|
|
|
|
|
To summarize, my research interests are in wireless communications and networking. I also like digital signal processing and adaptive filters. Topics that interest me: Wireless communication system design, network architecture and administration, protocols and services for mobile ad hoc wireless networks, smart MAC protocols for WLANs, IEEE WLAN standards, energy-aware media access control protocols for smart sensor networks, reliable multicasting in MANETs, OFDM/multi-carrier communications for multiuser systems, digital signal processing, adaptive modulation, and MIMO / smart antenna systems.
We propose a new Adaptive Allocation Algorithm for Multiuser OFDM (MOAAA), which outperforms existing multiple access schemes like OFDM-TDMA and OFDM-FDMA. The experiments have been conducted with reference to the powerline environment with multipath fading, exponential noise spectrum and narrow-band interference. We have presented the results obtained by Monte-Carlo simulation using Matlab under different channel and noise conditions. We have quantified the improvement in terms of overall transmit power and BER. Our new algorithm achieves a significant gain of 3 to 5 dB in SNR over OFDM-FDMA and TDMA. Hence, the proposed algorithm can be used for efficient multiple access for broadband multimedia communications over frequency selective fading environments.
PATENT:Multi-user Indoor Wireless Access Systems based on the OFDM Transmission Technique - under submission for Indian Patent.
Lego Mindstroms are tiny robots with an embedded micro-controller, motion controllers, sensors and an IR port for downloading firm-wares. LegOS is an embedded operating system for the Mindstroms based on Linux. This is an implementation project and a lightweight communication stack (MAC and routing support) was implemented in the Linux kernel to enable multi-hop communication between the robots through the IR port. To communicate with host PC, the stack was also implemented in host PC (as a daemon). Sample applications involving multi-hop communication were demonstrated over the stack developed.
In this project, a location estimation system that estimates the location of a mobile user relative to the location servers was implemented and demonstrated. The main advantage of the system is it is developed with off the shelf components: laptops and 802.11 based wireless cards. The location servers are laptops fixed in known locations and a user roams around with a laptop. All the laptops are equipped with wireless cards and the distances of the mobile user to the location servers are measured using the signal strength of the packet received from the servers. A triangulation algorithm is used to find the exact location based on the distance information. This was done in Linux platform. An adaptive learning algorithm is used reduce the estimation errors caused due to the highly varying signal strength in indoor environment.
In this project various system level power management schemes are analyzed. The power management model was developed in CSIM and analyzed extensively by simulation.
