Add performance report and fix learning rate parameter in DQN model

Author: csshafiqahmed

Research Article/AIDAS__AI_Enhanced_Intrusion_Detection_and_Authentication_for_Autonomous_Vehicles.pdf

@@ -0,0 +1,31 @@
+{
+  "protocol_performance": {
+    "total_sessions": 50,
+    "successful_authentications": 38,
+    "success_rate": 76.0,
+    "average_latency_ms": 90.78857421875,
+    "median_latency_ms": 77.34155654907227,
+    "latency_std_ms": 65.13708701691993,
+    "min_latency_ms": 72.3731517791748,
+    "max_latency_ms": 539.2971038818359
+  },
+  "security_metrics": {
+    "detection_accuracy": 97.8,
+    "false_positive_rate": 1.2,
+    "security_incidents": 12,
+    "puf_reliability": 97.3,
+    "encryption_strength": "AES-256 + ECC-256"
+  },
+  "computational_efficiency": {
+    "overhead_reduction": "31.25%",
+    "communication_overhead_bits": 2176,
+    "inference_latency_ms": 4.2,
+    "energy_efficiency_improvement": "53.2%"
+  },
+  "ai_enhancement": {
+    "dqn_convergence_time_s": 1.2,
+    "adaptive_threshold_adjustments": "\u00b117.8%",
+    "policy_update_frequency_ms": 250,
+    "learning_rate": 0.001
+  }
+}

aidas_performance_report_1749007231.json

@@ -125,7 +125,7 @@ def _build_model(self):
             tf.keras.layers.Dense(32, activation='relu'),
             tf.keras.layers.Dense(self.action_dim, activation='linear')
         ])
-        model.compile(optimizer=tf.keras.optimizers.Adam(lr=self.learning_rate), loss='mse')
+        model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=self.learning_rate), loss='mse')
         return model
 
     def update_target_model(self):