Convert Signal from Time Domain to Frequency Domain and then again back to Time domain

I have sinusoidal respiration signal of 1 min, which has the respiration rate of 18 breaths/min. I am trying to simulate the respiration signal by using curve fitting technique and using respiration rate (RR) as a control parameter. For, example if I if enter the RR as 24 breaths/min, then respiration signal with 24 breaths/min should be simulated.

I am having problem in generating simulated signal. In the code given below, the simulated signal has the same RR as that of original signal despite providing it different RR.

It would be really grateful, if anyone can help me out with this.

Here's my code

clc clear all

load 'resp.mat'; % Respiration Signal

t = [1:1:18000]./300;

Fs = 300;

resp_signal = resp(1:18000,:); % Respiration Signal for 60 sec

[resp_f,gof,output] = fit(t',resp_signal,'fourier8'); % Curve fitted respiration signal

y_resp = resp_f(t);

psig1 = y_resp - mean(y_resp); % Remove dc component

[pxx1,w1] = pwelch(psig1,[],[],[],Fs); % PSD of the signal

figure

plot(w1,pxx1);

axis([0 10, 0 100]);

[~,loc1] = max(pxx1);

freq1 = w1(loc1);

resp_rate1 = freq1 * 60;

bp = input('Enter Respiration Rate: ');

bps = bp * 17.068;

Fss = bps; % Frequency of the signal to be simulated

psig2 = y_resp - mean(y_resp);

[pxx2,w2] = pwelch(psig2,[],[],[],Fss); % PSD of the simulated signal

figure

plot(w2,pxx2);

axis([0 10, 0 200]);

[~,loc2] = max(pxx2);

freq2 = w2(loc2);

resp_rate2 = freq2 * 60;