The performance of the proposed scheme is compared to the theoretical limit given by the source entropy (considering both the temperatures and the differences of the temperatures) and to the performance of the LEC  and ALFC  algorithms.
In fact, because those datasets tend to contain long sequences of measurements that convey very little information, none of the methods under consideration can achieve high efficiency for sets of extremely low entropy (ALFC achieves approximately 10% lower efficiency than the proposed approach for Sets 8 and 9).
ALFC assumes that the measurements are transmitted in fixed-length packets which contain enough information so that the measurements within the packet can be decoded regardless of communication failures that may have caused previous packet losses.
We compare the performance of our approach to that of ALFC using the same set of parameters employed in the experimental evaluation presented in .
Figure 4 shows the average symbol length (L) after compression using ALFC and our proposed approach.
It also outperformed ALFC's best case compression in every scenario, except when the dictionary generated by Set 3 is used to compress the symbols in sets with very low entropy.
As we did for the temperature datasets, we also compared the performance of our approach to that of ALFC using the same set of parameters used in Section 4.2.