API tools faq
paste
Advertisement
Guest User

Untitled

a guest
Oct 14th, 2019
90
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 15.72 KB | None | 0 0
raw download clone embed print report
  1. clear all;
  2. close all;
  3. clc;
  4.  
  5.  
  6. % Regular Gradient descent Synapses
  7. R = 1000*[0 30 30 30; 0 0 30 30; 0 0 0 30; 0 0 0 0];
  8.  
  9. R12 = 30000;
  10. R13 = 30000;
  11. R14 = 30000;
  12. R23 = 30000;
  13. R24 = 30000;
  14. R34 = 30000;
  15.  
  16.  
  17.  
  18. % Define constants
  19. tau = 400e-12; % RC constant
  20. fs=100e3;
  21. fi = 1/(160e-6);
  22. cycles=512;
  23. N=fs/fi*cycles;
  24. ts = (0:1/fs:(50e-3)-1/fs);
  25. Ns= 5e3;
  26.  
  27. % CLK
  28. tr=1/fs*(1:N);
  29. tf=tr+1/(2*fs);
  30. dt=1/(2*fs);
  31.  
  32.  
  33. prompt = 'Enter non_ideal parameter value ';
  34. non_ideal = input(prompt);
  35.  
  36.  
  37. %%%%%%%%%%%%%%%%
  38. % Noise sources
  39. %%%%%%%%%%%%%%%%
  40. % Comparator mismatch
  41. kVth=5e-3;
  42. area_tran=100;
  43. sVth=kVth/sqrt(area_tran);
  44.  
  45. % There are two transistors in a differential comparator
  46. sVth=sqrt(2*sVth^2);
  47. offset=random('norm',0,sVth,1,length(ts));
  48.  
  49. % Comparator input noise
  50. sNoise=(10e-9)*sqrt(6.25e3);
  51. %noise=random('norm',0,sNoise,1,length(ts));
  52.  
  53. %Thermal noise
  54. thNoise=10e-16;
  55.  
  56. % Resistance mismatch
  57. sR = 0.05/sqrt(32);
  58.  
  59. frec=1;
  60. % Jitter
  61. sJit=20e-3; %frequency dependent
  62. vJit=random('norm',0,sJit,1,length(ts));
  63.  
  64.  
  65. % Resistance ladder
  66. Rf= normrnd(45e3,sR*(45e3),1,6);
  67. %Rf = 90e3;
  68. Rin= normrnd(45e3,sR*(45e3),1,4);
  69. RRef = normrnd([45e3 22.5e3 11.25e3 5.625e3], sR*(22.5e3),1,4);
  70.  
  71.  
  72. %Input
  73. %vin = 1.8/2*(1 + sawtooth(ts*fi*(2*pi)));
  74. alpha = 1.8/16;
  75. %vin=linspace(0,1.8-alpha, length(tr));
  76. vin = [0:alpha:1.8 - alpha];
  77.  
  78.  
  79. % Cap response
  80. vC=zeros(1,length(ts));
  81. %for i=1:length(ts)
  82. % vC(i)=vin(i)*(1-exp(-dt/tau))+noise(i);
  83. %end
  84.  
  85. %learning parameters
  86. Vref = 0.1125;
  87. Y = zeros(1,4);
  88. E_threshold = 1e-8;
  89. T_write = 5e-6;
  90. dt = 0.54e-8;
  91. %RMOS = 450;
  92. %K = 1e-3;
  93. Vp = 0.5;
  94. %Ron = 100;
  95. %Roff = 200e3;
  96. RMOS = normrnd(700,sR*(700),1,100);
  97. Yvar = normrnd(1,0.1*(1),1,4);
  98. Kon = normrnd(-4.8e-3,0.01*(4.8e-3),1,100);
  99. Koff = normrnd(2.8125e-3,0.01*(2.8125e-3),1,100);
  100. Von = -0.3;
  101. Voff = 0.4;
  102. %Von = normrnd(0.3,0.033*(0.3),1,6);
  103. %Vp = normrnd(0.5,0.1*0.5);
  104. %Vpgd = normrnd(0.4,0.1*0.4);
  105. Roff = normrnd(100e3,sR*(100e3),1,100);
  106. Ron = normrnd(2000,sR*(2000),1,100);
  107. VR_parasite = 0.015; %frequency dependent
  108.  
