The power system operation involves regular monitoring of various parameters including voltage, current, system frequency etc. at regular intervals. This huge volume of data storage and transfer is a challenging task, particularly when lower level microcomputers are employed as field devices. The increased computational burden of entropy based compression algorithms and requirement of the probability distribution of symbols makes it difficult to implement for data acquisition. Differential Binary Encoded Algorithm (DBEA) gives a high compression ratio with repetitive, slow varying data array. But it is not suitable for the GSM communication scheme where printable character transmission is possible. An improved algorithm, Modified DBEA (M-DBEA) is developed which not only extends the range of DBEA but also gives a character string containing printable characters only. The results obtained by M-DBEA with practical data values are encouraging and thus implemented at the microcontroller level to develop a handshaking based compressed temperature data transfer system using Arduino UNO microcontrollers. When suitable sensors are used, it is possible to extend the work to develop a smart data management system for compressing real time parameter values collected over the finite time duration and transmit the compressed information through any suitable wired or wireless communication scheme.