//VerilogA for modulator,neffmz,veriloga
//Xuanqi Chen, Zhifei Wang, Yi-Shing Chang, Jiang Xu, Jun Feng, Peng Yang, Zhehui Wang, Luan H. K. Duong, ”Modeling and Analysis of Optical Modulators Based on Free-Carrier Plasma Dispersion Effect,” IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD), vol. 39, no. 5, pp. 977-990, May 2020.

`include "constants.vams"
`include "disciplines.vams"

`define MAX_MODES 200
`define MAX_ITER 10000
`define IDX_NUM 10000
`define TOLERANCE 0.0000000001
`define MAX_TABLE_SIZE 300000

module neffmz(Ipow,Iphase,Ilam,Opow,Ophase,Olam,Vbias,Gnd);
    parameter real design_lam = 1.55e-6 from (0:inf);
    parameter real n_ps = 3.45 from (0:inf);
    parameter real n_sub = 1.45 from (0:inf);
    parameter real width_MMI = 3e-6 from (0:inf);
    parameter real width_port = 0.45e-6 from (0:inf);
    parameter real port_offset = 0.75e-6 from (0:inf);
    parameter real thickness = 0.22e-6 from (0:inf);
    parameter real alpha_mmi = 80 from (0:inf);
    parameter real l_ps = 0.1e-6 from (0:inf);
    parameter real l_arm = 40e-6 from (0:inf);
    parameter real alpha_phaseshift_ref = 80 from [0:inf);
    parameter real alpha_waveguide = 80 from [0:inf);
    parameter real alpha_absorp = 1 from (0:inf);
    parameter real alpha_wd = 1 from (0:inf);
    parameter real rib_width = 0.4e-6 from (0:inf);
    parameter real pn_offset = 0 from [0:inf);
    parameter real ni = 1e16 from (0:inf);
    parameter real N_A = 5e23 from (0:inf);
    parameter real N_D = 1e24 from (0:inf);
    parameter real Lp = 2e-6 from (0:inf);
    parameter real Ln = 1e-6 from (0:inf);
    parameter real epsr_Si = 11.7 from (0:inf);
    parameter real Is = 1e-14 from (0:inf);
    parameter real IBV = 1000u from (0:inf);
    parameter real BV = 40 from (0:inf);
    parameter real V0 = 1 from (0:inf);
    parameter real Cj0 = 1.5p from (0:inf);
    parameter real tau = 0.5n from (0:inf);
    parameter real Rs = 55 from (0:inf);
    parameter real emi = 1 from (0:inf);
    parameter real Vt = 0.0259 from (0:inf);
    parameter string neff_ps_filename = "../../../data/EIM/outputPN1.54-1.56-500.txt";

    input Ipow,Iphase,Ilam,Vbias;
    output Opow,Ophase,Olam;
    inout Gnd;
    electrical Ipow,Iphase,Ilam,Vbias,Opow,Ophase,Olam,Nint,Gnd;

    branch (Gnd,Nint) Iint;
    branch (Nint,Gnd) CCj,CCd,Idd,RRs;

    real ulam, ulam0, uwidth_in, uwidth_MMI, uoffset, uthickness, alpha0;
    real k1_ref, kc_ref;
    integer mode_num, mode_num_ref;
    real beta_x_MMI, beta_in, beta_x_in;
    real neff_MMI[0:`MAX_MODES];
    real beta_MMI[0:`MAX_MODES];
    real length_MMI;
    real ky_in, gamma_in;
    real ky_MMI[0:`MAX_MODES];
    real gamma_MMI[0:`MAX_MODES];
    real c0;
    real c0_dis[0:`MAX_MODES];
    real icoff[0:`MAX_MODES];
    real delta_beta;
    real ocoff_r[0:`MAX_MODES];
    real ocoff_i[0:`MAX_MODES];
    real c1r, c1i, c2r, c2i;
    real inr, ini, out1r, out1i, out2r, out2i;
    real sum_outfield1, sum_outfield2, sum_outfield3, sum_outfield4;
    real sum_singlemode1,sum_singlemode2,sum_singlemode3,sum_singlemode4;
    real border;
    integer i, j;
    real p0=1;
    real h;

