Simulation Results and Observations Part 1: Time-delay Estimation Problem: A: When x(n) = -0.6*x(n-1) + w(n): LMS Algorithm: Observation: The estimators converges to a value near to true value. 17
Sep 26, 2015
Simulation Results and Observations
Part 1: Time-delay Estimation Problem:
A: When x(n) = -0.6*x(n-1) + w(n):
LMS Algorithm:
Observation: The estimators converges to a value near to true value.
Observation: The graph exponentially decreases to a value near to zero.
NLMS Algorithm:
Observation: The values are neaarly converged.
Observation: Initially increased to a value 2.98 and become constant at 1.19.
RLS Algorithm:
Observation: Nearly converges to true values.
Observation: Initially increased to 2.02 and stays constant at 1.009.
Observation: Converges to a value 0.45.
B: When x(n) = 4*sin(2*pi*n/17) and (n) = 0:
LMS Algorithm:
Observation: Exactly converges to true values.
Observation: nearly converges to zero.
NLMS Algorithm:
Observation: Exactly converges to true value.
Observation: nearly converged to zero.
RLS Algorithm:
Observation: Exactly converges to the true values.
Observation: converged to zero.
Observation: Converged to zero.
Part 2: Noise Cancellation Problem:
LMS Algorithm:
Observation: Approximatly converging to true value but fluctuating.
Observation: Initially increased to 60 and stays constant at 58.
Observation: Fluctuates between -20 and 20.
Observation: Fluctuates between -12 and 12.
RLS Algorithm:
Observation: Converges nearly to true values.
Observation: Initially increases to 61 and stays constant at 51.2
Observation: Fluctuates at 2.5
Observation: Fluctuates between -20 and 20.
Observation: Fluctuates between -10 and 10.
Part 3: Noise Cancellation Problem:
LMS Algorithm:
Observation: Nearly converged to true values.
Observation: Exponentially decreased to 0.2
Observation: Fluctuates between -6 and 6
Observation: Converges neaarly to zero.
NLMS Algorithm:
Observation: Converges nearly to true values.
Observation: Initially increased to 16.3 and reduces to 6.1
Observation: Fluctuates to different values between -5 to 5.
Observation: Fluctuates between -3 and 3.
RLS Algorithm:
Observation: Converges to exact values.
Observation: Converges to zero.
Observation: Fluctuates between -7 and 7.
Observation: Converged to zero.
Observation: Nearly equal to 0.5
Part 4: RLS-Identification Problem:
A:
Observation: Exactly converged to true values.
Observation: Converged to zero.
Observation: Converged to zero.
B:
Observation: Nearly converged to true value.
Observation: Converged to 1.
Observation: Converged to 0.2
C:
Observation: Exactly converged to true values.
Observation: Converged to zero.
Observation: Converged to zero.
Part 5:
LMS Algorithm:
Observation: Converged to -3.6 which is near to true value -3.5
NLMS Algorithm:
Observation: Converged to -3.4 which is near to true value -3.5
RLS Algorithm:
Observation: Converged to -3.44 which is near to true value -3.5
39