Special Session 1:
Cross Disciplinary Research in Signal Processing,
RF Communication
and VLSI Circuit Issues.
ORGANIZERS
Masud H. Chowdhury, University of Illinois at Chicago, USA and A.B.M. Harun-ur Rashid, Bangladesh University of Engineering and Technology, Bangladesh.
SYNOPSIS
Many in the industry believe that the scaling of the minimum feature sizes down to nanometer range and the spiraling frequencies in GHz scale will lead to the system-on-a-chip (SOC) in many ASIC application areas. On the other hand, the system-on-package (SOP) paradigm, which may not always achieve the performance and form factors of an SOC, will facilitate rapid reengineering via reuse libraries. As a result SOP is finding increasing favor, and promises a high return on investment at a lower risk, compared to complete SOC paradigm. Many IC implementations will follow multi-core strategies instead of single-core implementation. Network-on-chip (NOC) will find its way into many applications. In the coming decades, the market competition among different design paradigms and implementation strategies will resolve itself as their technical and economical costs and benefits are being widely investigated and documented. However, all the recent studies and roadmaps like ITRS have indicated that one of the grand challenges that will be encountered by all of these future IC design paradigms is the limitation of metallic interconnect due to increased propagation delay and noise interference. At this cross-road of technology it is imperative to find ways to improve performance of traditional interconnects, and at the same time look beyond the metallic interconnects. In recent time various cross disciplinary research initiatives have been undertaken to explore well know algorithms, tools and techniques in other areas, such as, signal processing and RF communication, to be used in VLSI realm to take full advantage of what nanoscale IC technologies can offer. One of the radical alternatives to overcome the limitation of metal interconnect based IC technologies is to adopt RF communication scheme for on-chip global signaling. RF/wireless interconnect is basically the merger of existing microwave technology with integrated circuits. Some initial investigations show that RF/wireless interconnects is the best candidate for signal communication in next generation nanometer integrated circuits. However, the research on the merger of RF technology with IC design is still in incipient stage. The purpose of this session is to explore the feasibility of this merger, and identify research needs and challenges in this regard. While exploration for long term alternative is necessary, it is very important to keep looking for ways to better handle on-chip signal communication and signal integrity in current IC environment. Therefore, this session will include papers that explore both near-term and long-term solutions for the above mentioned issues.
|