    real beta_ps;
    real k_ps_r,k_ps_i;
    real p_ps_r,p_ps_i,p_wg_r,p_wg_i;
    real E_in1_r,E_in1_i,E_in2_r,E_in2_i;
    real E_out1_r,E_out1_i,E_out2_r,E_out2_i;
    real alpha_phaseshift;
    real W,Wdp,Wdn;
    real eps,phi_bi,phi;
    real N_avr;
    real gd,Id,Cj,Cd;
    real pn00,np00;
    real lam;
    real neff_phaseshift_ref,neff_phaseshift;

    real cp_tr11_r,cp_tr11_i,cp_tr12_r,cp_tr12_i,cp_tr21_r,cp_tr21_i,cp_tr22_r,cp_tr22_i;
    real ar_tr11_r,ar_tr11_i,ar_tr12_r,ar_tr12_i,ar_tr21_r,ar_tr21_i,ar_tr22_r,ar_tr22_i;
    real t1_11_r,t1_11_i,t1_12_r,t1_12_i,t1_21_r,t1_21_i,t1_22_r,t1_22_i;
    real t2_11_r,t2_11_i,t2_12_r,t2_12_i,t2_21_r,t2_21_i,t2_22_r,t2_22_i;

    //1550nm
    real sigma_ne=-8.8E-22;
    real sigma_nh=-8.5E-18;
    real sigma_ae=8.5E-18;
    real sigma_ah=6.0E-18;

    real xlength=0.6u;
    integer xnum=10001;
    real x[0:10000];
    real np[0:10000];
    real pp[0:10000];
    real pn[0:10000];
    real nn[0:10000];
    real xx[0:20001];
    real n[0:20001];
    real p[0:20001];
    real wi,wf;
    real dalpha_e[0:20001];
    real dalpha_h[0:20001];
    real dalpha[0:20001];
    real sum_dalpha;
    real dalpha_avr;

    integer ps_file;
    real width_start, width_end, width_step;
    real lam_start, lam_end, lam_step;
    integer width_num, lam_num;
    real tmp;
    real lam_old=0;
    real width_old=0;
    real neff_table_ps[0:`MAX_TABLE_SIZE] = {0};
    real neff_lamtable_ps[0:`MAX_TABLE_SIZE] = {0};
    real neff_widthtable_ps[0:`MAX_TABLE_SIZE] = {0};
    integer read_table_ps=1;

    analog function real findzeropoint_rect_x;
        input x_lower_input, x_upper_input, Vx;
        real x_lower_input, x_upper_input, Vx;
        real iter_count;
        real x_lower, x_upper;
        real y_mid;

        real y_lower, x_mid;

        begin
            findzeropoint_rect_x = -1;
            x_lower = x_lower_input;
            x_upper = x_upper_input;
            iter_count = 0;
            y_mid = 1;
            while (iter_count<`MAX_ITER && abs(y_mid) > `TOLERANCE) begin
                x_mid = (x_lower + x_upper) / 2;
                y_lower = tan(x_lower / 2) - sqrt(pow(Vx,2) - pow(x_lower,2)) / x_lower;
                y_mid = tan(x_mid / 2) - sqrt(pow(Vx,2) - pow(x_mid,2)) / x_mid;

                if (abs(y_mid) < `TOLERANCE) begin
                    findzeropoint_rect_x = x_mid;
                end

                if (y_mid * y_lower < 0)
                    x_upper = x_mid;
                else
                    x_lower = x_mid;

                iter_count = iter_count + 1;
            end
            if (findzeropoint_rect_x == -1)
                $display("No zero point found in the given range. Please try again with different range.");
        end
    endfunction

    analog function real findzeropoint_rect_y;
        input x_lower_input, x_upper_input, Vy, m;
        real x_lower_input, x_upper_input, Vy, m;
        real iter_count;
        real x_lower, x_upper;
        real y_mid;

        real y_lower, x_mid;

        begin
            x_lower = x_lower_input;
            x_upper = x_upper_input;
            iter_count = 0;
            y_mid = 1;
            findzeropoint_rect_y = -1;
            while (iter_count<`MAX_ITER && abs(y_mid) > `TOLERANCE) begin
                x_mid = (x_lower + x_upper) / 2;
                y_lower = tan(x_lower / 2 - m * `M_PI / 2) - sqrt((pow(Vy,2) - pow(x_lower,2))) / x_lower;
                y_mid = tan(x_mid / 2 - m * `M_PI / 2) - sqrt((pow(Vy,2) - pow(x_mid,2))) / x_mid;

