↳
Made no mistakes on this question, but had done similar things before
↳
//reverses bit between two indices of a number. #include using namespace std; int main() { unsigned int no=90,i=2,j=6,tmp=0,tmp1=0; tmp=no>>i; cout>1; } cout>1; if(x=i) tmp=tmp|0; else tmp=tmp|0x80000000; } cout< Less
↳
#include main(int arg, char *argv[]) { unsigned char num, hi_indx, lo_indx, mask, num_tmp, mask_c, num_rv, iter,i; num = atoi(argv[1]); hi_indx = atoi(argv[2]); lo_indx = atoi(argv[3]); mask = ((0xFF > (7-hi_indx+lo_indx); mask = (mask << lo_indx); mask_c = ~mask; num_tmp = num & mask_c; printf("%x %d %d %x",num, hi_indx, lo_indx, mask); num_rv = 0x0; iter = hi_indx-lo_indx+1; i = 0; while(i<=iter) { int flg; flg=(num & (1<<(i+lo_indx)))?1:0; if(flg) num_rv=num_rv|(1<<(hi_indx-i)); i++; } num = num_rv | num_tmp; printf("\nnum=%x ",num); } Less
↳
I realized this wasn't this job for me. I didn't study EE and the headhunter should have seen that. Yes I have a PhD in psychoacoustics and have done a fair amount of programming of filters and frequency domain analysis, but my theoretical knowledge is (extremely) minimal. 30 minutes after the interview and remembering that x[n-1] = z^-1, and then revisiting the textbook, I realized that the z-transform is indeed 1 + 2z^-1. *This would have been a very easy interview for someone with a BSEE.* Less
↳
1/(1+2z^-1)
↳
These qtns where to test the DSP, Communication Theory background.
↳
OFDM - Orthogonal Frequency Division Multiplexing. This is a multi-carrier modulation scheme, where the data is transmitted as large amount of orthogonal sub-carriers on parallel data streams. Since they are orthogonal, they do not affect each other sub-carrier. Each sub-carrier is modulated using QAM/PSK. This is very efficient when used with MIMO and in DSL systems. Less
↳
Ideally you can have as many beamformers as you can, but you need to reduce the number for practical applications. There is no specific formula. However they expect a fixed number. Seems they are just academic theoretical knowledge with no practical experience. Less
↳
I answered the feature extraction and dsp based techniques to create new features and machine learning algorithm to classify those signals. Less
↳
To reduce complexity of resampling through different stages