JP4560717B2 - Correlator - Google Patents

Description

  The present invention relates to a correlator necessary for detecting and demodulating a spread spectrum signal. In particular, the present invention relates to a correlator that can calculate a correlation value with a pseudo-random number (PN code) for performing synchronization acquisition on a received spread spectrum signal at high speed and does not increase the circuit scale.

Ｍ(t)は擬似乱数で＋１もしくは−１の値をとる。sinωtは搬送波を表す。ここではデータを送信することについては考えないので、送信データによる変調は含まれていない。

The spread spectrum signal is a signal obtained by spreading and modulating a carrier wave with a pseudo random number, and an example thereof is shown in FIG. The example of FIG.

M (t) is a pseudo random number and takes a value of +1 or -1. sinωt represents a carrier wave. Here, since transmission of data is not considered, modulation by transmission data is not included.

In order to detect such a spread spectrum signal from the received signal R (τ), a despreading process represented by Equation 2 is performed. t represents time, and T _M represents the time of one period of a pseudorandom number. This despreading process is to calculate a correlation value using the same pseudo-random number M (τ + p) and carrier wave sin (ωt + φ) used when creating a signal to be detected. φ of sin (ωt + φ) is a parameter for matching the phase with the carrier wave included in the signal to be detected, and this is adjusted to match the phase. Note that the method of matching the phases has a small relationship with the present application, and the description thereof will be omitted. If the parameter p for controlling the phase of the pseudo random number is changed in this state, the correlation value takes a large value only when the phases of the pseudo random number coincide with each other, so that the spread spectrum signal can be detected. it can. Rather, since the spread spectrum signal can be detected only by detecting a large correlation value, it is required to perform a calculation each time while changing the parameter p and to re-examine the correlation value, and to detect the signal. In order to perform it quickly, the correlation value must be calculated as fast as possible.

  The correlation value can be calculated by converting the received signal into a digital value by A / D conversion and performing a product-sum operation.

  FIG. 4 shows a simplest circuit example of a digital circuit that performs such product-sum operation. The correlator shown in FIG. 4 includes one multiplier, one adder, and one register other than the multiplier that multiplies the value of the carrier wave sin (ωt + φ), and has the smallest circuit scale. The number of clocks required for the calculation is the same as the number of data to be subjected to the product-sum operation, and the calculation time increases in proportion to the increase in the number of data.

  On the other hand, FIG. 5 shows an example of a circuit that can perform correlation operation at high speed by pipelining the circuit. The number of registers constituting this circuit is about twice the number of data to be multiplied and summed, and the number of adders is about the number of data. Therefore, the circuit scale becomes enormous as the number of data increases. On the other hand, the number of clocks required for the calculation is one clock because it is pipelined including the addition processing part.

JP 10-31373 A Yamane, Iyoda, Kaoru, Kubota, Watanabe: Application to spread spectrum sound wave distance measurement using pseudo-random M series, Transactions of the Society of Instrument and Control Engineers, Vol.39 No.10, 879/885, 2003

  The circuit shown in FIG. 4 described above minimizes the circuit scale, and the circuit shown in FIG. 5 is configured to minimize the calculation time. When calculating the correlation value, the number of data to be multiplied and summed and the required calculation time vary depending on the purpose and application, but here the number of data is assumed to be 1 to 20,000. Consider the circuits of FIGS. 4 and 5 with the goal of keeping the calculation time within a few μs.

  In the case of the circuit shown in FIG. 4, the circuit scale is very small, and the circuit scale is easy to create. However, since the calculation time is proportional to the number of data, it is necessary to set the operation clock to at least several GHz level in order to suppress the calculation time to several μs. In order to realize such an operation speed, it is necessary to create a dedicated IC using the latest semiconductor technology, but the cost becomes enormous and difficult to realize.