                if (abs(y_mid) < `TOLERANCE)
                    findzeropoint_rect_y = x_mid;

                if (y_mid * y_lower < 0)
                    x_upper = x_mid;
                else
                    x_lower = x_mid;
                iter_count = iter_count + 1;
            end
            if (findzeropoint_rect_y == -1)
                $display("No zero point found in the given range. Please try again with different range.");
        end
    endfunction

    analog function real eim_rect;
        input lam, width, thickness, n_ps, n_sub, numflag, mode_index;
        real lam, width, thickness, n_ps, n_sub;
        integer numflag, mode_index;
        real k1, Vx, yx, hx, beta_x, neff_x_MMI, Vy, yy, hxy;
        real beta_xy[0:`MAX_MODES];
        real neff_xy[0:`MAX_MODES];
        real beta_xy_out[0:`MAX_MODES];
        real neff_xy_out[0:`MAX_MODES];
        integer i, m, mode_num;

        begin
            k1=2*`M_PI*n_ps/lam;
            Vx=2*`M_PI/lam*thickness*sqrt((pow(n_ps,2)-pow(n_sub,2)));
            yx=findzeropoint_rect_x(1e-6,min(`M_PI,Vx),Vx);
            hx=yx/thickness;
            beta_x=sqrt(pow(k1,2)-pow(hx,2));
            neff_x_MMI=lam*beta_x/(2*`M_PI);
            Vy=2*`M_PI/lam*width*sqrt((pow(neff_x_MMI,2)-pow(n_sub,2)));
            i=0;
            m=0;
            while (m*`M_PI<Vy) begin
                yy=findzeropoint_rect_y(m*`M_PI+1e-6,min((m+1)*`M_PI-1e-6,Vy),Vy,m);
                m=m+1;
                // if m*`M_PI>=Vy
                //     break
                hxy=yy/width;
                beta_xy[i]=sqrt(pow(beta_x,2)-pow(hxy,2));
                neff_xy[i]=lam*beta_xy[i]/(2*`M_PI);
                i=i+1;
            end
            mode_num=m-1;
            if (numflag==1 || numflag==2) begin
                for (i=0;i<mode_num;i=i+1) begin
                    beta_xy_out[i]=beta_xy[i];
                    neff_xy_out[i]=neff_xy[i];
                end
                if (numflag==1)
                    eim_rect=neff_xy_out[mode_index];
                else
                    eim_rect=beta_xy_out[mode_index];
            end
            else if (numflag==3) begin
                eim_rect=beta_x;
            end
            else if (numflag==4) begin
                eim_rect=mode_num;
            end
        end
    endfunction

    analog function real fieldfunc;
        input y,width,offset,ky,gamma,m,c0;
        real y,width,offset,ky,gamma,m,c0;
        real fai;

        begin
            fai = m * `M_PI / 2;
            if (((y - offset) <= (width / 2)) && ((-width / 2) <= (y - offset)))
                fieldfunc=c0 * cos(ky * (y - offset) - fai);
            else if ((y - offset) > width / 2)
                fieldfunc=c0 * cos(ky * width / 2 - fai) * exp(-gamma * ((y - offset) - width / 2));
            else
                fieldfunc=c0 * cos(ky * width / 2 + fai) * exp(gamma * ((y - offset) + width / 2));
        end
    endfunction

    analog function real normpower_fieldfunc_quad;
        input width,offset,ky,gamma,m,c0,width_MMI,border;
        real width,offset,ky,gamma,m,c0,width_MMI,border;
        real n,h;
        // real border=width_MMI/2;
        integer i;
        real sum;
        begin
            n=`IDX_NUM;
            h=border*2/n;
            sum=0.5*(pow(abs(fieldfunc(-1*border,width,offset,ky,gamma,m,c0)),2)+pow(abs(fieldfunc(border,width,offset,ky,gamma,m,c0)),2));
            for (i=1;i<n;i=i+1) begin
                sum=sum+pow(abs(fieldfunc(-1*border+i*h,width,offset,ky,gamma,m,c0)),2);
            end
            normpower_fieldfunc_quad=1/sqrt(h*sum);
        end
    endfunction

