Soft-decision metric generation for higher order modulation

1. A method of reducing calculations in the decoding of a M-ary modulated convolutionally coded signal in a radio communication system, the method comprising the steps of:

a) determining a single function for a soft-decision metric for each bit in a symbol by restricting a set of all possible Gray-coded constellation points to those closest to a boundary between a bit value of 0 and 1 for each bit in an input symbol and applying a predetermined function corresponding to a range of restricted constellation points to an entire possible range of symbols;

b) inputting the symbol;

c) setting the soft-decision metric for each bit in the symbol using the predetermined function from the determining step;

d) outputting the soft-decision metric for each bit of the symbol to a turbo decoder;

e) decoding the symbol in the turbo decoder; and

f) repeating steps a) through e) until all symbols to be input are decoded.

2. The method of claim 1 , wherein the setting step includes a substep of scaling the soft-decision metric.

3. The method of claim 2 , wherein the scaling substep includes scaling the soft-decision metric by a factor of βA d /A p , wherein β is the squared magnitude of a filtered pilot signal, and A d and A p are the data and pilot signal gains, respectively.

4. The method of claim 1 , wherein the predetermined function of the determining step is defined by the difference between squares of the distances between the set of all possible Gray-coded restricted constellation points having 0 and 1 bit values and a hypothetical symbol failing within the range of restricted constellation points.

5. A method of reducing calculations in the decoding of an 8-PSK modulated convolutionally coded signal in a radio communication system, the method comprising the steps of:

a) providing a set of eight possible Gray-coded symbols for the 8-PSK modulated convolutionally coded signal in a constellation;

b) defining radial boundaries in the constellation;

c) inputting a symbol;

d) plotting the location of the symbol in the constellation;

e) locating two nearest constellation points to the symbol having a 0 value and a 1 value for each bit;

f) setting a soft-decision metric for each bit using the two nearest constellation points from the locating step;

g) outputting the soft-decision metric for each bit of the symbol to a turbo decoder;

h) decoding the symbol in the turbo decoder;

i) repeating steps a) through h) until all symbols to be input are decoded;

wherein the setting step further comprises the step of scaling the soft-decision metric, and

wherein the scaling step includes scaling the soft-decision metric by a factor of βA d /A p , wherein β is a squared magnitude of a filtered pilot signal, and A d and A p , are data and pilot signal gains, respectively.

6. The method of claim 5 , wherein the inputting step includes phase shifting the symbol by e jπ/8 .

7. A method of reducing calculations in the decoding of an 8-PSK modulated convolutionally coded signal in a radio communication system, the method comprising the steps of:

a) inputting a symbol having real part, x, and an imaginary part, y;

b) setting a soft-decision metric of a first bit of the symbol equal to the value of the imaginary part, y, of the symbol;

c) setting a soft-decision metric of a second bit of the symbol equal to the value of the real part, x, of the symbol;

d) setting a soft-decision metric of a third bit of the symbol equal to

(



x



-



y



)

⁢

1

2

e) outputting the soft-decision metrics of the first, second and third bits of the symbol to a turbo decoder;

f) decoding the symbol in the turbo decoder, and

g) repeating steps a) through t) until all symbols to be input are decoded.

8. The method of claim 7 , wherein the inputting step includes phase shifting the symbol by e jπ/8 .