  In the case of the pipelined circuit of FIG. 5, since the time required for the product-sum operation is one clock regardless of the number of data, the calculation time can be kept within a few μs. However, since the scale of the circuit increases in proportion to the number of data, when the number of data is 1 to 20,000, the number of D-FFs (flip-flops) constituting the register is several hundred thousand to three hundred thousand. In addition to this, at least several hundred thousand gates constituting the adding circuit are required. At such a scale, it is difficult to produce a circuit using a large-scale PLD (programmable logic device) even though the degree of integration is improved. If it can be realized, the cost will be high, and if it is a dedicated IC, the cost will be further increased.

  An object of the present invention is to solve such a problem, and an object thereof is to realize a spread spectrum signal correlator capable of reducing the circuit scale as much as possible and shortening the calculation time. An object of the present invention is to provide a spread spectrum signal correlator capable of reducing the number of data necessary for calculating a correlation value by reducing useless recalculation and shortening the calculation time.

  In the present invention, the spread spectrum signal is modulated with a pseudo-random number M (t), and the modulation period (one chip) corresponds to several carrier periods, during which the value of M (t) changes. Focused on not doing. Since the value of the pseudo-random number does not change during this one chip, the signal for one chip is digitally expressed by Nc data, so such data is added (integrated or accumulated). Is performed in advance, the data amount when calculating the correlation value is reduced to 1 / Nc. The sum of the data for one chip is called one-chip data.

Next, based on one chip data, a correlation value for L chips is calculated in advance (this will be referred to as preliminary calculation). Although there are a plurality of Ls, the number is small compared to the above-mentioned tens of thousands of data. In order to calculate the correlation value, the value of the pseudo-random number M (t) used in the calculation needs to be determined, but cannot be determined at the stage of calculation in advance. Therefore, calculations are made for all possible pseudorandom numbers M (t). When calculating the correlation value for L chips, there are 2 ^{L possible} values of M (t), so all of them are calculated, and the value (this is called L chip data) is stored in the memory. Like that. The point to be noted here is that the pattern of the value of M (t) increases exponentially as the value of L increases, so the calculation time and the memory size increase exponentially. For this reason, the value of L is determined in consideration of them.

That is, the present invention provides a correlator for calculating a correlation value between the received spread spectrum signal and the pseudo-random number, a plurality Nc integrating into parallel signals of one chip period of the received signal by shifting the summing start timings a number of one-chip accumulation means, the data of the first memory chip data storing respective integral values from the 1-chip accumulation hand stage, L-number stored in the first memory chip data (L is a plural number) storing the L chips precalculation means for outputting the respective product-sum operation result by performing product-sum operation of the combination values of the pseudorandom L chip component reads the computed L chips preliminary calculation results, respectively L chip and data memory, L chip stored in said memory for L chip data corresponding to the value corresponding to the value and time of L pseudo-random number that is part of the spreading code Sequentially adding taken out a preliminary calculation result, characterized in that a correlation calculation means for obtaining a correlation value of the spread code one period.

  The A / D converter that converts the received signal into a digital signal, and the A / D converted spread spectrum reception signal is multiplied by a carrier signal and a signal whose frequency and phase match, and the one-chip accumulation is performed. And a multiplier input to the means.

In addition, the present invention can be realized as software, and when installed in an information processing apparatus, as a function of a correlator in the information processing apparatus, a carrier wave, a frequency, and a phase coincide with a spread spectrum reception signal converted into a digital signal. A multiplication function for multiplying signals, a signal of one chip period of the multiplied signals, a plurality of Nc one- chip accumulation functions for integrating in parallel at different start timings, and this one-chip accumulation function of the function of storing each integrated value to 1 memory chip data, the 1-chip L number stored in the data memory (L is a plural number) of the data by reading the combination value of the pseudorandom L chips worth of and L chip precalculated function of outputting a respective product-sum operation result by performing product-sum operation, the computed L chips preliminary calculation result A function of storing each of the chip data memory, wherein L L chips stored in the memory chip data precomputed imaging corresponding to the value corresponding to the value and time of L pseudo-random number that is part of the spreading code A correlation calculation function for obtaining the correlation value for one cycle of the spread code is realized by sequentially extracting and adding the results .

