An Optimal Information-theoretic Design of Myoelectric Prosthetic Under BPSK Modulation

Author(s)

Sparsho DeFollow

School Name

Spring Valley High School

Grade Level

10th Grade

Presentation Topic

Mathematics

Presentation Type

Non-Mentored

Written Paper Award

2nd Place

Abstract

Myoelectric prosthetics are a relatively modern technology where hand movements are determined by electrical signals sent my muscles in the upper arm. While extremely promising, little progress has been made in the way of efficiency. The goal of this experiment was to determine whether Reed Solomon or Hadamard error correcting codes were more efficient at reducing errors caused during the functioning of a 20-state myoelectric prosthetic. It was hypothesized that Reed Solomon codes would perform better than Hadamard codes because of less excessive usage of checksum bits in a channel with relatively low noise. Each of the 20 states of the prosthetic were assigned a binary representation based on frequency using Huffman Compression. Mathematica was then used to simulate the effect errors had on the number of bits necessary for accurate transmission within the AWGN channel. Errors were given a binomial distribution, with variance related to entropy of the system and mean directly related to the bit error rate. Data is presented as mathematical truth, and there are no significance tests. It was determined that the hypothesis was correct. Reed Solomon codes were more efficient than Hadamard codes at all signal to noise ratios (SNR). Their efficiency, in comparison to the Hadamard codes, improved as SNR increased. The control also proved more efficient than Hadamard codes after SNR exceeded 0.4. In particular, if a crude myoelectric prosthetic were used, it would be prudent to use Reed Solomon codes rather than Hadamard codes are none at all.

Location

Founders Hall 140 B

Start Date

3-30-2019 9:15 AM

Presentation Format

Oral and Written

Group Project

No

COinS
 
Mar 30th, 9:15 AM

An Optimal Information-theoretic Design of Myoelectric Prosthetic Under BPSK Modulation

Founders Hall 140 B

Myoelectric prosthetics are a relatively modern technology where hand movements are determined by electrical signals sent my muscles in the upper arm. While extremely promising, little progress has been made in the way of efficiency. The goal of this experiment was to determine whether Reed Solomon or Hadamard error correcting codes were more efficient at reducing errors caused during the functioning of a 20-state myoelectric prosthetic. It was hypothesized that Reed Solomon codes would perform better than Hadamard codes because of less excessive usage of checksum bits in a channel with relatively low noise. Each of the 20 states of the prosthetic were assigned a binary representation based on frequency using Huffman Compression. Mathematica was then used to simulate the effect errors had on the number of bits necessary for accurate transmission within the AWGN channel. Errors were given a binomial distribution, with variance related to entropy of the system and mean directly related to the bit error rate. Data is presented as mathematical truth, and there are no significance tests. It was determined that the hypothesis was correct. Reed Solomon codes were more efficient than Hadamard codes at all signal to noise ratios (SNR). Their efficiency, in comparison to the Hadamard codes, improved as SNR increased. The control also proved more efficient than Hadamard codes after SNR exceeded 0.4. In particular, if a crude myoelectric prosthetic were used, it would be prudent to use Reed Solomon codes rather than Hadamard codes are none at all.