Method for generating soft decision signal from hard decision signal in a receiver system
View Patent ↗A receiver system, which generates a soft decision signal from a hard decision signal, includes a hard output receiver for determining a received bit to generate a hard decision signal. A hard input soft output receiver determines an estimated probability of symbol data corresponding to the received bit based on the hard decision signal and generates a soft decision signal represented by a log-likelihood ratio from the estimated probability.
1. A receiver system for generating a soft decision signal from a hard decision signal, the receiver system comprising:
a hard output receiver for determining a received bit to generate the hard decision signal; and
a hard input soft output receiver for determining an estimated probability of symbol data corresponding to the received bit based on the hard decision signal and for generating the soft decision signal represented by a log-likelihood ratio from the estimated probability,
wherein the hard input soft output receiver includes:
an adder, connected to the hard output receiver, for receiving the hard decision signal;
a bit counter connected to the adder, wherein the bit counter and the adder oversample the received bit using a predetermined sampling number to calculate a Hamming distance for the sampling sequence of the received bit; and
a probability-log likelihood ratio converter, connected to the bit counter, for determining the estimated probability of the symbol data based on the Hamming distance and the sampling number and generating the soft decision signal from the estimated probability.
2. The receiver system according to claim 1 , wherein the hard input soft output receiver determines a first estimated probability indicating the probability of the symbol data being positive and a second estimated probability indicating the probability of the symbol data being negative, and the hard input soft output receiver generates the soft decision signal by calculating the log-likelihood ratio of the difference between the first estimated probability and the second estimated probability.
3. The receiver system according to claim 1 , wherein the hard input soft output receiver generates the soft decision signal based on the equation of L c y k =2a tan h(D h /N s ), where L c y k represents the soft decision signal of the received bit, a tan h is a hyperbolic tangent, D h represents the Hamming distance, and N s represents the sampling number.
4. The receiver system according to claim 1 , wherein the hard input soft output receiver further generates the soft decision signal based on a noise signal which generates white noise.
5. The receiver system according to claim 1 , wherein the probability-log likelihood ratio converter includes:
a first decision unit for determining the sign of the Hamming distance to generate a sign decision signal;
a second decision unit for determining the absolute value of the Hamming distance to generate an absolute value decision signal;
a scaler, connected to the second decision unit, for scaling the absolute value decision signal in accordance with the sampling number to generate a scaled signal;
a log-likelihood ratio calculator, connected to the scaler, for calculating the absolute value of the soft decision signal based on the scaled signal; and
a multiplier for multiplying the absolute value of the soft decision signal by the sign decision signal to generate the soft decision signal.
6. The receiver system according to claim 5 , wherein the log-likelihood ratio calculator includes a pre-recorded table of a plurality of soft decision values calculated in accordance with the Hamming distance, the sampling number, and a scaling value of the scaler, the log-likelihood ratio calculator obtaining one of the plurality of soft decision values as the soft decision signal based on the scaled signal.
7. The receiver system according to claim 5 , wherein the log-likelihood ratio calculator functions as a linear approximator for obtaining the soft decision signal that corresponds to the scaled signal.
8. A method for generating a soft decision signal from a hard decision signal, the method comprising:
determining a received bit to generate the hard decision signal;
oversampling the received bit by a predetermined sampling number to calculate a Hamming distance for the sampling sequence of the received bit;
determining an estimated probability of symbol data corresponding to the received bit based on the hard decision signal; and
generating the soft decision signal represented by a log-likelihood ratio from the estimated probability of the symbol data, the Hamming distance and the sampling number, and wherein said generating the soft decision signal includes:
determining the sign of the Hamming distance to generate a sign decision signal;
determining the absolute value of the Hamming distance to generate an absolute value decision signal;
scaling the absolute value decision signal in accordance with the sampling number to generate a scaled signal;
calculating the absolute value of the soft decision signal based on the scaled signal; and
multiplying the absolute value of the soft decision signal by the sign decision signal to generate the soft decision signal.
9. The method according to claim 8 , wherein:
said determining the estimated probability includes:
determining a first estimated probability indicating the probability of the symbol data being positive; and
determining a second estimated probability indicating the probability of the symbol data being negative; and
said generating the soft decision signal includes:
calculating the log-likelihood ratio of the difference between the first estimated probability and the second estimated probability.
10. The method according to claim 8 , wherein said generating the soft decision signal includes generating the soft decision signal based on the equation of L c y k =2a tan h(D h /N s), where L c y k represents the soft decision signal of the received bit, a tan h is a hyperbolic tangent, D h represents the Hamming distance, and N s represents the sampling number.
11. The method according to claim 8 , wherein said generating the soft decision signal further includes generating the soft decision signal based on a noise signal which generates white noise.
12. The method according to claim 8 , wherein said calculating the absolute value of the soft decision signal includes obtaining the soft decision signal corresponding to the scaled signal from a table.
13. The method according to claim 8 , wherein said calculating the absolute value of the soft decision signal includes obtaining the soft decision signal corresponding to the scaled signal through linear approximation.