    analog function real infidx_modecoeffcient_quad;
        input width1,offset1,ky1,gamma1,m1,c01,width2,offset2,ky2,gamma2,m2,c02,border;
        real width1,offset1,ky1,gamma1,m1,c01,width2,offset2,ky2,gamma2,m2,c02,border;
        real n,h;
        // real border=20;
        integer i;
        real sum;
        begin
            n=`IDX_NUM;
            h=border*2/n;
            sum=0.5*(fieldfunc(-1*border,width1,offset1,ky1,gamma1,m1,c01)* fieldfunc(-1*border,width2,offset2,ky2,gamma2,m2,c02)+fieldfunc(border,width1,offset1,ky1,gamma1,m1,c01)* fieldfunc(border,width2,offset2,ky2,gamma2,m2,c02));
            for (i=1;i<n;i=i+1) begin
                sum=sum+fieldfunc(-1*border+i*h,width1,offset1,ky1,gamma1,m1,c01)* fieldfunc(-1*border+i*h,width2,offset2,ky2,gamma2,m2,c02);
            end
            infidx_modecoeffcient_quad=h*sum;
        end
    endfunction

    analog initial begin
        ulam0=design_lam*1e6;
        uwidth_in=width_port*1e6;
        uwidth_MMI=width_MMI*1e6;
        uoffset=port_offset*1e6;
        uthickness=thickness*1e6;
        alpha0=alpha_mmi*1e-6;
        k1_ref = 2 * `M_PI * n_ps / ulam0;
        kc_ref = 2 * `M_PI * n_sub / ulam0;

        mode_num_ref = eim_rect(ulam0, uwidth_MMI, uthickness, n_ps, n_sub, 4, 0);
        for (i=0;i<mode_num_ref;i=i+1) begin
            beta_MMI[i] = eim_rect(ulam0, uwidth_MMI, uthickness, n_ps, n_sub, 2, i);
        end
        length_MMI = p0*1.5*`M_PI/(beta_MMI[0]-beta_MMI[1]); //0.375Lpi?
        // $display("length_MMI=%f",length_MMI);
        border=20;

        pn00=pow(ni,2)/N_D;
        np00=pow(ni,2)/N_A;

        eps=epsr_Si*`P_EPS0;
        phi_bi=`P_K*$temperature/`P_Q*log(N_A*N_D/pow(ni,2));

        N_avr=N_A*N_D/(N_A+N_D);
    end

    analog begin
        ulam = V(Ilam) * 1e6;
        beta_in=eim_rect(ulam, uwidth_in, uthickness, n_ps, n_sub, 2, 0);
        // $display("beta_in=%f",beta_in);
        beta_x_in=eim_rect(ulam, uwidth_in, uthickness, n_ps, n_sub, 3, 0);
        // $display("beta_x_in=%f",beta_x_in);
        beta_x_MMI=eim_rect(ulam, uwidth_MMI, uthickness, n_ps, n_sub, 3, 0);
        // $display("beta_x_MMI=%f",beta_x_MMI);
        mode_num = eim_rect(ulam, uwidth_MMI, uthickness, n_ps, n_sub, 4, 0);
        // $display("mode_num=%d",mode_num);
        ky_in = sqrt(pow(beta_x_in,2) - pow(beta_in,2));
        // $display("ky_in=%f",ky_in);
        gamma_in = sqrt(pow(beta_in,2) - pow(kc_ref,2));
        // $display("gamma_in=%f",gamma_in);
        for (i=0;i<mode_num;i=i+1) begin
            neff_MMI[i] = eim_rect(ulam, uwidth_MMI, uthickness, n_ps, n_sub, 1, i);
            beta_MMI[i] = eim_rect(ulam, uwidth_MMI, uthickness, n_ps, n_sub, 2, i);
            ky_MMI[i] = sqrt(pow(beta_x_MMI,2) - pow(beta_MMI[i],2));
            gamma_MMI[i] = sqrt(pow(beta_MMI[i],2) - pow(kc_ref,2));
            // $display("neff_MMI[%d]=%f, beta_MMI[%d]=%f, ky_MMI[%d]=%f, gamma_MMI[%d]=%f",i,neff_MMI[i],i,beta_MMI[i],i,ky_MMI[i],i,gamma_MMI[i]);
        end

