Multi Rate Sigal Processing (ECE Department @ LAQSHYA)

Multi Rate Sigal Processing

Have you ever heard about conversion of audio tape to CD (compact disc)? we know that both are storage devices, though both are for storage purpose it needs a conversion process because there is change in their sampling rates. Sampling rate of CD is 44.1kHz, audio tape is 48kHz.

Normally TV broadcasting can be done in any one of the following schemes.

1)      PAL (Phase Alternating Line) European standard.               

2)       NTSC (National Television System Committee)American standard.
TV receiver set made to receive programs transmitted on PAL scheme butcannot receive NTSC programs and vice-versa. That is one who want to watch an European program in America on TV it is not possible, because Europe and America adapt different schemes of transmission. To avoid this TV receiver sets provided with sampling rate converter so that one can watch programs comfortably.
From the above example we can know that we need signal processing system that can process varying sampling rates such systems are known as ‘MULTIRATE SIGNAL PROCESSING SYSTEMS’.
The process of converting a signal in one sampling rate to another sampling rate is known as ‘SAMPLING RATE CONVERSION’. 
This type of conversion needs in many applications. Some are
1) High quality data acquisition and storage systems.
2) In trans multiplexers.
3) In speech processing to reduce the storage space or the transmitting rate of the speech data.
4) Narrow band filtering for fetal ECG and EEG.etc.
The sampling rate conversion can be done in two ways.
1)      D/A conversion and resampling at required rate.
2)      Sampling rate conversion in digital domain( Multi rate sampling )
Broadly sampling rate conversion can be of two types
1)      Down sampling                              
 2) Up sampling
   Down sampling : In this, sampling rate decreases by a factor M. If x(n) is a discrete time signal then it’s down sampled version that can be denoted by x(Mn). Down sampled signal by a factor ‘M ‘can be formed by taking every M^thvalue of x(n). In this sampling rate of the resultant signal decreases.
Up sampling : In this,sampling increased by a factor L. If x(n) is a discrete time signal then it’s up sampled version can be denoted by x(n/L). 
Up sampled signal by a factor ‘L’ can be formed by placing (L-1) zeros between each pair of samples of x(n). In this sampling rate of the resultant signal increases
...........................................................................................................................................

Article Presented By 

--------------------------

P.Koteswar Rao
Associate Professor
ECE Department
LAQSHYA Institute of Technology & Sciences