Fixed Readme English

Author: JovenSoh

README.md

@@ -10,7 +10,7 @@ BPSK       |Change in phase of signal by a magnitude of 𝜋
 LFM        |Constant change in frequency
 
 The neural network will learn to classify 3 different input signals, Continuous Wave (CW), BPSK and LFM.
-This is achieved through Keras 1D convolutional networks.  <br>
+This is achieved through time-series analysis using Keras 1D convolutional networks.  <br>
 
 ### Test Conditions
 Using Scipy and Numpy, signals of various phase shifts and sweep rates were synthesised.
@@ -30,13 +30,14 @@ After 200 epochs, the accuracy was around 90%, this may be attributed to the ide
 ![alt text](Accuracy.png "Accuracy of Neural Net") ![alt text](Loss.png "Is this loss?")
 
 The best weights from model training was saved and used to evaluate against a test set of signals with similar conditions. <br>
-The model's performance is optimal. However, there is some attrition in misclassifying a BPSK signal as a CW signal. <br>
-This may be due to the similarities of the two signals in the time-domain. Hence, it is recommended to explore the performance of the model in the frequency domain using FFT. <br> 
+The model's performance is optimal. However, there is some attrition in differentiating a BPSK signal from a CW signal. <br>
+This may be due to the similarities of the two signals in the time-domain. Hence, it is recommended to study the performance of the model by training it with data in the frequency domain. <br> 
+
 ![alt text](Confusion-Matrix.png "Confusion matrix of model evaluation")
 
 ## Conclusion
-In conclusion, a neural net to classify the modualtion of the signal is developed.
-Further improvements can be made by training the data on more unique data such as real signal environments.
+In conclusion, a neural net to classify the modulation of the signal was developed.
+Further improvements can be made by training the model on more unique data such as real signal environments.
 Work could also be done to measure the sensitivity of the neural net and measures its performance over 
 different SNR and sweeps rates to study the neural net's limitations and suggest improvements. <br>