comm.RayTracingChannel

Show the impact of not forcing the smallest propagation delay to be zero for a multipath channel model. Filter signals through a multipath ray tracing channel between two sites in Hong Kong, China. Build two multipath channel models by using the result from ray tracing. For the first ray tracing channel model, force the minimum propagation delay to zero. For the second ray tracing channel model, do not force the minimum propagation delay to zero.

Create a Site Viewer map display of buildings in Hong Kong. For more information about the OSM file, see [1].

sv = siteviewer("Buildings","hongkong.osm");

tx = txsite( ...
    "Latitude",22.2789, ...
    "Longitude",114.1625, ...
    "AntennaAngle",30, ...        % Azimuth angle
    "AntennaHeight",10, ...
    "TransmitterFrequency",28e9);
rx = rxsite( ...
    "Latitude",22.2799, ...
    "Longitude",114.1617, ...
    "AntennaAngle",120, ...       % Azimuth angle
    "AntennaHeight",1);

Create a ray tracing propagation model, which MATLAB® represents using a RayTracing object. Configure the model to use the image method and to find paths with up to 2 surface reflections. Perform ray tracing to find rays by using the propagation model.

pm = propagationModel("raytracing", ...
    "Method","image", ...
    "MaxNumReflections",2);
rays = raytrace(tx,rx,pm);

Create a channel model using the calculated rays in between the transmitter and receiver sites. The default configuration forces zero minimum propagation delay. Show the temporal and spatial profiles of the channel.

rtchan = comm.RayTracingChannel(rays{1},tx,rx);
rtchan.SampleRate = 50e6;
showProfile(rtchan);

Create a clone of the ray tracing channel model and reconfigure it to not force zero minimum propagation delay. Show the temporal and spatial profiles of the channel. The angle of departure and arrival plots do not change, but the power delay profile plot shows the updated delay profile result when the minimum delay profile is not forced to zero.

rtchandelayed= clone(rtchan);
rtchandelayed.MinimizePropagationDelay = false;
showProfile(rtchandelayed);

Filter randomly generated 16-QAM signals through the channel models. Display the leading 15 elements of y and ydelayed, which are output by the ray tracing channel objects rtchan and rtchandelayed, respectively. The leading samples in the delayed signal, ydelayed, are all zeros. When you model your communications system, you must account for this signal delay to avoid losing trailing signal data.

M = 16;       % Modulation order
frmLen = 1e3; % Frame length
numTx = rtchan.info.NumTransmitElements;

x = qammod(randi([0,M-1],frmLen,numTx),M);
y = rtchan(x);
numTxdelayed = rtchandelayed.info.NumTransmitElements;
x = qammod(randi([0,M-1],frmLen,numTxdelayed),M);
ydelayed = rtchandelayed(x);
y(1:15)

ans = 15×1 complex

  -0.0000 - 0.0000i
   0.0000 + 0.0000i
  -0.0000 - 0.0000i
   0.0000 + 0.0000i
  -0.0000 - 0.0000i
   0.0003 + 0.0011i
  -0.0022 - 0.0111i
   1.0158 + 0.0042i
  -1.0347 - 2.0177i
  -3.0922 + 0.0258i
      ⋮

ydelayed(1:15)

ans = 15×1

     0
     0
     0
     0
     0
     0
     0
     0
     0
     0
      ⋮

Appendix

[1] The OSM file is downloaded from https://www.openstreetmap.org, which provides access to crowd-sourced map data all over the world. The data is licensed under the Open Data Commons Open Database License (ODbL), https://opendatacommons.org/licenses/odbl/.

Filter signals through a multipath ray tracing channel model between two sites in a conference room. Build the multipath channel model by using the result from ray tracing.

