发布日期:2025-04-22 07:16 点击次数:62
又到了三部曲系列了,离上一次的三部曲有一段时辰了
图片
图片
图片
,这回的三部曲主要分为:swerling 0(Ⅴ)和swerling Ⅰ模子检测概率、swerling Ⅱ和swerling Ⅲ模子检测概率、swerling Ⅳ0(Ⅴ)检测概率和四种swerling模子检测概率的对比。话未几说,我们参预正题。由于这一系列的著述是上一篇著述的蔓延,是以会援用到上文的公式,提到的公式群众不错参考上一篇著述。检测概率相对于单个脉冲的SNR关系弧线以及MATLAB仿真
白夜念念凡,公众号:白夜尚好检测概率相对于单个脉冲的SNR关系弧线以及MATLAB仿真一、检测概率的计较
对于np>1的情况,Marcum界说的虚警概率为:
图片
np为累积脉冲数,nfa为Marcum虚警数。对于非报复主义,单个脉冲的检测概率由公式(2)给出(参考),当np>1时,使用Gram-Charlier级数计较检测概率。此时检测概率为:
图片
图片
其中,常数C3、C4和C6是Gram-Charlier级数的所有这个词,变量V为:
图片
其中VT是检测门限值,C3、C4、C6和
图片
的数值把柄主义报复类型变化。二、swerling V(0)型主义的检测
对于这类主义的报复,使用公式15计较检测概率。此时,Gram-Charlier级数的所有这个词为:
图片
例子:计较swerling Ⅴ型主义的检测概率:
主函数
clcclear allclose all%swerling 0型恶果图%败露np=1,10时检测概率相对于SNR的弧线pfa = 1e-9; %虚警率b = sqrt(-2.0 * log(pfa));index =0 ;for snr =0:0.1:20 index = index + 1; a = sqrt(2.0 * 10^(snr/10)); pro(index)=marcumsq(a,b); prob(index) = pd_swerling5(pfa,1,10,snr);endx=0:0.1:20;plot(x,pro,'k',x,prob,'k:');axis([0 20 0 1]);xlabel('SNR(dB)');ylabel('检测概率');legend('np=1','np=10');grid on title('检测概率相对于SNR的弧线,pfa=10^-9,非关联累积')marcumsq函数
function PD = marcumsq(a,b)%该方程期骗parls步履来计较检测概率max_test_value = 5000;if(a<b) alpha0 =1.0; dn = a/b;else alpha0 = 0; dn = b/a;endalpha_1 = 0.0;beta0 = 0.5;beta_1 = 0.0;D1 = dn;n=0;ratio = 2.0/(a*b);r1 = 0.0;beta = 0.0;alpha = 0.0;while beta<1000 n = n+1; alpha = dn + ratio * n*alpha0+alpha; beta = 1.0 + ratio * n*beta0 +beta; alpha_1 = alpha0; alpha0 = alpha; beta0 = beta; dn = dn * D1;endPD = (alpha0/(2.0*beta0)) * exp(-(a-b)^2/2.0);if(a>=b) PD = 1.0 - PD;endreturn
pd_swerling5函数
function pd = pd_swerling5(input1,indicator,np,snrbar)%该函数是用于计较swerling Ⅴ 玩忽 0 模子下np>1的主义的检测概率%input1 Pfa或nfa % Pfa:虚警概率,nfa:Marcum虚警数%indicator: 当input1 = Pfa时,为1,当input1 = nfa时,为2%np:脉冲累积数%snr:信噪比%pd: 检测概率if(np == 1) 'stop ,np must be greater than 1' return endformat longsnrbar = 10.0.^(snrbar./10);eps = 1e-8;delmax = 1e-5;delta = 1e4;%计较门限值Vt%接纳fehlner将虚警数界说的公式。if(indicator ~= 1) nfa = input1; pfa = np * log(2) / nfa;else pfa = input1; nfa = np * log(2) / pfa;endsqrtpfa = sqrt(-log10(pfa));sqrtnp = sqrt(np);vt0 = np - sqrtnp + 2.3*sqrtpfa * (sqrtpfa + sqrtnp -1.0);vt = vt0;while(abs(delta) >= vt0 ) igf = incomplete_gamma(vt0,np); num = 0.5^(np/nfa)-igf; temp = (np - 1)*log(vt0+eps)-vt0-factor(np - 1); deno = exp(temp); vt = vt0 + (num / (deno +eps)); delta = abs(vt - vt0)*10000.0; vt0 = vt;end%计较Gram-Charlier所有这个词temp1 = 2.0.*snrbar+1.0;omegabar = sqrt(np.*temp1);c3 = -(snrbar+1.0/3.0)./(sqrt(np).*temp1.^1.5);c4 = (snrbar + 0.25)./(np.*temp1.^2);c6 = c3 .* c3./2.0;V = (vt - np.*(1.0+snrbar))./omegabar;Vsqr = V.*V;val1 = exp(-Vsqr./2.0)./sqrt(2.0*pi);val2 = c3 .* (V.^2-1.0) + c4.*V.*(3.0-V.^2)-c6.*V.*(V.^4-10.*V.^2 + 15.0);q = 0.5 .* erfc(V./sqrt(2.0));pd = q - val1.*val2;incomplete_gamma函数