        c0=normpower_fieldfunc_quad(uwidth_in,0,ky_in,gamma_in,0,1,uwidth_MMI,border);
        // $display("c0=%f",c0);
        for (i=0;i<mode_num;i=i+1) begin
            c0_dis[i]=normpower_fieldfunc_quad(uwidth_MMI,0,ky_MMI[i],gamma_MMI[i],i,1,uwidth_MMI,border);
            // $display("c0_dis[%d]=%f",i,c0_dis[i]);
            icoff[i]=infidx_modecoeffcient_quad(uwidth_in,uoffset,ky_in,gamma_in,0,c0,uwidth_MMI,0,ky_MMI[i],gamma_MMI[i],i,c0_dis[i],border);
            // $display("icoff[%d]=%f",i,icoff[i]);
        end

        delta_beta = beta_MMI[0] - beta_MMI[1];
        // $display("delta_beta=%f",delta_beta);
        for (i=2;i<mode_num;i=i+1) begin
            beta_MMI[i]=beta_MMI[0] - i* (i + 2) * delta_beta / 3;
            // $display("beta_MMI[%d]=%f",i,beta_MMI[i]);
        end
        for (i=0;i<mode_num;i=i+1) begin
            ocoff_r[i]=icoff[i]*cos(-beta_MMI[i]*length_MMI);
            ocoff_i[i]=icoff[i]*sin(-beta_MMI[i]*length_MMI)*-1.0;
            // $display("ocoff_r[%d]=%f, ocoff_i[%d]=%f",i,ocoff_r[i],i,ocoff_i[i]);
        end

        h=border*2.0/`IDX_NUM;
        sum_outfield1=0;
        sum_outfield2=0;
        sum_outfield3=0;
        sum_outfield4=0;
        for (j=0;j<mode_num;j=j+1) begin
            sum_singlemode1=0.5*(ocoff_r[j]*fieldfunc(-1*border,uwidth_MMI,0,ky_MMI[j],gamma_MMI[j],j,c0_dis[j])*fieldfunc(-1*border,uwidth_in,uoffset,ky_in,gamma_in,0,c0)+ocoff_r[j]*fieldfunc(border,uwidth_MMI,0,ky_MMI[j],gamma_MMI[j],j,c0_dis[j])*fieldfunc(border,uwidth_in,uoffset,ky_in,gamma_in,0,c0));
            sum_singlemode2=0.5*(ocoff_i[j]*fieldfunc(-1*border,uwidth_MMI,0,ky_MMI[j],gamma_MMI[j],j,c0_dis[j])*fieldfunc(-1*border,uwidth_in,uoffset,ky_in,gamma_in,0,c0)+ocoff_i[j]*fieldfunc(border,uwidth_MMI,0,ky_MMI[j],gamma_MMI[j],j,c0_dis[j])*fieldfunc(border,uwidth_in,uoffset,ky_in,gamma_in,0,c0));
            sum_singlemode3=0.5*(ocoff_r[j]*fieldfunc(-1*border,uwidth_MMI,0,ky_MMI[j],gamma_MMI[j],j,c0_dis[j])*fieldfunc(-1*border,uwidth_in,-uoffset,ky_in,gamma_in,0,c0)+ocoff_r[j]*fieldfunc(border,uwidth_MMI,0,ky_MMI[j],gamma_MMI[j],j,c0_dis[j])*fieldfunc(border,uwidth_in,-uoffset,ky_in,gamma_in,0,c0));
            sum_singlemode4=0.5*(ocoff_i[j]*fieldfunc(-1*border,uwidth_MMI,0,ky_MMI[j],gamma_MMI[j],j,c0_dis[j])*fieldfunc(-1*border,uwidth_in,-uoffset,ky_in,gamma_in,0,c0)+ocoff_i[j]*fieldfunc(border,uwidth_MMI,0,ky_MMI[j],gamma_MMI[j],j,c0_dis[j])*fieldfunc(border,uwidth_in,-uoffset,ky_in,gamma_in,0,c0));
            for (i=1;i<`IDX_NUM;i=i+1) begin
                sum_singlemode1=sum_singlemode1+ocoff_r[j]*fieldfunc(-1*border+i*h,uwidth_MMI,0,ky_MMI[j],gamma_MMI[j],j,c0_dis[j])*fieldfunc(-1*border+i*h,uwidth_in,uoffset,ky_in,gamma_in,0,c0);
                sum_singlemode2=sum_singlemode2+ocoff_i[j]*fieldfunc(-1*border+i*h,uwidth_MMI,0,ky_MMI[j],gamma_MMI[j],j,c0_dis[j])*fieldfunc(-1*border+i*h,uwidth_in,uoffset,ky_in,gamma_in,0,c0);
                sum_singlemode3=sum_singlemode3+ocoff_r[j]*fieldfunc(-1*border+i*h,uwidth_MMI,0,ky_MMI[j],gamma_MMI[j],j,c0_dis[j])*fieldfunc(-1*border+i*h,uwidth_in,-uoffset,ky_in,gamma_in,0,c0);
                sum_singlemode4=sum_singlemode4+ocoff_i[j]*fieldfunc(-1*border+i*h,uwidth_MMI,0,ky_MMI[j],gamma_MMI[j],j,c0_dis[j])*fieldfunc(-1*border+i*h,uwidth_in,-uoffset,ky_in,gamma_in,0,c0);
            end
            sum_outfield1=sum_outfield1+h*sum_singlemode1;
            sum_outfield2=sum_outfield2+h*sum_singlemode2;
            sum_outfield3=sum_outfield3+h*sum_singlemode3;
            sum_outfield4=sum_outfield4+h*sum_singlemode4;
        end
        c1r=sum_outfield1*exp(-alpha0*length_MMI);
        c1i=sum_outfield2*exp(-alpha0*length_MMI);
        c2r=sum_outfield3*exp(-alpha0*length_MMI);
        c2i=sum_outfield4*exp(-alpha0*length_MMI);

