Project 1-3 (USRP project : 2-4 students)

802.11-like OFDM physical layer

Background

Communication systems with multiple carrier frequencies, such as systems based on orthogonal frequency division multiplexing (OFDM), have recently become increasingly popular. Examples of the use of OFDM include both wireless LANs (802.11agn, 802.16, 802.20, HIPERLAN/2), digital audio broadcasting (DAB) as well as terrestrial digital video broadcasting (DVB), and cellular such as 3GPP-LTE.

The goal of this project is to send and receive data over a real wireless link using analog and digital hardware based on the USRP2 with xcvr2450 daugtherboards (www.ettus.com). While signals are transmitted over the air in a realistic fashion the processing will be done off-line in matlab. Thus the data sequences to be transmitted are created in matlab, downloaded to the hardware, and sent on the channel (at the actual data-rate) of the transmitting node. The signal is captured on a receiving node, and saved for post-processing.

Two USRPs with imaginary antennas.

Specification

Implement a digital OFDM communication system using the 802.11a or 802.11n specification as a basis (the number of subcarriers and/or spacing may need to be altered due to limitations in the available sample-rate). Implement algorithms for synchronization (time of arrival estimation), frequency offset estimation, channel estimation and detection. A pure simulation environment should also be created in order to compare the results obtained in reality with that of the theory. In order to do so you must be able to estimate the signal to noise ratio of the channel. The difference between measurements and simulations should be investigated. You will also need to learn how to conduct experiments.

The basic requirements should be fulfilled at the mid-term evaluation. They should be proved by a draft report and demonstration of the “prototype” (prototype here means collecting measurements and running the receiver algorithms in matlab).

Either of the following three areas should be explored further (based on the chosen project #1-#3)

1. Advanced channel estimation by applying smoothing over subcarriers and sub-symbols see e.g. section 5 of [4].
2. Channel coding using convolutional, LDPC [9] or turbo-codes, see e.g. [8].
3. Inter-carrier interference suppression techniques to combat the influence of phase-noise, see e.g [7]. This may also involve using other hardware with more phase-noise.

The performance improvement of the techniques should be carefully assessed. These advanced requirements should be demonstrated at the grand-final, and documented in a final report, poster and oral presentation.