plot_soft.py

import matplotlib.pyplot as plt
import pandas as pd
import os
import numpy as np
from scipy.special import erfc
from datetime import datetime  # 新增：用于获取时间

# ================= 用户配置区域 =================

# 路径配置
csv_dir = "csv" 
img_dir = "pictures"

# 自定义绘图参数
PLOT_CONFIG = {
    # 1. 输出文件名
    # 若设为 None，则自动生成格式： "曲线名1_曲线名2_日期_时间.png"
    # 若设为字符串 (如 "my_plot.png")，则直接使用该文件名
    "output_filename": "Soft Decision Viterbi",

    # 2. 图表标题
    # 软判决，消息长度65536，帧数10000 比较不同卷积码的性能
    "title": "Soft Decision Viterbi Decoding for Different Convolutional Codes\nMessage Length = 65536, Frames = 10000",
    
    # 3. X轴范围 (min, max)，设置为 None 则自动调整
    "xlim": None, 
    
    # 4. Y轴范围 (min, max)，设置为 None 则自动调整
    "ylim": None  
}

# 要画的曲线 (Label : Filename)
curves = {
    'Conv. $(7,5)_8$ Soft Viterbi': "soft_75.csv",
    'Conv. $(15,13)_8$ Soft Viterbi': "soft_1513.csv",
    'Conv. $(1,5/7)_8$ Soft Viterbi': "soft_rsc.csv",
    "Uncoded BPSK (Theory)": "theory"
}

# =================================================

def set_academic_style():
    """设置学术画图风格 (Times New Roman, Serif)"""
    plt.rcParams.update(plt.rcParamsDefault)
    plt.style.use('classic')
    
    plt.rcParams['font.family'] = 'serif'
    plt.rcParams['font.serif'] = ['Times New Roman']
    plt.rcParams['font.size'] = 14
    plt.rcParams['axes.unicode_minus'] = False
    plt.rcParams['lines.linewidth'] = 1.5
    plt.rcParams['axes.linewidth'] = 1.5
    plt.rcParams['grid.linestyle'] = ':'
    plt.rcParams['grid.linewidth'] = 0.5
    plt.rcParams['grid.alpha'] = 0.5

def get_data_bounds(curve_dict, csv_directory):
    """预读取数据以确定自动坐标轴范围"""
    min_snr, max_snr = 0, 10
    min_ber = 1e-5
    
    snr_values = []
    ber_values = []
    
    for label, filename in curve_dict.items():
        if filename == "theory":
            continue
        filepath = os.path.join(csv_directory, filename)
        if os.path.exists(filepath):
            try:
                df = pd.read_csv(filepath)
                valid_df = df[(df['BER'] > 0) & (df['BER'] <= 1.0)]
                if not valid_df.empty:
                    snr_values.extend(valid_df['SNR'].values)
                    ber_values.extend(valid_df['BER'].values)
            except:
                pass
    
    if snr_values:
        min_snr = np.min(snr_values)
        max_snr = np.max(snr_values)
    if ber_values:
        min_ber = np.min(ber_values)
    
    return (min_snr, max_snr), min_ber

def plot_ber():
    set_academic_style()
    
    colors = ['black', 'black', 'black', 'black', 'black']
    markers = ['o', '^', 's', 'D', 'v', 'x']
    linestyles = ['-', '--', '-', '--']
    
    idx = 0 
    plotted_curves_names = [] # 记录实际画了哪些线，用于文件名生成

    fig, ax = plt.subplots(figsize=(8, 6))

    # --- 1. 确定坐标轴范围 ---
    (data_min_snr, data_max_snr), data_min_ber = get_data_bounds(curves, csv_dir)
    
    if PLOT_CONFIG["xlim"] is None:
        x_start = np.floor(data_min_snr)-0.2
        x_end = np.ceil(data_max_snr) + 1
        current_xlim = (x_start, x_end)
    else:
        current_xlim = PLOT_CONFIG["xlim"]

    if PLOT_CONFIG["ylim"] is None:
        y_bottom = pow(10, np.floor(np.log10(data_min_ber)) - 1)
        current_ylim = (y_bottom, 1.0)
    else:
        current_ylim = PLOT_CONFIG["ylim"]

