Difference between revisions of "Shared Correlation Accelerator for an RF SoC"
From iis-projects
Line 16: | Line 16: | ||
[[File:Lte_pss_corr_3d.png|400px|thumb|Example correlation output for the eMTC network synchronization (PSS).]] | [[File:Lte_pss_corr_3d.png|400px|thumb|Example correlation output for the eMTC network synchronization (PSS).]] | ||
− | + | <!-- | |
===Status: Available === | ===Status: Available === | ||
: Looking for 1 Semester student | : Looking for 1 Semester student | ||
Line 23: | Line 23: | ||
: VLSI I | : VLSI I | ||
: VLSI II/III (for an ASIC tapeout) | : VLSI II/III (for an ASIC tapeout) | ||
+ | ---> | ||
<!-- | <!-- | ||
===Status: Completed === | ===Status: Completed === | ||
Line 28: | Line 29: | ||
: Matthias Baer, Renzo Andri | : Matthias Baer, Renzo Andri | ||
---> | ---> | ||
− | + | ||
===Status: In Progress === | ===Status: In Progress === | ||
: Student: Sascha Giger | : Student: Sascha Giger | ||
: Supervision: [[:User:Lstefan | Stefan Lippuner]] | : Supervision: [[:User:Lstefan | Stefan Lippuner]] | ||
− | + | ||
===Character=== | ===Character=== |
Revision as of 17:29, 7 March 2018
Contents
Introduction
We are currently working on designing Radio Frequency (RF) Systems on a chip (SoCs) for the Internet of Things (IoT). Requirements for these devices often include data connectivity and location information. A drone may, for instance, use both GPS and cellular positioning to increase the accuracy of the location estimate. We want to integrate both functionalities on a single chip.
Positioning and network synchronization rely on performing a large number of cross-correlations in real-time. Due to the large throughput requirements, large accelerators are required. Instead of using a dedicated circuit for each standard, we would like to share a single, flexible accelerator for all these tasks.
Project Description
The goal of this project is, to develop a shared correlation accelerator for our RF SoC. This unit should support both positioning, as well as the synchronization for NB-IoT and eMTC. To do this, you will study the algorithms to quickly calculate a large number of cross-correlations. Using this knowledge, you will develop the accelerator in VHDL, or using High-Level Synthesis (HLS). It should be able to meet the throughput requirements of 100M correlation results/s, while keeping the area as low as possible.
Once you have implemented and verified the unit, you may tape out your own chip, or test it in our FPGA testbed.
Status: In Progress
- Student: Sascha Giger
- Supervision: Stefan Lippuner
Character
- 20% Theory, Algorithms and Simulation
- 50% Implementation (HLS/VHDL)
- 30% ASIC Implementation / FPGA Integration