People

Homepage of Harish Kundur Subramaniyan

PhD. students ( / AIO )

NameH. Kundur Subramaniyan, MSc (Harish) Kundur Subramaniyan, MSc H.  (Harish)
E-mail H. KundurSubramaniyan@utwente.nl
DepartmentEEMCS / Electrical Engineering
AddressCarré, 2728
 P.O. Box 217
 7500 AE Enschede
 The Netherlands
Phone+31 53 489 4831
Secretary+31 53 489 2644
 
 
Hi All, Welcome to my home page.
 
Aim of my project is to make a low power, low cost, low noise, highly linear and wide band receiver with flexible radio frequency and bandwidth. In short, the aim is: "implementation of a wide-band SDR" (Software Defined Radio).

In a traditional radio, performance metrics like noise, power, cost, linearity trade-off with each-other. A simple example of this trade-off is as follows. Say, you want to receive a weak radio signal with minimum corruption in the presence of a strong unwanted signal. To do this, you amplify the signal with a high gain using a low-noise amplifier. In this amplification process, the unwanted signal is also amplified leading to non-linear distortion and hence corruption of the wanted signal. This beats the whole purpose of receiving the signal without corruption. This is an example of a trade-off between noise and linearity. We could solve the problem by using an exotic/expensive filter that removes the interferer before the amplification but this makes our receiver expensive. Thus cost is also involved in the trade-offs.

The limitations of the trade-offs are dictated by physical parameters like supply voltage, transistor properties, cost and the amount of power we are willing to burn. Some of the parameters involved are shown in the famous design hexagon in [1] (see Figure 1). A good radio design takes into account all such trade-offs and is always a "good working compromise" among all the performance metrics.





Figure 1 Design Hexagon.

Recent research in SDR has lead to new techniques to push the limits of these trade-offs involved in radio design. The aim of my research is to study the performance improvement techniques that are already used and to propose new techniques which have benefits.

References:

[1] B.Razavi, "RF Microelectronics", Prentice Hall, Upper Saddle River, NJ, 1998.