  In this way, if the correlation value for L chips is calculated by preliminary calculation and stored in the memory as L chip data, the number of data to be handled can be obtained by using the correlation value when actually calculating the correlation value. It becomes 1 / (Nc × L) of the number of original data, and the calculation time can be greatly reduced. As a result, the frequency of the clock signal for operating the circuit can be lowered. In addition, since a memory element is used instead of a register to store data used for correlation value calculation, the hardware scale for other calculation processing can be made relatively small and available on a small scale. A low-cost correlator can be realized using a possible PLD circuit.

  FIG. 1 shows a circuit example of a correlator according to an embodiment of the present invention.

  In this embodiment, an A / D converter 1, a multiplier 2, a 1-chip accumulator 3, a 1-chip data memory 4, an L-chip preliminary calculator 5, an L-chip data memory 6, and a correlation calculator 7 are provided. It is the structure connected sequentially.

The received spread spectrum signal is converted into digital data by the A / D converter 1. This spread spectrum received signal is multiplied by a signal sin (ωt + φ) whose phase and frequency are matched with the carrier contained in the signal to be detected by the multiplier 2. This data is input to the one-chip accumulator 3, and one chip is added by each of the _Nc accumulators 31 _{1 to} 31 _Nc and stored in the one-chip data memory 4. In the signal to be detected, since the division of the spread modulation period (one chip) by the pseudo random number M (t) is not known, the accumulator 31 in the one-chip accumulator 3 is the number of data included in one chip. The same Nc are prepared, and the start timing of the addition calculation of each accumulator 31 is shifted, so that it is possible to cope with any timing of chip separation.

Further, L pieces of data stored in the one-chip data memory 4 are read out by the L-chip preliminary calculator 5 and inputted to the _L registers 51 _{1 to} 51 _L , and the L multipliers 52 _1. in the L-number of pseudorandom numbers that have been set storage elements (registers) 54 ₁ pseudorandom values from through 54 _L (L number of +1 or -1 value) adder 53 on multiplied by the to 52 _L Is added. Here, since calculations are performed for all possible patterns for the values of L pseudorandom numbers, 2 ^L pieces of L chip data are obtained from L pieces of 1 chip data. That is, here, as shown in FIG. 1, the L chip spare calculator 5 is prepared in parallel with 2 ^L pieces of ^L combinations of pseudo random numbers, or one L chip spare calculator 5 uses pseudo random numbers. The product sum calculation is performed 2 ^L times while changing the value of. However, in the case of 2 ^L times product-sum calculation with one L-chip preliminary calculator 5, the number of circuits is reduced, but the clock frequency of the L-chip preliminary calculator 5 needs to be increased. The 2 ^L pieces of data preliminarily calculated by the L chip preliminary calculator 5 are stored in the L chip data memory 6.

  The correlation calculator 7 gives the value of the pseudo-random number and the time corresponding to the time to the L-chip data memory 6 so that the corresponding L-chip data is immediately taken out and added. Thereby, the correlation calculator 7 calculates and outputs the correlation value. Since the value of the pseudo random number used in the correlation calculator 7 can be arbitrarily selected, the correlation value corresponding to the spread spectrum signal to be detected can be calculated by changing the type and phase of the pseudo random number as necessary. it can.

  FIG. 2 is a diagram for explaining the correlation calculation of the spread spectrum signal in the circuit configuration shown in FIG. 1. From the top, sin (ωt + φ) having the same frequency and phase as the carrier wave is added to the received spread spectrum signal. Multiply it and accumulate 1 chip data. L pieces of this one-chip data (here, L = 3) are taken out and preliminary calculation of the L-chip data is performed. Since there are eight combinations for the three pseudorandom numbers, eight preliminary calculations are performed, and preliminary calculation values of eight L chip data are stored in the L chip data memory. The correlation calculator 7 inputs a pseudo-random number for performing correlation calculation and a value corresponding to the time to the L chip data memory, and obtains a correlation value by extracting and adding the corresponding L chip data. Can do.

