RENEW Project Documentation

Version 1.0

Reconfigurable Ecosystem for Next-generation End-to-end Wireless

OFDM Uplink Unchained Mode

In this tutorial, we extend the simple Matlab script of the previous tutorial to transmit and receive in the Uplink an OFDM packet in the unchained mode. A series of random bits are QAM-modulated and put on subcarriers according to the 802.11g WiFi standard. Short-term and long-term training sequences (STS and LTS) are also included in the time-domain OFDM packet. This script creates two Iris node objects, one acting as base station (BS) and one as UE. Then, using the function set offered by the Matlab driver, the script will configure the TDD schedule, initiate the beacon transmission on the BS side and start transmitting the OFDM packet on the UE side. After reading the received data, it will detect the start of the packet through correlation, extract the payload, perform channel estimation and phase correction and finally plot the received constellation and the error vector metric.

Click here for the complete code.

Define pilot sequence

We follow the WiFi convention and define the pilots as frequncy domain long-term training sequences (LTS) and convert them to time domain samples through an IFFT.

 % LTS for fine CFO and channel estimation
lts_f = [0 1 -1 -1 1 1 -1 1 -1 1 -1 -1 -1 -1 -1 1 1 -1 -1 1 -1 1 -1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 ...
    1 1 -1 -1 1 1 -1 1 -1 1 1 1 1 1 1 -1 -1 1 1 -1 1 -1 1 1 1 1];
lts_t = ifft(lts_f, 64); %time domain  

Define and configure Iris nodes

Here, we first define the experiment’s parameters such as the TDD schedule, the IDs of the boards we are going to use, the sample rates etc. These parameters are kept in a struct. Then, we create instances of the Iris board objects (at the Matlab level) and initialize them with the parameter struct we created. This initialization will, in turn, create instances of Python objects in the Matlab driver with the same parameters defined here.

 b_sched = 'PGGGGGRG';               % BS schedule
    u_sched = 'GGGGGGPG';               % UE schedule
    % Create a vector of node objects
    sdr_params = struct(...
        'id', 'RF3C000007', ...
        'txfreq', 2.6e9, ...
        'rxfreq', 2.6e9, ...
        'txgain', 37, ...
        'rxgain', 30, ...
        'sample_rate', 5e6, ...
        'n_samp', length(tx_vec_iris), ...     % number of samples per frame time.
        'tdd_sched', b_sched, ...     % number of zero-paddes samples
        'n_zpad_samp', (N_ZPAD_PRE + N_ZPAD_POST) ...
    sdr_params(2) = sdr_params(1);
    sdr_params(2).id =  'RF3C000028';
    sdr_params(2).rxfreq = 2.6e9;
    sdr_params(2).txfreq = 2.6e9;
    sdr_params(2).tdd_sched = u_sched;
    node_bs = iris_py(sdr_params(1));
    node_ue = iris_py(sdr_params(2)); 

Modulation IFFT

These are done following standard Matlab practice and can be found in the body of the script. Same as demodulation upon reception of the signal.

Set up transmission

The following lines reset the correlator on the UE, hence, preparing it to synchronize based on the peak it will have to detect, set up reading streams on both nodes before the transmission starts, and configure the schedules on the BS and the UE in unchained mode, respectively.

chained_mode = 0;

Burn data on the BS and the UE

The BS will initiate the writing of the predefined beacon on the ram for transmission during the ‘P’ slots of the BS schedule. The beacon sequence is hardcoded in the Python driver and the Matlab user should not be concerned with it. The UE in turn also writes its pilots onto the ram so that they will be transmitted during the ‘P’ slots of its schedule.


Set up reception

The reading stream is activated on the BS and the correlator is enabled on the UE.


Start frame

By pulling the trigger on BS the frame starts. The first slot is defined as ‘P’ in the BS’ schedule, thus it will start transmitting the beacon stored onto its ram.



The BS reads and stores the received data from the receiver stream set up and activated earlier.

 [rx_vec_iris, data0_len] = node_bs.sdrrx(); 

End streams

The streams on both nodes are closed and the Iris nodes are rest.


Plot received samples

It is evident that the interaction with Iris nodes is identical to the first simple tutorial. What the experimnt script of this tutorial shows is how we can extend the range of what we are able to do with the transmitted and received data on Matlab. Below are some examples of figures we can show on the received data.

Received OFDM waveform

Peak detection with LTS correlation

Channel Estimate Magnitude anf Phase



Last updated on 20 Mar 2019 / Published on 18 Mar 2019