  109. E = 1 ;
  110. E_rate = 1;
  111.  
  112.  
  113. %Digital data set
  114. D = 1.8*[0 0 0 0; 0 0 0 1; 0 0 1 0; 0 0 1 1; 0 1 0 0; 0 1 0 1; 0 1 1 0; 0 1 1 1;
  115. 1 0 0 0; 1 0 0 1; 1 0 1 0; 1 0 1 1; 1 1 0 0; 1 1 0 1; 1 1 1 0; 1 1 1 1;];
  116.  
  117. n = 0;
  118.  
  119. t = 1;
  120. A = 0;
  121. DAC = zeros(1, length(vin));
  122. Neurons = zeros(4, length(vin));
  123. vout = zeros(1, length(vin));
  124.  
  125.  
  126.  
  127. %binary-weighted gradient descent Training loop
  128. W_prev = 1/2.+zeros(1,100);
  129. gamma = 1;
  130. for l = 1:1 %2000 samples 20ms
  131. Ptot_neuron = 0;
  132. Ptot_net = 0;
  133. Ptot_comparator = 0;
  134. for x = 1: length(vin)
  135. %k = randi([1,length(vin)]); %random dataset
  136. k = x;
  137.  
  138. noise=random('norm',0,sNoise);
  139. thermnoise = random('norm',0,thNoise);
  140. vJit=random('norm',0,sJit);
  141.  
  142. if vin(k)> 1.793
  143. vin(k)=0;
  144. end
  145. indx = int16(vin(k)/alpha) + 1;
  146.  
  147. n = n+1;
  148. vC(n)=vin(k)*(1-exp(-dt/tau));
  149. if non_ideal == 1
  150. Y(4) = activation(vin(k) - 8*Vref);
  151. Y(3) = activation(vin(k) +vJit - 4*Vref - (Rf(3)/(R(3,4)+RMOS(3*10+4) + thermnoise))*Y(4)*Yvar(4)/18 + noise +VR_parasite);
  152. Y(2) = activation(vin(k) +vJit - 2*Vref - (Rf(2)/(R(2,4)+RMOS(2*10+4)+ thermnoise))*Y(4)*Yvar(4)/18 - (Rf(2)/(R(2,3)+RMOS(2*10+3)+ thermnoise))*Y(3)*Yvar(3)/18 + noise +VR_parasite);
  153. Y(1) = activation(vin(k)+vJit - 1*Vref - (Rf(1)/(R(1,4)+RMOS(1*10+4)+ thermnoise))*Y(4)*Yvar(4)/18 - (Rf(2)/(R(1,3)+RMOS(1*10+3)+ thermnoise))*Y(3)*Yvar(3)/18 - (Rf(1)/(R(1,2)+RMOS(1*10+2)+ thermnoise))*Y(2)*Yvar(2)/18 + noise +VR_parasite);
  154. else
  155. % Y(5) = activation(vin(k) - 16*Vref);
  156. % Y(4) = activation(vin(k) - 8*Vref - 16*Y(5)/18);
  157. % Y(3) = activation(vin(k) - 4*Vref - 8*Y(4)/18 - 16*Y(5)/18);
  158. % Y(2) = activation(vin(k) - 2*Vref - 4*Y(3)/18 - 8*Y(4)/18 - 16*Y(5)/18);
  159. % Y(1) = activation(vin(k) - 1*Vref - 2*Y(2)/18 - 4*Y(3)/18 - 8*Y(4)/18 - 16*Y(5)/18);
  160.  
  161. Y(4) = activation(vin(k) - 8*Vref)
  162. Y(3) = activation(vin(k) - 4*Vref - 8*Y(4)/16)
  163. Y(2) = activation(vin(k) - 2*Vref - 4*Y(3)/16 - 8*Y(4)/16)
  164. Y(1) = activation(vin(k) - 1*Vref - 2*Y(2)/16 - 4*Y(3)/16 - 8*Y(4)/16)
  165. end
  166.  
