GNU Radio Manual and C++ API Reference
3.9.1.0
The Free & Open Software Radio Ecosystem
cvsd_decode_bs.h
Go to the documentation of this file.
1
/* -*- c++ -*- */
2
/*
3
* Copyright 2007,2011,2013 Free Software Foundation, Inc.
4
*
5
* This file is part of GNU Radio
6
*
7
* SPDX-License-Identifier: GPL-3.0-or-later
8
*
9
*/
10
11
#ifndef INCLUDED_VOCODER_CVSD_DECODE_BS_H
12
#define INCLUDED_VOCODER_CVSD_DECODE_BS_H
13
14
#include <
gnuradio/sync_interpolator.h
>
15
#include <
gnuradio/vocoder/api.h
>
16
17
namespace
gr
{
18
namespace
vocoder {
19
20
/*!
21
* \brief This block performs CVSD audio decoding. Its design and
22
* implementation is modeled after the CVSD encoder/decoder
23
* specifications defined in the Bluetooth standard.
24
* \ingroup audio_blk
25
*
26
* \details
27
* CVSD is a method for encoding speech that seeks to reduce the
28
* bandwidth required for digital voice transmission. CVSD takes
29
* advantage of strong correlation between samples, quantizing the
30
* difference in amplitude between two consecutive samples. This
31
* difference requires fewer quantization levels as compared to
32
* other methods that quantize the actual amplitude level,
33
* reducing the bandwidth. CVSD employs a two level quantizer
34
* (one bit) and an adaptive algorithm that allows for continuous
35
* step size adjustment.
36
*
37
* The coder can represent low amplitude signals with accuracy
38
* without sacrificing performance on large amplitude signals, a
39
* trade off that occurs in some non-adaptive modulations.
40
*
41
* The CVSD decoder effectively provides 1-to-8 decompression.
42
* More specifically, for each incoming input bit, the decoder
43
* outputs one audio sample. If the input is a "1" bit, the
44
* internal reference is increased appropriately and then
45
* outputted as the next estimated audio sample. If the input is
46
* a "0" bit, the internal reference is decreased appropriately
47
* and then likewise outputted as the next estimated audio sample.
48
* Grouping 8 input bits together, the encoder essentially
49
* produces 8 output audio samples for everyone one input byte.
50
*
51
* This decoder requires that output audio samples are 2-byte
52
* short signed integers. The result bandwidth conversion,
53
* therefore, is 1 byte of encoded audio data to 16 output bytes
54
* of raw audio data.
55
*
56
* The CVSD decoder module must be post-fixed by a down-converter
57
* to under-sample the audio data after decoding. The Bluetooth
58
* standard specifically calls for a 8-to-1 decimating
59
* down-converter. This is required so that so that output
60
* sampling rate equals the original input sampling rate present
61
* before the encoder. In all cases, the output down-converter
62
* rate must be the inverse of the input up-converter rate before
63
* the CVSD encoder.
64
*
65
* References:
66
*
67
* 1. Continuously Variable Slope Delta Modulation (CVSD) A Tutorial,
68
* Available:
69
* http://www.eetkorea.com/ARTICLES/2003AUG/A/2003AUG29_NTEK_RFD_AN02.PDF.
70
*
71
* 2. Specification of The Bluetooth System
72
* Available: http://grouper.ieee.org/groups/802/15/Bluetooth/core_10_b.pdf.
73
*
74
* 3. McGarrity, S., Bluetooth Full Duplex Voice and Data Transmission. 2002.
75
* Bluetooth Voice Simulink Model, Available:
76
* http://www.mathworks.com/company/newsletters/digest/nov01/bluetooth.html
77
*/
78
class
VOCODER_API
cvsd_decode_bs
:
virtual
public
sync_interpolator
79
{
80
public
:
81
// gr::vocoder::cvsd_decode_bs::sptr
82
typedef
std::shared_ptr<cvsd_decode_bs>
sptr
;
83
84
/*!
85
* \brief Constructor parameters to initialize the CVSD decoder.
86
* The default values are modeled after the Bluetooth standard
87
* and should not be changed, except by an advanced user
88
*
89
* \param min_step Minimum step size used to update the internal reference.
90
* Default: "10"
91
* \param max_step Maximum step size used to update the internal reference.
92
* Default: "1280"
93
* \param step_decay Decay factor applied to step size when there is not a run of J
94
* output 1s or 0s. Default: "0.9990234375" (i.e. 1-1/1024) \param accum_decay Decay
95
* factor applied to the internal reference during every iteration of the codec.
96
* Default: "0.96875" (i.e. 1-1/32)
97
* \param K Size of shift register; the number of output bits remembered
98
* by codec (must be <= to 32). Default: "32" \param J Number of bits in
99
* the shift register that are equal; i.e. the size of a run of 1s, 0s. Default: "4"
100
* \param pos_accum_max Maximum integer value allowed for the internal reference.
101
* Default: "32767" (2^15 - 1 or MAXSHORT)
102
* \param neg_accum_max Minimum integer value allowed for the internal reference.
103
* Default: "-32767" (-2^15 + 1 or MINSHORT+1)
104
*/
105
static
sptr
make(
short
min_step = 10,
106
short
max_step = 1280,
107
double
step_decay = 0.9990234375,
108
double
accum_decay = 0.96875,
109
int
K = 32,
110
int
J = 4,
111
short
pos_accum_max = 32767,
112
short
neg_accum_max = -32767);
113
114
virtual
short
min_step() = 0;
115
virtual
short
max_step() = 0;
116
virtual
double
step_decay() = 0;
117
virtual
double
accum_decay() = 0;
118
virtual
int
K() = 0;
119
virtual
int
J() = 0;
120
virtual
short
pos_accum_max() = 0;
121
virtual
short
neg_accum_max() = 0;
122
};
123
124
}
/* namespace vocoder */
125
}
/* namespace gr */
126
127
#endif
/* INCLUDED_VOCODER_CVSD_DECODE_BS_H */
gr::vocoder::cvsd_decode_bs::sptr
std::shared_ptr< cvsd_decode_bs > sptr
Definition:
cvsd_decode_bs.h:82
gr::vocoder::cvsd_decode_bs
This block performs CVSD audio decoding. Its design and implementation is modeled after the CVSD enco...
Definition:
cvsd_decode_bs.h:78
api.h
gr::sync_interpolator
synchronous 1:N input to output with history
Definition:
sync_interpolator.h:25
sync_interpolator.h
VOCODER_API
#define VOCODER_API
Definition:
gr-vocoder/include/gnuradio/vocoder/api.h:18
gr
GNU Radio logging wrapper for log4cpp library (C++ port of log4j)
Definition:
basic_block.h:29
gr-vocoder
include
gnuradio
vocoder
cvsd_decode_bs.h
Generated by
1.8.17