Algorithm for Model

Phase 1 : Estimation of path loss exponents-

1. Use terrain info from cells A, B, C, and D to obtain K    values for each of the cells A, B, C, and D.

2. Loop through each of cells A, B, C and D’s received power measurement, each time around the loop checking if the current square lies within conical azimuth range for the cell being used. Simultaneously form two separate pairs of summations based on the azimuth check, one pair for areas lying within range, and one pair for those lying outside range.

3. From the two summations formed, find two path-loss exponents for each cell using equation 5. Here each summation pair refers to the numerator and the denominator summation in equation 5.

 

 

 

 

         Equation 5. Estimation of N using least squares          method (for K = Ka for cell A)

         This equation is obtained by minimizing the variance of          the predicted received power versus the measure          received power. Consult me if you need a mathematical          proof.

4. Average each of the two path-loss exponent across the 4 cells to obtain ready to use path-loss exponents
for prediction of power in any cell.

Phase 2 : Generating predicted received power maps for                     a given cell-

1.  Use terrain values to find K for the cell.

2. Loop through a matrix and fill up values received power values using equation 2 for areas in azimuth range using the path-loss exponent found for areas in azimuth range.

3. Do the same for areas outside the azimuth, except use the other path-loss exponent.