  167. Rfed=45e3;
  168. P1_neuron = (vin(k) - 8*Vref )^2/Rfed ;
  169. P1_net = vin(k)^2/Rfed + Vref^2/Rfed/8;
  170. P2_neuron = (vin(k) - 4*Vref - (Rfed/(R(3,4)+450))*Y(4)/18)^2/Rfed ;
  171. P2_net = vin(k)^2/Rfed + Vref^2/Rfed/4 + (Y(4)/18)^2/(R(3,4)+450);
  172. P3_neuron = (vin(k) - 2*Vref - (Rfed/(R(2,4)+450))*Y(4)/18 - (Rfed/(R(2,3)+450))*Y(3)/18)^2/Rfed ;
  173. P3_net = vin(k)^2/Rfed + Vref^2/Rfed/2 + (Y(4)/18)^2/(R(2,4)+450) + (Y(3)/18)^2/(R(2,3)+450);
  174. P4_neuron = (vin(k) - 1*Vref - (Rfed/(R(1,4)+450))*Y(4)/18 - (Rfed/(R(1,3)+450))*Y(3)/18 - (Rfed/(R(1,2)+450))*Y(2)/18)^2/Rfed ;
  175. P4_net = vin(k)^2/Rfed + Vref^2/Rfed/1 + (Y(4)/18)^2/(R(1,4)+450) + (Y(3)/18)^2/(R(1,3)+450) + (Y(2)/18)^2/(R(1,2)+450);
  176. %Ptot_net = Ptot_net + P1_net + P2_net + P3_net + P4_net ;
  177. Ptot_neuron = Ptot_neuron + P1_neuron + P2_neuron + P3_neuron + P4_neuron ;
  178. %Ptot_comparator = Ptot_comparator + 4*(1e-8);
  179.  
  180. fliplr(Y);
  181. D(ceil(indx),:);
  182. E(end + 1) = (0.125/(1.8^2))*((Y(1) - D(ceil(indx),4))^2 + (Y(2) - D(ceil(indx),3))^2 + (Y(3) - D(ceil(indx),2))^2 + (Y(4) - D(ceil(indx),1))^2 + 0.01*rand()^2) + E(end);
  183. if E(end) == 0 %false positive case
  184. continue;
  185. end
  186. DAC(n) = 1/16*(1*Y(1)+ 2*Y(2) + 4*Y(3) + 8*Y(4));
  187. % Neurons(1,n) =Y(5);
  188. Neurons(1,n) =Y(4);
  189. Neurons(2,n) =Y(3);
  190. Neurons(3,n) =Y(2);
  191. Neurons(4,n) =Y(1);
  192. vout(n) = vin(k);
  193.  
  194. E_rate(end+1) = (1/n)*E(end);
  195. t(end+1) = t(end) + 1;
  196. if E_rate(end) < E_threshold
  197. break;
  198. end
  199. for i = 1:4
  200. for j = 1:4
  201. if i >= j
  202. R(i,j) = 0;
  203. else
  204. x = 10*i +j;
  205. W_f = W_prev(x)*(1-W_prev(x));
  206. if (D(ceil(indx),5-i) - Y(i))>=0
  207. Vw = -1*Vp*(R(i,j)/(R(i,j)+ RMOS(x)));
  208. dW = -1*Kon(x)*((Vw/Von -1)^3)*W_f;
  209. else
  210. Vw = 1*Vp*(R(i,j)/(R(i,j)+ RMOS(x)));
  211. dW = Koff(x)*((Vw/Voff -1)^1)*W_f;
  212. end
  213.  
  214. if n < 1600
  215. gamma = 1;
  216. elseif n>=1600 && n<2400
  217. gamma = 0.5;
  218. elseif n>=2800 && n<3200
  219. gamma = 0.25;
  220. else
  221. gamma = 0.03;
  222. end
  223.  
  224. beta = gamma*0.09*(Roff(x)-Ron(x))*dW*T_write/dt;
  225.  
