No abstract

This paper gives performance evaluation of wearable wireless body area networks (WBANs) during walking motion. In order to evaluate the performance, received signal strength (RSS), packet error rate (PER), and bit error rate (BER) are measured in an anechoic chamber and an office room. This measurement is conducted in the frequency band of 444.5 and 2450 MHz by using GFSK signal with symbol rate of 1 MHz. The results show that in the anechoic chamber the WBAN using the 444.5 MHz enables to provide error-free communication, on the other hand, the WBAN operated in the 2450 MHz faces packet errors. Measurement results in the office room give comparable performance between these frequencies. From these observations, the use of 2450 MHz for wearable WBANs needs reflection waves in order to compensate a shadowing effect caused by the human body using the WBAN.

No abstract

A wireless technology is required to realize robust transmission of medical images like a radiography image over noisy environment. The use of error correction technique is essential for realizing such a reliable communication, in which a suitable channel coding is introduced to correct erroneous bits caused by passing through a noisy channel. However, the use of a channel code decreases its efficiency because redundancy bits are also transmitted with information bits. This paper presents a joint source-channel coding which maintains the channel efficiency during transmission of medical images like a radiography image. As medical images under the test, we use typical radiography images in this paper. The joint coding technique enjoys correlations between pixels of the radiography image. The results show that the proposed joint coding provides capability to correcting erroneous bits without increasing the redundancy of the codeword.

This paper proposes a joint source-channel coding technology to transmit periodic vital information such as an electrocardiogram (ECG). It shows the iterative decoding method using a correlation value which can be obtained from ECG periodicity as Side-Information. The improvement is shown when bit strings are transmitted with different encoding rates. Because the proposed method has an error correcting system and makes a code processing decrease, some miniaturization and energy-saving can be expected of the equipment. Finally, the effectiveness of the proposed method is shown by comparisons with Differential Pulse Code Modulation, which is a typical compression method of ECG, and with the no coding method.