Circuits and methods for frequency offset estimation in FSK communications

#In frequency shift key (FSK) communication systems, the transmitting frequency and the receiving frequency have to be the same for optimal performance. Due to anomalies, such as the crystal differences between a transmitter and a receiver, the frequencies may be different, which results in communication errors.

#A frequency shift key (FSK) receiver includes a phase compensator for compensating for frequency differences in a transmitter that transmits a FSK signal to the receiver. The phase compensating is performed before the FSK signal is demodulated by the receiver.

#FIG. 1A is a graph showing an example of data that is to be transmitted by FSK modulation.

#FIG. 1B is a graph showing an example of a carrier signal for transmitting the data of FIG. 1A.

#FIG. 1C is a graph showing the data of FIG. 1A after FSK modulation.

#Communication systems require a transmitter node and a receiver node operating on the the same frequencies. When the frequencies are offset from one another, errors in the communications will occur. In some examples, crystals in the transmitter and receiver nodes are slightly different, which results in frequency offsets. Frequency compensation is applied in the receiver in many situations to compensate for the frequency offsets. For example, components within the receiver estimate the frequency offset and compensate for the frequency offset accordingly.

#One communication system where frequency compensation is applied is in frequency shift key (FSK) signaling or modulation. FIGS. 1A-1C illustrate an example of FSK modulation. FIG. 1A is a digital signal 100 that is to be transmitted by FSK communications. The digital signal 100 consists of logic one bits 102 and logic zero bits 104. FIG. 1B is a carrier signal 110 for the digital signal 100. In the examples described herein, the carrier signal 110 is a sine wave having a frequency that meets the Nyquist requirements for the digital signal 100. FIG. 1C is a FSK signal 120 of the digital signal 100. The FSK signal 120 has an increased or positive frequency relative to the carrier signal 110 representing the zero bits 104 and a decreased or negative frequency relative to the carrier signal 110 representing the one bits 102. In some alternative examples, the mapping is reversed so that a zero bit corresponds to a negative frequency and a one bit corresponds to a positive frequency.

#During FSK communications, a preamble is transmitted to the receiver. The preamble contains repetitive “01010101” bit patterns that can be used to perform frequency offset estimations. The bits are sometimes referred to as symbols in FSK communications. In the following example, the input sample rate is four times that of the FSK symbol rate. More specifically, the symbols or bits are sampled at a rate that is four times faster than the symbol or bit rate, which is referred to as a four times oversampling. Due to the four times oversampling and because the input data can start from anywhere inside a preamble repetition, the receiver needs to search eight consecutive locations in the preamble to find the best sample phase for a zero bit. In order to distinguish between a zero bit and a one bit, there are eight possible sampling phases (eight samples on two bits). It is noted that due to pulse-shaping in the transmitter, the phase difference between two samples may not be the same even for the 4 samples in a single bit. The best sample phase is detected by the largest phase difference between samples. In some examples, the average of a certain number of preamble bits is averaged to find the largest phase difference for a zero bit or a one bit.