  226. %if R(i) + beta*(D(ceil(indx),5-i) - Y(i))*D(ceil(indx),5-j)*Yvar(i)/1.8 >= Roff(x)
  227. % continue
  228. %end
  229. %if R(i) + beta*(D(ceil(indx),5-i) - Y(i))*D(ceil(indx),5-j)*Yvar(i)/1.8 <= Ron(x)
  230. % continue
  231. %end
  232.  
  233. %R(i,j) = R(i,j) + beta*random('norm',(D(ceil(indx),5-i) - Y(i))*D(ceil(indx),5-j)/(1.8^2), thNoise) + thermnoise;
  234. R(i,j) = R(i,j) + beta*(D(ceil(indx),5-i) - Y(i)+ 0.0001*rand())*D(ceil(indx),5-j)/(1.8^2);
  235. W_prev(x) = R(i,j)/(Roff(x) - Ron(x));
  236. end
  237. end
  238. end
  239. R12(end+1) = R(1,2);
  240. R13(end+1) = R(1,3);
  241. R14(end+1) = R(1,4);
  242. R23(end+1) = R(2,3);
  243. R24(end+1) = R(2,4);
  244. R34(end+1) = R(3,4);
  245.  
  246.  
  247. end
  248. if E_rate(end) < E_threshold
  249. break;
  250. end
  251. P(l) = (Ptot_comparator + Ptot_net + Ptot_neuron )/2^4;
  252. end
  253.  
  254.  
  255. %%
  256. figure(1);
  257. hold on;
  258. ylabel('Training Error','FontSize',18);
  259. xlabel('#Samples','FontSize',18);
  260.  
  261.  
  262. plot(t,E_rate,'LineWidth',4);
  263. grid on;
  264. leg = legend('Algorithm');
  265. set(leg,'FontSize',12);
  266. set(gca,'FontSize',12);
  267.  
  268.  
  269. figure(3);
  270. hold on;
  271. plot(t,(45e3)./R12);
  272. plot(t,(45e3)./R13);
  273. plot(t,(45e3)./R14);
  274. plot(t,(45e3)./R23);
  275. plot(t,(45e3)./R24);
  276. plot(t,(45e3)./R34);
  277.  
  278. figure;
  279. plot(P);
  280.  
  281.  
  282. tf = t(1:length(t)-1);
  283. figure(4);
  284. hold on;
  285. stairs(tf,DAC), xlim([n-300 n]);
  286. stairs(tf,vout), xlim([n-300 n]);
  287.  
  288. figure(5);
  289. subplot(4,1,1);
  290. stairs((n-48:n),Neurons(1,n-48:n),'b','linewidth',2);
  291. subplot(4,1,2);
  292. stairs((n-48:n),Neurons(2,n-48:n),'r','linewidth',2);
  293. subplot(4,1,3);
  294. stairs((n-48:n),Neurons(3,n-48:n),'g','linewidth',2);
  295. subplot(4,1,4);
  296. stairs((n-48:n),Neurons(4,n-48:n),'y','linewidth',2);
  297. % subplot(4,1,5);
  298. % stairs((n-48:n),Neurons(5,n-48:n),'y','linewidth',2);
  299.  
  300.  
  301. figure(6);
  302. subplot(4,1,1);
  303. stairs((n-3848:n-3800),Neurons(1,n-3848:n-3800),'b','linewidth',2);
  304. subplot(4,1,2);
  305. stairs((n-3848:n-3800),Neurons(2,n-3848:n-3800),'r','linewidth',2);
  306. subplot(4,1,3);
  307. stairs((n-3848:n-3800),Neurons(3,n-3848:n-3800),'g','linewidth',2);
  308. subplot(4,1,4);
  309. stairs((n-3848:n-3800),Neurons(4,n-3848:n-3800),'y','linewidth',2);
  310. % subplot(4,1,5);
  311. % stairs((n-3848:n-3800),Neurons(5,n-3848:n-3800),'y','linewidth',2);
  312. %
  313. figure(7);
  314. subplot(4,1,1);
  315. stairs((1:48),Neurons(1,1:48),'b','linewidth',2);
  316. subplot(4,1,2);
  317. stairs((1:48),Neurons(2,1:48),'r','linewidth',2);
  318. subplot(4,1,3);
  319. stairs((1:48),Neurons(3,1:48),'g','linewidth',2);
  320. subplot(4,1,4);
  321. stairs((1:48),Neurons(4,1:48),'y','linewidth',2);
  322. % subplot(4,1,5);
  323. % stairs((1:48),Neurons(5,1:48),'y','linewidth',2);
  324.  