        // $display("c1r=%f, c1i=%f, c2r=%f, c2i=%f",c1r,c1i,c2r,c2i);

        lam=V(Ilam);
        gd=(-Is*(exp(5)-1)+IBV)/(-5*emi*Vt + BV);
        Id=Is*(exp(V(Nint)/(emi*Vt))-1)+V(Nint)*gd;
        if (V(Nint) >= V0) 
            Cj=0;
        else
            Cj=Cj0/sqrt(1-V(Nint)/V0);
        //Cd=Id*tau/(emi*Vt);
        Cd=Is*(exp(V(Nint)/(emi*Vt))-1)*tau/(emi*Vt);

        I(Iint) <+ V(Vbias)/Rs;
        I(CCj) <+ Cj*ddt(V(CCj));
        I(CCd) <+ Cd*ddt(V(CCd));
        I(Idd) <+ Id;
        V(RRs) <+ I(RRs)*Rs;

        phi=phi_bi-V(Nint);
        // $display("V(Nint)=%e",V(Nint));
        if (phi<0) 
            phi=0;

        // $display("phi=%e",phi);
        W=alpha_wd * sqrt(2.0*eps*phi/(`P_Q*N_avr));
        Wdp=N_D * W/(N_A+N_D);
        Wdn=N_A * W/(N_A+N_D);

        for (i=0; i<xnum; i=i+1) begin
            x[i]=i*xlength/(xnum-1);
            np[i]=np00*(exp(`P_Q*V(Nint)/(`P_K*$temperature))-1)*exp(-(x[i]-Wdp)/Ln)+np00;
            if (x[i]>Wdp) begin
                pp[i]=N_A;
            end
            else begin
                pp[i]=0;
            end
            pn[i]=pn00*(exp(`P_Q*V(Nint)/(`P_K*$temperature))-1)*exp(-(x[i]-Wdn)/Lp)+pn00;
            if (x[i]>Wdn) begin
                nn[i]=N_D;
            end
            else begin
                nn[i]=0;
            end
            if (np[i]<0) np[i]=0;
            if (pn[i]<0) pn[i]=0;
        end