  Note that the present invention can also be realized as software that, when installed in an information processing apparatus, causes the information processing apparatus to function as a digital correlator that performs correlation calculation of spread spectrum received signals. That is, as a function of the digital correlator, a multiplying function for multiplying a spread spectrum received signal converted into a digital signal by a signal whose frequency and phase match with a carrier wave, and a signal of one chip period of the multiplied signal are integrated. Based on a plurality of one-chip accumulation functions, a function for storing the integrated value integrated in the one-chip data memory 4, and data stored in the one-chip data memory, L chips (L is a plurality) L-chip preliminary calculation function for performing a product-sum operation with a combination value of pseudo-random numbers, a function for storing the calculated L-chip preliminary calculation result in the L-chip data memory 6, and storing in the L-chip data memory A program for realizing a correlation calculation function for calculating a correlation value between a spread spectrum signal and a pseudo-random number based on the data of the L chip preliminary calculation result It may also be implemented as a digital correlator by installing the information processing apparatus. This program can be recorded in a recording medium or installed in the information processing apparatus via a communication line.

  In the present invention, in the position detection using the spread spectrum signal such as the position detection by GPS, the synchronization acquisition for the spread spectrum signal can be performed at a high speed and with a small circuit scale, so that the position detection using the spread spectrum signal can be performed. Can be used. In particular, it can be used to detect the position of an autonomously traveling robot that needs to know its own position in a building using ultrasonic waves.

The block diagram which shows the structure of the correlator of this invention Example. The figure explaining the correlation detection by an Example. The figure explaining spread spectrum. The simplest correlator circuit example. The example of a correlator of a pipeline structure.

Explanation of symbols

1 A / D converter 2 Multiplier 3 1-chip accumulator 4 1-chip data memory 5 L-chip preliminary calculator 6 L-chip data memory 7 Correlation calculator

Claims (3)

In a correlator that calculates a correlation value between a received spread spectrum signal and a pseudo-random number,
A plurality of Nc 1-chip accumulating means for adding the signals of the 1-chip period of the received signals and integrating them in parallel at different start timings ;
A first memory chip data storing integrated values from the 1-chip accumulation hand stage respectively,
L for outputting each of the 1-chip memory L number stored in the data (L multiple) data read out by performing product-sum operation of the combination values of the pseudorandom L chip partial sum operation result of Chip preliminary calculation means;
A memory for L chip data storing the computed L chips preliminary calculation results, respectively,
The L chips sequentially removed L chips preliminary calculation result stored in the data memory by adding, spreading code 1 cycle corresponding to the value corresponding to the value and time of L pseudo-random number that is part of the spreading code And a correlation calculating means for obtaining a correlation value of minutes . An A / D converter that converts the received signal into a digital signal, and the A / D converted spread spectrum reception signal is multiplied by a carrier signal and a signal whose frequency and phase match, and the one-chip accumulation means The correlator according to claim 1, further comprising an input multiplier. By installing the information processing device, the correlator function of the information processing device
A multiplication function for multiplying a spread spectrum received signal converted into a digital signal by a signal whose frequency and phase match with a carrier wave;
A plurality of Nc 1-chip accumulation functions for adding the signals of the multiplied signals in one chip period and integrating them in parallel at different start timings ;
A function of storing each integrated value from the 1-chip accumulate function in 1 memory chip data,
L for outputting each of the 1-chip memory L number stored in the data (L multiple) data read out by performing product-sum operation of the combination values of the pseudorandom L chip partial sum operation result of Chip preliminary calculation function,
A function of storing each said computed L chips preliminary calculation result in the memory for L chip data,
The sum is taken out the L L chips preliminary calculation result stored in the memory chip data corresponding to the value corresponding to the value and time of L pseudo-random number that is part of the spreading code sequence, the spreading code 1 A program for realizing a correlation calculation function that obtains a correlation value for a period .

Images