  325.  
  326.  
  327.  
  328. %% %% Static Evaluation
  329.  
  330. %maxDNL =zeros(1,length(R12));
  331. %maxINL =zeros(1,length(R12));
  332. %for id = 1:length(R12)
  333. %Input
  334.  
  335. VFS = 1.8;
  336. vin = [0:0.001:15/16*VFS];
  337.  
  338.  
  339. index = length(R12);
  340. R_mean_f = [mean(R14(index - 48:index-1)), mean(R24(index - 48:index-1)),mean(R34(index - 48:index-1)),mean(R13(index - 48:index-1)), mean(R23(index - 48:index-1)),mean(R12(index - 48:index-1))];
  341. vout_f=zeros(length(tr),4);
  342. vout_c=zeros(length(tr),4);
  343. vout_i=zeros(length(tr),4);
  344. out_f = zeros(length(vin),1);
  345. out_i = zeros(length(vin),1);
  346. out_c = zeros(length(vin),1);
  347. for i=1:length(vin)
  348. for j=1:1
  349. noise=random('norm',0,sNoise);
  350. thermnoise = random('norm',0,thNoise);
  351. vJit=random('norm',0,sJit);
  352. vout_f(i,1) = activation(vin(i) - 8*Vref);
  353. vout_f(i,2) = activation(vin(i) +vJit - 4*Vref - (Rf(3)/(R_mean_f(3)+RMOS(3*10+4) + thermnoise))*vout_f(i,1)*Yvar(4)/18 + noise +VR_parasite);
  354. vout_f(i,3) = activation(vin(i) +vJit - 2*Vref - (Rf(2)/(R_mean_f(2)+RMOS(2*10+4)+ thermnoise))*vout_f(i,1)*Yvar(4)/18 - (Rf(2)/(R_mean_f(5)+RMOS(2*10+3)+ thermnoise))*vout_f(i,2)*Yvar(3)/18 + noise +VR_parasite);
  355. vout_f(i,4) = activation(vin(i)+vJit - 1*Vref - (Rf(1)/(R_mean_f(1)+RMOS(1*10+4)+ thermnoise))*vout_f(i,1)*Yvar(4)/18 - (Rf(2)/(R_mean_f(4)+RMOS(1*10+3)+ thermnoise))*vout_f(i,2)*Yvar(3)/18 - (Rf(1)/(R_mean_f(6)+RMOS(1*10+2)+ thermnoise))*vout_f(i,3)*Yvar(2)/18 + noise +VR_parasite);
  356.  
  357. for j=1:4
  358. out_f(i) = out_f(i) + (vout_f(i,j)/1.8)*power(2,4-j);
  359. end
  360. end
  361. end
  362.  
  363.  