        for (i=0; i<2*xnum; i=i+1) begin
            if (i<xnum) begin
                xx[i]=-x[xnum-1-i];
                n[i]=np[xnum-1-i];
                p[i]=pp[xnum-1-i];
            end
            else begin
                xx[i]=x[i-xnum];
                n[i]=nn[i-xnum];
                p[i]=pn[i-xnum];
            end
            if (abs(xx[i]+rib_width/2) < 0.001u) begin
                wi=i;
            end
            if (abs(xx[i]-rib_width/2) < 0.001u) begin
                wf=i;
            end
            dalpha_e[i]=n[i]*sigma_ae*1e-4;
            dalpha_h[i]=p[i]*sigma_ah*1e-4;
            dalpha[i]=dalpha_e[i] + dalpha_h[i];
        end

        sum_dalpha=0;
        for (i=wi; i<=wf; i=i+1) begin
            sum_dalpha=sum_dalpha+dalpha[i];
        end
        dalpha_avr=sum_dalpha/(wf-wi+1.0);
        alpha_phaseshift = alpha_phaseshift_ref + dalpha_avr;

        if (read_table_ps==0 || lam_old==0 || width_old==0) begin
            read_table_ps=1;
        end

        if (read_table_ps == 1) begin
            ps_file=$fopen(neff_ps_filename,"r");
            if (ps_file==0) begin
                $display("Error: cannot open ps_file %s",neff_ps_filename);
                $finish;
            end
            $fscanf(ps_file,"%f",width_start);
            $fscanf(ps_file,"%f",width_end);
            $fscanf(ps_file,"%d",width_num);
            width_step = (width_end-width_start)/(width_num-1);
            // $display("width_start=%f, width_end=%f, width_num=%f, width_step=%f",width_start,width_end,width_num,width_step);
            $fscanf(ps_file,"%f",lam_start);
            $fscanf(ps_file,"%f",lam_end);
            $fscanf(ps_file,"%d",lam_num);
            lam_step = (lam_end-lam_start)/(lam_num-1);
            // for (i=0; i<width_num; i=i+1) begin
            //         $fscanf(ps_file,"%f",tmp);
            // end
            for (i=0; i<width_num; i=i+1) begin
                for (j=0; j<lam_num; j=j+1) begin
                    $fscanf(ps_file,"%f",tmp);
                    // $display("%f",tmp);
                    neff_widthtable_ps[i*lam_num+j] = width_start+i*width_step;
                    neff_lamtable_ps[i*lam_num+j] = lam_start+j*lam_step;
                    neff_table_ps[i*lam_num+j] = tmp;
                    //$display("neff_widthtable_ps[%d]=%f, neff_lamtable_ps[%d]=%f, neff_table_ps[%d]=%f",i*lam_num+j,neff_widthtable_ps[i*lam_num+j],i*lam_num+j,neff_lamtable_ps[i*lam_num+j],i*lam_num+j,neff_table_ps[i*lam_num+j]);
                end
            end
            $fclose(ps_file);
            neff_widthtable_ps[width_num*lam_num+1] = 0;
            neff_lamtable_ps[width_num*lam_num+1] = 0;
            neff_table_ps[width_num*lam_num+1] = 0;
            neff_widthtable_ps[width_num*lam_num+2] = 0;
            neff_lamtable_ps[width_num*lam_num+2] = -1;
            neff_table_ps[width_num*lam_num+2] = 0;
            read_table_ps=2;
        end

        if (read_table_ps == 2) begin
            if (lam*1e6<lam_start || lam*1e6>lam_end) begin
                $display("Warning: lam=%f is out of range [%f,%f]",lam*1e6,lam_start,lam_end);
            end
            if (Wdp*1e6<width_start || Wdp*1e6>width_end) begin
                $display("Warning: Wdp=%f is out of range [%f,%f]",Wdp*1e6,width_start,width_end);
            end
            neff_phaseshift = $table_model(Wdp*1e6, lam*1e6, neff_widthtable_ps, neff_lamtable_ps, neff_table_ps);
            // $display("neff=%f",neff_phaseshift);
        end

        beta_ps = 2.0*`M_PI*neff_phaseshift/lam;

        k_ps_r=-alpha_phaseshift*alpha_absorp;
        k_ps_i=beta_ps;

        p_ps_r=exp(k_ps_r*l_ps)*cos(k_ps_i*l_ps);
        p_ps_i=exp(k_ps_r*l_ps)*sin(k_ps_i*l_ps);

