KR100284698B1 - Multiple FSK Demodulation Device and Method - Google Patents

Abstract

Disclosed are a multiple FSK demodulation device and method for demodulating two or more FSK modulated signals. The present invention provides a first process of detecting output energy of a filter having a bandwidth including at least one tone, and a bandwidth including a number of tones included in the bandwidth of the first process and smaller than the number of tones included in the filter. In the second process of detecting the output energy of the filter having a filter, the energy detected in the first process and the energy detected in the second process are subtracted, which is smaller than the frequency band of the filter having a bandwidth including one or more tones. A third process of detecting the energy of the tones included in the band excluding the frequency band of the filter having the number of tones, and demodulating by comparing the energy detected in the second process with the magnitude of the energy detected in the third process And a fourth step of determining the data.

Description

Multi-FSX Demodulation Apparatus and Method

The present invention relates to a frequency shift keying (FSK) demodulation device and a method thereof, and more particularly, to a multiple FSK demodulation device and a method capable of demodulating two or more FSK modulation signals.

In general, the FSK modulation scheme is one of the digital modulation schemes commonly used in communication systems. In the existing communication system, only one specific level of FSK modulation is mainly used, but as the necessity of various services such as multimedia services increases, the necessity of supporting two or more data rates is increasing. For example, the Home RF system and the IEEE 802.11 wireless LAN system simultaneously support 1 Mbps and 2 Mbps, respectively, using 2-FSK and 4-FSK modulation schemes.

1 is a block diagram showing a conventional digital FSK demodulation device.

First, only the channel band is filtered through the first bandpass filter 110 and converted into a digital signal through an analog-to-digital converter (ADC) 120. This digital FSK signal has e1 (t) and e2 (t) depending on the value of data "0" or "1", and is a sine wave (eg 1200Hz and 1800Hz) with a constant frequency added and subtracted from the carrier frequency. )to be. The analog-to-digital converted FSK signal is simultaneously input to the first bandpass filter 130 and the second bandpass filter 140 through the A and B lines. The first band pass filter 130 and the second band pass filter 140 are, for example, filters for filtering only frequency components of 1200 Hz and 1800 Hz, respectively. The FSK signals b1 (t) and b2 (t) passing through the first bandpass filter 130 and the second bandpass filter 140 are multiplied by the multipliers 150 and 160 to detect the magnitudes of the respective frequency components. Generate a squared FSK signal c1 (t), c2 (t). The FSK signals c1 (t) and c2 (t) obtain a signal s (t) calculated by c1 (t) -c2 (t) in the subtractor 170. The FSK signal s (t) is outputted through the low pass filter 180 to exclude high frequency noise, and is generated as demodulated data via the determiner 190. That is, the determiner 190 is calculated by the subtractor 170 as c1 (t) -c2 (t), and if the result value is greater than or equal to "0", the data corresponding to the 1200 Hz component is "0", and "0". If less, it is determined as data "1" corresponding to the 1800 Hz component.

As shown in FIG. 1, the apparatus of FIG. 1 is implemented for one specific level of FSK modulated signal. In order to process both levels, an additional number of branches, such as A and B, is required to increase complexity and cost. There is a disadvantage. For example, to implement a 2FSK / 4FSK demodulator, a total of six independent branches are required, so six bandpass filters are required.

The technical problem to be achieved by the present invention is to calculate the output energy difference of the narrower filter and the narrower filter in the wider bandwidth filter while the output energy and frequency response of the relatively wider bandwidth are within the output frequency range of the filter. The present invention provides a multilevel FSK demodulation device and a method for demodulating a plurality of FSK signals modulated at different levels using a method for detecting energy in the remaining bands.

1 is a block diagram showing a conventional digital FSK demodulation device.

2 is a block diagram of a multilevel FSK demodulation device according to the present invention.

FIG. 3 shows frequency characteristics of each filter of FIG. 2.

4 is a block diagram showing another embodiment of a multilevel FSK demodulation device according to the present invention.

In order to solve the above technical problem, the present invention comprises a plurality of filters in two or more frequency modulated signal demodulation method,

Detecting an output energy of a filter having a bandwidth including at least one tone;

A second step of detecting output energy of a filter included in the bandwidth of the first step and having a bandwidth including a number of tones smaller than the number of tones included in the filter;

The frequency band of the filter having a bandwidth including a smaller number of tones in the frequency band of the filter having a bandwidth including one or more tones by subtracting the energy detected in the first process and the energy detected in the second process. Detecting the energy of the tone included in the band excluding the third step;

And a fourth process of determining demodulation data by comparing the energy detected in the second process with the magnitude of the energy detected in the third process.