  364. R_mean_c = [mean(R14(index - 3000:index-2600)), mean(R24(index - 3000:index-2600)),mean(R34(index - 3000:index-2600)),mean(R13(index - 3000:index-2600)), mean(R23(index - 3000:index-2600)),mean(R12(index - 3000:index-2600))];
  365. for i=1:length(vin)
  366. for j=1:1
  367. noise=random('norm',0,sNoise);
  368. thermnoise = random('norm',0,thNoise);
  369. vJit=random('norm',0,sJit);
  370. vout_c(i,1) = activation(vin(i) - 8*Vref);
  371. vout_c(i,2) = activation(vin(i) +vJit - 4*Vref - (Rf(3)/(R34(200)+RMOS(3*10+4) + thermnoise))*vout_c(i,1)*Yvar(4)/18 + noise +VR_parasite);
  372. vout_c(i,3) = activation(vin(i) +vJit - 2*Vref - (Rf(2)/(R24(200)+RMOS(2*10+4)+ thermnoise))*vout_c(i,1)*Yvar(4)/18 - (Rf(2)/(R23(200)+RMOS(2*10+3)+ thermnoise))*vout_c(i,2)*Yvar(3)/18 + noise +VR_parasite);
  373. vout_c(i,4) = activation(vin(i)+vJit - 1*Vref - (Rf(1)/(R14(200)+RMOS(1*10+4)+ thermnoise))*vout_c(i,1)*Yvar(4)/18 - (Rf(2)/(R13(200)+RMOS(1*10+3)+ thermnoise))*vout_c(i,2)*Yvar(3)/18 - (Rf(1)/(R12(200)+RMOS(1*10+2)+ thermnoise))*vout_c(i,3)*Yvar(2)/18 + noise +VR_parasite);
  374.  
  375. for j=1:4
  376. out_c(i) = out_c(i) + (vout_c(i,j)/1.8)*power(2,4-j);
  377. end
  378. end
  379. end
  380.  
  381.  
  382. for i=1:length(vin)
  383. for j=1:1
  384. noise=random('norm',0,sNoise);
  385. thermnoise = random('norm',0,thNoise);
  386. vJit=random('norm',0,sJit);
  387. vout_i(i,1) = activation(vin(i) - 8*Vref);
  388. vout_i(i,2) = activation(vin(i) +vJit - 4*Vref - (Rf(3)/(R34(1)+RMOS(3*10+4) + thermnoise))*vout_i(i,1)*Yvar(4)/18 + noise +VR_parasite);
  389. vout_i(i,3) = activation(vin(i) +vJit - 2*Vref - (Rf(2)/(R24(1)+RMOS(2*10+4)+ thermnoise))*vout_i(i,1)*Yvar(4)/18 - (Rf(2)/(R23(1)+RMOS(2*10+3)+ thermnoise))*vout_i(i,2)*Yvar(3)/18 + noise +VR_parasite);
  390. vout_i(i,4) = activation(vin(i)+vJit - 1*Vref - (Rf(1)/(R14(1)+RMOS(1*10+4)+ thermnoise))*vout_i(i,1)*Yvar(4)/18 - (Rf(2)/(R13(1)+RMOS(1*10+3)+ thermnoise))*vout_i(i,2)*Yvar(3)/18 - (Rf(1)/(R12(1)+RMOS(1*10+2)+ thermnoise))*vout_i(i,3)*Yvar(2)/18 + noise +VR_parasite);
  391.  
  392. for j=1:4
  393. out_i(i) = out_i(i) + (vout_i(i,j)/1.8)*power(2,4-j);
  394. end
  395. end
  396. end
  397.  
  398.  
  399.  
  400.  
  401.  
  402. LSB=1/16;
  403. out_step_unique = unique(out_f);
  404. out_step = zeros(1,16);
  405. for j=1:length(out_step_unique)
  406. out_step(out_step_unique(j)+1) = out_step_unique(j);
  407. end
  408.  
  409. w = length(out_step);
  410. s=zeros(1,w);
  411. in=zeros(1,w);
  412. for i=2: w+1
  413. s(i-1) = find(out_f == out_step(i-1), 1, 'first');
  414. in(i-1) = vin(s(i-1));
  415. end
  416. %x= diff(in);
  417. steps =zeros(1,15);
  418. last = 0;
  419. for k=1:15
  420. if in(k) == 0
  421. steps(k) = 0;
  422. else
  423. steps(k) = in(k) - last;
  424. last = in(k);
  425. end
  426. end
  427. step_av=mean(steps)
  428. DNL_f= steps/step_av - 1;
  429. DNL_f(1) = 0;
  430. maxDNL_f=max(DNL_f)
  431.  
  432. INL_f=zeros(1,length(out_step));
  433. for i = 1:length(steps)
  434. INL_f(i) = sum(DNL_f(1:i-1));
  435. end
  436.  
  437.  
  438. maxINL_f=max(INL_f);
  439.  
