What is the equivalent for a Hidden Markov Model in the WEKA toolkit?

问题

I need to classify a datastream which comes from a sensor network consisting of 8 accelerometers. Each accelerometer gives me a X Y and Z value. Thus at each sample i have 8 x 3 = 24 acceleration values. I sample at about 30 hz and the performance time is about 0.5 seconds.

At first i thought of using a Hidden Markov model for this but it seems that the WEKA toolkit does not provide such a thing. What is the WEKA equivalent for this?

Thank you.

EDIT: how to format data?

I have collected data and now i want to use the HMMWeka for classification. On the website it states that

Data instances must have a single, Nominal, class attribute and a single, relational, sequence attribute. The instances in this relational attibute may either consist of single, nominal data instances (in the case of discrete HMMs) or multivariate, numeric attributes (in the case of gaussian HMMs).

but i am confused how i would need to present my data in order to meet this requirement.

My data looks like this:

"GESTURE A",[{407 318 425};...{451 467 358};{427 525 445};][{440 342 456}...;{432 530 449};]
"GESTURE A",[{406 318 424};...{450 467 357};{422 525 445};][{440 342 456}...;{428 531 449};]
"GESTURE B",[{407 318 424};...{449 466 357};{423 524 445};][{440 342 456}...;{429 530 449};]
"GESTURE B",[{380 299 399};...{424 438 338};{404 500 426};][{433 337 449}...;{429 529 449};]

So for every gesture i have several examples. The data between {} represents X Y Z values of one accelerometer. The overaching [ ... ] represents one sample containing 8 accelerometers.

I don't understand how i can create an ARFF file from this?

Thank you