固定模式编码

题目概述

设信源可能输出的符号是26个字母，且每个字母出现的概率为：a, b, c, d, e, f 均为0.1，其它是等概的，试编写程序可以对任意字母序列（如presentation）进行固定模式的算术编码，并进行相应的译码。

题目分析

通过对题目分析可知，此题目是对（a~z）26个字母进行固定模式编码，译码。可以分为三部分，输入字符，编码，译码。

输入字符

可以创建一个一维数组（可以选择大小确定的数组，也可以使用指针创建动态数组，为了减小难度，选择大小确定的数组），通过cin输入字符。
循环数组，当为0时（注意不同于‘0’），结束循环，得到数组长度。
保证程序的健壮性，要对输入进行合法性判断，通过if语句依次判断是否在a~z， A-Z之间。

编码

判断每次的字符是什么，然后每次该字符的概率依次叠加，得到该字符串最后的概率pp。
通过公式求得最低位数digit。
并通过一个for循环，每次划分新的区间，得到最后的范围。
选取上界值然后减去pp的0.01，得到近似概率。
不断*2，得到二进制编码，直到达到最低位数digit。

译码

先将二进制编码转为十进制。
依次从0倒1选取区间，判断是否在相应区间内，如果在就输出相应字符。
知道字符数与输入相同。

问题分析：

1.精度问题，当输入过长，long double的精度以及不够用了，所以我选择使用二维数组存储字符，编译译码也对每一组进行编译译码，理论上可以达到无穷精度。
2.方法复杂度是n²,待改进。
3.耦合度过高，所以我将运算封装到函数中，来降低耦合度。

运行图片

代码

#include <iostream>
#include <string>
#include <cstdlib>
#include <math.h>
using namespace std;
/*
	created by Yu Zhao;
	date:2019.04.30;
	this project can decode from zero to 1600 in threoty;
	but when have too many 'z', it will has a error in result;
	contrast with old one:
		this project's code method is sample;

		but include the method to function;
		increase the length of code from 10 to 1600;
		reduce the length of code from 378 to 135;
*/
char str[1000];
char str_end[200][8];
char str_se_end[200][50];
int len_test[200];
int len;
int digit_end[200];
long double pp_end;
//char se_str_start[1000];								//输出显示的长度
long double p_abc[26] = {0.1,0.1,0.1,0.1,0.1,0.1,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02};
long double scale[27] = {0.0,0.1,0.2,0.3,0.4,0.5,0.60,0.62,0.64,0.66,0.68,0.70,0.72,0.74,0.76,0.78,0.80,0.82,0.84,0.86,0.88,0.90,0.92,0.94,0.96,0.98,1.00};
char letter[26] = {'a','b','c','d','e','f','g','h','i','j','k','l','m','n','o','p','q','r','s','t','u','v','w','x','y','z'};
bool input(){
	cout<<"********************输入********************"<<endl;
	int i = 0;
	cout <<"please input the liter(from a to z)\n";
	cin >> str;													//输入目标字符串
	while(str[i]!=0 ){											//读取长度——有一个问题
		len++;
		i++;	
	}
	for(i=0;i<len;i++){											//判断输入是否合理
		if(str[i]>='a'&&str[i]<='z'||str[i]>='A'&&str[i]<='Z'){
			
		}
		else {
			cout << "your liter is wrong"<<endl;
			exit(1);
			return false;
		}
	}
	i = 0;
	cout <<"the liters you put is :";
	for(int j = 0;j<((len/8)+1);j++){							//转二维数组
		
		for(int k = 0;k<8;k++){
			if(j<len/8){
				str_end[j][k] = str[i];
				i++;
				cout<<str_end[j][k];
				len_test[j]=8;
			}else{
				str_end[j][k] = str[i];
				i++;
				cout<<str_end[j][k];
				len_test[j]=len%8; 
				if(i ==len%8 )
					break;
			}		
		}
	}
	cout<<endl;	
	cout<<"your liter is right"<<endl;
	cout <<"the liters' length is :"<< len<<endl;
	return true;
}
void decode(int len,char abc[],int count){
	long double top = 0;
	long double bottom = 0;
	long double pp = 1;
	int digit;
	int flag;
	long double value = 0;
	char se_str_start[1000];
	for(int i = 0;i<len;i++){									//计算最后概率及区间
		flag = (int)(abc[i]-97);
		bottom = bottom + scale[flag]*pp;
		pp = pp*p_abc[flag];
		top = bottom + pp;
	}
	digit = ceil((log10(1/pp))/0.301);							//确定最低位数
	pp_end = pp_end*pp;
	digit_end[count] = digit;
	value = top-0.001*pp;												//取值
	for(i = 0;i<digit;i++){											//转为二进制
		value = value*2;
		se_str_start[i] = (char)((int)value+48);								//转为字符数组
		str_se_end[count][i]=(char)((int)value+48);
		if(value>=1)
			value = value-(int)value;
		cout<<se_str_start[i];
	}
}
void record(int se_len,char se_str[],int count){
	int i = 0;
	long double start = 0;
	int j = 0;
	long double value = 0;
	long double pp = 1;
	long double bottom = 0;
	long double top = 0;
	pp_end = 1;
	top = bottom+scale[j+1]*pp;
	bottom = bottom+pp*scale[j];
	for(i=0;i<se_len;i++)														//转为十进制
		value = value+((int)(se_str[i])-48)*pow(2,(-(i+1)));
	pp_end = pp_end*value;
	for(i = 0;i<len_test[count];i++)														//搜寻区间并输出————有问题
		for(j=0;j<26;j++){
			top = start+scale[j+1]*pp;
			bottom = start+pp*scale[j];
			if((value>=bottom)&&(value<top)){
				cout<<letter[j];
				pp = top-bottom;
				start = bottom;	
				break;
			}
		}
}
int main(){
	pp_end = 1;
	input();
	int digit_length=0;
	cout<<"********************编码********************"<<endl;
	cout<<"the end of jiema is :";
	for(int i=0;i<=len/8;i++){
		decode(len_test[i],str_end[i],i);
		digit_length=digit_length+(int)digit_end[i];
		if(len_test[i]==0)
			break;
	}
	cout<<endl;
	cout<<"the digit is :"<<"	"<<digit_length<<endl;
	cout<<"the length of the section is :"<<"	"<<pp_end<<endl;
	cout<<"********************译码********************"<<endl;
	cout<<"the end is :";
	for(i = 0;i<=len/8;i++)
		record(digit_end[i],str_se_end[i],i);
	cout<<endl;
	cout<<"the value is :"<<pp_end<<endl;
	return 0 ;
}

其他

另：因为代码是自己一个人写的，所以有很多不完善的地方，请谅解。切忌照抄
转载请说明出处