention, a reference pulse in which a plurality of pulses corresponding to a plurality of pulse positions are superimposed is generated, and a correlation detection value is obtained in a plurality of pulse position units using the reference pulse. Is included in a specific plurality of pulse positions, the plurality of pulse positions are divided to obtain a correlation detection value of each pulse, thereby obtaining a true pulse position. In any case, the number of necessary shifts of the position of the reference pulse is reduced and the pulse position of the received pulse signal is efficiently detected, as compared with the case where the reference pulse corresponding to the conventional single pulse position is generated. It becomes possible.
[0053]
The correlator may be composed of a plurality of detectors, in which case each detector detects the correlation between the received pulse signal and the reference pulse at different pulse positions simultaneously and in parallel, and at least one detector has By detecting the correlation between the received pulse signal and the reference pulse in a plurality of pulse position units, the processing speed can be further increased.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a schematic configuration of a pulse modulation transmission device.
FIG. 2 is a block diagram illustrating a schematic configuration of a pulse modulation receiving device.
FIG. 3 is an explanatory diagram of a pulse waveform (a) for pulse modulation and an autocorrelation (b) used in the embodiment of the present invention.
FIG. 4 is a block diagram illustrating a schematic configuration of an example of a searcher according to the signal detection device of the present invention.
FIG. 5 is a timing chart for explaining pulse superposition and parallel search in the conventional and the embodiment of the present invention.
FIG. 6 is a diagram illustrating a configuration example of a three-parallel correlator according to the embodiment of the present invention.
FIG. 7 is a diagram illustrating a configuration example of a two-parallel correlator according to the embodiment of the present invention.
FIG. 8 is a diagram showing a configuration example of a serial correlator (serial output) according to the embodiment of the present invention.
FIG. 9 is a diagram illustrating a configuration example of a serial correlator (parallel output) according to an embodiment of the present invention.
FIG. 10 is a timing chart showing an example of initial synchronization when the three parallel correlators of FIG. 6 are used.
FIG. 11 is a diagram showing search pulse positions in the example of FIG. 10;
FIG. 12 is a timing chart showing an example of initial synchronization when the two parallel correlators of FIG. 7 are used.
FIG. 13 is a diagram showing search pulse positions in the example of FIG. 12;
FIG. 14 is a timing chart for explaining application of the present invention to PPM.
[Explanation of symbols]
101: Scrambler, 102: Pulse generator, 103: Multiplier, 104: RF transmitting section, 105: Transmitting antenna, 201: Receiving antenna, 202: RF receiving section, 203: Pulse generator (Pulse generator), 204: Multiplier, 205: Integral & dump, 206: De-scrambler, 207: Searcher, 208: Tracking loop, 209: Voltage Control oscillator (VCO), 401 Multiplier, 402 Correlator, 402a Three parallel correlator, 402b Two parallel correlator, 402c Serial correlator (serial output) ), 402d: serial correlator (parallel output) 403: square operation unit, 404: pulse position estimator (pulse position estimator), 405: adder (adder), 406: register (register), 407: comparator ( Comparator, 408 ... Selector, 410 ... Pulse generator, 412 ... Window controller

Claims (11)

A signal detection device that detects a pulse position of a reception pulse signal in a pulse modulation reception device that receives a pulse-modulated signal,
For each symbol period, a pulse generator that generates a reference pulse by superimposing a plurality of pulses corresponding to a plurality of pulse positions so that pulses in the symbol section do not overlap,
A multiplier for multiplying the received pulse signal by the reference pulse,
A correlator that outputs a detection signal indicating a correlation between the reference pulse and the received pulse signal based on an output of the multiplier;
Calculating means for squaring the output of the correlator;
Adding means for generating a correlation detection value by adding the squared signal a predetermined number of times n;
Control means for controlling the pulse generator to shift the position of the reference pulse,
Storage means for storing the correlation detection value obtained for all positions of the reference pulse,
Means for comparing a plurality of correlation detection values stored in the storage means and selecting a pulse position having the largest value. The correlator has a plurality of detectors that output a detection signal indicating a correlation between a received pulse signal and the reference pulse, and each detector calculates a correlation between the received pulse signal and the reference pulse at different pulse positions. At the same time, the arithmetic means and the adding means process a plurality of detection signals in parallel, and at least one detector detects a correlation between the received pulse signal and the reference pulse in a plurality of pulse position units. 2. The signal detection device according to claim 1, wherein: 3. The method according to claim 2, wherein when the correlation detection value in the plurality of pulse position units is larger than another correlation detection value, the plurality of pulse positions are divided and assigned to the plurality of detectors, and the correlation detection value is obtained again. The signal detection device according to claim 1. 4. The signal detecting device according to claim 3, wherein the predetermined number of times n is reduced to a smaller number of times n 'when obtaining the correlation detection value again. The correlator has a single detector that outputs a detection signal representing a correlation between a received pulse signal and the reference pulse, and the detector detects the received pulse signal and the reference in units of the plurality of pulse positions. 2. The signal detection device according to claim 1, wherein a correlation with a pulse is detected. After detecting the plurality of pulse position units having the largest correlation detection value, the single detector sequentially calculates correlation detection values for the plurality of pulse positions, and determines the pulse position having the largest correlation detection value as a true pulse position. The signal detection device according to claim 5, wherein The pulse generator generates the reference pulse based on a gate signal corresponding to one or a plurality of pulse positions in a symbol section, and the detector generates a reference pulse according to the gate signal during a period designated by the gate signal. 7. The signal detection device according to claim 2, further comprising an integrator for integrating an output of the multiplier. 2. The pulse modulation according to claim 1, wherein the pulse modulation is a bi-phase pulse modulation in which a first pulse is assigned to symbol 1 and a second pulse having a polarity opposite to that of the first pulse is assigned to symbol 0 at the same pulse position. Signal detection device. The signal detection device according to claim 1, wherein the pulse modulation is a pulse position modulation (PPM) pulse modulation in which different pulse positions are assigned to symbols 1 and 0. A signal detection method for detecting a pulse position of a received pulse signal in a pulse modulation receiving device that receives a pulse-modulated signal,
For each symbol period, generating a reference pulse by superimposing a plurality of pulses corresponding to a plurality of pulse positions so that pulses in the symbol section do not overlap;
Multiplying the received pulse signal by the reference pulse;
Based on the output of the multiplier, outputting a detection signal representing the correlation between the reference pulse and the received pulse signal in a plurality of pulse position units for at least a part of all pulse positions,
Squaring the output of the correlator;
Generating a correlation detection value obtained by adding the squared signal a predetermined number of times n for a plurality of pulse position units for at least a part of all pulse positions;
Controlling the pulse generation to shift the position of the reference pulse;
Storing the correlation detection value determined for all positions of the reference pulse,
Comparing the stored plurality of correlation detection values and selecting a pulse position having the largest value;
When the correlation detection value of the plurality of pulse positions is larger than the other correlation detection values, dividing the plurality of pulse positions into a plurality of detectors and assigning the divided pulse positions to obtain a correlation detection value again. Signal detection method. 11. The signal detection method according to claim 10, wherein when the second correlation detection value is obtained, the predetermined number n is reduced to a smaller number n '.

Images