Look-up table for converting N-bit digitized to floating point numbers. More...

#include <BitTable.h>

Inheritance diagram for dsp::BitTable:

Public Types
enum	Type { OffsetBinary, TwosComplement, SignMagnitude }
	Interpretation of the bits in each value.

enum	Order { MostToLeast, LeastToMost }
	The order of values in each byte (bit significance)

Public Member Functions
	BitTable (unsigned nbit, Type type, bool reverse_bits=false)
	Constructor.

virtual	~BitTable ()
	Destructor.

void	set_effective_nbit (unsigned bits)
	Set the effective number of bits. More...

unsigned	get_effective_nbit () const

void	set_order (Order)
	Set the order of the samples in each byte.

Order	get_order () const

void	set_zero_is_a_threshold (bool)
	Set to true if zero is a threshold; set to false if zero is a value. More...

bool	get_zero_is_a_threshold () const

unsigned	get_values_per_byte () const
	Get the number of floating point values per byte, 8/N.

unsigned	get_unique_values () const
	Get the number of unique values of an N-bit integer, 2^N.

double	get_scale () const
	Get the scale factor used to normalize the variance to unity.

Type	get_type () const
	Return the digitization convention.

const float *	get_values (unsigned byte=0) const
	Returns pointer to values_per_byte floats represented by byte.

void	generate (float *table) const
	Generate a look-up table for conversion to floating point. More...

virtual void	generate_unique_values (float *values) const
	Generate a look-up table of unique_values floats.

virtual unsigned	extract (unsigned byte, unsigned i) const
	Extract the ith sample from byte. More...

virtual double	get_optimal_variance () const
	Return the optimal variance of normally distributed samples.

virtual double	get_nlow_threshold () const
	Return the optimal threshold closest to and less than unity.

virtual void	get_nlow_lookup (char *nlow_lookup) const
	Return the number of low voltage states in each of 256 bytes.

Public Member Functions inherited from Reference::Able
	Able (const Able &)

Able &	operator= (const Able &)

unsigned	get_reference_count () const

Public Member Functions inherited from Reference::HeapTracked
	HeapTracked (const HeapTracked &)

HeapTracked &	operator= (const HeapTracked &)

bool	__is_on_heap () const

Static Public Attributes
static const unsigned	bits_per_byte = 8
	Number of bits per bytes.

static const unsigned	unique_bytes = 1 << bits_per_byte
	Number of unique 8-bit combinations.

Protected Member Functions
void	build ()
	Build the lookup table.

void	destroy ()
	Destroy the lookup table.

Protected Member Functions inherited from Reference::Able
Handle *	__reference (bool active) const

void	__dereference (bool auto_delete=true) const

Protected Attributes
float *	table
	For each unsigned char, values of the two output voltage states.

Type	type
	Digitization convention.

const unsigned	nbit
	Number of bits.

unsigned	zero_is_a_threshold
	Zero is a threshold.

unsigned	effective_nbit
	The effective number of bits.

const bool	reverse_bits
	Reverse the order of the bits.

Order	order
	The order of the samples in each byte.

const unsigned	values_per_byte
	Number of N-bit values per byte.

const unsigned	unique_values
	Number of unique N-bit values.

const unsigned	nbit_mask
	N-bit mask.

double	scale
	The scale factor used to normalize the variance to unity.

Additional Inherited Members
Static Public Member Functions inherited from Reference::Able
static size_t	get_instance_count ()

Static Public Member Functions inherited from Reference::HeapTracked
static void *	operator new (size_t size, void *ptr=0)

static void	operator delete (void location, void ptr)

static void	operator delete (void *location)

static size_t	get_heap_queue_size ()

Detailed Description

Look-up table for converting N-bit digitized to floating point numbers.

Member Function Documentation

◆ extract()

unsigned dsp::BitTable::extract	(	unsigned	byte,
		unsigned	i
	)		const

virtual

Extract the ith sample from byte.

Each byte is treated as unique_values consecutive values, from most significant bit to least significant bit, or vice versa (depending on order).

Parameters

byte	the byte pattern containing unique_values values
i	the index of the value to extract from the byte

Reimplemented in dsp::EmerlinTwoBitTable, dsp::CODIFTwoBitTable, and dsp::VDIFTwoBitTable.

◆ generate()

void dsp::BitTable::generate ( float * table ) const

Generate a look-up table for conversion to floating point.

Parameters

table pointer to space for at least unique_bytes * values_per_byte floating point values

Referenced by dsp::TwoBitFour::get_lookup_block().

◆ set_effective_nbit()

void dsp::BitTable::set_effective_nbit ( unsigned bits )

Set the effective number of bits.

For example, a digitizer may set 8-bit thresholds to effect a 6-bit digitizer, leaving head room for RFI

◆ set_zero_is_a_threshold()

void dsp::BitTable::set_zero_is_a_threshold ( bool flag )

Set to true if zero is a threshold; set to false if zero is a value.

If zero is a threshold, then there are an equal number of +ve and -ve states and no representation of the value 0; e.g. in signed 8-bit values, the range [-128,127] is shifted by +0.5 to [-127.5,127.5] and there is no zero, only -0.5 and +0.5 on either side of it.

If zero is a value, then [-128,127] remains [-128,127] and zero remains zero.

The documentation for this class was generated from the following files:

BitTable.h
BitTable.C

Public Types

Public Member Functions

Static Public Attributes

Protected Member Functions