  440.  
  441. out_step_unique = unique(out_c);
  442. out_step = zeros(1,16);
  443. for j=1:length(out_step_unique)
  444. out_step(out_step_unique(j)+1) = out_step_unique(j);
  445. end
  446.  
  447. w = length(out_step);
  448. s=zeros(1,w);
  449. in=zeros(1,w);
  450. for i=2: w+1
  451. s(i-1) = find(out_f == out_step(i-1), 1, 'first');
  452. in(i-1) = vin(s(i-1));
  453. end
  454. %x= diff(in);
  455. steps =zeros(1,15);
  456. last = 0;
  457. for k=1:15
  458. if in(k) == 0
  459. steps(k) = 0;
  460. else
  461. steps(k) = in(k) - last;
  462. last = in(k);
  463. end
  464. end
  465. step_av=mean(steps)
  466. DNL_c= steps/step_av - 1;
  467. DNL_c(1) =0;
  468. maxDNL_f=max(DNL_c)
  469.  
  470. INL_c=zeros(1,length(out_step));
  471. for i = 1:length(steps)
  472. INL_c(i) = sum(DNL_c(1:i-1));
  473. end
  474.  
  475.  
  476. maxINL_c=max(INL_c);
  477.  
  478.  
  479.  
  480. out_step_unique = unique(out_i);
  481. out_step = zeros(1,16);
  482. for j=1:length(out_step_unique)
  483. out_step(out_step_unique(j)+1) = out_step_unique(j);
  484. end
  485.  
  486. w = length(out_step);
  487. s=zeros(1,w);
  488. in=zeros(1,w);
  489. for i=2: w+1
  490. s(i-1) = find(out_i == out_step(i-1), 1, 'first');
  491. in(i-1) = vin(s(i-1));
  492. end
  493. %x= diff(in);
  494. steps =zeros(1,15);
  495. last = 0;
  496. for k=1:15
  497. if in(k) == 0
  498. steps(k) = 0;
  499. else
  500. steps(k) = in(k) - last;
  501. last = in(k);
  502. end
  503. end
  504. step_av=mean(steps)
  505. DNL_i= steps/step_av - 1;
  506. DNL_i(1) =0;
  507. maxDNL_i=max(DNL_i)
  508.  
  509. INL_i=zeros(1,length(out_step));
  510. for i = 1:length(steps)
  511. INL_i(i) = sum(DNL_i(1:i-1));
  512. end
  513.  
  514.  
  515. maxINL_i=max(INL_i);
  516.  
  517.  
  518.  
  519.  
  520.  
  521. %end
  522. %close all
  523. figure
  524. hold on
  525. plot((0:14),DNL_f,'b-*','linewidth',2);
  526. plot((0:14),DNL_c,'r--','linewidth',2);
  527. plot((0:14),DNL_i,'g--','linewidth',2);
  528. hold off
  529. ylabel('DNL (LSB)');
  530. xlabel('Input code')
  531. legend('Gradient Descent')
  532. xlim([0 14])
  533. box on
  534. set(gca,'linewidth',3)
  535. set(findall(gcf,'-property','FontSize'),'FontSize',44)
  536. set(findall(gcf,'-property','FontName'),'FontName','Calibri')
  537. set(findall(gcf,'-property','FontWeight'),'FontWeight','bold')
  538.  
  539. figure
  540. hold on
  541. plot((0:15),INL_f,'b-*','linewidth',2);
  542. plot((0:15),INL_c,'r--','linewidth',2);
  543. plot((0:15),INL_i,'g--','linewidth',2);
  544. ylabel('INL (LSB)');
  545. xlabel('Input code')
  546. hold off
  547. box on
  548. xlim([0 15])
  549. legend('Gradient Descent')
  550. set(gca,'linewidth',3)
  551. set(findall(gcf,'-property','FontSize'),'FontSize',44)
  552. set(findall(gcf,'-property','FontName'),'FontName','Calibri')
  553. set(findall(gcf,'-property','FontWeight'),'FontWeight','bold')
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement
Public Pastes
Advertisement
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!