Beamforming

.github/workflows/github-actions.yml

@@ -18,7 +18,7 @@ jobs:
     steps:
     - uses: actions/checkout@v3
     - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v4
+      uses: actions/setup-python@v5
       with:
         python-version: ${{ matrix.python-version }}
         cache: 'pip'

.github/workflows/publish_documentation.yml

@@ -15,7 +15,7 @@ jobs:
           persist-credentials: false
 
       - name: Set up Python 3.11
-        uses: actions/setup-python@v4
+        uses: actions/setup-python@v5
         with:
           python-version: 3.11
 

mapc_sim/experimental/antenas.py

@@ -0,0 +1,21 @@
+import chex
+import jax.numpy as jnp
+
+
+def isotropic(pos: chex.Array) -> chex.Array:
+    """Calculate gain matrix for each point to point communication.
+
+    Parameters
+    ----------
+    pos: Array
+        An array of 2d positions for `n` points
+
+    Returns
+    -------
+    Array
+        Antena gain for each point in pos.
+
+    """
+    npoints = pos.shape[0]
+
+    return jnp.zeros((npoints, npoints))

mapc_sim/experimental/walls.py

@@ -0,0 +1,20 @@
+import chex
+import jax.numpy as jnp
+
+
+def free_space(pos: chex.Array) -> chex.Array:
+    """Calculate free space loss
+
+    Parameters
+    ----------
+    pos: Array
+        An array of 2d positions for `n` points
+
+    Returns
+    -------
+        Transmission loss for every pair of points in `pos`.
+
+    """
+    npoints = pos.shape[0]
+
+    return jnp.zeros((npoints, npoints))

mapc_sim/sim.py

@@ -12,7 +12,7 @@
 
 
 def network_data_rate(key: PRNGKey, tx: Array, pos: Array, mcs: Array, tx_power: Array, sigma: Scalar,
-                      walls: Array, return_sample: bool = False) -> Union[Scalar,tuple]:
+                      loss_gain: Array, return_sample: bool = False) -> Union[Scalar,tuple]:
     r"""
     Calculates the aggregated effective data rate based on the nodes' positions, MCS, and tx power.
     Channel is modeled using TGax channel model with additive white Gaussian noise. Effective
@@ -41,9 +41,9 @@ def network_data_rate(key: PRNGKey, tx: Array, pos: Array, mcs: Array, tx_power:
         Transmission power of the nodes. Each entry corresponds to the transmission power of the transmitting node.
     sigma: Scalar
         Standard deviation of the additive white Gaussian noise.
-    walls: Array
-        Adjacency matrix of walls. If node i is separated from node j by a wall,
-        then `walls[i, j] = 1`, otherwise `walls[i, j] = 0`.
+    loss_gain: Array
+        Adjacency matrix of combined loss and gain. If node `i` is separated from node `j` by a wall or the antena is directional,
+        then `loss_gain[i, j] = val`, otherwise `walls[i, j] = 0`.
     return_sample: bool
         A flag indicating whether the simulator returns raw number of transmitted frames.
 
@@ -59,7 +59,7 @@ def network_data_rate(key: PRNGKey, tx: Array, pos: Array, mcs: Array, tx_power:
     distance = jnp.sqrt(jnp.sum((pos[:, None, :] - pos[None, ...]) ** 2, axis=-1))
     distance = jnp.clip(distance, REFERENCE_DISTANCE, None)
 
-    signal_power = tx_power[:, None] - tgax_path_loss(distance, walls)
+    signal_power = tx_power[:, None] - tgax_path_loss(distance, loss_gain)
 
     interference_matrix = jnp.ones_like(tx) * tx.sum(axis=0) * tx.sum(axis=1, keepdims=True) * (1 - tx)
     a = jnp.concatenate([signal_power, jnp.full((1, signal_power.shape[1]), fill_value=NOISE_FLOOR)], axis=0)

mapc_sim/utils.py

@@ -1,19 +1,18 @@
 import jax
-import jax.numpy as jnp
 from chex import Array
 
 from mapc_sim.constants import *
 
 
-def tgax_path_loss(distance: Array, walls: Array) -> Array:
+def tgax_path_loss(distance: Array, loss_gain: Array) -> Array:
     r"""
     Calculates the path loss according to the TGax channel model [1]_.
 
     Parameters
     ----------
     distance: Array
         Distance between nodes
-    walls: Array
+    loss_gain: Array
         Adjacency matrix describing walls between nodes (1 if there is a wall, 0 otherwise).
 
     Returns
@@ -27,7 +26,7 @@ def tgax_path_loss(distance: Array, walls: Array) -> Array:
     """
 
     return (40.05 + 20 * jnp.log10((jnp.minimum(distance, BREAKING_POINT) * CENTRAL_FREQUENCY) / 2.4) +
-            (distance > BREAKING_POINT) * 35 * jnp.log10(distance / BREAKING_POINT) + WALL_LOSS * walls)
+            (distance > BREAKING_POINT) * 35 * jnp.log10(distance / BREAKING_POINT) + loss_gain)
 
 
 def logsumexp_db(a: Array, b: Array) -> Array:

pyproject.toml

@@ -17,11 +17,11 @@ authors = [
 
 requires-python = ">=3.9"
 dependencies = [
-  "chex~=0.1.85",
-  "jax~=0.4.23",
-  "jaxlib~=0.4.23",
-  "matplotlib~=3.8.2",
-  "tensorflow-probability[jax]~=0.23.0"
+  "chex~=0.1.86",
+  "jax~=0.4.30",
+  "jaxlib~=0.4.30",
+  "matplotlib~=3.9.1",
+  "tensorflow-probability[jax]~=0.24.0"
 ]
 
 [build-system]