JP3888220B2 - Code point detection method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a device suitable for controlling (radio control) an object to be controlled at a remote location based on control data transmitted by radio waves, and in particular, a carrier wave frequency is frequency-modulated for multiple purposes. The present invention relates to a code point detection method suitable for transmitting / receiving value data.
[0002]
[Prior art]
Radio control (hereinafter referred to as radio control) technology for manipulating devices or devices that move at remote locations with control information on radio waves has become widespread, and as controlled devices, moving objects such as model cars and ships can be used. It is common practice to steer.
Since such a control device generally uses a narrow-band modulated signal permitted by the Radio Law, it is usually performed with binary pulse data. However, in order to improve the responsiveness of the controlled device, data transmission When the speed is increased, the bandwidth of the carrier frequency is widened, causing intersymbol interference with the control signal of the adjacent channel, causing a problem.
[0003]
[Problems to be solved by the invention]
For example, in a conventional 4-level FSK decoded signal, a preamble signal with a signal duty of 50% is inserted into a part of a transmission modulated wave at an appropriate period, and a zero-crossing point of this code is detected, and from that point, T / 2 The point shifted by this is determined as the sampling point of the code detection point.
However, a PLL circuit or the like is required to detect a zero-crossing point of the signal waveform, which complicates the circuit and adds a preamble signal, thereby deteriorating code transmission efficiency.
[0004]
[Means for Solving the Problems]
The code point detection method of the present invention was made to improve such a problem,
For example, for a 1/8 shift 4FSK modulated signal obtained by differentially encoding a quaternary signal, a multi-level modulation method is adopted in which frequency modulation is performed by converting the signal locus into a cosine curve at a symbol period for band limitation. Yes.
Then, the zero output point is detected as a symbol point by passing the demodulated signal through a differentiating circuit.
[0005]
In the above four-value FSK modulation method, a modulation signal is obtained from difference data from the previous symbol value, and the obtained four-value differential mapping data takes a specific value of eight values on the frequency plane and takes each symbol value. The inter-symbol distance between them has a quaternary amplitude.
[0006]
DETAILED DESCRIPTION OF THE INVENTION
First, a multi-level modulation method proposed by the present applicant and applicable to the present invention will be described.
FIG. 1 is a block diagram showing an embodiment to which the code point detection method of the present invention can be applied. In this figure, reference numeral 21 denotes a differential encoder (decoder) to which transmission data is inputted, and is constituted by a table in which the next transmission data is mapped based on the previous transmission data, as will be described later.
Reference numeral 22 denotes a frequency modulator that frequency-modulates the modulation signal formed by the differential encoder 21, reference numeral 23 denotes a high-frequency power amplifier that amplifies the power of the modulated carrier frequency, and reference numeral 24 denotes a transmission antenna.
[0007]
Reference numeral 25 denotes a receiving antenna that receives information transmitted from the transmitting side, 26 denotes a high frequency amplifier, and 27 denotes a frequency detector including a discriminator.
Reference numeral 28 denotes a differential decoder (decoder) for converting the data decoded by the mapping on the transmission side differential encoder into the original array. The control information of the controlled device is obtained by the data of the differential decoder 28. It is made so that you can get.
[0008]
FIG. 2 is a weight shift amount table showing the shift amount of the data to be output next by the four-value current data Y1 and Y2 input to the transmitter.
In this figure, Y1 and Y2 indicate intervals (code weights) in which the shift amount is mapped with data having 2 bits as one symbol.
[0009]
FIG. 3 shows how the code weight of data changes as a weight transition table.
In this figure, the symbol position of the previous symbol is indicated by ■, the left column indicates the code weight and the shift amount, and the horizontal axis indicates the data transition.
As can be seen from these figures, in the present invention, the weight of the 4-level FSK signal maps a 2-bit signal for each symbol onto an 8-level amplitude plane by differential encoding.
[0010]
From the above table, when the code weights of 8 values are (−7), (−5), (−3), (−1), (+1), (+2), (+3), (+5), and (+7), respectively, For each of the data (00) (10) (01) (11), +2. -2. +6. It can be seen that an amplitude differential value of −6 is given.
[0011]
Hereinafter, the case of transition 4 in FIG. 3 will be described as an example. In the case of transition 4, when the previous data weight position is (−1) and the data input at the time of forward shift is (00), the shift in FIG. The amount is +2. Therefore, from (-1) + (+ 2), the code weight becomes a level having a code weight of +1. Also, in the case of transition 4, when the 2-bit symbol to be input next is (10), the code weight is obtained from FIG. 2 (the shift amount is -2, so (-1) + (-2) =-3. Becomes a level of -3.
[0012]
Similarly, at the position of transition 4, when the next input data is (01), the shift amount is +6, so the sign weight is +5 level. When the data is (11), the shift amount is -6. The level of the code weight of −7.
In addition, when it falls below -7, it is changed to +7, and the weight distance between +7 and -7 is ± 2.
[0013]
The code weight is a form in which the code weight is shifted by 1/8 for each symbol, and the symbol weight always has a different code weight. Therefore, run length 1 is guaranteed.
The weight distance of each symbol value is 4, which is the same distance as that of the normal 4-FSK system, and the error rate characteristics are the same.
Further, the only bit with a different sign adjacent to one symbol is 1, which is gray-coded.
Therefore, it is possible to prevent a large numerical value from jumping due to a 1-bit error.
[0014]
As can be seen from the above code table, when the symbols (00) and (10) are consecutive, the code weight is shifted by 1/8. Therefore, the modulation system of the present invention is called a 1/8 shift 4FSK modulation system.
Demodulation normally performs frequency detection in the same manner as the 4-FSK modulation method, returns to the baseband region, and enables determination of 2-bit data from the amplitude difference between the received data and the previous data.