In order to solve the above other technical problem, the present invention provides two or more frequency modulated signal demodulation device,

First bandpass filter means for filtering one or more tones of a received signal;

Second bandpass filter means for filtering again the following number of tones from one or more tones filtered by said first bandpass filter means;

First calculation means for detecting energy of tones output from said first band pass filter means and said second band pass filter means;

The energy of the tones detected by the first and second computing means is subtracted to reduce the energy of the tones included in the band except the frequency band filtered by the second band pass filter means from the frequency band within the first band pass filter means. First computing means for detecting energy;

And a judging means for judging data by comparing magnitudes of energy detected from said first and second computing means.

Hereinafter, exemplary embodiments of the present invention will be described with reference to the accompanying drawings.

2 is a block diagram of a multilevel FSK demodulation device according to the present invention. The apparatus of FIG. 2 includes a first broadband pass filter 211 and a second broadband pass filter 212, the first broadband pass filter 211 and a second broadband pass filter for inputting an FSK signal to filter a strong band. First and second multipliers 215 and 217 that square the output of 212, and a first narrow band pass filter that filters back output signals of the first and second broadband pass filters 211 and 212. 213 and the third and fourth multipliers 216 and 218 square the outputs of the second narrow band pass filter 214, the first narrow band pass filter 213, and the second narrow band pass filter 214. The first subtractor 219 and the second subtractor 220 for calculating the difference between the first multiplier 215 and the third multiplier 216 and the difference between the second multiplier 217 and the fourth multiplier 218, respectively. First, second, third, and fourth low pass filters 223,221, 222 and 224 that filter only DC components, and determine bits from signals output from the first, second, third and fourth low pass filters 223, 221, 222 and 224. Judgment machine (22 5) consists of.

FIG. 3 shows frequency characteristics of each filter of FIG. 2. Referring to Figure 3, 301 is the frequency axis, 308 is the tone corresponding to "1" of 2-FSK, 309 is "0" of 2-FSK, 310 is "10" of 4-FSK, 311 is "11" of 4-FSK, 312 is "1" of 4-FSK, and 313 are tones corresponding to "0" of 4-FSK. 302 is a frequency band of the first broadband pass filter 211 that filters the tones of 310, 308, and 211, and 303 is a frequency band of the second broadband pass filter 212 that filters the tones of 312, 309, and 313.

Referring to the operation of the present invention in combination with Figure 2 and 3 as follows.

First, the FSK modulated signal input to the apparatus of FIG. 2 may be an intermediate frequency signal or a baseband digital signal that has passed through an RF stage (not shown).

Also, asynchronous receivers that typically demodulate 4-FSK should have a filter with a frequency response such as 304, 305, 306, 307 of FIG. 2-FSK tones are when "1" is in the band of 304 or 305 of FIG. 3 and "0" is in the band of 306 or 307 depending on the modulation index. Therefore, in the case of 2-FSK, the first wideband filter 211 and the second wideband filter 212 filter the tones of 308 and 309 of FIG. 3, respectively, and in the case of 4-FSK, the first wideband filter 211 Filters the tones of 310 and 311 and the second wideband filter 212 filters the tones of 312 and 313. The first multiplier 215 and the second multiplier 217 square each to detect energy for the tones filtered from the first and second broadband pass filters 211 and 212, respectively. The first narrowband pass filter 213 and the second narrowband pass filter 214 have a frequency response of 305 and 306 of FIG. 3, respectively, in the first wideband pass filter 211 and the second wideband pass filter 212. The output tones are narrow-filtered. For 4-FSK, tones such as 311 and 312 are filtered. For 2-FSK, only unwanted noise is present in the filter band. The third multiplier 216 and the fourth multiplier 218 square each to detect energy for the tones filtered from the first narrowband pass filter 213 and the second narrowband pass filter 214. The first subtractor 219 subtracts the energy detected by the first multiplier 215 and the second multiplier 217, and the second subtractor 220 detects the third multiplier 216 and the fourth multiplier 218. Subtracted energy. The second low pass filter 221 and the third low pass filter 222 may include a direct current component h corresponding to an energy difference from which the high frequency portion is removed from the energy subtracted from the first subtractor 219 and the second subtractor 220. ₂ , h ₃ ) only. In other words, taking only the upper branches as an example, the energy existing in the band of 305 is subtracted from the energy existing in the band of 302 of FIG. In addition, the first low pass filter 223 and the fourth low pass filter 224 remove high frequency components from energy output from the second multiplier 217 and the fourth multiplier 218, and the direct current components h ₁ and h. ₄ ) Only detect it. In the case of 2-FSK, the determination unit 225 determines the tone of 308 of FIG. 3 when the DC component h ₂ of the second low pass filter 221 is larger than the DC component h ₃ of the third low pass filter 222. Is determined to be "1" because it is present. Otherwise, it is determined to be "0" because tones of 309 exist. In case of 4-FSK, the direct current of the first, second, third and fourth low pass filters 221, 222, 223 and 224 Comparing all the components (h ₁ , h ₂ , h ₃ , h ₄ ), select the largest one. The largest DC component (h ₁ ) is “11”; the largest DC component (h ₂ ) is “10”. If the direct current component h ₃ is largest, it is determined as "0". If the direct current component h ₄ is largest, it is determined as "1".