    # --- 2. 绘制理论线 ---
    if "Uncoded BPSK (Theory)" in curves:
        snr_db_theory = np.linspace(current_xlim[0] + 0.2, current_xlim[1], 300)
        snr_linear = 10 ** (snr_db_theory / 10.0)
        ber_theory = 0.5 * erfc(np.sqrt(snr_linear))
        
        ax.semilogy(snr_db_theory, ber_theory, 
                    color='black', linestyle='-', linewidth=1.5, 
                    label='Uncoded BPSK (Theory)', zorder=1)
        # 理论线一般不计入文件名，除非你想加

    # --- 3. 绘制仿真线 ---
    for label, filename in curves.items():
        if filename == "theory":
            continue
            
        filepath = os.path.join(csv_dir, filename)
        if os.path.exists(filepath):
            try:
                df = pd.read_csv(filepath)
                df = df[df['BER'] > 0]
                
                c = colors[idx % len(colors)]
                m = markers[idx % len(markers)]
                ls = linestyles[idx % len(linestyles)]
                
                ax.semilogy(df['SNR'], df['BER'], 
                             marker=m, 
                             markersize=7, 
                             markeredgewidth=1.0,
                             markerfacecolor='none', 
                             markeredgecolor='black',
                             linestyle=ls, 
                             linewidth=1.5,
                             color=c, 
                             label=label, 
                             zorder=10,
                             clip_on=False) 
                
                idx += 1
                plotted_curves_names.append(label) # 记录成功绘制的曲线名
                
            except Exception as e:
                print(f"Error reading {filename}: {e}")

    # --- 4. 装饰 ---
    if PLOT_CONFIG["title"]:
        ax.set_title(PLOT_CONFIG["title"], pad=15)
        
    ax.set_xlabel(r'$E_b/N_0\ (\mathrm{dB})$')
    ax.set_ylabel('Bit Error Rate (BER)')
    
    ax.minorticks_on()
    ax.grid(which='major', linestyle='-', linewidth=0.5, color='gray', alpha=0.5)
    ax.grid(which='minor', linestyle=':', linewidth=0.5, color='gray', alpha=0.3)

    ax.set_xlim(current_xlim)
    ax.set_ylim(current_ylim)

    legend = ax.legend(loc='best', frameon=True, fontsize=12)
    legend.get_frame().set_edgecolor('black')
    legend.get_frame().set_linewidth(0.8)
    legend.get_frame().set_alpha(1.0)

    # --- 5. 保存文件 (处理文件名逻辑) ---
    if not os.path.exists(img_dir):
        os.makedirs(img_dir)
    
    # 核心逻辑：判断是否自定义文件名
    if PLOT_CONFIG["output_filename"]:
        final_filename = PLOT_CONFIG["output_filename"]
    else:
        # 默认：曲线名拼接 + 时间戳
        # 去除空格，防止文件名中有空格
        clean_names = [name.replace(" ", "") for name in plotted_curves_names]
        # 如果曲线太多，名字会很长，这里可以截断或者直接连接
        names_str = "_".join(clean_names)
        
        # 获取当前时间
        time_str = datetime.now().strftime("%Y%m%d_%H%M%S")
        
        # 拼接
        if names_str:
            final_filename = f"{names_str}_{time_str}.png"
        else:
            final_filename = f"BER_Plot_{time_str}.png" # 防止没画任何曲线的情况

    save_path = os.path.join(img_dir, final_filename)
    plt.savefig(save_path, dpi=600, bbox_inches='tight')
    print(f"Plot saved to: {save_path}")
    
    plt.show()

if __name__ == "__main__":
    plot_ber()