I am taking temperature, pressure and humidity data from weather sensor want to predict the data by applying adaptive filter as predictor model using Least Mean Square (LMS) algorithm to separately predicting these quantities.
I have just started working on adaptive filter so any suggestions would the helpful to me.
Thanks in advance.
Yes you can and it's very easy and simple. Since you are new to adaptive filtering you may need to know why it's useful to use LMS algorithm for wind prediction.
The nature of the wind data is non-stationary ( time series, the values of the data keep changing with respect to the time).The LMS algorithm have the ability to adjust it's coefficients according to the changes of the statistics of the input signal.
Now the question is how many input signal you want to process at the same time? Because the LMS algorithm have the ability to process only one input at the same time. if you want to process two or three you can use CLMS and QLMS respectively. If you want to know why the LMS is not efficient for prediction more than one input at the same time let me know.
Haydar Imad Mohammed , Thank you for the answer.
I am only using one quantity at a time.
For now I have used temperature data of total 44 samples as input d(n) , delay factor is 5 , step size is = 0.001 , using single tap adaptive filter. In Figure 1 attached, it can be seen that around 22nd sample I am getting optimized co-efficient w and the my output y(n) is nearly converged and error e(n) is minimized but here I am facing a problem that when I reach 44th sample I further get no output seen in Figure 2. Here I want adaptive filter to predict 45th sample and so on ( if possible ).
How can I predict future values from this system ? You previously answered that it can be done but here I am stuck.
Any help would be beneficial for me.
Thanks.
The temerature is similar to any other data you can use adaptive filter
Zainab Noori Yes I agree to your answer but can you share an example how any data is being predicted using adaptive filter.
- Badges
- Science topic
- Similar topics
- Engineering
- Electrical Engineering
- Signal Processing