Meanwhile, as another embodiment of the apparatus of FIG. 2, the positions of the low pass filters 221, 222, 223 and 224 may be located after calculating the difference between the two squared signals, and the low pass after the tones of the two band pass filters 213 and 214 are squared, respectively. The difference may be calculated after passing through the filters 221, 222, 223, and 224. In addition, in the case of 2-FSK, the frequency response of the first narrowband pass filter 213 uses a band equal to 304 in order to detect the tone of 308 in FIG. 3, and in case of 4-FSK, the frequency response of 301 is 302. We can subtract the energy of 304 from the energy of, but if the tone of 308 is present at the boundary of the filter, such as 304, there is a high probability of error in the determination.

4 is a block diagram showing another embodiment of a multilevel FSK demodulation device according to the present invention.

First Wideband Filter 211 and Second Wideband Filter 212, First Narrowband Filter 213 and Second Narrowband Filter 214, First, Second, Third, and Fourth Multipliers 215, 217, 216, 218 is the same as the apparatus of FIG. The energy output from the second and fourth multipliers 217 and 218 is directly input to the second low pass filter 420 and the third low pass filter 421 to be output as DC components (h ₃ and h ₄ ). The energy output from the first and third multipliers 215 and 216 is directly input to the first low pass filter 419 and the fourth low pass filter 422 and output as DC components h ₁ and h ₆ . (h ₁ ) and the direct current component (h ₃ ) are subtracted from the first subtractor 423 and output as the direct current component (h ₂ ), and the direct current component (h ₄ ) and the direct current component (h ₆ ) are the second subtractor (424). ) is subtracted from and output to a DC component (h _5). determining section 425 in the 2-FSK for comparing the direct current component (h ₁₎ and the DC component (h _6), the DC component (h ₁₎ is large. " 1 "and when the direct current component h ₆ is large," 0 ". That is, unlike the apparatus of FIG. 2, the energy passing through the band 302 of FIG. 3 and the energy passing through the band 303 are compared. Also for 4-FSK direct current If _{(h 2, h 3, h} 4, h 5) is determined by selecting data corresponding to the largest to compare.

As described above, according to the present invention, since two or more FSK modulated signals can be demodulated by a relatively simple implementation, the cost is reduced.

Claims (2)

In the method of demodulating two or more frequency modulated signals with a plurality of filters, Detecting an output energy of a filter having a bandwidth including at least one tone; A second step of detecting output energy of a filter included in the bandwidth of the first step and having a bandwidth including a number of tones smaller than the number of tones included in the filter; The frequency band of the filter having a bandwidth including a smaller number of tones in the frequency band of the filter having a bandwidth including one or more tones by subtracting the energy detected in the first process and the energy detected in the second process. Detecting the energy of the tone included in the band excluding the third step; And a fourth process of determining demodulation data by comparing the energy detected in the second process with the magnitude of the energy detected in the third process. In at least two frequency modulated signal demodulators, First bandpass filter means for filtering one or more tones of a received signal; Second bandpass filter means for filtering again the following number of tones from one or more tones filtered by said first bandpass filter means; First calculation means for detecting energy of tones output from said first band pass filter means and said second band pass filter means; The energy of the tones detected by the first and second computing means is subtracted to reduce the energy of the tones included in the band except the frequency band filtered by the second band pass filter means from the frequency band within the first band pass filter means. First computing means for detecting energy; And determining means for comparing the magnitude of energy detected from said first computing means and said second computing means to determine data.