生成10Hz时间序列：寻求高效简洁的插值方法

需求：将时间序列插值为精确的10Hz（即每0.1秒一个时间戳），原方案存在时间戳精度问题，且大数据集处理效率低。原尝试代码如下：

library(tidyverse) 
library(zoo)

options(digits.secs = 3, pillar.sigfig = 6)

data <- tribble(
  ~timestamp, ~value, 
  "09/12/2024 00:05:35.677", 139.664,
  "09/12/2024 00:05:35.776", 138.706,
  "09/12/2024 00:05:35.876", 143.348,
  "09/12/2024 00:05:35.975", 141.516,
  "09/12/2024 00:05:36.074", 136.731,
  "09/12/2024 00:05:36.174", 138.275,
  "09/12/2024 00:05:36.273", 143.015) %>%
  mutate(timestamp = mdy_hms(timestamp))

start <- min(data$timestamp) %>% round_date("0.1 sec")
end   <- max(data$timestamp) %>% round_date("0.1 sec")

data_10Hz <- data %>%
  complete(timestamp = seq.POSIXt(start, end, by = .100)) %>%
  arrange(timestamp) 

data_10Hz <- data_10Hz  %>%
  mutate(value = na.approx(value)) %>%
  filter(timestamp == round_date(timestamp, "0.1 sec"))

核心改进思路 #

• 规避POSIXct浮点精度问题：将时间转换为整数毫秒偏移量（从起始时间算起），用整数运算保证0.1秒间隔的精准性
• 跳过冗余中间步骤：直接针对目标时间点插值，避免生成大量含NA的中间行，大幅提升大数据集处理效率

高效实现代码 #

library(tidyverse)
library(zoo)

options(digits.secs = 3, pillar.sigfig = 6)

# 原始数据处理
data <- tribble(
  ~timestamp, ~value, 
  "09/12/2024 00:05:35.677", 139.664,
  "09/12/2024 00:05:35.776", 138.706,
  "09/12/2024 00:05:35.876", 143.348,
  "09/12/2024 00:05:35.975", 141.516,
  "09/12/2024 00:05:36.074", 136.731,
  "09/12/2024 00:05:36.174", 138.275,
  "09/12/2024 00:05:36.273", 143.015) %>%
  mutate(timestamp = mdy_hms(timestamp))

# 1. 转换时间为起始点的整数毫秒偏移量
start_time <- min(data$timestamp)
data <- data %>%
  mutate(ms_offset = as.integer(difftime(timestamp, start_time, units = "secs") * 1000))

# 2. 生成精确的10Hz目标毫秒序列（每100ms一个点）
target_ms <- seq(
  from = round(min(data$ms_offset), -2),  # 取整到最近的100ms倍数
  to = round(max(data$ms_offset), -2),
  by = 100
)

# 3. 直接插值，无需生成冗余行
interpolated_values <- na.approx(data$value, x = data$ms_offset, xout = target_ms)

# 4. 转换回标准POSIXct时间戳
data_10Hz <- tibble(
  timestamp = start_time + target_ms / 1000,
  value = interpolated_values
)

# 查看结果
data_10Hz

方案优势 #

• 精度绝对可靠：整数毫秒运算彻底避免POSIXct浮点秒数的内部表示误差，生成的时间戳完全是0.1秒的整数倍
• 处理效率翻倍：跳过complete生成大量中间NA行的步骤，直接针对目标点插值，大数据集下速度提升明显
• 逻辑简洁清晰：减少不必要的数据转换，代码可读性与维护性更强

可选优化：使用xts包（专业时间序列工具） #

如果频繁处理时间序列，xts包的插值逻辑更贴合时间序列特性：

library(xts)

# 转换为xts时间序列对象
data_xts <- xts(data$value, order.by = data$timestamp)

# 生成精确的10Hz目标时间序列
target_times <- seq(
  round(min(index(data_xts)), "0.1 sec"),
  round(max(index(data_xts)), "0.1 sec"),
  by = 0.1
)

# 插值并转换回tibble格式
data_10Hz_xts <- na.approx(data_xts, xout = target_times)
data_10Hz <- as_tibble(data_10Hz_xts) %>%
  rename(timestamp = index, value = coredata)

内容的提问来源于stack exchange，提问作者ckluss