TECHNICAL FIELD
The present invention relates to a wireless communication receiver device, and particularly relates to a binary frequency-shift keying demodulator.
BACKGOUND OF THE INVENTION
Among wireless communication devices, the demodulator is often seen as one of the most important component in receiver end. Demodulators are utilized at the rear end of the receiver to demodulate the signal modulated from the front end, matching the initial information signal. Bit Error-Rate (BER) is the important key to rate such demodulators. Current Binary Frequency-Shift Keying (BFSK) signal demodulation methods can be categorized as coherent demodulation and incoherent demodulation, where the incoherent demodulation has lower resistance to noise. Among several coherent demodulation implementations, differential demodulator is easily structured and it also provides lower Bit Error-Rate. In addition, differential demodulator does not require local carrier wave, and it also demands lower precision of the resonator while it has lower phase error caused by the carrier signal. It is one of the most common demodulation methods, as it is referred in FIG. 1. However, the disadvantages of the differential demodulator are the larger circuit board design, which causes power consumption, and a necessary addition of an external phase-shifting circuit. Such large board and external circuit design is more affected by fabrication processes. Therefore under the circumstances when the information signal is similar to carrier signal, the demand on the filter is higher, such that a differential demodulator is not appropriate anymore.
SUMMARY OF THE INVENTION
The present invention is to provide a BFSK demodulator that is without external support and has a simpler but more condensed circuit structure design. The present invention provides a BFSK demodulator comprising a frequency-to-voltage converter, a differentiator circuit and a sampling selector circuit, wherein a BFSK information signal passes through the frequency-to-voltage converter and becomes a voltage input into a differentiator circuit. The sampling selector circuit receives an output produced by the differentiator circuit and reproduces a demodulated signal after filtering possible noises.
The present invention does not require any external support elements. The present invention has a simpler, smaller circuit board design, and has a lower power consumption rate.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a block diagram of a prior art Binary Frequency-Shift Keying Demodulator
FIG. 2 is a block diagram of a Binary Frequency-Shift Keying Demodulator in present invention
FIG. 3 is a circuit illustration of a frequency-to-voltage converter in present invention
FIG. 4 is a circuit illustration of a differentiator circuit in present invention
FIG. 5 is a circuit illustration of a sampling selector circuit in present invention
DETAILED DESCRIPTION OF THE INVENTION
As is seen in FIG. 2, the present invention provides a Binary Frequency-Shift Keying Demodulator comprising a frequency-to-voltage converter, a differentiator circuit, and a sampling selector circuit. A BFSK signal is input to a frequency-voltage converter. A square wave output signal is produced by the frequency-voltage converter as shown on step A. Frequencies fc+Δf and fc−Δf (fc is carrier wave frequency) match voltage V1 and V2, respectively (V1<V2). Since the voltage difference at step A may be too narrow to active a logic circuit to transform this demodulated signal at step A into digital, the signal at step A is differentiated by a differentiator circuit, having an output signal at step B. The signal at step B is filtered by a sampling selector circuit to produce a voltage signal output C. The signal at step C is the demodulated signal.
The output voltage wave from the frequency-to-voltage converter has various spikes from the discreteness caused by charge injection. When the wave passes through a differentiator, the spikes would be exposed along with rising and falling edges. Since the ΔV signal at BFSK is considerably high compared with the narrow amplitude at the spikes, the output from the differentiator is a series of pulse signals with small influence from the spikes. Through proper voltage limitation in sampling selector circuit, demodulated digital signals can be retrieved by filtering the pulse signals.
FIG. 3 illustrates a frequency-to-voltage converter adopted by the present Binary Frequency-Shift Keying Demodulator. When the input signal Fin is LOW, transistors Mp1 Mp7 are ON and transistor Mn2 is OFF, while signals φ1 φ2 are both LOW. Capacitor C1 is being charged by Iin and Capacitor C3 is being charged by Ic. When the voltage on capacitor C3 is lower than Vref, D is HIGH turning transistor Mn6 ON, voltage on capacitor C1 is zero. When the voltage on capacitor C3 is higher than Vref, D is LOW, turning transistor Mn6 OFF, voltage on capacitor C1 is therefore rising. When the input signal Fin is HIGH, transistors Mp1 Mp7 are both OFF and transistor Mn2 is ON, φ2 turns HIGH first while φ1 stays LOW, charges being rearranged on capacitors C1 and C2. Then φ2 turns LOW, φ1 turns HIGH, capacitors C1 C3 start discharging until the voltage reach zero, and therefore φ1 φ2 both turn back to LOW, until next signal period. The voltage on capacitor C2 is referring to the voltage of the information frequency signal. It is well known that with a smaller capacitor C2, more frequent capacitor C1 charges, the voltage on capacitor C2 is closer to the initial voltage on capacitor C1. The purpose of adding a charging time control circuit here is to reduce the BER while operating at a better signal to noise ratios by enlarging the BFSK differential voltage Δ V under a limited source voltage. Transistor Mp5 is implemented to reduce the charge injection effect caused from the ON/OFF actions of transistor Mn4.
FIG. 4 illustrates the differentiator circuit adopted in the present invention. The differentiator comprises a voltage-to-current converter, a current mode differentiator, and a current-to-voltage converter.
The present invention adopts the current mode differentiator because a simple structured differentiator does not employ a traditional feedback circuit, and it not only has a lower power consumption rate, but also has a broader bandwidth.
The transmitting function of the differentiator circuit is as follows.
gin is the output admittance from the last level.
Since the output from the frequency-to-voltage converter is a voltage signal, a converting process conducted by a voltage-to-current converter is necessary to produce a current signal input to the current mode differentiator, as it is designed at the first level in the present differentiator circuit. It is therefore needed a current-to-voltage converter to convert the current output from the differentiator into a voltage signal.
FIG. 5. illustrates a sampling selector circuit. The output from the differentiator is a pulse signal, as shown on FIG. 2, and the output signal is passed into first and second comparator B1 and B2 respectively. The reference voltage of comparator B1 is Vref1. The reference voltage of comparator is Vref2. When the portion of pulse signal Fin is higher than Vref1, the output at point E produces a square wave signal from the comparator B1; when the portion of pulse signal Fin is lower than Vref2, the output at point F also produces a square wave signal from the comparator B2. Combining signals from point E and F produces a clock signal at point G. A D flip-flop therefore produces a demodulated signal output by sampling the output at point E and the clock signal at point G.
Properly selecting reference voltage Vref1 and Vref2 can filter out numerous high frequency noises, especially the discreteness caused by the charge injection effect from the frequency-to-voltage converter.