Abstract
Quantization of signals is required for many transmission, storage and compression applications. The original signal is quantized at the encoder side. At the decoder side, a replica of the original signal that should resemble the original signal in some sense is recovered. Present quantizers make an effort to reduce the distortion of the signal in the sense of reproduction fidelity. Consider scenarios in which signals are generated from multiple classes. The encoder focuses on the task of quantizing the data without any regards to the class of the signal. The quantized signal reaches the decoder where not only the recovery of the signal should take place but also a decision is to be made on the class of the signal based on the quantized version of the signal only. In this paper, we study the design of such scalar quantizer that is optimized for the task of classification at the decoder. We define the distortion to be the symmetric Kullback-Leibler (KL) divergence measure between the conditional probabilities of class given the signal before and after quantization. A high-rate analysis of the quantizer is presented and the optimum point density of the quantizer for minimizing the symmetric KL divergence is derived. The performance of this method on synthetically generated data is examined and observed to be superior in the task of classification of signals at the decoder.
| Original language | English |
|---|---|
| Title of host publication | 2011 Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2011 - Proceedings |
| Pages | 277-282 |
| Number of pages | 6 |
| DOIs | |
| State | Published - Apr 21 2011 |
| Event | 2011 Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2011 - Sedona, AZ, United States Duration: Jan 4 2011 → Jan 7 2011 |
Other
| Other | 2011 Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2011 |
|---|---|
| Country | United States |
| City | Sedona, AZ |
| Period | 1/4/11 → 1/7/11 |
Fingerprint
Keywords
- Classification
- High Rate Theory
- Kullback-Leibler Divergence Measure
- Quantization
ASJC Scopus subject areas
- Computer Networks and Communications
- Signal Processing
- Education
Cite this
Quantization for classification accuracy in high-rate quantizers. / Dogahe, Behzad Mohammadi; Murthi, Manohar.
2011 Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2011 - Proceedings. 2011. p. 277-282 5739225.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Quantization for classification accuracy in high-rate quantizers
AU - Dogahe, Behzad Mohammadi
AU - Murthi, Manohar
PY - 2011/4/21
Y1 - 2011/4/21
N2 - Quantization of signals is required for many transmission, storage and compression applications. The original signal is quantized at the encoder side. At the decoder side, a replica of the original signal that should resemble the original signal in some sense is recovered. Present quantizers make an effort to reduce the distortion of the signal in the sense of reproduction fidelity. Consider scenarios in which signals are generated from multiple classes. The encoder focuses on the task of quantizing the data without any regards to the class of the signal. The quantized signal reaches the decoder where not only the recovery of the signal should take place but also a decision is to be made on the class of the signal based on the quantized version of the signal only. In this paper, we study the design of such scalar quantizer that is optimized for the task of classification at the decoder. We define the distortion to be the symmetric Kullback-Leibler (KL) divergence measure between the conditional probabilities of class given the signal before and after quantization. A high-rate analysis of the quantizer is presented and the optimum point density of the quantizer for minimizing the symmetric KL divergence is derived. The performance of this method on synthetically generated data is examined and observed to be superior in the task of classification of signals at the decoder.
AB - Quantization of signals is required for many transmission, storage and compression applications. The original signal is quantized at the encoder side. At the decoder side, a replica of the original signal that should resemble the original signal in some sense is recovered. Present quantizers make an effort to reduce the distortion of the signal in the sense of reproduction fidelity. Consider scenarios in which signals are generated from multiple classes. The encoder focuses on the task of quantizing the data without any regards to the class of the signal. The quantized signal reaches the decoder where not only the recovery of the signal should take place but also a decision is to be made on the class of the signal based on the quantized version of the signal only. In this paper, we study the design of such scalar quantizer that is optimized for the task of classification at the decoder. We define the distortion to be the symmetric Kullback-Leibler (KL) divergence measure between the conditional probabilities of class given the signal before and after quantization. A high-rate analysis of the quantizer is presented and the optimum point density of the quantizer for minimizing the symmetric KL divergence is derived. The performance of this method on synthetically generated data is examined and observed to be superior in the task of classification of signals at the decoder.
KW - Classification
KW - High Rate Theory
KW - Kullback-Leibler Divergence Measure
KW - Quantization
UR - http://www.scopus.com/inward/record.url?scp=79954548820&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79954548820&partnerID=8YFLogxK
U2 - 10.1109/DSP-SPE.2011.5739225
DO - 10.1109/DSP-SPE.2011.5739225
M3 - Conference contribution
SN - 9781612842271
SP - 277
EP - 282
BT - 2011 Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2011 - Proceedings
ER -