function [value] = incomplete_gamma(vt,np)%该函数主要计较不澈底Gamma函数%vtformat longeps = 1.000000001;%测试np=1的情况if(np == 1) value1 = vt * exp(-vt); value = 1.0-exp(-vt);endsumold = 1.0;sumnew = 1.0;calc1 = 1.0;calc2 = np;xx = np * log(vt + 0.0000000001) - vt - factor(calc2);temp1 = exp(xx);temp2 = np /(vt + 0.0000000001);diff = .0;radio = 1000.0;if(vt >= np) while(radio >= eps) diff = diff + 1.0; calc1 = calc1 *(calc2 - diff)/vt; sumnew = sumold + calc1; radio = sumnew / sumold; sumold = sumnew; end value = 1.0 - temp1*sumnew*temp2;else diff = .0; sumold =1.; radio = 1000; calc1 = 1; while(radio >= eps) diff = diff +1.0; calc1 =calc1*vt/(calc2+diff);sumnew = sumold+calc radio = sumnew / sumold; sumold =sumnew; end value = temp1 * sumnew;end
factor函数
function [val] = factor(n)%使用对数计较n的阶乘来幸免溢出format longn = n+9.0;n2 = n * n;temp = (n-1)*log(n)-n+log(sqrt(2.0 * pi*n))+((1.0-(1.0/30+(1.0/105)/n2)/n2)/12)/n;val = temp - log((n-1)*(n-2)*(n-3)*(n-4)*(n-5)*(n-6)*(n-7)*(n-8));图片
上图败露了np=1,10时检测概率相对于SNR的弧线,从图中不错看出,为获取雷同的检测概率,10个脉冲非关联累积需要更少的SNR。
三、 Swerling Ⅰ型主义的检测
五月桃色网Swerling Ⅰ型主义的检测概率准确公式:
图片
图片
例子1:主函数
clcclear allclose allpfa = 1e-9;nfa = log(2)/pfa;b = sqrt(-2.0 * log(pfa));index = 0;for snr = 0:0.1:22 index = index + 1; a = sqrt(2.0*10^(snr/10)); pro(index)=marcumsq(a,b); prob(index) = pd_swerling1(nfa,1,snr);endx=0:0.1:22;plot(x,pro,'k',x,prob,'k:');axis([2 22 0 1]);xlabel('SNR - dB');ylabel('检测概率');legend('swerling Ⅴ','swerling Ⅰ')grid on
图片
图中败露的是np=1和
图片
,对于Swerling Ⅰ和Ⅴ型报复,检测概率行动SNR函数的弧线。不错看出,为了获取与无报复情况雷同的检测概率,在有报复时,需要更高的SNR。例子2:主函数
clcclear allclose allpfa = 1e-11;nfa = log(2) / pfa;index = 0;for snr = -10:0.5:30 index = index + 1; prob1(index) = pd_swerling1(nfa,1,snr); prob10(index) = pd_swerling1(nfa,10,snr); prob50(index) = pd_swerling1(nfa,50,snr); prob100(index) = pd_swerling1(nfa,100,snr);endx = -10:0.5:30;plot(x,prob1,'k',x,prob10,'k:',x,prob50,'k--',x,prob100,'k-.');axis([-10 30 0 1]);xlabel('SNR-dB');ylabel('检测概率');legend('np=1','np=10','np=50','np=100');grid on;图中败露了np=1,10,50,100时,检测概率相对于SNR弧线,其中
图片
图片
pd_swerling1函数
function pd = pd_swerling1(nfa,np,snrbar)%计较主义报复模子为swerling Ⅰ型的检测概率format longsnrbar = 10.0^(snrbar/10);eps = 0.00000001;delmax = .00001;delta = 10000;%计较阈值Vtpfa = np*log(2)/nfa;sqrtpfa = sqrt(-log10(pfa));sqrtnp = sqrt(np);vt0=np - sqrtnp + 2.3 * sqrtpfa * (sqrtpfa + sqrtnp - 1.0);vt = vt0;while(abs(delta) >= vt0) igf = incomplete_gamma(vt0,np); num = 0.5^(np/nfa) - igf; temp = (np - 1)*log(vt0+eps)-vt0-factor(np-1); deno = exp(temp); vt = vt0+(num/(deno+eps)); delta = abs(vt - vt0)*10000; vt0 = vt;endif(np == 1) temp = -vt/(1.0+snrbar); pd = exp(temp); return;endtemp1 = 1.0 + np*snrbar;temp2 = 1.0/(np*snrbar);temp = temp2 + 1;vall = temp^(np-1);igf1 = incomplete_gamma(vt,np-1);igf2 = incomplete_gamma(vt/temp,np-1);pd = 1.0 - igf1 + vall*igf2 * exp(-vt/temp1);
四、追想
以上是对于swerling 0(Ⅴ)和swerling Ⅰ模子有关的检测概率和模子的一都实质,其中包括了有关模子的数学公式和matlab仿真。其中swerling 0(Ⅴ)模子先容中对照单脉冲和非相参累积,发现非相参累积只需要很小的SNR就能达到单脉冲的恶果。swerlingⅠ型中对比了swerling 0和swerlingⅠ型,有报复和无报复比拟,要达到雷同的检测概率,需要更大的SNR,另外还对比了不同np要求下,swerlingⅠ模子,检测概率和SNR之间的关系。
五、参考实质
雷达系统策画MATLAB仿真
如有侵权,请磋议删除:
大沢佑香作品
本站仅提供存储行状,整个实质均由用户发布,如发现存害或侵权实质,请点击举报。Powered by 萝莉 胜利女神:nikke @2013-2022 RSS地图 HTML地图
Copyright Powered by365建站 © 2013-2024