        p_wg_r=exp((k_ps_r+alpha_waveguide-alpha_phaseshift_ref)*(l_ps+l_arm))*cos((k_ps_i)*(l_ps+l_arm));
        p_wg_i=exp((k_ps_r+alpha_waveguide-alpha_phaseshift_ref)*(l_ps+l_arm))*sin((k_ps_i)*(l_ps+l_arm));
        // $display("p_ps_r=%g, p_ps_i=%g, p_wg_r=%g, p_wg_i=%g",p_ps_r,p_ps_i,p_wg_r,p_wg_i);
        ar_tr11_r=p_ps_r;
        ar_tr11_i=p_ps_i;
        ar_tr12_r=0;
        ar_tr12_i=0;
        ar_tr21_r=0;
        ar_tr21_i=0;
        ar_tr22_r=p_wg_r;
        ar_tr22_i=p_wg_i;
        // $display("p_ps_r=%f, p_ps_i=%f, p_wg_r=%f, p_wg_i=%f",p_ps_r,p_ps_i,p_wg_r,p_wg_i);

        t1_11_r=c1r*ar_tr11_r-c1i*ar_tr11_i+c2r*ar_tr21_r-c2i*ar_tr21_i;
        t1_11_i=c1r*ar_tr11_i+c1i*ar_tr11_r+c2r*ar_tr21_i+c2i*ar_tr21_r;
        t1_12_r=c1r*ar_tr12_r-c1i*ar_tr12_i+c2r*ar_tr22_r-c2i*ar_tr22_i;
        t1_12_i=c1r*ar_tr12_i+c1i*ar_tr12_r+c2r*ar_tr22_i+c2i*ar_tr22_r;
        t1_21_r=c2r*ar_tr11_r-c2i*ar_tr11_i-c1r*ar_tr21_r+c1i*ar_tr21_i;
        t1_21_i=c2r*ar_tr11_i+c2i*ar_tr11_r-c1r*ar_tr21_i-c1i*ar_tr21_r;
        t1_22_r=c2r*ar_tr12_r-c2i*ar_tr12_i-c1r*ar_tr22_r+c1i*ar_tr22_i;
        t1_22_i=c2r*ar_tr12_i+c2i*ar_tr12_r-c1r*ar_tr22_i-c1i*ar_tr22_r;

        t2_11_r=t1_11_r*c1r-t1_11_i*c1i+t1_12_r*c2r-t1_12_i*c2i;
        t2_11_i=t1_11_r*c1i+t1_11_i*c1r+t1_12_r*c2i+t1_12_i*c2r;
        t2_12_r=t1_11_r*c2r-t1_11_i*c2i+t1_12_r*c1r-t1_12_i*c1i;
        t2_12_i=t1_11_r*c2i+t1_11_i*c2r+t1_12_r*c1i+t1_12_i*c1r;
        t2_21_r=t1_21_r*c1r-t1_21_i*c1i+t1_22_r*c2r-t1_22_i*c2i;
        t2_21_i=t1_21_r*c1i+t1_21_i*c1r+t1_22_r*c2i+t1_22_i*c2r;
        t2_22_r=t1_21_r*c2r-t1_21_i*c2i+t1_22_r*c1r-t1_22_i*c1i;
        t2_22_i=t1_21_r*c2i+t1_21_i*c2r+t1_22_r*c1i+t1_22_i*c1r;
        // $display("t2_11_r=%f, t2_11_i=%f, t2_12_r=%f, t2_12_i=%f, t2_21_r=%f, t2_21_i=%f, t2_22_r=%f, t2_22_i=%f",t2_11_r,t2_11_i,t2_12_r,t2_12_i,t2_21_r,t2_21_i,t2_22_r,t2_22_i);

        inr=sqrt(V(Ipow))*cos(V(Iphase)/360.0*2*`M_PI);
        ini=sqrt(V(Ipow))*sin(V(Iphase)/360.0*2*`M_PI);
        out1r=t2_11_r*inr-t2_11_i*ini;
        out1i=t2_11_r*ini+t2_11_i*inr;
        out2r=t2_21_r*inr-t2_21_i*ini;
        out2i=t2_21_r*ini+t2_21_i*inr;

        V(Olam) <+ lam;
        V(Opow) <+ pow(out1r,2)+pow(out1i,2);
        V(Ophase) <+ atan2(out1i,out1r)*360.0/(2*`M_PI);
    end
endmodule