[0015]
FIG. 4 shows the transition (modulation waveform) of the code weight with respect to the symbol value when the code weight -7 is set as an initial value.
The four-value code sequence used in this embodiment is “00111100010101011000011000101110”.
As can be seen from this modulated waveform, there is no portion that is continuous in the frequency plane for each symbol.
[0016]
By the way, in the present invention, for example, a band limitation is applied to the 1/8 shift 4FSK modulated signal so that the intersymbol interference becomes zero.
That is, signal processing for converting a 1/8 shift 4FSK modulated wave waveform having a pulse interval of T as shown in FIG. 5 into a signal locus of COS (πt / T) (T is a symbol period) was performed. The thick line is the pulse waveform, and the thin line is the waveform converted to COS (πt / T).
[0017]
This signal conversion can be easily realized by referring to the cosine table for the signal level.
In the waveform conversion of the present invention, the signal locus is a cosine curve, and its impulse response becomes zero when t ≧ ± T. Therefore, the eight-value code point locus is always constant, and the intersymbol interference has a zero sign.
Then, the modulated waveform is transmitted by frequency modulation as seen in FIG.
[0018]
FIG. 6 is a block diagram in which a code point detection circuit is attached to a multi-level modulation circuit (binary-quaternary conversion), where 31 is a binary-quaternary conversion circuit, and 32 is a cosine table or the like. A band limiting circuit 33 is a frequency modulator, and 34 is a high frequency power amplifier.
35 is a high-frequency amplifier that amplifies the signal received by the receiving antenna, 36 is a frequency detector comprising a discriminator, etc. 37 is a differentiation circuit that differentiates the detection output, and 38 is a differential waveform output from the differentiation circuit 37. The zero point detection circuit 39 detects the zero point of the code point, and 39 is a code point detection circuit for detecting the code point using the timing of the zero point detection circuit as a sampling point.
[0019]
Thus, the receiver demodulates the transmission modulation signal by frequency-detecting the transmission signal.
FIG. 7 shows an example of a multi-value signal to be transmitted (indicated by a thick line) and a waveform (indicated by a thin line) obtained by converting the multi-level signal level into a COS (πt / T) locus. The impulse response of the filter that performs this waveform conversion is set to zero when t ≧ ± T.
Moreover, the waveform shown with a dotted line shows the differential waveform which differentiated the waveform converted into the locus | trajectory of COS ((pi) t / T).
Then, when the multi-value signal subjected to frequency detection is passed through the differentiating circuit 37, a sine waveform (dotted line) is formed so that the point where no code interference occurs is zero as shown in FIG. Can be detected.
Therefore, if the frequency detection voltage is detected at this point, the detection output voltage can be determined very accurately.
[0020]
That is, data sampling must be performed at a point where the intersymbol interference is zero.
An example of reproducing a so-called Nyquist point where the intersymbol interference is zero is shown below.
(A) Differentiating the received signal (16 times oversampling value difference).
(B) The Nyquist point has a differential value of zero. This point is averaged on the time axis through the LPF.
The detection output voltage is converted into a PCM code to form operation information of the steered device. In the above embodiment, a simple binary-to-fourth conversion circuit is used as the multi-level modulation circuit. However, as previously proposed by the present applicant, a differential encoder and a differential decoder are used. Needless to say, the present invention is applicable to an 8-shift 4FSK multi-level modulation circuit.
[0021]
【The invention's effect】
As described above, according to the present invention, the code point can be detected only by differentiating the demodulated signal, and the circuit configuration can be simplified.
Since the code point can be detected at the timing when the zero point of the differential output is detected, high-speed bit synchronization is possible.
Since the sign which becomes the 1/8 shift 4FSK signal does not take the same symbol, the differential output passes through the zero point for each symbol. Therefore, it is possible to detect the symbol point at all times, and there is an effect that the addition of the preamble is not necessary and the data transmission efficiency is improved.
[Brief description of the drawings]
FIG. 1 is a block diagram of a multi-value modulation type transceiver to which the code point detection method of the present invention is applied.
FIG. 2 is a diagram of code weights for differential encoding.
FIG. 3 is an explanatory diagram showing a state in which current data is mapped based on previous data.
FIG. 4 is a waveform diagram showing a part of a modulation waveform when shifted by mapping.
FIG. 5 is a waveform diagram showing a frequency detection output when a transmission pulse is waveform-converted by a COS (πt / T) curve.
FIG. 6 is a block diagram of a circuit to which the code point detection method of the present invention is applied.
FIG. 7 is a waveform diagram when a signal demodulated by a cosine curve is passed through a differentiating circuit.
[Explanation of symbols]
31 Binary to 4-value conversion circuit, 32 Band limiter, 33 Frequency modulator, 34 High frequency amplifier, 35 High frequency amplifier, 36 Frequency detector, 37 Differentiation circuit, 38 Zero point detection circuit, 39 Code point detection circuit

Claims (4)

For the modulation signal, take a multi-level modulation method that performs frequency modulation by converting the signal trajectory into a cosine curve or a curve equivalent to it in the symbol period for band limitation,
A code point detection system characterized in that a zero output point is detected as a symbol point by passing the demodulated signal through a differentiating circuit. The modulation signal is a 1/8 shift 4FSK modulation signal, the modulation signal is obtained from difference data from the previous symbol value, and the obtained multi-value differential mapping data is an eight-value specific value on the frequency plane. The code point detection method according to claim 1, wherein the inter-code distance between the respective symbol values has a four-value amplitude. The code point detection method according to claim 1, wherein the modulation signal is radio controlled PCM code data. 2. The code point detection method according to claim 1, wherein the mapped data is constituted by a Gray code.

Images