Gaussian filter with uneven sampled data - r

Question

I have a time series which is unevenly sampled and I want to downsample it to 20Hz. I have made a moving average by binning the data points when in 0.05s time windows (20Hz) and applying an arithmetic mean to it. The data frame looks like this:
Time Right Left
1 0.000000000 18.21980 30.98789
2 0.009222031 22.15157 37.18590
3 0.022511959 25.63218 42.49231
4 0.029854059 28.43851 46.57811
5 0.039320946 30.43885 49.29414
6 0.052499056 31.60561 50.67852
7 0.059612036 32.01045 50.92879
8 0.076606989 31.80335 50.34975
9 0.082647085 31.18134 49.29151
10 0.090698957 30.35415 48.09110
And the code I used for the moving average was this:
data$group_num <- floor(data$Time/0.05)
data2<-NULL
data2$Right = aggregate(data$Right,
list(group_num=data$group_num), mean)
data2$Left = aggregate(data$Left,
list(group_num=data$group_num), mean)
data2$Time = aggregate(data$Time,
list(group_num=data$group_num), mean)
However, for optimizing it I want to make rather a Gaussian filter so that the data points in the middle of the bin have more weight. I couldnt find any function that could deal with uneven sampling. Thus, I started writting a script, where I managed to give them weights.
data$weight <- ((data$Time-data$group_num*0.05)*((data$group_num+1)*0.05-data$Time))^5
I have to normalize these weights by the mean of the weights in their own bin (for instance). By trying to normalize these weights to the mean of their own group I ran into problems with too slow functions. Could anybody give me a hand with it??

Answer 1

I finally managed to do what I wanted. The key function that helped me to overcome this one was ave(). Thank god. So this is what I basically did:
data$weight <- (abs(data$Time-(data$group_num*0.05+0.025)))^(-1)
data$Norm<-ave(data$weight,data$group_num,FUN=function(x) x/sum(x))
data$Time2<- data$Time*data$Norm
data$Right2<- data$Right*data$Norm
data$Left2<- data$Left*data$Norm
data2$Time<- tapply(data$Time2, data$group_num, sum)
data2$Right<- tapply(data$Right2, data$group_num, sum)
data2$Left<- tapply(data$Left2, data$group_num, sum)
Thanks Marat Talipov for the help. From what I see in your code this could also work. But because this is just working fine and quick enough I will stay with this.