Define a 3-D map for a conference room with one table and four chairs by using a standard tessellation language (STL) data file. Define a transmitter site close to the wall and a receiver site above the table.

mapFileName = "conferenceroom.stl";

tx = txsite("cartesian", ...
    "AntennaPosition",[-1.45; -1.4; 2.3], ...
    "TransmitterFrequency",2.8e9);
rx = rxsite("cartesian", ...
    "AntennaPosition",[.6; .2; 1.0]);

Use the siteviewer object and the show object function to visualize the 3-D scenario. The transmitter and receiver sites are colored in red and blue, respectively.

siteviewer(SceneModel=mapFileName);
show(tx,"ShowAntennaHeight",false)
show(rx,"ShowAntennaHeight",false)

Create a ray tracing propagation model, which MATLAB represents using a RayTracing object. Configure the model to find paths with up to 3 surface reflections. By default, the model uses the shooting and bouncing rays (SBR) method.

pm = propagationModel("raytracing", ...
    "CoordinateSystem","cartesian", ...
    "MaxNumReflections",3);

Perform ray tracing to find rays by using the propagation model.

rays = raytrace(tx,rx,pm,"Map",mapFileName);

Extract the computed rays from the returned cell array, and then plot the rays. Each ray is colored based on its path loss value.

rays = rays{1,1};
plot(rays)

Create a channel model using the calculated rays in between the transmitter and receiver sites. Show the temporal and spatial profiles of the channel.

rtchan = comm.RayTracingChannel(rays,tx,rx);
showProfile(rtchan);

Show the filtered signal in a constellation diagram.

M = 16;       % Modulation order
frmLen = 1e3; % Frame length

numTx = rtchan.info.NumTransmitElements;
x = qammod(randi([0,M-1],frmLen,numTx),M);
y_samprate10e6 = rtchan(x);
constellationdiag = comm.ConstellationDiagram( ...
    ...NumInputPorts=2=1, ...ChannelNames={
    ..."Frequency response channel"}, ...
    XLimits=[-5 5], ...
    YLimits=[-5 5], ...
    ReferenceConstellation=qammod(0:M-1,M));
constellationdiag(y_samprate10e6(:));

At 10e6 sample rate, the diagram shows well-defined and tightly clustered points around the ideal 16-QAM constellation positions. This indicates that the channel has not significantly altered the signal and that the channel introduces minimal ISI and distortion. In a conference room where the primary focus is on maintaining a stable and predictable channel behavior this delay profile is ideal.

For the selected sample rate (10e6), the delay spread of the multiple rays is too low. A higher sample rate would capture the temporal variations and also introduce significant ISI and distortion.

To modify the sample rate of the ray tracing channel, you can set the SampleRate property by using a name-value argument when you create the object or you can create a channel model by using the rays and site and set the SampleRate property after you create the object.

rtchan1 = comm.RayTracingChannel(SampleRate=2e7)

rtchan1 = 
  comm.RayTracingChannel with properties:

                      SampleRate: 20000000
                 PropagationRays: [1×1 comm.Ray]
        MinimizePropagationDelay: true
                   TransmitArray: [1×1 arrayConfig]
    TransmitArrayOrientationAxes: [3×3 double]
                    ReceiveArray: [1×1 arrayConfig]
     ReceiveArrayOrientationAxes: [3×3 double]
         ReceiverVirtualVelocity: [3×1 double]
       NormalizeImpulseResponses: true
         NormalizeChannelOutputs: true
                ChannelFiltering: true

tx = txsite( ...
    Latitude=22.2789, ...
    Longitude=114.1625, ...
    AntennaAngle=30, ...        % Azimuth angle
    AntennaHeight=10, ...
    TransmitterFrequency=28e9);
rx = rxsite( ...
    Latitude=22.2799, ...
    Longitude=114.1617, ...
    AntennaAngle=120, ...       % Azimuth angle
    AntennaHeight=1);
pm = propagationModel("raytracing", ...
    Method="sbr", ...
    MaxNumReflections=3);
rays = raytrace(tx,rx,pm);

rtchan2 = comm.RayTracingChannel(rays{1},tx,rx);
rtchan2.SampleRate = 2e7

rtchan2 = 
  comm.RayTracingChannel with properties:

                      SampleRate: 20000000
                 PropagationRays: [1×2 comm.Ray]
        MinimizePropagationDelay: true
                   TransmitArray: [1×1 arrayConfig]
    TransmitArrayOrientationAxes: [3×3 double]
                    ReceiveArray: [1×1 arrayConfig]
     ReceiveArrayOrientationAxes: [3×3 double]
         ReceiverVirtualVelocity: [3×1 double]
       NormalizeImpulseResponses: true
         NormalizeChannelOutputs: true
                ChannelFiltering: true

modOrd = 16;
frmLen = 1e3;
numTx = rtchan2.info.NumTransmitElements;
x = qammod(randi([0,modOrd-1],frmLen,numTx),modOrd);
y = rtchan2(x);